Rules and standards discussions/Archive/Archive08

From ISFDB

Jump to: navigation, search

This is an archive page for the Rules and standards discussions page. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from August 2009 - December 2010‎.

Archive Quick Links
Archives of old Rules and standards discussions.


1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16


Expanded archive listing


Contents

Normalize artist name when entering it from a signature?

{{PublicationFields:Artist}} is silent on if there are desires as far as how an artist name should get entered when the credit is based on the signature. It came up here where it had been primary-verified as "Faragasso" and another updated the record to be "Jack Faragasso." I tend to do "as stated" but know I've also normalized at times.

Maybe it does not matter as long as the notes are accurate. At some point someone may want to know if Faragasso or Jack Faragasso should be canonical. At present they can review all of the publication notes to see if a trend can be spotted. --Marc Kupper|talk 06:51, 30 August 2009 (UTC)

Use of periods after a title

I'm holding a submission which wants to make variants of two publications of a story where the only difference is a period after "Mrs". I'm unable to find any stated policy about how to handle this. I personally feel the titles should be merged because British publishers often drop periods after Mr, Mrs, Dr, etc. Should a printing preference trump the author's title? There was a previous discussion which concluded (I think) that double or single quotes would not be a factor in creating variants. Would not this be a similar situation? Thanks. MHHutchins 16:09, 7 September 2009 (UTC)

Help:Screen:EditPub says:
"Suffixes such as "Jr" should follow a comma and space, and be followed by a period if they are abbreviations. This should be regularized if they are not presented this way in the publication. E.g. "Sam Merwin Jr" should be entered as "Sam Merwin, Jr."; similarly, it's "Edward Elmer Smith, Ph.D."; or "Frederick C. Durant, III". Other prefixes and suffixes should follow analogous rules."
I take this to mean that "Mrs" should be regularized to "Mrs." (and "Mr" to "Mr.") regardless of what the pub has, and thous should not be a variant, just as we regularize periods after initials. -DES Talk 17:11, 7 September 2009 (UTC)
True, but that rule applies to the Author field while the submission in question wants to make "Mrs Rinaldi's Angel" into a VT of "Mrs. Rinaldi's Angel".
The original reason to enter titles as they appear on the title page was to ensure that users who know nothing about ISFDB could find information by simply entering the title in the search box (or via Google), so that would seem to be an argument in favor of creating a VT in this case. I suppose it's possible that some users will automatically add a period after "Mrs", but that's a very subjective standard. By the way, a quick search shows that we have over 230 occurrences of "Mrs." and over 25 occurrences of "Mrs". Ahasuerus 17:23, 7 September 2009 (UTC)
For what it's worth, Google doesn't seem to care - "Mrs Rinaldi's Angel" and "Mrs. Rinaldi's Angel" return the same number of expected results with the same apparent ranking. Regularise, variant, or change our searching? I think I'd quite like to add better search options here (exact, versus slightly looser punctuation-free, or case-insensitive, searching), but we're already experiencing some performance problems there. BLongley 18:05, 7 September 2009 (UTC)
Fuzzier searches would be nice and we could certainly benefit from using Google's search capabilities directly, but Google wants $2,000/year for sites with up to 100,000 pages and we have hundreds of thousands of pages. We could try using MySQL's built-in capabilities instead, but rumor has it that performance could be an issue. Something to discuss on the Development page, I suppose... Ahasuerus 18:46, 7 September 2009 (UTC)
Sorry, I misunderstood. I thought this was an author variant, not a title variant. Fuzzy searches would be nice, if they don't hurt performance too much, but we surely don't have them today. I think I would favor a VT, which could be found and possibly reguarlized when and if we have good fuzzy searches, for the reasons Ahasuerus cites above. -DES Talk 23:56, 7 September 2009 (UTC)

Novel or Collection for fix-up

I just posted a question on the Help Desk about a technical issue, but there is also a question on whether the Moorcock Elric fix-ups should be treated as novels or collections in this collection/omnibus (currently listed as novel). A more complete explanation is here.

Thanks. --Rtrace 04:12, 9 September 2009 (UTC)

That is generally a judgment call. Your description on the help desk sounds a lot like collection to me, but I don't know the books directly. -DES Talk 08:02, 9 September 2009 (UTC)

Serials reprinted as Project Gutenberg titles

Triplanetary for example which reprints the parts of the original serial. Shouldn't the title date be the date the Project Gutenberg title was published? Because of the author name this is a variant title. But if the names matched there would be a tendency for someone to want to merge the titles and use the date when the serial was first published.--swfritter 15:24, 12 September 2009 (UTC)

I (probably) was the one who did it (I verified it, anyway), & I just assumed that the date of original publication of the serial was the correct date. I'm fairly sure I did several that way; as it didn't even occur to me that this was an issue, I didn't think to ask. Didn't think about someone doing a merge. You're probably right. I'm not quite sure how to find all the ones I did, though. (I'll fix this one, anyway. I can always fix it back if someone else jumps in arguing for what I originally did.) -- Dave (davecat) 21:27, 12 September 2009 (UTC)
I want to argue for what you originally did. If we look at 1197, we see the first publication of the title was the Amazing Stories serialization. So the broader question really is, should the title's date be that of the serialization (1934) or the first book publication (1948). If it were short fiction and published just in a single magazine issue, it seems pretty clear the "standard" is to use the date of the magazine publication. If that is indeed the case, I don't see why a serialization should be treated any differently. --MartyD 01:15, 13 September 2009 (UTC)
There is quite a bit of history associated with this issue -- take a look at Help:Use of the SERIAL type and the links at the bottom of that page. We may also revisit the issue of VT dating now that the Publication display page shows VT dates when they differ from canonical VTs, but that's a somewhat different issue. Ahasuerus 01:40, 13 September 2009 (UTC)
P.S. I have updated Help:How to connect serials to titles to reflect the changes to Serials that have been implemented over the last few weeks. Unreprinted Serials are still somewhat awkward, primarily because there is no way to link multiple serializations, so we will probably want to revisit this issue once we all get used to the new system. Ahasuerus 02:09, 13 September 2009 (UTC)
I have to admit When cataloging short fiction, the date of the original publication is always used as the "publication date" regardless of whether the original appearance was in a magazine, anthology, collection, or a chapbook (which we call "chapterbooks" at the moment). However, when cataloging novel length fiction, the date of the original serialization is not used and the date of the first book publication is used instead. is pretty clear (and obviously contradicts what I said). I was unaware of the precedent. --MartyD 09:54, 13 September 2009 (UTC)

Placeholders for serial only titles

See Charles Willard Diffin. Note that someone has created a NOVEL parent for The Pirate Planet, Brood of the Dark Moon, and Blue Magic. This gives the impression that there was an actual book publication and the process is not supported in Help. The alternate method I used with Land of the Lost seems to make more sense. In this case the serial section would look really messy if all the the various serial episodes were unlinked. I don't think either process is supported in Help. It has been a significant issue for me in entering serials reviewed by Bleiler in S-F: The Early Years. In order to avoid another interminable discussion about what I perceive to be a relatively straight forward issue I have just gone ahead and entered the data in clunker mode. Is the method I used for Land of the Lost valid within the current framework? If anyone thinks not I will submit a proposal although by the time it gets considered it probably will be too late to be of any value to me. Please also not Two Thousand Miles Below which was recently published by Project Gutenberg. I am seeing this as the first "book" publication so give the title the date of the Project Gutenberg publication.--swfritter 13:43, 14 September 2009 (UTC)

Land of the Lost looks like the best compromise solution that I have seen so far. It avoids creating the impression that a Serial has appeared in book form, but at the same time it helps group related Serial records together. The later can be important when stories have been serialized multiple times, especially when one or more serializations appeared pseudonymously. Ahasuerus 14:40, 14 September 2009 (UTC)
I would tend to think that having no placeholders at all would be the best option, but if a placeholder must be there I can see arguments on either side. -DES Talk 14:45, 14 September 2009 (UTC)
There are certain cases when we really need to link related Serial Titles in some fashion. For example, if you run an Advanced Title Search on "Title includes 'Cosmos' and Title Type = 'SERIAL'", you will find 29 installments of a Serial strewn across 20 Authors' bibliographies. If a user comes across "Part 7 of 29", he will have no obvious way of finding the other installments :( Ahasuerus 19:05, 14 September 2009 (UTC)
The method used for Land of the Lost avoids making it look like there was a book pub, but creates a single SERIAL record for something that wasn't a "Complete Novel". -DES Talk 14:45, 14 September 2009 (UTC)
True, a Serial record that doesn't use "Part X of Y" or "Complete Novel" in its title would be a departure from the current standard. Since the lexical match logic is no more and we no longer rely on this convention to maintain data integrity, I think the benefit of having related Serial records linked will outweigh the disadvantage of further complicating the naming convention.
Having said that, while I believe that the impact on the Summary page will be positive, I am not so sure about the impact on the Publication page. Take a look at Astounding Stories, January 1934 where it appears as follows:
• 112 • Land of the Lost (Part 2 of 2) • serial by Charles Willard Diffin  (aka Land of the Lost 1933 ) 
I don't think this will be very clear, especially when a serial straddles the year boundary -- in this case 1933-1934. I suppose we could suppress the display of VTs when the canonical title is a Serial and the Author/Title data is the same as what the Pub has, but it's kind of ugly. Ahasuerus 19:05, 14 September 2009 (UTC)
All serials appear that way now in publication records, regardless of whether it's a variant of a novel, serial or shortfiction title record. I don't like it. (Does anyone?) Look at this issue of Astounding. And look at how Two Thousand Miles Below is displayed in this issue of Astounding, just because it was only published in "book" form in 2009 by Project Gutenberg. MHHutchins 02:59, 15 September 2009 (UTC)
In some cases the additional VT information that is now displayed for Serials in the Publication screen is quite valuable, e.g. here is how Garrett/M. Janifer's "Occasion for Disaster" appears in [1]:
• 8 • Occasion for Disaster (Part 1 of 4) • [Psi-Power] • serial by Randall Garrett and Laurence M. Janifer  (aka Supermind 1963 ) [as by Mark Phillips ] 
It tells the user that this Serial was reprinted in book form in 1963 and provides the name of the book, which is definitely useful. Other times it's not particularly informative, e.g. the contents section of Astounding Science Fiction, February 1956 includes the following line:
• 8 • Double Star (Part 1 of 3) • serial by Robert A. Heinlein  (aka Double Star) 
which, if you know how to interpret it, means that the Serial was reprinted in book form the same year and that the title was the same. And finally there is If, November 1962:
• 8 • Podkayne of Mars (Part 1 of 3) • serial by Robert A. Heinlein  (aka Podkayne of Mars 1963 )
which means that the novel appeared in book form in 1963.
So, given all of the above, how about the following display change: when a Serial appears in the Contents section of the Publication display screen and that Serial is a VT of a Novel record, display the words "book publication as" instead of "aka"? That way the three examples above would like as follows:
• 8 • Occasion for Disaster (Part 1 of 4) • [Psi-Power] • serial by Randall Garrett and Laurence M. Janifer  (book publication as Supermind 1963 ) [as by Mark Phillips ] 
• 8 • Double Star (Part 1 of 3) • serial by Robert A. Heinlein  (book publication as Double Star) 
• 8 • Podkayne of Mars (Part 1 of 3) • serial by Robert A. Heinlein  (book publication as Podkayne of Mars 1963 )
I think that would help when dealing with Novels, but I am not sure what to do about Serials that are VT'd to Shortfiction. Ahasuerus 14:56, 15 September 2009 (UTC)
I don't feel particularly strongly about this, but if we are going to agree to adopt the "Serial placeholder" method the help should be edited to describe it. -DES Talk 14:45, 14 September 2009 (UTC)
Oh, absolutely! Ahasuerus 19:05, 14 September 2009 (UTC)

(unindent)Now that we can create Titles with more than 15 authors, you can see what a Cosmos installment looks like in Perry Rhodan #48. I am not sure I like it, but I can't think of any other way to link the various parts at the moment. Ahasuerus 02:27, 15 September 2009 (UTC)

I believe I may have created the NOVEL record for the Diffin serials when I was working on the serial project. I think we're forgetting that a novel is not a book, and a book is not a novel. The Pirate Planet is a novel which was published in four issues of a magazine. It has never been published as a book, and the display clearly states "only appeared as". Placing it into a serial category with Land of the Lost diminishes its stature. Why should Two Thousand Miles Below be classified differently when it only appeared as a Project Gutenberg "book"? Haven't we changed more than a few serials that are novella length as variants of the shortfiction title? Look at Like Unto the Locusts and then see how it appears on Pohl's summary page; and Gulf on Heinlein's page. Both are linked to shortfiction titles and not listed under a serial category. Ideally, IMHO, there would be no such display category as SERIAL, just as there are no display categories for NOVELLA or SHORT STORY. Are we seeing an awakening of the schism between the bookists and the lengthists? (You veterans know what I talking about.) MHHutchins 02:59, 15 September 2009 (UTC)
Suggestion retracted.--swfritter 12:32, 15 September 2009 (UTC)
The original idea behind the Serial "title type" was to clearly distinguish between texts that have appeared in book form and the texts that have not appeared in book form, which is what Clute/Nicholls and other encyclopedias do. At the time, we didn't want to create placeholder Novel Titles for unreprinted Serials for fear of confusing the issue -- that was one of the reasons for the "lexical match" logic, which was later found wanting on other accounts.
The fact that the Summary display now says "only appeared as" is a big step forward and removes one of the main objections to creating placeholder Novel Titles. However, the other objection -- which has to do with dates -- is still outstanding. As we have often discussed over the last 3 years, bibliographers and collectors use the first "book publication" date (as opposed to the first serialization date) to date Novels and we use the same approach. If we are to start creating placeholder Novel Titles for unreprinted Serials, how will we date them? If you look at the Novel section of Jack Williamson's Summary Bibliography page, you will see that, say, Golden Blood is dated 1964, the year of the first book publication, rather than 1933, the year it was first serialized. If we were to create placeholder Novel Titles for Williamson's unreprinted Serials like Islands of the Sun and date them using the year when they were serialized, it would make the ordering inconsistent. Ahasuerus 15:17, 15 September 2009 (UTC)

[unindent] This is really two issues both having to do with display only. Adding a novel placeholder for serials only affects the author's summary page (I think.) Serial display in a pub record is another matter. On the first issue, I very much agree that the problem arises from the arbitrary (and arcane) belief that serialized novels should not be considered published until they're in book form. I personally do not like the fact that Golden Blood is dated 1964, but I bow to the standards of bibliophiles of the past who held these elitists beliefs. (I'm only half joking here.)
Onto the second issue: I wonder how the average 15-year-old trying to find info on a Heinlein juvenile would think of

• 8 • Double Star (Part 1 of 3) • serial by Robert A. Heinlein  (aka Double Star)

Your suggestion changing it to (book published as Double Star) helps, but I wonder if it's possible to suppress the variant info entirely in cases where the titles and year of publication are exact matches. All one would need to do is click on the serial record as it's displayed in the pub record and be sent to the title record which shows the parent record, and even better, a link to the parent. A link to the parent record doesn't currently exist in a pub record display. Would that be possible, or is that putting just one more unnecessary link on a pub display where almost everything else is linked? Just ruminating here. Since working on the serial project I've not added any further novel placeholder records for serials, and will refrain from doing so until a consensus has been reached. MHHutchins 16:02, 15 September 2009 (UTC)

It certainly would be possible to suppress the non-serial parent title in this situation if the titles match and the authors are the same. I also like that idea: there's already a lot going on in that display, and adding more text is probably not the best way to go. --MartyD 17:11, 15 September 2009 (UTC)
OK, so it sounds like we want to suppress the display of canonical titles in the Contents section of the Publication screen if and only if:
  • The VT is a Serial and the canonical title is a Novel or Shortfiction
  • The authors of the two Titles are identical
  • The titles are also identical up to the first left parenthesis
We will also want to change "aka" to "book publication as" when the VT is a Serial and the canonical title is a Novel. Come to think of it, a similar approach may also work for placeholder Serials that started this discussion. Ahasuerus 19:47, 15 September 2009 (UTC)
OK, the change has been implemented and deployed in patch r2009-28 except that, based on local testing, the display of Novel/Shortfiction VTs for identical titles is suppressed even if the VT author(s) are different. The examples listed above now appear as follows:
• 8 • Occasion for Disaster (Part 1 of 4) • [Psi-Power] • serial by Randall Garrett and Laurence M. Janifer (book publication as Supermind 1963 ) [as by Mark Phillips ]
• 8 • Double Star (Part 1 of 3) • serial by Robert A. Heinlein
• 8 • Podkayne of Mars (Part 1 of 3) • serial by Robert A. Heinlein

plus:

• 3 • Star Lummox (Part 1 of 3) • serial by Robert A. Heinlein (book publication as The Star Beast) 
Any other Serial-related display tweaks that we want to implement while we are at it? (I am also exploring different ways to show both the magazine publication date and the first book publication date on the Summary page, but that may be tricky.)
P.S. Also note how much shorter and more meaningful our Rules discussions are when we can make suggested software changes quickly :) Ahasuerus 03:34, 16 September 2009 (UTC)
Quickly, my ass. That was phenomenally fast. Less than 12 hours after the suggestion that the display be changed, and it's done. We should have taken software changes into our hands years ago (sorry, Al, just expressing an honest opinion here.) MHHutchins 04:42, 16 September 2009 (UTC)
A minor problem with such a quick discussion/change cycle: I started reading. I clicked on Michael's links. "What's the problem with that?", I thought. I had to read through quite a lot to discover that the display didn't show whatever he was complaining about, since the change got made before I ever saw the item.
I'm assuming that this change is an improvement. I certainly have no problem with what I saw (& with what I understand from the discussion). But it might be well to paste in copies of the offending displays in future discussions (as Ahasuerus did later on). -- Dave (davecat) 14:17, 16 September 2009 (UTC)
That's a good point. It's easier to link directly to "problem records" during these types of discussions, but it's safer to copy them to this page in order to make sure that the context is preserved and the arguments are still clear 6 months from now, long after the data was changed. Ahasuerus 15:49, 16 September 2009 (UTC)
Re: Michael's point, Al laid the groundwork for distributed development (with Roglo's and Marc's help) some time last year (2008), but we didn't have anyone on board who knew enough about Python to set up a development environment at home. Well, except Roglo, who was unavailable during much of 2008 as well. It wasn't until Marty volunteered his services and wrote a "How to Configure an ISFDB Development Environment for Dummies" guide that our tech-savvy editors were able to re-start development in May. We are slowly getting better at it, so it takes us less time to implement bug fixes and new features.
Naturally, our developers' availability fluctuates over time and lately I have been the only one with the time to work on the code consistently, with Marty and Bill contributing as circumstances permit. I am trying to do as much as I can while I am available (at least 1 minor bug fix/feature per night or one major feature in 3 days), but there are limits to what one developer/tester/installer can do and still keep his sanity. For example, I was so tired on Monday night when I finished implementing that night's patch that I accidentally wiped out the document that I use to create our weekly backups and one of the backup scripts. They are not hard to recreate, but it raises a red flag. If I start deploying bug-ridden software (or worse) because I am no longer 18 and can't go on for months with little sleep, I may burn out the way Al did. Ahasuerus 16:14, 16 September 2009 (UTC)
For a random change, this turnaround would perhaps have been TOO fast -- need to give people time to comment -- but we've been trying to make the display of variants and serials better for some time now, so this was more like a tweak. I wanted to second the point about the development activity: It's a lot easier to make progress when multiple people can chip in, and we're fortunate the project's source code has been made available to the community and that we have a champion of improvement who's willing to put the effort into getting improvements and fixes deployed. --MartyD 17:05, 16 September 2009 (UTC)
Regardless of how it's done, kudos is deserved to everyone involved. A virtual toast to them all. MHHutchins 17:21, 16 September 2009 (UTC)

Radio dramatization of Siodmak's Donovan's Brain

Someone has gone to the trouble of adding the radio dramatization of Donovan's Brain. I'm pretty sure that the only audio that's IN is readings, not dramatizations. Are there any objections to these being removed as a pub record. I can add the links that are now on the pub records to the title record. MHHutchins 15:34, 19 September 2009 (UTC)

I agree this should be deleted. This has been discussed before. -DES Talk 19:12, 19 September 2009 (UTC)
But the Pub Notes seem to suggest it IS a reading? BLongley 09:57, 20 September 2009 (UTC)
I would treat a reading with multiple cast members as a dramatization, and in many if not most cases i expect it will actually be a dramatization by anyone's standards. -DES Talk 17:38, 20 September 2009 (UTC)
I agree with DES (five readers?). Despite the note, the person who entered the record call it a "dramatization". That was my first clue. I'll do further research before deleting it. MHHutchins 19:49, 20 September 2009 (UTC)
It's a dramatization, available on Archive.org. MHHutchins 19:58, 20 September 2009 (UTC)
The phrase "Welles' parody of Donovan's Brain" does it for me. Delete. BLongley 17:27, 21 September 2009 (UTC)

Dating Artwork

When an independently existing work of art is used by a publication, how should the art be dated? E.g., for Gary Kilworth's 2006 collection here http://www.isfdb.org/cgi-bin/title.cgi?933060 they used Abel Grimmer's 1604 painting "Tower of Babel" so should the date be 1604 or 2006? The former would be more accurate, but I can see other considerations. Theoretically someone else could decide they'd also like to use "Tower of Babel" for their new novel, so there'd still be multiple entries for that work of art (one 1604 cover art for Kilworth, one 1604 cover art for Joe Blow) at any rate -- unless we further complicate things by lumping the various cover art credits together under an artist credit for "Tower of Babel") Jonschaper 06:29, 28 September 2009 (UTC)

I use publication date, and am happy if people merge COVERART records where the art is the same. However, I know some people are resistant to that as the artwork may not be exactly the same - e.g. a wraparound cover on the hardback becomes front-cover only on the paperback, and then gets trimmed on later printings to give the author name more prominence, etc. However, as we only show covers on publication records the point is a bit moot - we can't record enough on the COVERART record to distinguish such. BLongley 17:40, 28 September 2009 (UTC)

Pseudonyms and "Writing as"

This help page (Author->Pseudonym) indicates that an author's non-pseudonymous name be used when presented as "Writing as" as it denotes a previous publication under a pseudonym (e.g. by Isaac Asimov, writing as Paul French). I'm currently working on my copies of this series. The first book uses only the pseudonym, "Daniel Moran". The second and third books are presented as "Robert E. Vardeman writing as Daniel Moran". (I'm ignoring the omnibus, which I suspect was actually published after the individual titles and uses Vardeman anyway.) Since the 2nd and 3rd books were never published under the pseudonym alone, should the name on the pub record be the actual name (Vardeman) or the pseudonym (Moran)? I'll also point out that the spine of both books only has Vardeman. The cover and title page (which matters), have both names. My personal feeling is that the non-pseudonym is what should be used, but thought I'd make sure others agree. ~Ron --Rtrace 13:36, 28 September 2009 (UTC)

Non-pseudonym (Vardeman). Pseudonyms are only used when the canonical name isn't shown at all. BLongley 17:33, 28 September 2009 (UTC)
Moran was a pseudonym that became confused with an actual author's name and was apparently only continued for the series run. Vardeman and the real Moran were/are both aware of the conflict it created for readers. A more true type pseudonym use is Alan Burt Akers which took over for the 'Dray Prescot' line and the real author "Kenneth Bulmer". So I prefer Vardeman as I have collected him, without realizing it, for years. The real problem is when a 'pseudonym' is used by an author and that author loses his readers, due to the camouflage. So I am not a fan of pseudonyms, and am a fan of their being 'aired' out. So Vardeman is my preference.
I think though if you could find a fan group or something to support the pseudonym and it's use, a different case could be made. 2¢ and not worth much. Thanks, Harry. --Dragoondelight 19:55, 28 September 2009 (UTC)
That's right, Vardeman and his publisher decided to use "Daniel Keys" on Keys to Paradise, presumably due to the fact that Vardeman had 13 (!) books come out in 1986-1987. What they didn't realize was that Daniel Keys Moran was an active SF writer and that there was potential for confusion. Once the first volume came out and the issue was brought to their attention, the credit was changed to "Robert E. Vardeman writing as Daniel Moran". As Bill pointed out, we use the canonical name when it is used along with the pseudonym, e.g. "Isaac Asimov writing as Paul French" or "Nora Roberts writing as J. D. Robb". Ahasuerus 20:21, 28 September 2009 (UTC)
As always, record what's stated and we'll fix the pseudonyms later. I only found George Kinley this week. And there'll always be some dispute about "canonical name" - e.g. John Wyndham for "John Wyndham Parkes Lucas Beynon Harris". BLongley 20:31, 28 September 2009 (UTC)
Another example might be Kevin Anderson versus Kevin J. Anderson. Different people. BLongley 20:40, 28 September 2009 (UTC)
I note the "In Death" series are all listed as by Robb, cover note of "Roberts" notwithstanding. -DES Talk 21:16, 28 September 2009 (UTC)
Feel free to go fix. I don't think we have any primary verifiers. I've tried a bit of a clean-up at times, but really it's of no interest to me. And there's lots of foreign stuff to deal with too. BLongley 21:56, 28 September 2009 (UTC)
Actually i have verified quite a few. -DES Talk 22:41, 28 September 2009 (UTC)
In the case of Robb vs. Roberts, she uses Roberts for the romance stuff and Robb for the mystery stuff. It seems more appropriate to use the pseudonym on the mysteries. Not that I care specifically about her case, but it's an illustration of why using the non-pseudonym from "writing as" may not always be the best choice. --MartyD 10:32, 29 September 2009 (UTC)
I tend to agree in this case. Moreover, the "Robb" stuff is the majority of her SF output -- they are all police procedurals set in the 2050s. Aside from that, a few of her works are paranormal romance, but most are straight contemporary romance, which should be in, if at all, as NONGENRE, and without the Robb books she might well be considered under the "certain threshold". -DES Talk 14:59, 29 September 2009 (UTC)
She's entirely below the "certain threshold" by me, and I'd be happy if only the Robb books remain. Not volunteering, just saying. BLongley 19:45, 29 September 2009 (UTC)
You really think so? 29 or 30 novels and several novellas unquestionably science fiction, plus some paranormal or horror romance? Granted it is less than 20% of her total bibliography, but by that standard, Asimov is below the threshold: his non-fiction outnumbers his SF by at least 5 to 1 in title count, and probably more than that in word count. Ah well, it is a tricky line to draw. I suspect the real issue is the kind of SF the "In Death" series is. The SFnal elements are used as settings for stories which are otherwise fairly contemporary police thrillers. I at one point started to write a review claiming that a few mechanical changes would convert these to present-day police novels. I started to try to specify what changes would do this, and found the speculative technology more deeply embedded in the plots than i had noticed. Ditto the future history. The books don't work without their setting and that setting is SFnal, although more subtly so than most SF, IMO. They do have a significant romance component, which is not to everyone's taste, but if we eliminate poor SF, or SF we don't like, from counting towards the threshold, well.... -DES Talk 20:51, 29 September 2009 (UTC)
I'd be quite happy with less Asimov too, although the difficulty there is the volume of Nonfiction published in Genre magazines. At least he didn't write more romance than SF. And never got into the sort of fiction that makes me want to avoid working on Cecilia Tan's bibliography for instance. Oh well, my threshold affects nobody else's threshold. BLongley 18:30, 30 September 2009 (UTC)

(unindent) We never came to a conclusion about this, and now there's a submission that would change the authorship of The Halvifar Cycle omnibus from Alan Burt Akers to Kenneth Bulmer. The cover credits this as "Kenneth Bulmer writing as Alan Burt Akers", and all sources I've found so far use "Akers" as the author. Furthermore, the items collected in the omnibus were published as by Akers, not as by Bulmer. Wouldn't it make most sense to capture this as by Akers and make that a variant of a Bulmer parent, rather than recording this as by Bulmer with no Akers reference? --MartyD 10:28, 3 July 2011 (UTC)

And there's another submission that wants to do the same with "Andrew North" over "(Andre Norton)", from the title page ("Andre Norton writing as Andrew North" on the cover) on Sargasso of Space. Here it's not even a writing as credit.... --MartyD 11:28, 4 July 2011 (UTC)
I'd stick with the current standard, in part because changing it would require a lot of rework, much of it non-trivial. Ahasuerus 18:51, 4 July 2011 (UTC)

What to do when the Foreign title is more English than the English one?

Here's a fun one. Sunnydale Park. I've put it under the English title "Paleo" for now, but I'm really tempted to make it a variant title, as "Paleo" is worse English than "Sunnydale Park". BLongley 19:42, 29 September 2009 (UTC)

Subtitles: Variant or No

I've noticed a practice with novels which I've perpetuated without being certain that it is the preferred method. That is, when encountering a novel with a subtitle that the ISFDB title record does not include, entering the publication with an extended title field on the pub, while leaving the shorter (no subtitle) title field on the title record and not creating any variants. For example, Domnei has titles with both the subtitle "A Comedy of Woman Worship" and without. However, this practice doesn't work if the novel ever gets contained in an omnibus or collection (same example, 1002750), and one wants to reflect the title, with subtitle, as it appears in the collection. For this reason, I suspect that I've been doing it wrong all along and that the addition of a subtitle should merit a variant title in all cases. How do others approach this? Thanks. ~Ron --Rtrace 03:30, 5 October 2009 (UTC)

There may be some controversy about the way subtitles are handled, but I personally feel you're doing it the right way. What we have to keep in mind, and I may have brought this up before, we have to distinguish between "subtitles" and "extended titles". Domnei: A Comedy of Woman Worship has a subtitle. Subtitles, IMHO, should not be part of the title record, but can be part of the pub record. Extended titles are actually the full title and usually contain the word "or", e.g. The Secret Ascension, or Philip K. Dick is Dead, Alas. This should be part of the title record, and if not titled that way in a pub, a variant should be created. I don't believe variants should be created based on subtitles, e.g. Soldiers of Paradise: A Starbridge Novel, even if this subtitle appears on the title page of the pub. To get to the second part of your question, there is no way to have the subtitle display in an omnibus content record, because contents are titles, not pubs, and only title records are displayed. Even if you create a variant so that it displays your preferred title, there will be an additional "aka" credit, e.g. "Domnei: A Comedy of Woman Worship (1913) novel by James Branch Cabell (aka Domnei)". That looks awkward, I think you'll agree. MHHutchins 03:52, 5 October 2009 (UTC)
I too prefer sub-titles (and worse, series or contents descriptions) not to be included in titles. I can live with them on Pub titles but wouldn't enter them that way myself. As for the problem with when they're entered with sub-titles in collections or omnibuses, I'd just leave it as a note. BLongley 08:17, 5 October 2009 (UTC)
I definitely prefer that series designations (e.g. "A Starbridge Novel" or "Book 3 of The Tale of the Five") not be included even at the publication level. I feel the same about generic subtitles (e.g. "A Novel", "A Romance", "A Novel of Terror" etc). True subtitles I have mixed feelings about. In some cases i think these should be treated as "extended titles". One example might be 1635: The Cannon Law and the other 163x novels. Another example might be ". . . And Call Me Roger": The Literary Life of Roger Zelazny. Other examples that seem to me to possibly warrant including a sub title in the title record are: 2001: A Space Odyssey, A Rogue's Life: From His Birth to His Marriage, A Voyage to the Moon: With Some Account of the Manners and Customs, Science and Philosophy of the People of Morosofia, and Other Lunarians, A. I: The Death of Evan Chan, 62: Pompeii A.D. 62: Pompeii, Acorna: The Unicorn Girl , Alien$: We're Only Here for the Money, Alva & Irva: The Twins Who Saved a City, Andre Norton: A Primary and Secondary Bibliography, Anne Mccaffrey: A Life With Dragons, Arthur: High King of Britain, City Pier: Above and Below, and Cyteen: The Betrayal. In these the sub-title carries enough of the meaning that the "main title" is less useful without it. In a few of these examples there is an argument for using only the sub-title (The Cannon Law, The Unicorn Girl and The Betrayal) where the main title primarily identifies a series, but in these cases the books are almost always known (and searched for) by the combined title. (There are many other cases where IMO a series prefix is not needed and could be removed). (These examples, by the way, all come from the first few pages of searching for novel titles with a colon in their names -- a few were actually NONFICTION and I have adjusted them. )
All that said, i think usually a subtitle should not be a reason for a variant, and the practice described above is fine. But when a sub-title is significant enough to be included in the main title record, then a version without the subtitle might be reason for a variant. If there are publications with different sub-titles, then I am more inclined to think a variant is warranted. -DES Talk 16:01, 5 October 2009 (UTC)
I updated Template:PublicationFields:Title to add a note about the series name. --Marc Kupper|talk 01:07, 11 October 2010 (UTC)

Variants based on diacriticals

I proposed making a variant relationship between Stableford's "Salome" and "Salomé", and a question was raised about whether these should instead be merged, with a note about the accent different placed in the appropriate pub. Search using either form of the vowel finds both forms of the title, and the point was raised that we do this with Philip José Farmer -- we don't have an alternate name of Philip Jose Farmer, and we don't create variant titles based on the difference in accent. My own experience was based on "Maelstrom" vs. "Maelström", where searching by one or the other does not find both.

In researching this, I found this section of Help:Screen:EditPub#Author, which is at odds with the non-variant practice cited above, even using Farmer as the example:

Accented characters. If you are entering a name such as "Philip José Farmer" that is printed with an accented e, that accented character should be reproduced in your entry of the name. Two versions of an author's name that are printed with and without accents are treated as variants; you should not convert one form to another.

The text in Help:Screen:EditPub#Title is similar, although it makes no explicit mention of variants:

Symbols and punctuation. Strange symbols should be entered if appropriate typographical characters exist. [...] Other characters should be entered in Unicode if possible; this includes accented characters, and symbols such as em-dashes.
I think this part of Help was written when we still assumed that we fully supported Unicode. We were young and naive back then... Ahasuerus 16:42, 27 October 2009 (UTC)

Clearly we'd like to avoid clutter, so should we in practice not create variants in situations where diacriticals are ignored by search / database text matching? I personally prefer consistency, so if variants due to diacritical differences are "necessary" in some cases, I would use variants for diacritical differences in all cases. But perhaps not everyone agrees that they're necessary, even in the umlaut case I cited. And I certainly don't want to be the one to try to make variants for all of those Farmer titles.... :-) --MartyD 10:52, 27 October 2009 (UTC)

I believe the underlying issue here is that we do not have full support for Unicode at this time. Our display and search software tries its best, but all database records are stored in Latin-1 (which supports West European languages like English, French, German and Spanish, but not much else), which imposes limits on what can be done. Moreover, our records use one of the Swedish collations (the MySQL default), which determines search behavior, e.g. whether a search on "Jose" finds "José" and a search on "Maelstrom" finds "Maelström".
Ideally, we would first convert all of our data from Latin-1 to full Unicode (aka UTF-8), teach our software how to work with it, then figure out how searching and everything else behaves under full Unicode and only then make policy decisions. Unfortunately, converting from Latin-1 to UTF-8 is not always easy and has been known to cause data corruption, major downtime, heart attacks and earthquakes. First, we'd need to touch base with Al to see whether he has investigated this issue over the years. Ahasuerus 16:42, 27 October 2009 (UTC)
I don't see this as a "Unicode" issue but rather policy and using it to help solve (or lower) the tension between "accuracy" and "clutter." In publications we try to document what is stated and from that can develop reliable, and hopefully fact-based, bibliographies. If one publication credits the cover artist as Ed Acuna and another as Ed Acuña then
  • If the goal is accuracy we designate one of those names as canonical and make the other a variant or pseudonym. (Note that at present ISFDB makes this difficult as attempts to enter "Ed Acuña" get mapped into "Ed Acuna" as that's the name in the database.)
  • If the goal is cleaner bibliographies then we we use one name or the other. Which name to use is murky as some of the help says to figure out the canonical name based on earliest publication, popularity, etc. and the "Accented characters" section of Template:TitleFields:Author implies you go straight for the accented name.
The artist's web site does not resolve Acuna/Acuña.
I personally would like to see a trend towards accuracy and to use VTs. We'd need a code change because at present the code that ensures author names have a consistent case, for example, converting data-entry of "Charles De Lint" into "Charles de Lint", is also converting diacritics meaning all of these are the same letter "Aa Āā Ăă Ąą Ǻǻ Άά Αα Àà Áá Ââ Ãã Ää Åå". To fix this we'd only be converting the upper/lower case pairs in that list. --Marc Kupper|talk 02:43, 23 August 2010 (UTC)

Author Data Fields- Modification for effect?/ Listing a birth name / note field

I am using [2] to build-up the biographical summaries on artists. The book gives some confidence as the articles have a "Sources" section, which show the editors contacts. I have noticed that several times there are biographies located at other sites than Wikipedia and IMDB. Can I use the web page entry for these, after making sure they are still up? I know we are not at this time, or probably ever, going to build true biographies, but listing these 'other' sources can give greater certainty (utility)to these DB author summaries. Thanks, Harry. --Dragoondelight 13:08, 2 November 2009 (UTC)

We can link to other sites in the "Web Page" field, e.g. we link Sherrilyn Kenyon to her official site. It's a multiple field, so we can list as many sites as we need to, although we favor official/authorized sites, especially for living persons, since they have to be more reliable and don't raise potential legal problems. Since authorized sites are rarely available for deceased authors and since dead people are unlikely to get upset if we link to a site that accuses them of assassinating Kennedy and Princess Diana, fan sites are OK to use. Ahasuerus 20:01, 2 November 2009 (UTC)
Add as many useful web pages about the author as you like, I'd say. You can add multiple ones in one edit, just separate them with semi-colons - the screen doesn't make that clear. BLongley 20:06, 2 November 2009 (UTC)
Great because interested parties' bios are generally what you get after the artist/author passes. At least it gives the user a shot, without committing to trying to right bios, which wikipedia has so much trouble with. Thanks, Harry. --Dragoondelight 21:01, 2 November 2009 (UTC)

Listing a birth name. Rick Berry was born Richard Riley. We list him as Richard R. Berry. I added (Riley, Richard) after his legal name. Is this acceptable? Is there a better method. I hate to ask for another field, as the changing of legal name is common. I feel confident of my entry as the "Sources" list a conversation with the author, but the article did not use "Richard Berry" only Rick Berry. As a 'double think' what about people who change their married names, even males sometimes. Should we not account for this? Thanks, Harry. --Dragoondelight 13:08, 2 November 2009 (UTC)

A 'note field' for inconsistencies or conflicting data? Something like that used under 'synopsis' on the book page? Thanks, Harry. --Dragoondelight 13:08, 2 November 2009 (UTC)

This recently came up in another conversation, and in the end "Birth Name" was moved to the Author's Biography Wiki page, presumably keeping "Legal Name" for current Legal name. BLongley 20:05, 2 November 2009 (UTC)
The downside being that nobody ever looks there.--swfritter 22:10, 2 November 2009 (UTC)
An author/artist can have a number of legal names over the course of his/her life (how dare they!). If we want to list all of them, we will need to make "Legal Name" a multiply occurring field. It's probably a good idea down the road and I will add it to the list of requested features. We'll have to figure out a way to distinguish between "Birth Name", "Current legal Name" and "Past Legal Name", though. Ahasuerus 20:10, 2 November 2009 (UTC)
I did find out that Brom is either Gerald or Jerrold, but it really does not matter as I read his personally inspired bio and it continually states "Brom", so he has reached the avatar stage. The only real reason to have a 'legal name' is for tracking, so I will desist on variants. Thanks, Harry. --Dragoondelight 21:01, 2 November 2009 (UTC)
And yes, we need a free text field for Author level bibliographic notes, e.g. "Not to be confused with the other 3 David Alexanders". Unfortunately, new fields are a pain to implement, so it will be some time before we get to it. Ahasuerus 20:10, 2 November 2009 (UTC)
I am sorry, but I had another duh pop out after I wrote this query. We need a text reference field, so we can note bio(s). That takes writing bios almost totally out of our hands. In other words here's our source, find a better. Sorry, for the extra work, but I have always been fascinated by bios. Thanks, Harry. --Dragoondelight 21:01, 2 November 2009 (UTC)

Help:Screen:AuthorData - possible example problem

Morning! This [3]. The last entry subject; Author Image. The last example is of an author image hosted at Wikipedia. I was distinctly informed on one occasion that images links were not allowed at that site. Thus I am conflicted by the example. Thanks, Harry. --Dragoondelight 13:33, 3 November 2009 (UTC)

Good catch, we shouldn't be linking directly to Wikipedia images since they consider it "bandwidth theft" and can block our links to their articles in retaliation. We'll need to change the example and clarify that we should only link to the sites that have given us permission to link to their images. Ahasuerus 23:13, 3 November 2009 (UTC)
Fixed. Ahasuerus 01:53, 5 November 2009 (UTC)

Premium paperback size

Some paperback publishers are doing something new: the premium paperback size. This apparently started back in 2005. Entries in the isfdb do not make note of this format, just calling it pp along with all the other mass market paperbacks.

The traditional mass market paperback measures 4 3/16 inches wide and 6 3/4 inches tall; the premium paperback measures 4 1/4 inches wide and 7 1/2 inches tall. The width difference is unnoticeable (even difficult to measure) but the extra 3/4 inch of height is VERY noticeable.

The two books I now have that are premium paperback but which are listed as "pp" are William Gibson's Spook Country, ISBN 0-425-22671-1 and Terry Goodkind's Confessor, ISBN 0-7653-5430-6. —The preceding unsigned comment was added by Evilsofa (talkcontribs) .

The current standards state "...any book that is at least 7.25 in (or 19 cm) tall, or at least 4.5 in (11.5 cm) wide/deep, and is not a hardback, is a tp." The size you describe should thus be entered as "tp". Perhaps in the future, if this size should become a more common practice, another category could be created. (Please sign your posts using four tildes: ~~~~). Thanks. MHHutchins 03:05, 8 November 2009 (UTC)
At the moment, it seems to be the direction that the publishing industry is moving in, but it's probably too early to tell. Del Rey experimented with unusually sized paperbacks back in the 1980s, but it didn't take. Ahasuerus 03:14, 8 November 2009 (UTC)
When Bill and I sparred over this issue a few months ago it was decided to call this new format a "tp". I suggested adding a new classification but was shot down. What's the point in revisiting the subject? Is time proving me right?Kraang 03:35, 8 November 2009 (UTC)
I only recall sparring with you over Alastair Reynolds pubs that were pb height and tp width? Both cause me shelving problems (7" high is OK, 5" depth/width is not), so I wanted to preserve "pb". I'm fine with sub-divisions of tp if those are found necessary though. Both "pp" and "tp" and whatever the Alastair Reynolds format get called are all fine by me, so long as I can avoid them. BLongley 21:58, 8 November 2009 (UTC)
Database design (whether enforced programmatically or as a matter of policy) reflects the field (publishing) that we are dealing with. As the field evolves, so do our cataloging standards: we now support e-books, audio books, etc which we didn't support 15 years ago. The only problem that we have is distinguishing between new formats that are here to stay (heck, pulps were a new and unfamiliar format once!) and fads that peter out after a couple dozen Publications. We want to accommodate the former but not the latter.
The fact that we now have new editors ask about this format suggests that it is becoming increasingly common and, if the trend continues, we may need to revisit the standard soon. Ahasuerus 03:49, 8 November 2009 (UTC)

Arthur C. Clarke's group titles, again

This has been discussed before, but I don't believe any firm policy was established. So here we go again... How should the group titles "Venture to the Moon" and "The Other Side of the Sky" be recorded? There is currently many ways in which they appear in publications (all of the examples are verified pubs]:

  1. Group title only here and here
  2. Story titles only here and here
  3. Group and stories titles Here
  4. Here and here, one set has the group title with individual records for the stories and the other only has the individual story titles.
  5. Here, the group title is given as an essay with an uncredited authorship.
  6. Here, it looks like the editor stopped in the middle of removing the group title.

I personally lean toward solution three (we do the same thing when an anthology or collection becomes a constituent of an omnibus). Whatever the outcome, it should be consistent. MHHutchins 18:12, 13 November 2009 (UTC)

I prefer option three, as we have to take care not to confuse the group titles if the entire group is not present. The group title is often the ToC entry and sometimes title page entry, and the short-stories are confused with 'chapters' or 'sections'. BLongley 19:13, 17 November 2009 (UTC)
I also prefer solution 3. And it would be better to have a GROUPTITLE (gp) tag.--ErnestoVeg 19:48, 18 November 2009 (UTC)

[This Publication Record]

Harry (Dragoondelight) started using this and 'explained' his reasoning to Moderators twice that I read (and got a headache both times). Now two other editors have "copy-edited" (sorry for the bad pun ;-) ) and are using it as well. Is there any bibliographical need for this statement? At worst it assumes that anyone reading the notes is too daft to realize the notes refer to the record of which they are a part. At best it assumes the same thing...... It's like adding 'etc.' to a list after listing the whole list. Perhaps only a peccadillo but I really don't think it should be common practice. ~Bill, --Bluesman 04:00, 17 November 2009 (UTC)

There are many cases where bibliographical data comes from secondary sources and we are not always good at documenting secondary sources. Although I don't do so myself, such statements make the source of bibliographical data unequivocal.--swfritter 15:27, 17 November 2009 (UTC)
It's not my favourite practice, but we had to get Harry to do something more when he started going into great detail about the prior printing statements. When you're used to the notes referring to the publication record in question, discovering that those details might be several lines down because he'd started at first hardcover, listed first tp and then first pb before getting on to this Nth pb reprint, "This" was an improvement. A complete rewrite to start with the current record, finish that, and then explain that there are also details of prior publications recorded would be clearer, but nobody seems to want to make that sort of effort. BLongley 19:23, 17 November 2009 (UTC)
[after edit conflict] I have no problem with the wording (as it was thoroughly discussed and the meaning became clear.) My preference is to place any printed statement in quotes with the word "stated" to indicate that the info comes directly from the book. If it's not in the book, I record the source in the notes. My approach and Harry's may be different, but I think we come to the same conclusion.
As Swfritter says, quite often notes (and even data fields) are completed with information from secondary sources. In a perfect world, the original editor (and subsequent editors) will indicate the source of that info. Our goal should be to encourage such practice, but we can't drive editors away who might feel it to be too demanding. Hopefully, someone will come along who will verify the book and make it clear which info is actually stated in the book. It's my understanding that Harry only uses this statement when he verifies a book. At least we have an editor to go to when questions arise about such records.
You might want to question those new editors that you've seen use this statement to determine if their definition of the term is the same as Harry's. And to point out that this is not a standard practice. Perhaps they were only cloning one of Harry's verified records. MHHutchins 19:30, 17 November 2009 (UTC)

(Unindent) Thanks all! Appreciate the feedback. ~Bill, --Bluesman 16:08, 18 November 2009 (UTC)

Replica editions of magazines as anthologies?

I think we are doing them this way. It seems to be the best way to do the Project Gutenberg Astounding Stories titles. And now Librivox is reading them and I have entered that as an anthology.--swfritter 16:30, 18 November 2009 (UTC)

I think Anthologies are the safe way to go since reprints can have new metadata (e.g. a new price) and can occasionally include new Essays and such. Besides, Anthologies support multiple printings and Magazines don't. Ahasuerus 23:24, 18 November 2009 (UTC)
The only unusual thing is the serials. Since they are reprints of the original installments the logic for listing serials separately does not seem to apply. There isn't anything in Help to cover this situation since the section in Help only applies to reprints in magazines.--swfritter 23:34, 18 November 2009 (UTC)

Weirdisms

I've started working on some issues of Weird Tales and I've encountered a series of items that appear both in this magazine and Whispers that I've seen treated a few different ways. The items are titled "Weirdisms" and they consist of a piece of artwork with a paragraph of text. They are only ever credited to the artist. I've uploaded an example. These items have been listed as both essays and interior art. Additionally, I've seen them titled as "Weirdisms" or Weirdisms followed by a colon and the title of the paragraph. They have also appeared with "(cartoon)" appended. A titles search on "Weirdism" will yield the variations. I tend to think of these more as interior art rather than essays. I also like the extended title, though without the "(cartoon)". I could also see going with only the title of the paragraph and using a series titled "Weirdisms" to collect them. Given that the example would be titled either "Weirdisms: Grave Robbing" or simply "Grave Robbing". Does people concur, or would they rather see these as essays? --Ron ~ RtraceTalk 22:47, 25 November 2009 (UTC)

Another option is two entries - one interior art and one essay entry. There are certainly many documented essays of that length or less and the art is more than significant enough to justify a separate entry. If most of them have a discernible title than that should be used but you could also put the title and date of the magazine in parenthesis as is often done with continuing essay series. In any case you might want to keep "Weirdisms" in the title. Eventually we will probably want to round them all up and put them in a series.--swfritter 01:27, 26 November 2009 (UTC)

Why is the series appearing on contents line for novels?

Morning! This. [4]. Suddenly the series appears in contents, I know this method is used on magazines, but this a change that will affect hundreds, ney thousands, of titles. Thanks, Harry. --Dragoondelight 12:39, 27 November 2009 (UTC)

If the only Title in a novel pub is a novel, then it's not displayed by the software. If a pub contains other Titles, e.g. Essays or Interior art, then all Titles, including the novel Title, are displayed. If any of the pub's Titles happen to be part of a series, then the name of the series will be displayed next to the Title (in square brackets.) That's the way it's worked for a long time; there have been no changes to the software in this area. Ahasuerus 15:37, 27 November 2009 (UTC)
Now I see it, but I missed it before. Thanks, Harry. --Dragoondelight 16:40, 27 November 2009 (UTC)

It's a Magazine? It's an Anthology? It's a Magthology!

If this is the wrong venue for this question, please nudge me to the right door.

I am working on back issues for Tales of the Unanticipated. TotU started its life as a long-period magazine, the period being 9 to 16 months. It was published in 8.5x11 stapled quarto format.

In 1999, Issue #20, it changed to a 7.25x9.25 perfect binding trade paperback. I contend that it morphed into a serial anthology at that time, even though it was still called a magazine on its cover through #25. From #26 on, it's been variously CALLED an anthology, a magthology, and an antholozine. It's an anthology. Some things that distinguish it from a magazine are:

  • The publishing date is the publishing date. We don't print the 2012 issue in 2009. 11/6/2008 mean 11/6/2008.
  • There's no display-until date. They should display them until they sell them already!
  • There's no letters the the editor section.
  • We slave over each issue like you'd slave over a book. We don't rush an issue to the stands.
  • We print back issues, back to #20, as needed.
  • Individual ISBNs.

So, my question: Would it break anything if I identify issues #20 on as Anthologies? If so, how about #26 on?

Oh, you can look at the covers (and some of the insides from #26 on) at www.totu-ink.com/ix WXRock 21:42, 27 November 2009 (UTC)

Also, it has a Magazine wiki page. Should I continue that as if it were a magazine? Remove #20 on from that page? Show the format break some other way? WXRock 21:47, 27 November 2009 (UTC)

The use of an ISBN would seem to me to indicate the anthology option. There is a precedent. Pohl's Star Science Fiction had both anthology and magazine incarnations. You can still put links to the anthologies on the magazine wiki page and put a note to indicate on that page to indicate the morph factor. Another perhaps more correct option would be to not document them in the wiki grid but put a link to the anthology series there.--swfritter 15:36, 28 November 2009 (UTC)

Entering a Review when was edited by...

When entering a review, how do I indicate that the title was edited by someone. Should I enter that name in the Author page? Leave Author blank and edit the Title? WXRock 17:56, 28 November 2009 (UTC)

I am assuming it is an anthology? Just enter the editor as the author. As with this title.--swfritter 19:52, 28 November 2009 (UTC)
Turns out it's this title and it all fits together OK. I thought it wasn't indexed because I (*blush*) searched by Name instead of Fiction Title. So much to learn for an old guy! WXRock 01:32, 29 November 2009 (UTC)

Series:Crossroads Adventure RPG

Morning! This. [5]. This is an RPG role playing game series. I have Dzurlord, bought it under the assumption it was something of a novel. It is entirely RPG. Thanks, Harry. --Dragoondelight 14:25, 30 November 2009 (UTC)

Yes, it's a bit of an anomaly, but I think the last time we discussed the series, we decided to keep it since it's linked to so many prominent SF universes. Ahasuerus 03:58, 2 December 2009 (UTC)

Credit for INTERNALART photos

I am entering publications that have internal art, mostly story illustrations. But we've also got photographs, mostly of interview subjects. The photographers get credit in the TOC. Should I enter them here too? See TOTU #20, pages 53, 57, and 60.

If I should include them, can I title them like "Photo: Neil Gaiman"? WXRock 00:32, 2 December 2009 (UTC)
I'm not sure if this has ever come up before, but I don't see that the value achieved would be worthy of the effort involved. If this were adopted, it could be logically argued that author photos on book jackets should be recorded. And frankly, I don't want to make yet another round of of my verifications. (I'm in the middle of my third go-round, and finding it tedious.) MHHutchins 15:30, 2 December 2009 (UTC)
In going through the 50's mags I generally did not enter author photos. I actually sort of regret that now. In many cases the photo may be the only one that exists of a certain author. Photos on books are of less significance since they have a tendency to be recycled and often represent the author at a much younger age than the publication of the book. The less important the author the more important the photo. There are two ways that photos seem to be in the database now. As WXRock suggests. Others have "(photo)" appended to the title. I would prefer the first method. I don't think it should be required to enter such data in order to justify verifying a pub but I see no problem with somebody adding data later. My feeling: change the data - ask me and the other verifiers first. Add data, notify me.--swfritter 16:26, 2 December 2009 (UTC)

Handling misspelled author name

Seems like this must have been discussed somewhere, but I can't find it. According to Jonschaper, in Fantasy Tales, Autumn 1989 the author on Wayland's Smithy is listed as "Jon Bye" in the TOC and "John Bye" on the story's title page. Locus credits that poem to "Jon Bye" and credits that same author name spelling to "The Farthing Lord" in Fantasy Tales #15 1986 (I found this pub also listed on iberlibro.com with enumerated contents, using the same "Jon" spelling). I also found a citation in a Google Books scan of The Supernatural Index that spells the name "Jon".

So it appears the story's title page misspells the author's name. As the name is spelled correctly elsewhere in the pub, should we simply spell it correctly and record the misspelling in the notes? Or should we capture it as spelled and make the misspelling a pseudonym for the correct spelling? --MartyD 11:45, 2 December 2009 (UTC)

I have done that on occasion and a few times with titles also. My own opinion is that if there is internal evidence in a pub as to the right title/author spelling it can be done as long as the discrepancies are adequately documented.--swfritter 14:06, 2 December 2009 (UTC)
Another waivering of the rules about recording credits as shown? I've always handled these situations by creating variants, as I assumed everyone else did. Maybe I read the rules too literally? Funny thing is, the first few months on the database I was correcting obvious errors in credits, but became indoctrinated by the ISFDB mantra to record "as is". It would be too hard for me to break that habit now. I can't see making an exception for an obscure author, when there are entries for popular authors for which variants have been created. (I've done dozens alone for the misspelled author credits as printed in the Italian magazines entered by ErnestoVeg.) MHHutchins 15:24, 2 December 2009 (UTC)
I would err on the side of variants but if an author is referred to correctly in the TOC, on the cover and possibly elsewhere it seems to me that there is a little flex room - as long as there is full documentation. Titles, especially in magazines, can be a bit problematic because what often appears on the title page has been played around with by the graphic designer. If the TOC and header pages agree I go with them - and document. Any apparent deviation from standards should always be documented.--swfritter 16:11, 2 December 2009 (UTC)
I am specifically asking about the case where the pub has both the correct spelling and an incorrect spelling. I guess the real question is do we consider an interior content-delimiter "title" page to be of equal weight to a book's main title page? I.e., if I have a novel, and the author's name, say, were misspelled on the title page but appeared correctly on the copyright page, it's pretty clear we'd record the misspelled name as-is and then do a variant. At least, I would. But are title pages on interior content like title pages or are they more like covers, where liberties may have been taken in the presentation? --MartyD 18:53, 2 December 2009 (UTC)
I was also referring to correct/incorrect spelling assuming that the title page of a story in a magazine had an incorrect spelling of author but the name is actually spelled correctly in multiple other places in the magazine. Books are different. In that case the name on the title page is used. In any case, a name should never be entered in a manner different from than what can be found in the pub. What does help actually says about crediting title page of story and table of contents: "Short fiction, essays and poems. For short stories, essays and poems, take the title from the heading on the page where the work begins, rather than from the table of contents, if there is one. This distinction is not too important, and if you know that one form of the title is the usual one (e.g. the contents page has "Night Fall" but the story heading is "Nightfall") then use the one you know is standard. You can also choose to use the table of contents version where the story heading gives a non-standard presentation of the title form--e.g. if the table of contents says "Bell, Book and Candle" and the story header says "Bell, Book & Candle", you can use the former. If both the table of contents and the story title agree, though, the form given should be used, even if it is different from the standard." Author is not specifically mentioned in the section but it seem the same logic would apply. --swfritter 20:52, 2 December 2009 (UTC)

Comic Strips in Pubs

I've been running into a lot of comic strips in pubs lately, mostly in older magazines. My inclination was to categorize them as short stories, or serials, since some of them were long running. It also seems to me that they need to be identified as other than cartoons or text, maybe as ":Comic Strip: title," and either "short story," "serial," and possibly an "interior art" credit if the artist and writer were not the same person, e.g., Jack Williamson and Lee Elias for "Beyond Mars." Any thoughts?--Rkihara 18:29, 2 December 2009 (UTC)

Fixer submissions - how low do we go?

As a new moderator, this is my first encounter with these submissions. I would like some parameters to work within. Tonight I looked at about a dozen, rejecting two as they were written for readers of the ages 2-4 (not my judgment, the actual description!). Of the 40-50 in the queue tonight, none are for even teenagers.

It's not quite as bad as it may appear. As I work Fixer submissions, I tend to go for the low hanging fruit first, so borderline cases accumulate for a while. I approved a few dozen today and most were supernatural romance novels of one kind or another, a completely different headache :-\ Ahasuerus 07:37, 5 December 2009 (UTC)

Regardless of content, at what point do we (or do we not) cut off? I realize much of this is a matter of judgment but some rules must apply? I'm still in shock that a category "paranormal romance" even exists (old school). If it was up to me I'd nix 'em all, but that's what this page is for.... ~Bill, --Bluesman 05:22, 5 December 2009 (UTC)

ISFDB:Policy provides some guidance:
Speculative fiction is defined to exclude [snip]:
  • Fairy tales with no known author (?)
  • Animal books for very young children, i.e. books for preschoolers which depict simple scenes from animal life featuring anthropomorphized animals
Admittedly, the explosion of juvenile fantasy in recent years took us by surprise and we haven't really figured out what to do with it. Up until now, Al's robot, Dissembler, was responsible for importing forthcoming books and he is a very discriminating robot, who grabs only top publishers and well known genre authors. Fixer is not nearly as discriminating and he grabs everything that walks or talks like SF, hence the recent glut of submissions. We are about 70% done with January 2010, so we probably want to let Fixer finish and then figure out what we want to do about it. Ahasuerus 07:37, 5 December 2009 (UTC)
Okay, so until then it's err on the side of inclusion? It just seems a lot of time to spend on 'fringe' works. One with no Amazon descriptions (I then decided to try a couple of the blogs listed) ended in about 10 minutes with the glowing recommendation of a five-year-old! I can hardly wait for the discussion!! LOLLOL!!! ~Bill, --Bluesman 18:18, 5 December 2009 (UTC)
I lean toward inclusion. Although it can be a lot of work - one Fixer submission led to all of R. D. Henham's entries for instance. Not stuff I'll ever read myself I think, but if it leads to a kid finding SF it's all in a good cause. BLongley 00:27, 6 December 2009 (UTC)
One thing that I discovered in late 2006 when I was processing forthcoming books manually (Dissembler and Fixer didn't exist at the time) was that the process got easier after a few months. Having cleaned up the bibliographies of the most prolific authors, I didn't need to spend nearly as much time on their subsequent books.
Having said that, the big problem with books for pre-teens is that they are very short (they have to be due to the target audience's attention span), but they make up for it in the sheer number of titles. The "Fairy" super-series of books by Daisy Meadows (a pen name shared by 4 writers) may contain fewer words than a single Dozois anthology, but there are 100+ titles in the series...
By the way, Bill [Bluesman], I have reviewed the "Robot and Rico" books that you rejected yesterday and it turns out that the target audience is "Grade 1-2" rather than age 2-4. It can get confusing since sometimes Amazon clerks enter the audience's age and other times they enter the grade level. Also, "Flyaway" by Lucy Christopher, a book for tweens, may be SF since the publisher's description ends with "a breathtakingly magical journey", which may or may not be metaphorical. Darn metaphors, always getting in the way! :-) Ahasuerus 02:03, 6 December 2009 (UTC)

Website or company as cover artist

In editing this publication the question was raised if websites or companies should be recorded as cover artist, or if in those cases the cobver artist should be left blank, using the notes for clarificarion. The database knows a number of websites as coverartist (here and here, and also companies like this. The help text is silent about this. Any thoughts? Willem H. 12:51, 6 December 2009 (UTC)

I have waffled at times on this issue but I think it is alright to enter the name of conglomerate artist if the actual artist is unknown. It doesn't seem to do a lot of harm and I have found situations where the same company will have multiple credits which is probably information someone might want to know about.--swfritter 16:53, 6 December 2009 (UTC)
My thinking behind raising the question is artists are like authors and editors. It seems to me authors should be people (or pseudo-people as the case may be), and if that's true then that standard should be applied to artists and editors. That said, I agree it does no harm. Except for perhaps some unexpected search results, the worst that could possibly happen is someone's deciding to fill out details and wiki pages with company information, using the excuse that they are "authors" or "artists" or "editors". --MartyD 18:24, 6 December 2009 (UTC)
An author search for the word "studio" indicates that it that it is not an unusual practice to credit entities although it is kind of interesting to note that Mondolithic Studios, Inc. has been give the legal name Mondolithic Studios and it looks like some one could do a variant title thing with it if they wanted to. If you give the editors enough rope . . .!--swfritter 18:55, 6 December 2009 (UTC)
Just a note that librarians have to deal with conference proceedings, government publications, etc all the time, which is why they have a separate field in the MARC21 standard for "Corporate Names". We have a somewhat similar issue with (a) magazines/anthologies edited by self-described "collectives" and (b) some children's books which do not state the author clearly and are often attributed to the publisher. I am not saying that we should implement something similar to what MARC21 does, but at least we are not alone in our struggles :) Ahasuerus 20:12, 6 December 2009 (UTC)
I dislike it, because it is usually the separate artist field of 'cover design'. The problem, which we probably not solve, is that they are the last artist talent on the field and put all those obscuring ornatments, banners, etc in place. Right now we have three 'artists' that probably are not doing the actual art. Carol Russo, Tony Destafano and Ray Lundgren. Lundgren obscures Carl Lundgren the artist. Destafano is notorious for boxing and obscuring the true artist, same Roy Lundgren. Rick Berry for one artist works out of a large studio/artist combo and I am positive that he and other artists there hand over some work to 'junior associates, etc.'. One associate, a partner at his firm, Darrell Andersen, gets credit for Neuromancer in one book only. I can not really dig up a profile for him, I think we will have some wait to see him bloom out because of his 'corporate status'. So, really, I guess it does not matter, except it detracts from the artist. I even dislike this kind of credit James Jay/Jay Studios. Who did it? Thoughtfully, an entry does give DB users somewhere to look for art. Jane Frank and Robert Weinberg do not recognize studios or cover design artists. I think there reasoning is that there are multitudes of artists housed in them and you can not separate the art/artist. So, I guess my belaboring is that we need someway to make sure that we parse the artist from the mill formating. So enter the only data we have and I will gnash my teeth till the artist springs forth like Athena. It is still better than trying to parse artist from Pinnacle and such. LOL Make it the last ditch choice for entry, behind signature recognition. Thanks, Harry. --Dragoondelight 13:53, 7 December 2009 (UTC)

Additional field for ISBN / Catalog numbers

Moved from MHHutchins talk page for a general discussion MHHutchins 01:22, 12 December 2009 (UTC)

Since we can now change our software, how about adding a new field to the Publication record? Something like "Corrected (or "derived"?) ISBN" which would be used for links to Amazon etc? Ahasuerus 18:04, 1 December 2009 (UTC)
Sounds like a good idea: one field that states exactly what is printed in the pub (ISBN, SBN, catalog number), and another field which could be used to link to other databases. MHHutchins 21:35, 1 December 2009 (UTC)
OK, Feature Request 2907284 has been created. Ahasuerus 05:08, 2 December 2009 (UTC)
I definitely like that. Maybe it should be just a "second ISBN" field rather than make it exclusively for derived/corrected ISBNs? I'm thinking of the publications where people have put the Canadian ISBN in the notes of the US edition, making them unsearchable: or the dust jacket I found which managed to contradict itself on the same piece of paper. Or the "rebound" editions which have a paperback ISBN internally and a "rebound" ISBN externally. This could kill a lot of arguments in one go! BLongley 21:33, 11 December 2009 (UTC)
Hm. "Alternative ISBN"? Ahasuerus 22:16, 11 December 2009 (UTC)
Well, we call it "ISBN / Catalog #" when editing, and seem to derive whether it's an "ISBN" or a "Catalog ID" dynamically at some point after entry and before display. It's probably our most over-worked field though, so maybe put it up on Rules and Standards (for hopefully the last time for many years). The people using it for ISSN or Issue numbers of Magazines should have some input in this too. And the LCC(C)N enterers. Maybe the BLIC users too if we ever find a good way of linking there... BLongley 23:12, 11 December 2009 (UTC)
Of course, we'll find somebody that wants to have three (or more) such fields for some reason or other, but we'll overcome one of the major hurdles. I don't think Mike and I have ever disagreed on "record the data as shown in the book", we've just disagreed on exactly where to record such at times: with only one field to play with, I've always preferred "most useful" to go in the one field, but that's always been debatable when some resources cope with invalid ISBNs and others don't. BLongley 23:12, 11 December 2009 (UTC)

[unindent] I think the implementation of this feature would solve many of the issues surrounding the "record as is" vs. "more useful data" debate. As Bill points out, there must be a discussion about what exactly goes into each of the fields. Without a narrow definition of what data to enter, I can see it becoming chaotic. MHHutchins 23:25, 11 December 2009 (UTC)

Just thought of a few more issues: a) we could use such to record ISBN-10 and ISBN-13, and show which are used on a single publication - one, or the other, or both. I'm not keen on that much rework though, now Marty has sorted out searches on either form. b) We could record SBN and the ISBN conversions separately. I'm not keen on that either - indicating that a leading zero wasn't there originally doesn't add much value. What I would be willing to do (in the long run) is c) record the instances where British publishers put their serial numbers into their assigned ISBN ranges eventually. (Panther and Sphere for instance.) But that will take months, so no promises about timescales, and I can't do such conversions for US publications. (Marc seems to know how to do DAW - I don't.) But if it's wanted, I'll try it. BLongley 23:41, 11 December 2009 (UTC)
Actually, "Ace" might be the US one to try first. They do/did seem to have a nice indication of printing number encoded in their Catalog IDs. BLongley 23:44, 11 December 2009 (UTC)
What about volume and issue numbers for magazines?--swfritter 14:20, 12 December 2009 (UTC)
Fine by me. That doesn't seem to need an additional field unless you want to record ISSN at publication level too. Or do you want Volume Number in one and Issue in the other? I suspect the tricky bit of coding is making sure all the possible reference links work from either field for books. Magazine uses of the field(s) shouldn't trigger such look-up capabilities. Or is there an ISSN database we should link to? BLongley 23:03, 12 December 2009 (UTC)

(unindent)Taking a step back, here are the types of IDs that we commonly deal with:

  1. Catalog ID, e.g. 12345 for SFBC or "H-123" for pre-ISBN books
  2. Alternative Catalog ID, e.g. DAW's "Collector numbers"
  3. ISBN-10
  4. ISBN-13 (introduced in 2007, but some sources, e.g. OCLC, also calculate it for pre-2007 books)
  5. Alternative ISBN, mostly used by US/UK and US/Canada "dual ISBN" books
  6. SBN for mid-1970s books
  7. "Corrected" ISBN which corrects publisher mistakes and lets us link to other Web sites
  8. Library of Congress' "LCCN"
  9. OCLC - more than one may exist per book
  10. SSN - an ID for magazines
  11. Magazine Volume/Issue number - at least two are possible due to "whole numbers"

At the moment, IDs 1-7 go into the "Catalog ID/ISBN" field and IDs 8-11 are generally recorded in Notes. Anything else that I am missing? Ahasuerus 16:49, 13 December 2009 (UTC)

I no longer enter ISSN numbers. It actually might make some sense to use the "Alternative Catalog ID" field as a text field for for Vol #, Issue #. The ISBN field is best reserved for those handful of magazines that have ISBN or other identifiers that specifically identify the magazine. Vol #, Issue # are best handled as text - there are a few really odd entries. Perhaps we could actually select the viewable title for these fields via a dropdown? That way only "Catalog ID", "ISBN", "Volume# Issues#", etc. would appear on the screen.--swfritter 17:19, 13 December 2009 (UTC)
I think you're a little off for the SBN usage dates, but as all such can be converted to ISBNs that's not a problem. Our software now seems good at spotting ISBN-10s and ISBN-13s, so as long as any other use of the fields doesn't trigger unwanted warnings I'm happy. Two fields that both work as links to other sites when they're ISBN-like is all I want. Explanations of why there are two such can be left to notes. If one is an ISBN and the other a LCCN, BLIC, Collector Number, whatever, it shouldn't be a problem. If they're both ISBNs, we can explain why one is from copyright page and the other from back-cover or whatever. Just implement it and we'll argue about usages later, please? BLongley 23:45, 13 December 2009 (UTC)
If this was a minor table like Tags, we could make the change quickly and then see if we liked the results. Unfortunately, Publications is one of the core ISFDB tables and it touches almost everything in the system. As you probably noticed when you were working on the code, database retrieval is mostly centralized, but database filing is not, so adding a new field to a major table may affect dozens of places in the code. That's why foreign language support hasn't been finished yet -- the Title table is every bit as popular as the Publication table and adding a new field to it affects all kinds of things.
Having said that, we can do it, but it will take a fair amount of time and code rewriting, so we might as well decide what exactly we want first. Hopefully, the improvements that I plan will make subsequent database changes easier, but we have plenty of time to decide what we want to do with IDs. Ahasuerus 04:14, 14 December 2009 (UTC)

Essays as part of Series

Is doing this okay? There are two submissions in the queue that want to add two essays to a fiction series. I just haven't run across this before. ~Bill, --Bluesman 20:58, 12 December 2009 (UTC)

Did I really add the series to Afterword (Tales of Mithgar)? That wasn't the plan, please reject the edit. I meant to change the date, and nothing else. I don't think essays belong in fiction series. Willem H. 21:35, 12 December 2009 (UTC)
And I accidentally approved one of them (Final Note [Tales of Mithgar])! Held the other. Might as well wait for some experience to chime in. ~Bill, --Bluesman 21:46, 12 December 2009 (UTC)
Waiting for experts is ok by me. Final Note (Tales of Mithgar) has not become part of the series, so I suppose I didn's add the series there (or maybe the software edits it out?). Anyway, Help sais nothing about this. Come to think about it, I once added some essays to Brian Aldiss's Enigma series. Can't recall a discussion about it then, so maybe it's good to have one now. Willem H. 22:08, 12 December 2009 (UTC)
And they show with an (ES). No smoke, either! :-) ~Bill, --Bluesman 22:18, 12 December 2009 (UTC)
Essays in series have always been a bit of a mystery to me but they seem to work for most people. I think Series support needs a little beefing up (and not just the ability to have publications/titles/editions in multiple series, which is the big problem, but ordering of sub-series would be a nice start) but we've improved it a bit recently. Once again, the Magazine editors' practices differ from the Book editors' ones, but no harm done on either side that I can see. Reiterating that the Canonical author version goes in the series rather than the Variant author might be worth a reminder occasionally - I've seen series where both are in, or the variant author gets the series and the canonical one doesn't, and "Show All Titles" won't help if it's for the variant, but it's one of the smaller problems here. BLongley 23:13, 12 December 2009 (UTC)

Should excerpts be merged?

It's quite common to see multiple records for excerpts of the same book. For some authors, it's up to 4+ records. Should these be merged? While excerpts are usually the first chapter, this is not always the case so it's hard to say for certain they are exactly duplicate. However, from my perspective, the duplicate records clutter the author's shortfiction listing and I don't see much value in knowing there were slightly different excerpts for a given book.

If this is already specified somewhere, you can just point me to it. A search of the wiki didn't turn up anything. Given how often I see this in the ISFDB, I wasn't certain if nobody was doing maintenance on them or if it was intentional. Thanks. --JLaTondre 14:15, 13 December 2009 (UTC)

If we know that two Titles, in this case excerpts, use identical text, then we should certainly merge them or set up Variant Titles if the author/title is different. Unfortunately, it can be hard to tell whether the text is indeed the same.
Consider the case of Frederick Marryat, a good half of whose Title records are various excerpts from The Phantom Ship. Some of them, e.g. "The Werewolf" and "Krantz's Narrative", have separate titles while others just say "Excerpt". Some of them are merged/VT'd, e.g. "The Phantom Ship (excerpt)", but none of the pubs are verified, so how do we know that they reprint the same chapter(s)/section(s)? Ahasuerus 16:38, 13 December 2009 (UTC)
Good point. I was thinking primarily of excerpts at the end of a series book. Those are almost always (but not definitely) the first chapter of the next book. However, I'm not sure what verification has do with it. Unless the same person has multiple of those books, the fact that they are interdependently verified isn't going to help sort it out. --JLaTondre 17:06, 13 December 2009 (UTC)
If we have multiple active verifiers, we can ask them to check the text :) Ahasuerus 17:13, 13 December 2009 (UTC)
I just had an "FR" thought.... What if we had EXCERPT as a title type, a la REVIEW, and provided the same sort of linking capabilities? --MartyD 11:49, 15 December 2009 (UTC)
REVIEWs and INTERVIEWs are handled via separate code sections and add a significant amount of overhead to the application. If we were to use the same approach to address every type of related Titles, the code would quickly grow and eventually become impossible to maintain.
On the other hand, we have another outstanding request to add a "relationship" field which will support values like "abridgment of", "expansion of", "based on", etc. Once this ability has been implemented, "excerpted from" would be easy to add to the list of supported "relationships". Ahasuerus 16:59, 15 December 2009 (UTC)
We should probably move this piece of the discussion elsewhere.... I was thinking something much more limited than the derivative work case. Like REVIEWs, there are things printed in other publications that are EXCERPTs. These excerpts are not really works/titles in their own right. Why record all of that stuff about them? Much simpler to do like REVIEW: record the fact that the excerpt appeared, when it appeared, on what page it appeared, and the title it's an excerpt from. Then, like reviews, a title could show the myriad places excerpts appear for anyone who cares to investigate them. Content presentation could also show these things as subordinate titles (per your experimenting with interiorart). --MartyD 18:09, 15 December 2009 (UTC)
From the design perspective, it may work, but from the implementation perspective it could be iffy. The way the software works at the moment, there are a lot of places where the code has to handle three separate cases: one for reviews, another one for interviews and a third one for everything else. We could add a fourth case, but it would be a fair amount of work, increase the complexity of the code and may well be made obsolete once "relationships" are implemented. BTW, "relationships" include derived titles, but are not necessarily limited to them, e.g. short stories <--> fixup novels. Ahasuerus 00:11, 16 December 2009 (UTC)
What is the main purpose of entering excerpts? Because they are there? or because they help determine editions? Does anybody collect a book because it has an excerpt? How much policy/programmer energy should be expended on an implementation that would make data entry more complex?--swfritter 00:19, 16 December 2009 (UTC)
I do not think excerpts have separate identities as we are only stating title with (excerpt) therefore I think they should be merged without the need to compare. It is a part, but not necessarily the complete part of any section of a novel. Also some of them give warnings that the excerpt portions may not be in the final publishing of the novel. Therefore we can gather them together because they only represent part of the parent and not necessarily have to be the same excerpt. Gathered together they then become a source for someone looking for some information on that title and can be compared later. Using the notes section they can then be differentiated if the opportunity presents. Thanks, Harry. --Dragoondelight 21:35, 16 December 2009 (UTC)
I like that line of reasoning. --MartyD 23:55, 16 December 2009 (UTC)
Title records are generally merged/VT'd only when the text is the same. With excerpts it's hard to be sure whether the text is the same or not, but I think we can usefully distinguish between "real excerpts", i.e. excerpted chapters which appear in collections and anthologies like the Phantom Ship examples above, and "glorified ads" for forthcoming novels.
"Real excerpts" often have separate titles like the referenced "The Werewolf" or "Krantz's Narrative", which can serve as clues as to whether they should be merged. If they don't have separate titles, I don't think they should be merged/VT'd (unless there is evidence that the text is the same) since they may contain different parts of the excerpted novel.
"Glorified ads", on the other hand, are very likely to be identical since they typically appear during the same 12 month period (except in cases of delayed publication), so I wouldn't be opposed to reversing the presumption and merging them unless there is evidence that the texts are different.
It may also be useful to come up with a different term to describe "glorified ads" to make it easier to tell when Titles should be merged. Ahasuerus 00:35, 17 December 2009 (UTC)
I like Harry's logic. "(promotional excerpt)" perhaps? Although I would prefer "(waste of paper - nobody reads these)".--swfritter 14:27, 17 December 2009 (UTC)

Alternate History vs. Historical Fiction

I'm not sure where best to ask this. Please feel free to move it to a more appropriate home if it shouldn't be here.

I have put on hold this Fixer submission of To Try Men's Souls (see also this edition, which has a Look Inside). While Amazon has it in its "Alternate History" node, I don't really buy that. But I also see we have a lovely Newt Gingrich bibliography with more such works, some even arranged in series.

None of these look like "alternate history" to me, but rather what I would call "historical fiction". Two questions:

  • Is everything in that bibliography alternate history?
  • Where is the line between alternate history and historical fiction?

Thanks. --MartyD 12:39, 27 December 2009 (UTC)

To Try Men's Souls is apparently a plain vanilla historical novel, which we may want to include as NONGENRE given Forstchen's long career within the genre. The rest, however, are all legitimate alternative history books, e.g. in Gettysburg Lee wins the day, which is as canonically alt. history as you can get. The Pacific War series is also alt. history starting with a much more effective Japanese attack on Pearl Harbor and leading to a very different match-up in December 1941.
Granted, minor liberties with our timeline do not an alternative history novel make -- or else we would have to include half the historicals out there -- but in this case the changes are so major that I don't think there is any doubt that they are legitimate alternative history.
Naturally, one often wonders whether professional historians like Forstchen, Gingrich and Turtledove add alternative fiction elements to what would otherwise be straight historical novels for purely commercial reasons, but that's a whole different issue :) Ahasuerus 22:01, 27 December 2009 (UTC)
As my formal History education ended in the middle of the Gunpowder Plot (and therefore I don't officially know if Guy Fawkes was guilty or not) I have to recuse myself from a lot of these decisions - e.g. I know nothing of Lee and Gettysburg. I guess I'd spot the difference if America was still part of the British Empire, or Germany ruled both, but things like L. Neil Smith's "North American Confederacy" stuff still confuse. BLongley 22:46, 27 December 2009 (UTC)
Fawkes confessed to Jack Bauer so I guess he must have been guilty.--swfritter 01:51, 28 December 2009 (UTC)
The key thing to remember is that any enterprise later dubbed a "Plot", a "Mutiny" or a "Conspiracy" was very likely unsuccessful. If, however, it is commonly referred to as a "Revolution", then chances are that it was at least partially successful. Ahasuerus 02:27, 28 December 2009 (UTC)

Internal Author Credits

I've just seen this discussion where it seems that "use title page credit" and "exactly as entered" principles should be ignored, according to Help (this help page maybe ?). I haven't read the help the same way, it seems: I think I've reworked several collections to use content title page credits, and before I edit them all back I'd like to make sure this is what we want. BLongley 21:51, 5 January 2010 (UTC)

I think the relevant sections are a) "Author - The name of the author of the work. This will be duplicated from the publication author field for novels, anthologies, collections and omnibuses. The name should be entered exactly as it actually appeared in the publication. This includes pseudonyms" and b) "Short stories and artwork. For short stories that appear in magazines and anthologies, the author's name should be taken from the story heading, rather than from the table of contents, if there is one."

A) doesn't make sense to me for Anthologies or we'd credit all contents to the Editor. And Omnibuses involving different Authors wouldn't work either. So B) makes sense for contents, but it doesn't specifically mention Collections. Should they be? BLongley 21:51, 5 January 2010 (UTC)

We've got people working in opposite directions and I'd rather have this agreed between all moderators and editors, and Help clarified. Better than me invoking "However, as with titles, this is not particularly important" from section B and just ignoring the problem. BLongley 21:51, 5 January 2010 (UTC)

I think the person who wrote the section excluding collections from the rule about internal credit understood that the title pages of stories in a collection are rarely credited. The only exceptions I can think of are collaborative stories. I think the rare cases where a pseudonym is credited in an author's story collection is simply there to indicate how the story was originally published and in no way is there to credit the story as it is being presented. The name on the title page trumps the credit given on the story's title page, in these rare instances (I can't emphasize that too much). Saying we're "working in opposite directions" over such a rare situation seems to me somewhat exaggerated. Yet I do agree that inconsistencies in the help pages should be repaired. Anyone up to the task?
One exception that I can think of is a facetious introduction/afterword written by the collection's author but attributed to someone else. I seem to recall a case where it wasn't obvious that the introduction was written by the author and not the (fictitious) editor. Ahasuerus 00:06, 6 January 2010 (UTC)
Perhaps this one? Remember how it riled up some controversy? (Enough that I stayed away from the database for more than a week.) Looking at the record now, I see it still wasn't resolved. It's an interesting tangent to this discussion, but probably not a good idea to bring it up now. :) MHHutchins 00:31, 6 January 2010 (UTC)
I remember trying to figure out an "Introduction" that when read definitely separated the identity of the editor from the author, though they were definitely the same person, they had not reached the same conclusions. Thus I think they/he achieved alternate separate identities. Most confusing. I remember thinking who will actually read and get it! Thanks, Harry. --Dragoondelight 13:19, 6 January 2010 (UTC)
BTW. Wasn't there a consistency project with the purpose of clearing up cases where pseudonyms were incorrectly credited in author collections? MHHutchins 22:39, 5 January 2010 (UTC)
Yes, there was. I have a set of scripts which check the latest backup for all kinds of internal inconsistencies and at one point we spent a few months cleaning them up. I stopped running the scripts a while back in part because there weren't many easily identifiable discrepancies left and in part because I was busy doing other things. I will try to re-run the script tonight and see how many new discrepancies we have introduced over the last 12-24 months :) Ahasuerus 00:06, 6 January 2010 (UTC)
If I had run into this situation, I would be inclined to apply the "anthology" rules to this collection, leaving it a collection, but using the credits on the interior title pages. Even though we know the pen-names to be aspects of the collection's author, there is some value in having that linking confirmed and documenting where it occurs. Suppose this book were the first/only publication by this author in our database. Note are an alternative, of course, but they don't support searching. --MartyD 12:03, 6 January 2010 (UTC)
My best example is probably Best of Kuttner 2, where I spent some time making sure the "exactly as stated" rule was applied as I understood it two and a half years ago. I certainly hadn't known of Lewis Padgett or Lawrence O'Donnell before I discovered ISFDB. And as we have few ways of recording evidence for pseudonymous attributions in the database itself, only in the Wiki, I do still like the idea that content title page authors are as stated for such evidential purposes - it's not as though we're hiding the original authors, they will display as "Henry Kuttner and C. L. Moore [as by Lewis Padgett]" or "Henry Kuttner and C. L. Moore [as by Lawrence O'Donnell]" or suchlike. The advantage is that we are documenting a source of a pseudonym (admittedly, not in the easiest way to find), the disadvantage is that we might be unable to spot when we have misattributed a content-entry author. BLongley 22:15, 6 January 2010 (UTC)
How did I start all this. Bah Humbug! I did not! I do think I also have become confused by the to me sudden change that title page (toc) trumped story title page. I have been using story title page for everything and if I have to revisit The Martian Chronicles I will have a Bradbury dimentia episode. I still have not figured out how to fill in that help page that Mike and Marc created.
Back to subject. The question as I see it is that a collection is a one person anthology. Ooops! Still I believe the impact of the discussion is that if a collection or any other publication states on the story title page a pseudonym, whether that should be used. After a vicious internal debate the answer I see is yes, because the pseudonym is actually a grouping/sub-grouping of the author's efforts which were 'hidden' and therefore the reader losses the impact of the why/what/where/how? when it is there and not used to bring it to the user/reader's attention. Therefore the pseudonym was an active component in the work's composition and when noted in a story title page, not a copyright statement, it is a statement by the editor/author that you the user/reader are treading in that world not the later (consolidated) author world of the publication. Thanks, Harry --Dragoondelight 13:13, 6 January 2010 (UTC)
How often in an anthology do you see a story in which the author credit is given parenthetically, as is the case for the stories in Early del Rey that began this discussion? The author credits for the stories that were originally published as by del Rey are also presented parenthetically. It's my opinion that the way the credits are presented recognizes the fact that Lester del Rey is the author of the stories in this book, and here, by the way, are the credits as the stories were originally published. I recognize valid reasons for keeping the pseudonyms on the pub record, but I also see the logic of recording the stories as being the work of the author who is credited on the title page. I believe the policy behind recording individual stories credits on their title page was established because of the differences that often crop up between table of contents and title page credits. How many author collections has anyone seen that give author credit on each story's title page other than collaborations? So should we record those as uncredited? Reductio ad absurdum, mea culpa. MHHutchins 17:44, 6 January 2010 (UTC)
I'm sure there's a few collections, other than the Kuttner example above, that I own with author credits to the pseudonym on each story. Kuttner/Moore collaborations are a particular nightmare of mine after a major boo-boo in my early editing days. But even with comparatively recent anthologies an old pseudonym may still appear - "Don A. Stuart" or "Anson MacDonald" may be clear to many here, but they weren't to me four or five years ago. If we regularise too much, we lose some sources of information. Of course, we do have a big gap in that we have no standard way to record what the copyright page reveals... :-/ I'm not keen on demoting this info to notes, but if that's what people have been doing then I lack the ambition or time to resurrect it. I'm sure most people can look at an "Author X" Collection and realise "Author X" is involved in all fiction contents even if we credit his or her pseudonyms on some content titles. BLongley 22:32, 6 January 2010 (UTC)
The question then becomes if you do NOT list the pseudonym that was in the book, then the db user loses information he may need to find the material or similar material. I agree with Bill that having the author as by does not hurt the entry, and it does give the user information that some work was under pseudonyms. As an example, I recently looked for works under the pseudonym and turned up a few. More is better than not. Thanks, Harry. --Dragoondelight 23:32, 6 January 2010 (UTC)
I also agree with Bill and it was my understanding that the story title page rule applied to collections as it does to anthologies and magazines. I can find several examples of August Derleth collections where I have entered under the story using the pseudonym because the story heading had "as by Stephen Grendon" (e.g. "Who Shall I Say is Calling?" and Other Stories). Another example would be Lovecraft's "revisions" (The Horror in the Museum and Other Revisions). I certainly wouldn't relish trying to change those to Lovecraft's canonical name while reflecting the pseudonym in the notes. --Ron ~ RtraceTalk 03:58, 7 January 2010 (UTC)
Sometimes you get surprising results from what seems like an innocent edit. Looks like there's a majority for equality in the treatment of collections and anthologies. I don't see any reason to treat "The Early Del Rey" different than any other collection, so I'll start re-entering the pseudonyms. Anyone up to clarifying the help page on this matter? Thanks, Willem H. 08:10, 10 January 2010 (UTC)
Help page modified, see if it looks good. -DES Talk 23:33, 1 August 2010 (UTC)

Definition of "shortfiction"

Working on this list has prompted me to realize that editors may have different definitions for "shortfiction". Some of the pieces that I've marked "FA" are obviously fictional articles, "in-universe" if you will. But there are some that creep closer to essay. For instance the pieces in this record. Should genealogies, glossaries, timelines, character lists, charts, appendices, etc, even maps be considered "shortfiction"? (Yes, there are some maps on the list as well.) Sorry to open this stale can of worms again. MHHutchins 18:19, 6 January 2010 (UTC)

For me: Maps and Charts and Graphical Timelines or Genealogies I'd class as INTERIORART, and the rest would be ESSAY. But I wouldn't go down to that level of detail for a book, as opposed to the magazine level of detail. If it affects the page count I'd just leave a note that the fiction ends at page X and the rest of the stuff takes it up to Y. Personally, I take such entries as warning signs that I probably don't want to read the book. (Yes, I know "Lord of the Rings" is a classic example that has all sorts of extra bumf in various editions - I'm just not interested in anything bar the fiction.) But moderating such is going to get awkward unless the people that DO care explain what they're doing. BLongley 21:47, 15 January 2010 (UTC)

Partial/Whole Interior Art Credits

Herzbube, MartyD and I have been having a discussion regarding the interior art credits for the omnibus The Tolkien Reader. As I see it, the illustrations break down into three categories:

  1. A horizontal sword that decorates the title page of most of the contained titles.
  2. A set of illustrations for "Farmer Giles of Ham" that contains most, but not all of the illustrations from the standalone edition.
  3. A set of illustrations for "The Adventures of Tom Bombadil" that appear to contain all of the illustrations for that title that appeared in a different omnibus.

The question has arisen on how to enter the titles for the interior art. My preference, at least for 2&3 is to have separate interior art titles for both "Farmer Giles" and "Tom Bombadil". That way the titles could be merged with the same interior art titles where they occur elsewhere and we could easily see where the art it repeated. However, I'm slightly hesitant for "Farmer Giles" as the stand-alone publication has a number of additional full page illustrations that are not included in The Tolkien Reader. Is it OK to have a single interior art title that includes more illustrations in one pub than it does in another (with appropriate notes)? A third interior art title could be added to cover the title page headings.

Patrick and Marty (who have done the lions share of the research) would like to add a single interior art credit for The Tolkien Reader.

Does anyone else have an opinion on how these should be entered?

Thanks. --Ron ~ RtraceTalk 04:36, 8 January 2010 (UTC)

I believe this comes down to personal preference, as there is no standards about the entry of interiorart in collections. I've done it both ways. Here, for example, I created a record for each piece. I know there have been times when I've created a single record for the entire collection. In most cases, it wouldn't make much difference, because 90% of the time these collections aren't reprinted. In the case of the Tolkien, which has many printings, it might be best to have all of them be the same. (I would lean towards individual records, but I don't have a copy of the title.) I guess I'm throwing the ball back to those who have verified copies. Something that just occurred to me: have any other these works been published as a standalone and carried the same artwork with them? If so, that would be the best case for creating individual records. Mhhutchins 18:48, 15 January 2010 (UTC)
I read your comment again and saw that you've already answered that last question of mine, which makes me agree with the method you propose of creating separate records. Mhhutchins 18:50, 15 January 2010 (UTC)
Just an opinion, but repeating title pages I would credit as ornament art (notation) with 1 interior art. If it is simple repetition art it is best not to break them down. I did some where the art became repeated on chapters time and again once I started. Not much value to that. The other art I would, if it was not too many like some illustrated books, make each one a separate item. If it is older art, that is reprinted from another source, I append (reprint) and give it the original date, that way others can get a look at the original art. Examples are A Princess of Mars, [6], where I used Swifritter's brackets and reprint and The Dragoman's Revenge, [7], where I was able to separate the art to the content title and artist, but had not been told about reprint. If it is original art to that time period, not a ripoff like mine are, I would not use reprint for the art. Thanks, Harry. --Dragoondelight 20:46, 15 January 2010 (UTC)

Page count for issues of Perry Rhodan

I'd accepted several submissions in which Don Erikson corrected the page count for issues of Perry Rhodan, when it occurred to me why there may have been a discrepancy in how the pages were counted initially. Should we be using the magazine method of counting pages, or the book method? I would be in favor of using the book method, since they are printed in book form. This is the method I used in entering Destinies and Far Frontiers. Are there any strong arguments in supporting the magazine method (counting all pages plus the covers)? Thanks. Mhhutchins 18:39, 15 January 2010 (UTC)

I so desperately want to avoid Perry Rhodan problems that the few examples I have are free to anyone that wants to take them on bibliographically. But I do feel that the British editions are books rather than magazines, and should probably be separated from magazines that may have contained the same shortfiction. BLongley 21:55, 15 January 2010 (UTC)
Bob Hall told me to use the magazine system of page numbering and almost all should have been corrected to that system. As for it being a magazine, I would incline against as many anthologies are just as varied. I personally have the feeling some of the letters and essays were just dummies for Ackerman. Still, if Podkayne of Mars Baen edition, [8], can have numerous essays from a contest then why not an anthology? The British PR actually cut most of the Ackerman material out. Thanks, Harry. --Dragoondelight 00:01, 16 January 2010 (UTC)

Playground of the Mind Froms?/Excerpts?/Dates?

This discussion was moved from Bluesman's talk page.

Afternoon! This. [9]. I just cloned the second printing but think there are many problems with things. First it is either "From The Green Lantern's Bible" or "The Green Lantern's Bible" neither you or the other copy verifier caught the The. Bigger problem is if it is an excerpt from what. That leads to all the other excerpts should be dated to their original publications. Otherwise Harry spends time looking for an Inferno short story excerpt while he needs to look for a novel Inferno excerpt. I could find no Green Lantern's Bible with or without the's or From's. I think that all three printings should be congruent. Your thoughts? Off to message Mhhutchins. My mind is bogling at the problem. Thanks, Harry. --Dragoondelight 23:42, 21 January 2010 (UTC)

In the hardcover edition all of the excerpts are titled "From xxx", so that's the way they were entered. Also "The Green Lantern's Bible" has never officially been published, nor was there a possibility that it ever would be. Niven was working on the Green Lantern comic book and was asked by the publisher to write a "bible" so that later writers would all be working in the same universe. What's printed here is his contribution to the bible and background for a story that I don't believe was ever commissioned. In the record I verified, it's recorded as an essay.
About the dates: it's my understanding (though I don't have the hard facts to back it up) that all excerpts are dated from the date of their appearance as an excerpt, not the date of the publication from which they were excerpted. It's similar to how preview excerpts are dated. I could be wrong and it might be a matter for discussion. Mhhutchins 00:03, 22 January 2010 (UTC)
Thanks, Mike! ~Bill, --Bluesman 00:07, 22 January 2010 (UTC)
Hokay! I will do mine as I was shown. BTW all my excerpts have been dated to the publication date of the novel, etc as per instructions since I have been here. I can see your argument, but in looking through this 'trash heap' collection, if the dates are not the 'origin' dates then it is confusing to find the origin source (the first time I found the usefulness of using the origin date). As for 'From' I know they are there, but Bill deleted most and that matches instructions I received on including "part", "book", etc in titles. By explanation, The Green Lantern's Bible, I am using "The" and not dropping as in the current examples. Use "From", but not "the"? Sorry, reviewing this looks confusing so I will correct my copy to my understanding and we will then have three unique contents to peruse. Sorry, if I am 'stranger than normal lately", the weather is cold, dreary, raining and apparently not just putting me to sleep, but causing some edginess. LOL. Thanks, Harry. --Dragoondelight 12:12, 22 January 2010 (UTC)
I have submitted my changes as proposed as I understand the need. Please check this and notation. Essentially I separated all "From" titles into essays and then used (excerpt) for those not totally reprinted. All dates are the dates as used for the original publication. As I believe I said this serves as a pointer to the original material. I have no idea if you wish to use hold to make a more complete comparison or not. Will make a moderator entry that you guys may wish to. Thanks, Harry. --Dragoondelight 15:28, 22 January 2010 (UTC)
Pardon, I had to pull my copy of Fallen Angels to redo that entry and see if the essay actually came from there. My copy says no. Thanks, Harry. --Dragoondelight 15:34, 22 January 2010 (UTC)
I've placed Harry's submission on hold pending the outcome of this discussion. Because of his addition of many new content records, that may require them being either merged, varianted, or deleted, this discussion might save us some unnecessary submissions.
I've conceded the point that excerpts be dated the same as the original publication (now what about preview excerpts?!?!?). And believe that the excerpts should be followed by the qualifier "(excerpt)". But they should retain their titles as given on the title page and recorded as "From XXX (excerpt)". These changes have been made in my verified record.
The biggest thing that pops out looking at Harry's submission is the separation of the "From XXX" titles into two records: the introduction becoming an essay and the fiction getting retitled as "XXX (excerpt)". I'm not in favor of creating individual essay records for story introductions in author collections, and would normally just place a note about them in the note field. Then there's the problem of interspersed commentary. Take "From Lucifer's Hammer" as an example: it begins with an introduction, the fiction excerpt begins, and is then interrupted several times with commentary, then concludes, and is then followed by an afterword. I believe one content record, as shortfiction and titled "From Lucifer's Hammer (excerpt)", should be able to handle all of this, with a note that several of the excerpts have interspersed commentary.
I also disagree with a couple other changes: retitling "Retrospective" as "Retospective (Dream Park) (excerpt)". This piece was published in Destinies about a year before the novel was published, was not billed as an excerpt, and is self-contained. The same situation applies to "Rammer" as regards to A World Out of Time. I also see Harry wants to give the Green Lantern piece the name of the first section ("The Ecocatastrophe Era"), but there is another section titled "Oa" which remains unrecorded. I don't see any benefit in retitling this piece. Mhhutchins 17:33, 22 January 2010 (UTC)
I'm thinking that I should accept Harry's submission, just so that we can get a real world perspective on how it would look for comparison purposes. What do you think, Harry? Mhhutchins 18:01, 22 January 2010 (UTC)
If you accept this, then every line before, after or during any story [interspersed commentary] in any collection/anthology could become a titled essay. Harry never stops with one.... The entirety of these extras could be covered in a line or two in the notes, though from the comments below I don't think that option is open. ~Bill, --Bluesman 20:41, 22 January 2010 (UTC)
So I learned it, but frankly sometimes it is satisfying, as when several authors comment on their work, but it rarely feels right when it is done by a single author or editor. I think I like my solution below. A non-fiction work with embedded fictional material that we note. Thanks, Harry. --Dragoondelight 00:56, 23 January 2010 (UTC)
Truthfully, Doing it clouded my thought processes about it. What I really think is that no one edited the galley proof. All the "From"'s IMO are just headers and should have been deleted. Then most of the pre-story essays would be interstitial commentary. Unfortunately, the editor or proof-reader was not doing his/her job. If the question of allowing it and then letting me see it to 'compromise' the set up, very well. I did not notice the interspersed commentary in Lucifer's Hammer, I suspected it though, expect there might be more elsewhere. I will check back tomorrow, if you wish, but today is 'Niven' overload. I am also suspicious of the artist credit. Did I say Niven has always been a difficult author for me to follow. Thanks, Harry. --Dragoondelight 20:18, 22 January 2010 (UTC)
Can we call it an interstitial collection by the author including segments embedded in story material. If we do that we do not have to create most of the essays. In essence all of the interstitial material and the stories are one thing as per copyright. We then only have to document the excerpt of novels, the stories and the titled previously printed essays. This we justify to the user/reader as source material used. What we have is a form of study guide with one author's personal commentary. The thought that I am having is it is really not a collection, but one big author non-genre nonfiction book. It is a how I and others did these things book. An explanation of his work with examples. Does this make sense to anyone? Note there is only one copyright given. Thanks, Harry. --Dragoondelight 00:56, 23 January 2010 (UTC)
I would appreciate it if you would accept my submission, so I can change it to 'Non-Fiction' and then delete a lot of entries. The main reason is I did a lot of work to correct dates and do not wish to do that again. Not being able to look at it in it's present state, I can not say for sure, but this proposal to recognize the author is writing about his work and using fiction elements and some previous essay's to illustrate it, will have several deletions of essays. The topic then may become if an essay is changed in anyway does it retain it's status upon incorporation into NF book. I think not, as it then becomes an element of a story, not the central thought of the book. Thanks, Harry. --Dragoondelight 13:06, 23 January 2010 (UTC)
I will accept the submission so you can see how it's displayed, but will not accept one that would change this publication to NONFICTION. (Other moderators can overrule me, of course.) It is a COLLECTION in every definition of the term. As currently defined in the rules section the NONFICTION type is "Used for books that are predominantly or completely non-fiction." Niven's book is predominantly fiction if a page-by-page count was made (I'm not going to do a count). After accepting the submission, I will ask that you not merge any of the content titles before the conclusion of these discussions. Also, to keep this discussion from overwhelming Bluesman's talk page, I will be moving it to the Rules and Standards discussions page. Thanks. Mhhutchins 16:57, 23 January 2010 (UTC)
I was typing the following at the same time you were putting in the above...... Non-fiction??? By rough count there are about 550 pages of fiction as opposed to 150 pages of non-fiction, almost 2/3 of which is in between pages 495-588. Plus 18 complete short stories with minimal commentary. Non-Fiction???? ~Bill, --Bluesman 17:00, 23 January 2010 (UTC)
Looks like a collection to me still, although I do understand that the lengthier notes and use of excerpts probably means there was more of an intention to educate us about the background of the stories than entertain. BLongley 20:21, 26 January 2010 (UTC)
Just chiming in to concur with the general direction here, especially with the decision to make it a Collection rather than Non-fiction. Ahasuerus 03:09, 27 January 2010 (UTC)

[unindent] The record with Harry's changes is now live. Any other editors with ideas please feel free to chime in. Mhhutchins 17:10, 23 January 2010 (UTC)

If the book had been by a playwright and they included material and then interjected commentary into it, you would still consider it as a whole. I think the fictional material is just the bonus of his story of life and writing (including commentary about other authors). You have to consider that putting extraneous non-fiction into portions of his work as changing it, then is it RW or fiction? Thanks, Harry. --Dragoondelight 20:13, 23 January 2010 (UTC)
Sorry for the delay in responding as I didn't notice your comment until I added another subject for discussion below. To answer "If the book had been by a playwright...", no. The nonfiction added to this volume is there to enhance the reader's enjoyment of the fiction. And I strongly suspect a large part of that readership would not even read the nonfiction. The interstitial material is only added to the excerpts, which themselves are a fictional oddity to begin with. I suspect excising pieces from a larger work leaves much to be desired and it's Niven's added material here (intros, afters, and innards) that justifies their inclusion in the collection. This collection is actually a sequel to N-Space, a collection which had the same format. I wish more editors would chime in, as it seems only the three of us with the actual book have an opinion. I wonder what it looks like from the outside. Mhhutchins 05:06, 26 January 2010 (UTC)
And I thought Scatterbrain would be a bad one to enter (it's been done already, lucky me). As I understand it, it's up to the editor to determine how much detail he wants to add to a publication. From this perspective I have no problem with Harry's method. It should not be a rule though. I wouldn't want to add separate entries for all the notes in The Collected Stories of Roger Zelazny, but I will someday add all the Ellison essays to the Dangerous Visions anthologies. Willem H. 19:52, 26 January 2010 (UTC)
Harry's version looks too cluttered to me, and has gone to too low a level of detail for my liking. I don't mind recording substantial essays but some of the interstitial material is obviously very short, e.g. "From A World Out of Time" doesn't even sort correctly - it's presumably on the same page as the start of the excerpt? (We ought to address that, I've found that pages of poems are becoming a bit of an annoyance when we can't sub-divide page numbers.) BLongley 20:17, 26 January 2010 (UTC)
I do think we ought to discourage too much detail of non fiction as actual content entries - when it comes down to mere paragraphs related to the following entry, just note it if the editor wishes it recorded. There's a slippery slope here: we could end up with, for instance, an INTERIORART entry for every instance of a credited graphic on chapter headings. If there needs to be a cut-off guideline, I'd say related essays should be two or three pages at least before they're worth recording, or should be by a notable different writer. For instance, in Star Trek: The New Voyages I was OK with adding the short introductions as they were probably a major selling point. But I rarely add author's introductions to pieces even if they have a title, they just clutter up searches. And I don't want to feel peer-pressured into doing so. BLongley 20:17, 26 January 2010 (UTC)

(Unindent) Actually, if such detail becomes the norm, then I want the [VERY Concise Listing] option I suggested implemented soon, just so I can still find the fiction. BLongley 20:44, 26 January 2010 (UTC)

Actually, I agree with Bill Longley, too much, but I did it before I had been apprised that it was a "Retrospective Collection". Which I infer means the author gets to do his 'homespuns' in the book. Actually, I like the little bits and comments, but when he dumps some into fiction text it is/was a bit much. Still, I wish to delete most of the essays and submit it differently tomorrow.
Bad news. Before the Golden Age has almost it's entire interstitial commentary listed as essays, while they really are not separate to themselves and are complete only as part of the whole collection. It is not pretty. I have learned aka Niven, that titles do not really make the essay, when it is part of the whole. Thanks, Harry. --Dragoondelight 23:00, 26 January 2010 (UTC)
Please note I just re-edited my submission to reflect the book as a "retrospective collection". All my desired deletions are marked "del" and the only effective change is to the notation. That being to point out what the book is and what is an essay in such a work. Thanks, Harry. --Dragoondelight 13:46, 27 January 2010 (UTC)
Thus we are open to specific changes/deletions/additions to make the material clearly understood. Thanks, Harry. --Dragoondelight 13:46, 27 January 2010 (UTC)
Well, I verified what is left and am Happy (LOL) with it as I am with the author (nyet?). Still open for tweaks. Thanks, Harry. --Dragoondelight 14:02, 29 January 2010 (UTC)

Neil Gaiman's graphic novels

I placed a submission on hold that wants to add a record for the single-volume edition of Neil Gaiman's Marvel 1602. (For background see this notice left on the submitter's talk page.) It's been my understanding that graphic novels (or, as in this case, a comic book limited series) aren't allowed into the database, and I would have rejected it with no problems. Trouble is, though, there is a problem. Currently there are many graphic novels (or comic books) in our bibliography for Neil Gaiman. Does his stature in the field allow an exception to be made, having obtained that certain threshold which allows records for nonfiction work? After all, this is speculative fiction (albeit with pictures) and much more closer to our raision d'etre than a book about rocketry, nanotechnology, or Star Trek jokes. Personally, I would be in favor of allowing select graphic works into the database, but how can we avoid the "slippery slope"? Thanks. Mhhutchins 04:54, 26 January 2010 (UTC)

In our recent fleshing out of the SFBC records, I had noticed the graphic 'novels' of Gaiman's as well. It's possible some of those were created just to link to an SFBC edition. I know I have created a number of "stub" records in the db for just such a link, though I can't recall doing any for Gaiman. As to graphic works themselves, my take on the little exposure I've had is that the fiction merely fleshes out the art, not the other way around. If we are strictly a printed word database, then they should stay out; but we allow a lot of peripheral works as well, as long as the written fiction of the author is above than ephemeral threshold. I have no issues with them in or out as long as they don't exist in the db as stand-alones. I really don't want to see the entire DarkHorse catalog included. Deciding where the threshold is .... now that's difficult. Almost have to be done author by author. Not very objective but how else? ~Bill, --Bluesman 17:14, 26 January 2010 (UTC)
I'd be quite happy if we allowed true graphic novels, but not collated comic book series. But it's not simple to define such - e.g. I'd include Watchmen as a graphic novel even though it was published first as a comic book mini-series, but exclude the Swamp Thing books as just bound volumes of consecutive issues of an ongoing series. An author-by-author exception might be easier to implement, so long as it doesn't allow every single comic: or Alan Moore would become a massive problem. How do we decide on when a threshold has been passed though? And where to record it? BLongley 19:51, 26 January 2010 (UTC)
As an aside, I seem to remember putting back Stardust by Gaiman (and sometimes Vess) as it seems to have been deleted entirely at one point, when it's definitely a novel with some 'Graphic Novel' or 'Illustrated' versions. Sometimes we go too far on the zapping. BLongley 19:51, 26 January 2010 (UTC)

Merging chapterbook records

I'm holding a submission which wants to merge this 1991 chapbook of Beggars in Spain with this 2000 e-book. Maybe it's my general unease about dealing with non-printed materials, but I'm hesitant about doing so. If there's a good reason for doing so other than "they're solo publications of the same novella-version of the title" please let me know. All of the published versions of the novella show up under it's title record so that should offset the need to have the chapterbook records merged. The only other reason I can think of is that it de-clutters the author's summary page. The best reason for not merging the two titles is that I believe chapbooks are unique publications, and that another publication of the work shouldn't be considered a reprinting of the earlier work. Of course, publications of novels are merged without there being misapprehension that a later printing is a reprint of an earlier one. Just throwing out some ideas, or trying to talk myself into approving the submission. If anyone can get me off the fence, please help. Mhhutchins 02:41, 1 February 2010 (UTC)

I don't see any problem in merging them. We could add a new categories, Chapter-ebook & eNovel. :-)Kraang 04:06, 1 February 2010 (UTC)
I had similar concerns but we also merge some vary disparate book publications. I would say merge also. The only situation that might be of concern is if a chapterbook contains more than just a single story - and there are differing opinions as to the borderline between collection/chapterbook as regards that issue.--swfritter 16:35, 1 February 2010 (UTC)
That's the reason I chose not to merge this chapbook with this one, although both a publications of the one story by Lovecraft. Mhhutchins 17:20, 1 February 2010 (UTC)
As I recall, I started out using similar criteria and merging Chapterbook titles only when they were more or less similar (for admittedly poorly defined values of "more or less" :-) ), but eventually became bolder about merging standalone chapbooks, e-books and audio books. Not that I have a whole lot of experience with them since Chapterbook titles are still relatively new. Ahasuerus 03:33, 2 February 2010 (UTC)
I'll accept the submission and merge the two records. When the contents of the publications are more disparate than these two, I might have a more convincing case not to merge them. :) Thanks. Mhhutchins 05:37, 2 February 2010 (UTC)
Similar to the situation where two collections with the same title have different contents in which case it becomes a judgment call as to whether there should be a merge. I don't think very many users are going to look at the chapterbooks. They are more likely to look at the story in the short fiction list so aggressive merging would seem in order.--swfritter 15:59, 2 February 2010 (UTC)
We could probably do with a user poll to see what people actually do use ISFDB for. I know I do look at Chapterbooks as that's the only way I know when I've collected all physical books by an Author - and having the e-books in there as well is a nuisance for me. Still, merging them reduces the number of times I have to look at them only to find them of no interest. BLongley 18:01, 2 February 2010 (UTC)

(unindent) we currently routinely place ebook editions of novels as publications of the same title as physical publications (that's what all the PG editions do AFAIK). It seem to em that all separate publications of a given text, whether on paper, ebook, audiobook, or graven in stone should be publications of the same chapterbook title. -DES Talk 22:56, 1 August 2010 (UTC)

Bravely suggesting update to Help concerning quarto mag size

Per this discussion. Dastardly rule breakers have made quarto a default size for magazines. If no one has any objections in the next couple of days I will add it to the list. Those who broke the rules in the past will still be subject to punishment; in this case an extra fifteen minutes a day of ISFDB editing per day for the next week. Also, we need a decision on tall paperbacks.--swfritter 16:40, 2 February 2010 (UTC)

Only fifteen minutes? I could do that with my eyes closed (and have!) Mhhutchins 17:55, 2 February 2010 (UTC)
Add quarto by all means. A definition would be good though, I only enter that if it's been called that elsewhere, e.g. at the BSFA Index. BLongley 18:12, 2 February 2010 (UTC)
According to this it is 10 in. × 8 in. which which puts it between pulp and bedsheet. Is that consistent with the titles that are in the db? Curiously Contento lists thru the August, 2006 issue of Weird Tales as quarto but with the following issue begins listing the size as 8" x 10½". I wonder if there is an actual size change?--swfritter 19:55, 2 February 2010 (UTC)
We might also want to state that quarto is preferable to A4 if there is not an exact size match.--swfritter 23:35, 2 February 2010 (UTC)
OK. Proposed update. Quarto -- 8" x 10" through 8.5" x 11". I will update Help in a couple of days. We might possibly need some minor tweaking at some time.--swfritter 16:44, 8 February 2010 (UTC)
Contento lists sizes up to 9" x 11" as quarto. Before I make the change to 8" x 10" through 9" x 11" I would really appreciate some positive input because this change may require some updates to existing data. Fortunately we have been entering invalid data on a consistent basis. There are about 1000 pubs entered as "quarto" and 500 as "bedsheet".--swfritter 15:15, 17 February 2010 (UTC)
According to some sources the difference between bedsheet and quarto is more than just size. It's the binding as well (after all, this is the binding field). Quarto, I believe, is center-stapled (sometime referred to as "saddle-stapled"), meaning that each 17" x 11" sheet is folded in half then bound through the center by staples. Bedsheet is cut sheets that are side-stapled, forming a square spine instead of a folded one. Should binding be considered when determining whether a magazine is bedsheet or quarto? Or am I opening another can of arbitrary, indeterminate, and inconclusive worms?
I would also argue that our use of the term "quarto" was perhaps non-standard, but never invalid. Those of us who used it have one of the most respected bibliographers in the field as our source. I think our problem here is that there are so many different sizes and bindings that it's difficult to pigeon-hole every publication into a category, especially when the difference comes down to fractions of an inch. Mhhutchins 16:46, 17 February 2010 (UTC)
If it's not a standard listed in Help it is invalid. Otherwise anybody can enter anything they want which at times seems to be pretty much the case. Size is pretty much the only basis in Help for determining a value. It would actually be better if we had much more generic terms that don't imply a binding type.--swfritter 17:57, 17 February 2010 (UTC)
Without a dropdown menu, people can, and will, enter anything they want in any field. The terms "hardcover" and "paperback" have nothing to do with size. It's the binding. If the field was intended only to denote size it would be easy, just enter the dimensions. As far as non-standard being equivalent to invalid, I refer you to Hamlet, Act 1, Scene 5: "There are more things in heaven..." So, if it's not listed in our Help pages, it's invalid. Hmmm. Maybe someone just never got around to listing an otherwise valid binding type. Sorry about the snarkiness. I'll bow out of the conversation and allow any other interested parties to have their say. Mhhutchins 19:11, 17 February 2010 (UTC)
I will leave it to editors who entered the data to find a way to define the term in Help. My only intent was to resolve a problem created by those who were entering undefined data which in turn led to the discussion about valid entries for this field. It would seem that a term which is used a thousand times in our database should have a definition.--swfritter 19:41, 17 February 2010 (UTC)
I'm in the Bill Longley camp: I have used it in the past, but only where the source I was working with used it. So I'm afraid that while this editor would like to see it officially allowed, he isn't able to contribute to formulating a definition.... --MartyD 22:44, 17 February 2010 (UTC)

Large Digest

Maybe we should also consider adding the other non-digest (large digest?) formats too? Asimov's and Analog have substantially different dimensions than the standard digests. Sheila Williams identified these in an editorial as "F-trim size" and "L-trim size." Asimov's and Analog were in "F-trim size" from 1998 through 2008, and Asimov's is now "L-trim size," not sure about the present Analog.--Rkihara 19:12, 2 February 2010 (UTC)

I agree. The older digests seem like dwarfs. Large digest as a generic term makes sense unless there is a significant difference between F and L. I get both Asimov's and Analog as ebooks so I can't compare them.--swfritter 19:55, 2 February 2010 (UTC)
Using a tape measure, the F-trim is 5-1/4" x 8-3/8", and the L-trim, 5-7/8" x 8-9/16".--Rkihara 20:24, 2 February 2010 (UTC)
If we suggest a size of 5-1/2" x 8-1/2" that would accommodate both. We could specifically mention in Help that F-trim and L-trim are included in this category. Started a new section to avoid confusion.--swfritter 23:35, 2 February 2010 (UTC)

Unwanted pseudonym link

A new editor has posed an interesting question, from a discussion [here].

" ... I've also noted a pseudonym that is not connected to the real name, but I know in this case the author does not wish her work under the pseudonym associated with her real name due to subject matter. How is a case like this handled? Thanks! --Ruadhan 16:28, 8 February 2010 (UTC)"

We've tried to accommodate writers' wishes before, and think this would be the correct way to go. ~Bill, --Bluesman 17:22, 8 February 2010 (UTC)

I think we only accommodate a writer's wishes until the time when the pseudonym becomes public information. We wouldn't create pseudonyms based on rumor or gossip anyway. But if a writer makes a public statement about the pseudonym, regardless of their wishes, I think the database should acknowledge the link. Where did Ruadhan get the information? If through the author herself, why even bring up the question? Mhhutchins 18:12, 8 February 2010 (UTC)
That's right, the test is generally "is this information publicly available?" (which eliminates rumors and private communications) rather than "does the writer want this information to be known?". For example, we had Nathan Archer properly attributed years before Lawrence Watt-Evans officially came out of the closet because the data was publicly available elsewhere. Ahasuerus 19:40, 8 February 2010 (UTC)
He/she is a writer, and seems to know a lot of others, from the discussion. The last part of which brought up a question of an accent that is in Ruadhan's real name, in the DB, and I'm not sure my response was correct. I know there is something in the software about accents but I can't remember exactly what. Would you mind looking at the thread? Thanks. ~Bill, --Bluesman 19:35, 8 February 2010 (UTC)

When does a Fanzine become a Magazine?

I just noticed that Paperback Inferno, Volume 4, Number 1 doesn't have an ISSN but Paperback Inferno, Volume 4, Number 2 does. They're still poorly-duplicated A4 corner-stapled editions though (but by now sometimes they use different-coloured paper). Looking ahead, by the mid-70s edition numbers they're A3 folded and stapled to make A4, and sometimes have proper cover-art. Is the criteria set down anywhere? BLongley 21:06, 22 February 2010 (UTC)

(Aside.... Never has such a small stack of paper looked so daunting. I'm renaming the Stableford Collection the Albatross. ) BLongley 21:06, 22 February 2010 (UTC)

I don't find anyplace where the term fanzine is defined. Now that we have magazine as a search option it is less important but the Editor records do need to be placed in a series for the magazine search to work. There are a couple of magafanzines that I have placed in both the magazine and fanzine wiki pages lists. SFRA Review, which seems to be similar in content, has its issues entered as magazines.--swfritter 15:28, 23 February 2010 (UTC)
All Locus issues before 1983 were entered as fanzines, because that's when the Hugo Awards committee determined that it was a semi-prozine, i.e. it was winning too many Hugos and shouldn't be in the same category as other fanzines. I would make the cutoff when a fanzine starts printing advertisements from major publishers. Richard Geis's Science Fiction Review never received advertising revenue so I entered them all as fanzines. Fantasy Review had paid advertising from the beginning, so I entered them all as magazines. I don't think the print run or paper quality should be a major factor, although they should be considered. At a certain point it all becomes subjective.
About the Stableford collection, Bill, do what I do. Keep them all in storage and only bring out about 10 issues at a time. Knock out an issue every few days until that stack is finished. Take a break of a week or so, then bring out ten more issues. It won't take more than a year or so to complete then. I have Tuck and Reginald looking down at me from the shelf everyday and when I get tired of the piercing glances I take down one, enter a page or two, then put them back. They're satisfied for a week or so. Then I feel those stares again... Mhhutchins 19:46, 23 February 2010 (UTC)
I think you either underestimate the size of the Stableford Collection, or you were only referring to the Paperback Parlour and Inferno issues... maybe the Vectors as well, in a year. Still, you've done most of the Science Fiction Review and Fantasy Review I have, so double-checking those would be comparatively quick. It's the items where I'll have to enter them from scratch with no copy'n'paste I fear most. BLongley 16:36, 25 February 2010 (UTC)
You're right, I don't recall the number of issues involved. How about just creating a stub record for each issue (title, editor, date)? What's that, surely less than five minutes each? Then start adding contents at your own pace, maybe a few pages at a time. You know the cliche: every journey begins with a single step. Mhhutchins 20:11, 25 February 2010 (UTC)
Stub records would actually hinder me, as I'm using their complete absence as a sign I need to do something. But if someone wants to add them and then asks me to verify them it would be a help - I did suggest such to Mike Cross, but at a bad time, so nothing has come of that yet. His BSFA data is still a good start though, even if I have to adjust Editor or Reviewer's names a lot. And concentrating on Paperback Parlour and Paperback Inferno at the moment is finding a lot of UK editions we're missing. I suspect Vector reviews will too, but those also include hardbacks that I have no expertise in. BLongley
See this wiki magazine page. Which leads to entries like these which have pubs like this. You can't get any stubbier than that. Using mag editor series to navigate through the magazines won't be as useful we don't document the existence of issues, even those for which data has not been entered.--swfritter 23:26, 26 February 2010 (UTC)
I agree with Stephen here. A stub at least points out the existence of the publication. Something's better than nothing. Mhhutchins 23:47, 2 March 2010 (UTC)

(Unindent} If advertising is the cut-off then this appears to be the first. I'm now up to some "Cover-Art" issues but don't know the first of those. I guess I'll continue with "Fanzine" for now and might redo later if we have a definite rule - I'm already reconsidering adding Interior Art and Letters to past issues as the contributors become more and more notable. BLongley 21:32, 2 March 2010 (UTC)

Multiple Uses of One Piece of Art

This is a three-part question:

1) I've begun doing edits to the Twilight Zone magazine series and (at least in the early issues) their common practise is to print one large piece of art on the story's titlepage, then reproduce details from that larger piece of art on other pages. In the case of Harlan Ellison's "Grail" in the first issue they actually do the reverse, revealing more of the one, identical piece of art on each page-spread until the full piece of art is finally revealed on the last page. So far I have only been listing art credits on those pages where the full piece of art is produced, including the unique case of "Grail". Is this correct practise?

2) Another thing Twilight Zone appears to do (less frequently though) is have one piece of art that runs continuously in a strip on every page of the story. The art on each page is different, but there is continuity (kind of like the Bayeux Tapestry) so it can be seen as a continuation of the art on the preceeding page. For example, in one story there are joggers running along the bottom of the page, becoming more crowded as the story progresses until they blur together into a black smear at the story's end. Should this only get one interiorart credit?

3) Again from Twilight Zone: There's at least one case so far (Phyllis Eisenstein's "The Fireman's Daughter" here) where one piece of art is reproduced multiple times, but each time there are minor changes (i.e. the amount of fire and smoke pouring out of a building's windows). One or more interiorart credits?

Personally, I would lean to only one credit for the first two scenarios, multiple credits for the third. Jonschaper 01:27, 25 February 2010 (UTC)

I would go with one credit for all three scenarios, unless the changes in the third are more drastic than described. Mhhutchins 00:15, 3 March 2010 (UTC)

Publication Date for Scripts

My last Twilight Zone magazine related question for now:

A regular feature in the magazine is the publication of scripts from the show. For example, Rod Serling's The Monsters Are Due on Maple Street.

The earliest date that can be pinned on this final version of the script is the date of its first broadcast as a TV show, on March 4, 1960. In contrast, the magazine reproducing this script is dated May 1981. (NOTE: I could not find earlier publications of the script itself. The only other publications of The Monsters appear to be a prose adaptation and a comic adaptation)

Should the broadcast date be used to approximate the date of the script's existence as a final product, or should its first publication as part of a collection be used? I personally like the idea of using the broadcast date since that gives a better sense of the chronology of Serling et al's work, and I think this is a different situation than that of a final draft of a book predating publication for a year or many years since 1) official copies of the script in its final form would have received some distribution before the broadcast (it's possible people own copies of them), and 2) (in contrast to the author who cannot find a publisher or the Stephen King draft that needed to be typeset and printed) the broadcast represents the script's raison d'etre. But I also acknowledge that 1) a note about the broadcast date can always be added to the title data, and 2) the broadcast itself is a different medium and may deviate from a script so the broadcasts themselves cannot be considered "publications".

Opinions? Thanks Jonschaper 02:02, 25 February 2010 (UTC)

I'd already told you my opinion in our discussion on your page, but it seems the topic hasn't rattled anybody else's cage. I see your point in wanting to date this as the time the production was broadcast or premiered, but movies and screenplays are two different animals. Being a collector of screenplays, I know the printed word more often than not diverges from the filmed image, sometimes more drastically than others. I feel the same way about first printings of taped interviews, transcripts of speeches, etc. There are plenty of essays that first appeared as speeches (at conventions especially) and later printed in fanzines, or even professional publications (see this for example). I believe the first time they see print should be the date of publication. Considering an actor's script as a screenplay is stretching it a bit, don't you think? Typing isn't publishing. Even if we did consider it, then do we date it the day production began, or the day the script was delivered to the director, or when the screenwriter pitched it to the producer, or the day it came out of his typewriter? Reductio ad infinitum ad absurdum. Mea culpa. Mhhutchins 00:01, 3 March 2010 (UTC)
I prefer publication date too. It's a slippery slope if we start using broadcast/release dates, where novelisations could suddenly get back-dated too - and when the book is a novelisation of the film of a book or short story, we could end up with some really weird results. For instance, Total Recall is September 1989, the film was June 1990, the story is April 1966, and they're apparently remaking the movie... BLongley 20:39, 3 March 2010 (UTC)
Hi, I just noticed the responses. I'll bow to the apparent consensus and use the publication dates from now on and change the ones already completed. I think novelisations and short stories are easier to date independently of films than scripts, so I see less of a slippery slope, but I also see the weaknesses on my end. Thanks to all for your opinions Jonschaper 05:02, 11 March 2010 (UTC)

James Redfield: In or Out

I'm about to enter my copies of James Redfield's The Celestine Prophecy and The Tenth Insight as I consider them as Fantasy titles. However, no online database (Locus1, Fantastic Fiction) except Ernesto Vegetti's Fantascienza mentions the author, and the ISFDB-database only has this excerpt. A Google search results in lots of Fantasy/SF references, so the question is, are they in or do they stay out? Willem H. 19:36, 2 March 2010 (UTC)

There's so much borderline junk already in the database, I don't see any problem in adding these titles. If we can include everything ever written by Robert Louis Stevenson, Anthony Burgess and Graham Greene because they wrote a couple of spec-fic titles, why not these true fantasies? Hey, Charles Fort is here. I'm surprised Chariots of the Gods isn't in the database (although there's a few reviews of it here.) Unlike Americans who buy this crap and believe it, perhaps the Italians realized it was bunkum from the start. Mhhutchins 00:12, 3 March 2010 (UTC)
R. L. Stevenson's[10] page is now much smaller after I axed a good bit of it.Kraang 02:26, 3 March 2010 (UTC)

Series: Sub-Series

A new editor's very first edit raises a question. Rookies!!! Hmmm, guess we all were one once.... Anyway, [this] is the discussion so far. I went and checked [Bova's site] and there is a chronology and a sub-series. I know we can't show a book in two series at once, but I was wondering if the numbering couldn't be interspersed? In other words have all the books currently in [the series] be numbered as they are by Bova whatever main or sub-set they may be in? Or is that too much mayhem? ~Bill, --Bluesman 01:24, 15 March 2010 (UTC)

Might have lost your point. Is it to look like this? [11]. Some caution though, I have seen several series figured out only to be broken and recombined later. Thanks, Harry. --Dragoondelight 11:20, 15 March 2010 (UTC)
General idea, though I think this one would look more broken up. Doesn't display as oddly as I thought it might. ~Bill, --Bluesman 13:59, 15 March 2010 (UTC)
Go for it and see how it displays. If it doesn't work out you can always change it back. A couple years ago many changes weren't so easily reverted. That's the beauty about most changes today. (And easier for a moderator than a non-mod.) Mhhutchins 16:53, 15 March 2010 (UTC)
I think there's a Feature Request for Sub-Series numbering within Series, I don't know if anyone's working on that though. (I'm not - haven't had coding time for ages.) BLongley 19:15, 15 March 2010 (UTC)
This is a related issue. The Doctor Who series. There are actually 2 sub-series of the Doctor Who universe, one by Bantam and the other by Pinnacle. There are only 10 Pinnacle books which are identical to the ones in the Bantam series. The sub-series as listed now is the Bantam listing with the individual pubs mentioning that they are also Pinnacle books. Suppose the user has some of the Pinnacle books and want to see what he is missing by doing a search for the Pinnacle series. Right now, the user will find nothing if searching by series. The only option is for the user to do a search for the Bantam series, then go through each pub individually to find out which ones also belong to the Pinnacle series. Major pain. P.S. I don't have same titled books from both series to do a comparison to see if they are actually identical. If they are not identical, then they should receive their own separate pub listing and a Pinnacle series be created. --Astromath 16:29, 28 May 2010 (UTC)

Handling a single credit that uses two different names

See User_talk:Hauck#artist_credit_on_The_January_Dancer. The credit in the book is to "Nicolas Bouvier/Sparth". We already have Nicolas Bouvier and Sparth (not yet linked to each other, although that will soon be rectified). Do we want to capture this dual-credit as a single name, then linked to one of the others via pseudonym, or do we want to capture this using one of the two names -- presumably whichever one we decide is canonical -- and perhaps just mention the full credit in the notes? --MartyD 12:46, 5 April 2010 (UTC)

From a practical level: If a user is looking for credits for Bouvier/Sparth they would not find the other credits if that is the string they typed in.--swfritter 16:03, 6 April 2010 (UTC)
Good point, thanks. Since there's no apparent groundswell of vehement opinion, looks like using the credit as-is and linking via pseudonym is the way to go. --MartyD 10:03, 7 April 2010 (UTC)

Separation British Perry Rhodan from American Perry Rhodan

[12] The 39 Orbit/Futura titles are shown above. All are translation reprints of the American Perry Rhodan printings. Little to no other material is included from the American Perry Rhodan series. All are novella length, but actually are the seventies standard for "short novel", this is in all printings, German, American, etc. Recognition of the differences needs to be established. Topic: Serial short novel or short novel recognition versus magazine/chapterbook Topic: minor material to be handled as current novel policy. Topic: Return to author recognition for single short novel. Topic: First five are double short novel best method of authorship recognition. Topic: Oddball Perry Rhodan printings not connected to Ace. Topic: Creation of the original German serial run. Thanks, Harry. --Dragoondelight 20:13, 6 April 2010 (UTC)

Yeesh, you'll never get an agreement with that many topics! :-/ I would like to get this sorted once and for all though, so what I want is summarised as:
a) Return all the British pubs to Book-Length categories: mostly NOVEL but the first few would have to be OMNIBUS, COLLECTION or ANTHOLOGY. Not Chapterbooks. As title-page doesn't mention Editor, credit title-page authors as Co-Authors or Co-Editors.
b) Deal with contents as the Magazine Editors wish - it looks as though we could have most of the British pubs containing a Canonical NOVEL title that the "(Complete Novel)" SERIALs in the magazines are variants of. These will eventually become variants of German titles, I think?
Just let us deal with those 39 British ones and then you can all go off and do the Dutch and German things. BLongley 20:04, 7 April 2010 (UTC)
If they are not going to be magazines then the "(Complete Novel)" SERIALs rule would not apply. It does look as though a lot of the titles are right on the novella/novel border.--swfritter 21:19, 7 April 2010 (UTC)
Fine, then if not magazine, no problems, then all it takes is to separate them, yet maintain connection to the American PR is to put the Orbit material as sub-series. That connects the material as coming through Ace. Never make a good politician. LOL Funny thought, but the short novel stories, really are more like large chapters bound into Cycles. Thanks, Harry. --Dragoondelight 15:16, 8 April 2010 (UTC)
We can't create a separate sub-series for the Orbit pubs since you can't list a Title (or its variants) in two separate series. There is an outstanding request to add support for "Publication Series" so that we could see, say, all Ace Doubles or Tor Doubles as a single list, but it hasn't been done yet. Ahasuerus 05:32, 11 April 2010 (UTC)
Some idea of what the Ace "Magazines" are going to end up as would be good though. For instance, Clark Darlton has a load of "complete novel" SERIALS that we could sometimes link back to the Orbit NOVEL if they're going to stay that way. BLongley 18:26, 8 April 2010 (UTC)
I have rearranged the first four titles a bit to see what we can do. I also spot checked a few original German pubs and they were mostly between 60 and 70 pages long although they may have gotten longer later. If we enter them as novellas, then the German editions will become chap(ter)books and the Summary Bibliography pages will list them as Shot Fiction Series, perhaps a counter-intuitive solution. Alternatively, we could make an executive decision and declare that all Perry Rhodan novellas should be considered novels.
Re: the Ace series, it's probably best to enter them as anthologies, which will also cut down on the number of "Complete Novel" records. We may also want to modify the software to display the series number next to the series name in the Content section of the Publication display screen. Ahasuerus 05:32, 11 April 2010 (UTC)
I do like the rearrangement, but I think there are a few problems.
- The original (first) German editions are magazines (pamphlets, since they are staple-bound), so according to the present rules, if they are to be novels all should have the "Complete Novel" extension (or can we make an exception in this case, can a magazine be entered as a chapterbook, can a magazine have only a novel as contents?).
- Walter Ernsting published almost his entire output as Clark Darlton (I think "Twilight of the Gods" was the only title published under his own name. I would suggest making Clark Darlton the canonical name.
- The Perry Rhodan series is divided in cycles of 50 to 100 titles. This should reflect somehow in the database (sub-series?), but I don't know how to get sub-series in the right order (show the sub-series alphabetical, or numbering them?).
There are probably a thousand other questions, that will come on entering, but this is all I can think of now. If there's some kind of agreement on how to do it, I think I have all the data needed to submit the original Perry Rhodans in large numbers. Willem 09:10, 11 April 2010 (UTC)
I only glanced at couple of later issues in German (and I hope my memory is not that bad). The ones I saw were longer than their height and were double column, thus the word count would be higher? I might get a copy for looking at in a couple of weeks, (means digging into a storage shed of a book dealer). I can not remember advertisements? I think, correct me please, that you are saying is that they are serial novels, short novels, stories, etc much like the Hardy Boys, but under a magazine weekly schedule? While the cycles seem serial, I remember from reading that the early books swung back and forth, Mutant versus Arkon type, yet each was linear, barring the Ace missed printings? Thus they appear serial in format.
Clark Darlton is fine with me if that is how it is in the German.
I think the cycles will subdivide reasonably, but number is probably important. 1-50, 51-100, etc.
I fumbled around and realized I do not have the original publication dates like 9 September 1961, but that got me thinking about two thinks I noticed on covers, not contents, would we not enter dates in German? I noticed the "Enterprise Stardust" title has on cover the asterisks around 'Stardust' but they are double asterisk in bottom position first and then top position ending. Is that of any importance? Thanks, baffled, dazzled, confounded, ignorant, and tuppled, Harry. --Dragoondelight 12:09, 11 April 2010 (UTC)
Haven't done a word count yet, but I could live with both novel and novella. The point is, they were (almost) all about the same length, so to me it looks ridiculous to have some as novels and some as novellas (sometimes even both for the same title) depending on what publication they are in.
I do have all the publication dates. The series started on September 8, 1961, and has been more or less weekly ever since.
I dug out #1 (one of the few early Rhodans I actually have), and yes, it's Unternehmen „Stardust" both on the cover and the title page. Wonder how that will look in the database. It might be better to have the title as Unternehmen "Stardust".
There were all kinds of "extra's" in the German editions, advertisements, essays about the authors, reader's letters, interior art (including detailed drawings of the spaceships from the series, call it German "gründlichkeit"), separately numbered pages in the center (Perry Rhodan Lexicon) etc. I won't be entering all of that, maybe I can point a real fanatic to ISFDB for that. --Willem 13:43, 11 April 2010 (UTC)
I think of them as short novels, but that is from reading translation (Ace/Orbit). Agreed the same standard for all.
Publication dates with title are the most important building blocks.
Personally, If this wiki supports the title that way, I think we should use it. I make many title searches with 'copy' because it gets better results.
I knew about some art from Ackerman. The rest sounds to me like a magazine. Thus we would end up with title (complete novel). I can understand not wishing to 'flesh out' the entries to the final stage. Once basic entries are made, some 'fanatic'/'funatic' will do that. We need a magazine strategist's advice on this, but it sounds like a magazine to me. Especially to make sure we have the basic format in line. The one thing I really hate is that 'Magazine' format does not allow follow-up printings. If I understand it some of the German editions were printed 5 times. Maybe, it is time to 'break' that block, unless there is a better solution. Thanks, Harry. --Dragoondelight 14:57, 11 April 2010 (UTC)
Let's wait what the moderators say about the length. Looks like Ahasuerus is in favour of novels, but that was assuming the German editions were paperbacks.
I think the first printing should be entered first, we can worry about the other printings later. Indeed, there were five magazine printings, but also several types of omnibus and nowadays there's even the e-book. Reprint magazines are allowed, but there's no cloning a magazine, so it's more work :-) --Willem 15:25, 11 April 2010 (UTC)
When we get around to adding the pulp magazines for The Shadow, Doc Savage and such we'll probably want to use the same sort of setup. We currently only have those in the book reprints but I believe each was a single magazine issue first. Dana Carson 23:23, 11 April 2010 (UTC)
Thanks for the pointers, folks, thinking about it... Ahasuerus 21:23, 12 April 2010 (UTC)

(unindent)I entered German Pr. 512 from a copy and "Perrypedia" to show what it looked like. Granted it is a 'pot luck' choice, but in it's case it looks less like a magazine than otherwise. The only non-story extras in that issue, other than interior art, did not seem to be relevant. A 1 page essay on James Lovell meeting Walter Ernsting and 1 page, not paginated, letters column for fans. Of course, someone will have to allow it in if everyone gets to look at it. Point is it is about 58 pages of written story material in double column, bigger than chapterbook, about short novel length, but not the modern novel length. It is not a magazine, omnibus, etc. Thanks, Harry. --Dragoondelight 18:59, 18 April 2010 (UTC)

The publication is now in the database. A one page essay and a page worth of letters is not a whole lot, but on the other hand, there have been reprint magazines which had no editorial content whatsoever.
Our choices appear to be as follows:
  • Enter all German originals as books. The length of the book will determine whether it's entered as a Novel or as a Chapterbook. Create a "Publication series" in the Wiki. The advantage is that it's a fairly close representation of reality and also supports multiple printings of individual issues if there were any (do we know this?). The disadvantages include a significant increase in the number of Chapterbooks in the database and a lot of clutter on prominent contributors' Summary pages since Chapterbooks, unlike Novels, require two Title records. In addition, our Publication series support is weak at the moment.
  • Enter all German originals as Magazines. This addresses "the Chapterbook problem" and the Publication series problem, but will we end up with a lot of "uncredited" EDITOR records?
  • Declare all PR originals to be Novels for our purposes. Addresses "the Chapterbook problem" and the reprint problem, but bends the rules a bit too much for my taste.
Unless I am missing some other option, I think the Magazine solution is the least of the three evils, but not by much. Ahasuerus 00:37, 19 April 2010 (UTC)
I resubmitted it as two interior arts dropped out, I shortened the note to the German Perrypedia site, and included the cover. I have seen statements that the series has been reprinted 5 times in German. If you opt for magazine, the (complete novel) will need adding. My personal fear is that magazine does not allow additional printings and there will be at least 11 foreign language editions eventually. Remember some issues had 'extras' added. Personally I dislike chapterbook, but can see 'serial' in the methodology used for the series creation, which kind of screams magazine to me. Still, it is out of my visualization league at this point, and up to those with experience, battle scars, and copies. Thanks, Harry. --Dragoondelight 11:05, 19 April 2010 (UTC)
About the choices, I have some thoughts.
  • Declare all PR originals to be Novels: I.m.o. this would stretch the rules too far. I did a rough wordcount on a few of the originals. With about 5 to 6 words per line and 52 lines per column, two columns per page and 55 pages of text, they are around 30.000 words, well below the standard for a novel.
  • Enter all German originals as books: I don’t like this solution. It would make all the originals chapterbooks (considering the length of the stories). The original series is published as a pulp magazine. I do agree with Harry about the digest size (they are 6 x 8½ inches).
  • Enter all German originals as Magazines: Looks to me as the only reasonable solution. I don’t think there will be a problem with the editor records. The storyline was guarded from the beginning by one or two of the authors (in German the “Exposé-Redaction”). From around #695 they were credited in the publications. For the first 673 issues the storyline was guarded by K. H. Scheer (except for a few when he was ill in 1971). The precise numbers are mentioned in Perrypedia, the German part of course.
To be short, I vote for magazine. --Willem H. 18:50, 19 April 2010 (UTC)
Fine. Should we take note of this from the Perry Rhodan article in Wikipedia under HISTORY; "Written by an ever-changing team of authors, Perry Rhodan is issued in weekly novella-size installments in the traditional German Heftroman (pulp booklet) format." Should we had Heftroman to the binding type to accomodate that type material? I defer to Willem as he has the material. Thanks, Harry. --Dragoondelight 12:03, 21 April 2010 (UTC)
I submitted #100, as I think it should be. If it's accepted, I will add the cover. If I missed anything, I would like to know about it. It's my first magazine. Thanks, --Willem H. 15:55, 22 April 2010 (UTC)
The pub is here, this is the image. --Willem H. 09:46, 25 April 2010 (UTC)
Looks great to me. I do think one of the magazine pros might need to establish how the series layout will look. Thanks, Harry. --Dragoondelight 12:07, 25 April 2010 (UTC)

Eastern names

See User_talk:GaborLajos#Gizella_N._Csath.C3.B3 for an editor's approach to handling Hungarian names, where our existing rules for canonical and legal names don't match the "Eastern" style of name presentation. The approach, in a nutshell is:

  • First Name + Last Name in "Canonical Name" field (backward relative to pub's presentation)
  • Last Name + First Name in "Legal Name" field (comma-less, in proper order relative to pub's presentation)
  • Last Name in "Last Name" field

I propose we either change the rules around canonical name to allow proper order (and keep legal name handling the way we have it, fixing the entries made using the above approach), or we change the rules around legal name to codify the approach above for dealing with Eastern-style names (as was done in entries using the above approach). Comments or other ideas? Thanks. --MartyD 10:11, 4 May 2010 (UTC)

To fulfil all the rules I entered now all the Hungarian names with the approach described above, but I think it would be nicer to display the names in the correct form everywhere and to achieve this we should use the "Last Name + First Name" in the canonical names. Is there any chance to allow this? I read somewhere in your guidelines that "Last Name + First Name" is not allowed in the canonical name and I noticed that in this case the "Last Name" field calculated incorrectly. But what if I correct the Last Name field manually -> the sort order will be correct and the displayed name will be correct as well also for name following the "Eastern" name order. GaborLajos 16:47, 5 May 2010 (UTC)
Are we talking only about doing this with Eastern writers? How about Western writers whose works appear in Hungarian translations?--swfritter 17:02, 6 May 2010 (UTC)
I think what we're really talking about is whether canonical name should be strictly First + Last and then what to do with capturing the "natural" usage when it differs from that versus using the canonical name field for the "natural" usage (which might be Last + First in some cases). From what I've seen of the Hungarian translations entered so far, western names are not altered on the publications (Isaac Asimov is still "Isaac Asimov"). I can't say whether that is true in, say, a Japanese translation. Speaking of Japanese, I know I've seen this same issue with Japanese names (where, as a died-in-the-wool provincial America, I can't tell which is the first name or the last name). I think my preference would be to make canonical be "natural" instead of strictly First + Last and leave Legal Name and Last Name as we currently have them specified. So an "Eastern" name would not have a First + Last representation. If some publication did do that, then we could make that representation a pseudonym of the canonical Last + First..... --MartyD 21:49, 6 May 2010 (UTC)
Canonical names are generally the most commonly credited name. If a name is credited in the Eastern fashion than it would seem that is the way to credit it in our database. Using our current standard for legal names will make it clear which is the last name. The last name field would have to be updated manually. My own preference would be that Western names in translated works be entered in the Western style even if they are credited in the Eastern style.--swfritter 22:01, 6 May 2010 (UTC)

(unindent)Checking Hungarian records in WorldCat, I see:

  • R. L. Stevenson ; [fordította Benedek Marcell, Göncz Árpád].
  • Hermann Hesse. [Ford. Kászonyi Ágota].
  • Stephen King ; [fordította, Szántó Judit].
  • Isabel Allende ; fordította, Egry Katalin.
  • Edgar Allan Poe ; fordította Bart István.
  • Hermann Hesse. [Ford. Kászonyi Ágota].
  • Robert Louis Stevenson ; [fordította, Horváth László].
  • Lian Hearn ; [translated by Tábori Zoltán].
  • Lewis Carroll.
  • Carlo Collodi ; [fordította és átdolgozta Rónay György ; Szecskó Tamás rajzaival].
  • Kenneth Grahame ; [fordította Szemlér Ferenc ; Horváth Mária rajzaival].
  • Ursula K. Le Guin ; [fordította Füssi-Nagy Géza].

and so on, so it looks like the ordering of English/German/Italian etc names is preserved. There are other issues here, but I won't be able to get to them until some time over the weekend :( Ahasuerus 04:46, 7 May 2010 (UTC)

Hi! The names of Western writes are not changed in the Hungarian editions, their original name order preserved, but on the same edition the Hungarian names follow the Eastern name order in case of Hungarian translators, illustrators, ... Exactly according to the examples queried from the WorldCat db by Ahasuerus. BR, GaborLajos 15:21, 8 May 2010 (UTC)
That negates any concerns I might have although there are others who have more experience with non-English publications who might have other concerns.--swfritter 14:54, 9 May 2010 (UTC)

Entering Braided Stories

The following is cut and pasted from my talk page (some irrelevant portions omitted). Issues regarding the handling of braided stories arose when I created a series listing for Poul Anderson's Flying Mountains stories collected here. This collection includes stories originally published in magazines plus previously unpublished stories and an interstitial sequence tying the various stories together. Since this involves issues beyond the one publication, I feel rules and standards is the most appropriate forum for this part of the discussion.Jonschaper 05:37, 10 May 2010 (UTC)


Do you know if the Interludes are each standalone pieces, if they are six parts of one piece, or if they are something else? I am mostly just curious (I was thinking about whether they should individually be in the series or if there is some more appropriate way to treat them). --MartyD 01:25, 4 May 2010 (UTC)

Hi, I wasn't entirely certain how to handle them either since unlike some of the other stories they were obviously written for the collection. I noticed that the Epilogue is specifically described as a "two page short story" here which suggested to me that the epilogue, interludes and prologue are somehow separate from it. Jonschaper
Yes, I saw that. I also saw that Prologue is classified as SHORTFICTION, so I assumed it may be similar. See this discussion about the handling of a "braided" story -- a single story broken into pieces and used to tie other stories together. I was thinking that approach might be better for the interludes, if it applies, but I do not know what the interludes are. --MartyD 11:28, 6 May 2010 (UTC)
I'll try contacting one of the verifiers of the collection. Cheers Jonschaper

(Unindent) Most interesting. I like the idea from the "braided story" discussion, but really don't know what this one story should be called. The present splitting away of the Prologue and Epilogue is just plain wrong, despite being so titled in the book. There is one flow, one arc through all the pieces, Prologue/Interludes/Epilogue are just one long meeting. Unfortunately there isn't one main recurring character [though a few do appear in more than one story] to hinge the series on [Missy Blades seems to be the focus of the 'braided' story and is in one story, her parents in another]. Inventing a title seems wrong, using the collection title [maybe minus the 's' on Tales???] might be the best way. Consistent. Or the full title with a bracketed addendum, like we do for (excerpt). There's enough to warrant a length of novelette, over 40 pages. Of course this pre-supposes creating one entity in the contents, which goes against the way the material is presented, not only in the TOC but in the text. Not keen on TFTFM: Prologue; TFTFM: Interlude 1; etc. either, but it might be a way to go. How would all eight parts get linked in that possible scenario? Like a mini-collection. I can't think of any way the DB can group eight short-fictions into one longer short-fiction. A mini-omnibus...¿¿? [kidding!!] This would be a beaut on the Standards page! ~Bill, --Bluesman 00:10, 7 May 2010 (UTC)

Help:Use_of_the_SERIAL_type#Other_uses has another idea that might work here. The parts could be represented as installments of a serial. Or a variation on that, I suppose these pieces could be made a sub-series (w/Prologue as #1 through Epilogue as #8). I will ping the participants in the earlier discussion to see if they have suggestions borne of experience. --MartyD 10:01, 7 May 2010 (UTC)
First, we have worse solutions to similar problems in the database. Compare the Zelazny story in this pub and that one (The Wild Cards series is a true cornucopia of different solutions to problems). I have been told that the level of detail for entering contents is often left to the editor, but I think there should be some form of guideline. In this case I think it should be something like "If the sections can stand on their own as stories, they should be entered seperately, if not, they should be as one entry, explained in the (title)notes", but that's my opinion. When I verified my edition of Tales of the Flying Mountains, the prelude/interludes/epilogue were there as they are now, and I left them as they were. I would do it different now, probably as Bluesman suggested, one entry as Tale of the Flying Mountains, explained in the notes (probably after a heavy discussion on the rules and standards page). The addition of (serial) to the title is an interesting idea. I think that would make things clearer. --Willem H. 12:35, 7 May 2010 (UTC)
I like the idea of creating one record if the story is continuous regardless of how many parts there are. The note field can be used to explain how it's presented in the book. I don't like making it into a serial, as all parts appear in one publication. I say this knowing it wasn't how I entered the braided stories in the Wild Card anthologies, but there was/is no standard for such oddities. Perhaps this discussion should be moved to the Rules & Standards page? Mhhutchins 13:54, 7 May 2010 (UTC)
Another example of a braided story are the Wild Card books edited by George R. R. Martin. --Astromath 16:09, 28 May 2010 (UTC)

Animal Stories for Grownup-ish Readers

I understand that talking animal stories for children are specifically excluded. I have this PublishAmerica classic up for deletion even though it targets older readers since I think the same reasoning applies to it (the author's other listed book has supernatural themes). Clearly not every talking animal book for adults should be excluded, but as in Simak's City series I would think there has to be additional elements that definitely place them under the definition of SF. From the description of Corman here, no, it isn't Conan under the sea but an anthropomorphised fish in a tale that appears to be no more SF than Disney's Incredible Journey or the many other variations about a lost trio of domestic pets. Of course there is likely a gray area given things like Disch's Brave Little Toaster stories, and it looks like Corman contains a bit of allegory. So, is allegory enough? Thoughts and opinions? Jonschaper 23:25, 12 May 2010 (UTC)

The Amazon.com description doesn't make it sound speculative at all. I couldn't argue for its inclusion in the db. BTW, $25 for a trade paperback??? Also don't forget Animal Farm, Watership Down, The Cockroaches of Stay More, The Secret of NIMH, the books of E. B. White and Kenneth Grahame, all talking-animal stories deserving their place in the db, regardless of the targeted audience. Mhhutchins 23:33, 12 May 2010 (UTC)
All good examples there of just how grey an area it is, and how we certainly have to be careful about avoiding decisions based on personal prejudice. On a tangent: According to Amazon $25 is a sales price from $30! (The author himself advertises it for $13). And this price is from the guys who published (correction: originally agreed to publish) "Travis Tea"'s Atlanta Nights because, according to them, Sci-Fi doesn't involve any standards. Maybe PublishAmerica actually publishes some good stuff (I haven't made the effort to find out) but I cringe whenever I look at their website and see what titles could qualify for inclusion here, like "Angus Grady: The Beginning: Enforcer of Allard Island: His Personal Journals". Jonschaper 00:28, 13 May 2010 (UTC)
OK, I just read the "Product Description" and boy is this a stinker! It's definitely a fantasy(deranged author) of some sort but 322 pages! Yikes! I say lose it!Kraang 02:02, 13 May 2010 (UTC)
$25-30 for a trade paperback is not uncommon for vanity publishers, which is basically what PublishAmerica is. As to whether we want to include it or not, well, the loss of a borderline book by a vanity publisher certainly won't be a disaster, but it may be difficult to come up with an unambiguous rule. How do we draw the line to include a "talking animals" book like Animal Farm and exclude a "talking animals" book like this one? Ahasuerus 03:11, 13 May 2010 (UTC)
The reason I put the deletion on hold and asked Jon to seek other opinions is pretty much exemplified in ' reaction. That it may be tripe and nothing any of us would willingly read is really beside the point. I assume the rules are there to help us put personal feelings aside and have some degree of uniformity in inclusion and exclusion. Our specific exclusion of talking animal books aimed at preschoolers is easy enough to apply -- this work fairly clearly does not fall under that -- and we generally have an inclusionist bias. What in the rules/definitions would make us decide that Watership Down is in but this is out? --MartyD 10:03, 13 May 2010 (UTC)
Unless anyone chooses to voice agreement with the err-on-the-side-of-keeping-it sentiment within the next 18 hours or so, I will un-hold and approve Corman's deletion.... --MartyD 15:13, 16 May 2010 (UTC)
It's so awful I vote for deletion, despite agreeing with Ahasuerus. Maybe we should have a "Vanity Press Rule," allowing for deletion or exclusion of such publications based on editorial judgment?--Rkihara 16:37, 16 May 2010 (UTC)
Perhaps it'll be a precedent we can apply to some of those awful submissions Fixer dredges up. Yeah, they have vampires in them, but.... --MartyD 18:18, 16 May 2010 (UTC)
A little bit of background information is probably in order. Back at the dawn of time, i.e. ca. 1995, when the first iteration of ISFDB eligibility rules was crafted, all vanity publishers were "out". At the time, there was a reasonably clear separation between "real" publishers and vanity publishers and we could easily draw the line between the two sets. Not only did we assume that there was no cross-over appeal, but the vanity stuff was typically too obscure for us to monitor reliably and, besides, most ISFDB users had no way to access vanity press-published books.
A few years later, with the rise of Amazon and other on-line booksellers, the situation changed. Self-publishing (and other forms of non-traditional publishing which often came close to vanity publishing) took off and a number of mainstream SF writers used them when they couldn't find a traditional publisher. The distinction between self-publishing, POD, fanzines, small presses and "standard" vanity publishers became increasingly blurred. Granted, a vast majority of vanity-published books still contain the same awful slush pile stuff that has always been the trademark of the field, but it was no longer a given. In addition, ISFDB users can now easily access most vanity/self-published books on-line.
At first we thought about adding "quality" or some other considerations, e.g. "Has this author been published by a mainstream publisher?", to the list of eligibility criteria, but there were so many different permutations that eventually we gave up and opened the flood gates.
Re: "Quality". It was not added to the list of eligibility criteria in part because traditional publishers have always produced their share of sub-literate stinkers (Badger, anyone? :-), in part because judging quality is inherently subjective and well outside of our shared area of expertise, and in part because most of the time we have no access to the text of the books that we catalog.
Now, nothing says that we can't change the rules again, but we will first need to address the concerns listed above. Ahasuerus 18:31, 16 May 2010 (UTC)
According to the latest Locus there were a MILLION books published last year including the non-traditional sources mentioned above. If only 1% are spec fiction that meet our criteria then that means 10,000 books. And Lulu.Com is going public. "Vanity" publishers like Lulu are now becoming the proving ground for new authors as the big-time publishers focus on proven names. Somewhat overwhelming especially when you consider that nearly all of our moderators, including myself, are mostly interested in legacy S-F.--swfritter 19:07, 16 May 2010 (UTC)
The impact of non-traditional publishing on the number of new pubs that we need to process can be indeed substantial. Certainly if we couldn't get to the books published by Gollancz and Ace because we were working on PublishAmerica, we would have a major problem.
By the way, Fixer can query Amazon in two modes: "major publishers" only and "all publishers", so I usually try to get the major publishers a few months ahead of time and the rest a few weeks in advance. Over the last month or so, Fixer has also been configured to auto-suspend all major vanity publishers, but he still finds and submits quite a few non-traditional books.
However, Fixer's logic aside, the main concern that I have at the moment is the consistency of our "Rules of Acquisition". If we make them too subjective, then we will have one set of editors entering borderline books and another set of editors deleting them. That's a recipe for frustration, conflict and gafiation. Ahasuerus 19:39, 16 May 2010 (UTC)

(unindent) Well, let's step back from the larger dreck issue and see if we can decide this one specific case. Watership Down is in; I imagine we're all content with that. Under the current rules, what is different about Corman the Carp that would warrant its exclusion? If it is in under current rules, then I think it should stay pending any applicable rule changes, right? --MartyD 20:46, 16 May 2010 (UTC)

Sounds like the safe thing to do. I also find that I feel much better about leaving borderline titles in the database if I assign a tag that lets other users know what it's all about, e.g. "techno-thriller" for near future shoot'em-ups with marginal speculative elements. That way our users are properly forewarned what they are getting themselves into :) Ahasuerus 00:31, 17 May 2010 (UTC)
Given the lack of support for keeping it and the more strident support for deleting it, I approved the deletion. Perhaps we can make expand the point about books with talking animals to cover it (books aimed at children featuring talking animals but no other speculative content?). --MartyD 10:02, 24 May 2010 (UTC)

Space Clusters graphic novel inclusion

According to the note here for Space Clusters, it is "only included because of artists other works." The artist in question is Alex Nino. Although 1) I have mixed feelings about excluding some graphic novels, 2) Space Clusters is firmly within the SF genre and 3) Alex Nino has done a bit of work for SF titles (most notably the Weird Heroes series), I think there's stronger cause for excluding it. If Nino's "other works" is sufficient cause for exception to the general "no graphic novels" rule then there'd be tons of Alex Nino work that should be included (all the Warren and DC horror anthology titles just to scratch his prolific surface), as well as tons of work by other artists like Wally Wood, Gray Morrow, Don Martin, Gene Colan, Mike Ploog, Dan Adkins, Jeffrey Jones, Berni Wrightson, Harry Harrison (as artist), and many dozens of others in order to be consistent, WHICH of course brings us to the reason for the general exclusion of graphic novels: to avoid turning this into a comic book database. On this basis I think Alex Nino's work as an artist on included titles is not a sufficient reason to treat Space Clusters as an exception from the "no graphic novels" rule. Thoughts? Jonschaper 03:52, 25 May 2010 (UTC)

I don't think notability of the artist is enough. Note that the author appears to be incorrect - it's Arthur Byron Cover, also already present here, but I don't think he's notable enough either. BLongley 12:42, 25 May 2010 (UTC)

Publisher rules question and suggestion

Is there a standardized way of entering the publisher in the publisher field? What I mean is that a lot of Del Rey are entered as just Del Rey and others as Del Rey / Ballantine. Others are similar. If there is no standardization I would like to put forth one: Imprint / Imprint's Owner. Another standardization would merge same publishers. What I'm talking about here is there are a lot of pubs that use Ace for the publisher and and other that use Ace Books. There are others out there. I'd like to see them merged into one, e.g. Ace Books. --Astromath 15:59, 28 May 2010 (UTC)

There is no standard, and past discussions show that we have two diametrically-opposed camps at least. I for instance prefer "Del Rey" rather than "Del Rey / Ballantine" as the "/ Ballantine" is redundant AFAIK (all Del Rey's by Ballantine) and would in fact make confusion with non-Del Rey Ballantine editions more common. And over a longer period the ownership of imprints changes, publishers become imprints or divisions, and the "imprint / publisher" may become "imprint / division / publisher" or something even more long-winded, with no real benefit. Fragmentation of the imprint over multiple Wiki pages also increases, as does the use of Publisher within the database: e.g. I'd like to standardise NEL and New English Library - preferably on the shorter name, to reduce typing. BLongley 16:22, 28 May 2010 (UTC)
The only long-term hope I can foresee is to have imprint AND publisher fields - start them off as the same data, but allow one to be the consolidated imprint and the other free-text for any other ownership junk people want to add (I don't want it at pub level, some people obviously do). OR make it even more complex and have the full hierarchy of imprints, publishers, division, publishing groups, etc, AND the changes over time. In the meantime, I'd just check to see whether the Publisher or Publisher Wiki page exists before you consider changing the publisher publication data - it indicates someone's been working with that name and probably won't appreciate the links being lost. BLongley 16:22, 28 May 2010 (UTC)
Ok. I'll accept that, but what about my other suggestion about merging Ace with Ace Books. One is definitely not an imprint of the other. This seems to me a difference that comes from taking the name of the publisher from the cover vs taking the name of the publisher from the title page. --Astromath 16:50, 28 May 2010 (UTC)
I'd be happy with simplifying Ace books to Ace, not sure how others feel. As to Title-page publisher - I find that that's often too high a level to capture the imprint, and usually go by spine. Especially after finding title-pages where the publisher is just a logo: e.g. a G or M in a star pattern for Gollancz and Millennium respectively. BLongley 19:28, 29 May 2010 (UTC)
Librarians rely solely on title page credits, and avoid the use of any information from the dustjacket (whether it be author, title or publisher credit). This may have come about because dustjackets were once considered ephemeral, simply marketing tools, even sometimes removed before the books were placed into circulation. Mhhutchins 19:56, 29 May 2010 (UTC)
I also would like to see some consistency about publisher credits, but am not ready to suggest that there be only one location in the book as the prime source to complete the publisher field. I've seen many books published within days of each other from the same publisher that aren't consistent about how the publisher is credited. I'm also unwilling to go back to check that each of the pub records that I have verified takes its publisher credit from the same location. An area that could be made consistent is the use of "Company", "Ltd.", "Books", "Publications", "Publishers" etc. at the end of the publisher name. This doesn't suggest that these be dropped entirely for every publisher, but that we could try to arrive at a consensus about how individual publishers should be entered. For example, the entries for "DAW" and "DAW Books" are evenly split, and if given a choice, I'd rather they all be "DAW Books". I've never seen a book published by DAW without the Books. Adding those bites of data to a thousand records won't make that much difference in the size of the database. But if the consensus comes to DAW, I could live with it. (Same as NEL and New English Library). There was an effort to arrive at canonical names for publishers several years ago, but fizzled out because the problem with imprints / publishers / corporations couldn't be overcome. If an additional field for imprint could be easily added to database records, it might revive the efforts to establish canonical publisher credits. Mhhutchins 19:50, 29 May 2010 (UTC)

Problem ISBN 812

I have several books that have ISBNs start with 0-812-5????-?. I've been using Readerware and its been rendering the ISBNs as 0-8125-????-?. I wrote to Readerware to inform them of the problem and they wrote back that there is no problem. The publisher code 812 does not exist. They pointed me out to this website: International ISBN Agency to back up their claim. Readerware customer service said that the ISBN rendering of 812 is an error on the publisher's part. I wish to put forth the suggestion that all 812-5 ISBNs be changed to 8125 for the correct ISBN standard for that publisher code. If you need an example, look at Geis of the Gargoyle --Astromath 20:23, 4 June 2010 (UTC)

I'd rather that we record what the publisher prints in their books. Imagine being a moderator and you have editors who make submissions wanting to "correct" the hyphens in the ISBN. Currently the system has the ability to automatically place the hyphens, and it's correct at least 99% of the time (a non-scientific estimate). Perhaps an exception has been programmed to account for Tor's mistake. Granted, Tor screwed up, but links to Amazon, Abebooks, WorldCat work. Twenty plus years and 2600 plus books later, I'd rather just leave 'em be. Mhhutchins 22:04, 4 June 2010 (UTC)
Two points somewhat unrelated to each other:
  1. Exceptions for 765 and 812 have been programmed to account for Tor's mistakes. See ISFDB:Community_Portal#ISBNs_display_corrections. The correct group length for 0-756... and 0-812... is four -- 7653 is assigned to Tor, 7656 is assigned to M. E. Sharpe; 8123 is assigned to Great Source Education Group, 8125 is assigned to Tor, 8129 is assigned to Modern Library. When I recently redid the ISBN formatting to handle the many groups that were not being handled correctly, I researched these encoded exceptions (which predated my involvement). Tor consistently misrepresents its numbers as three-digit. For one of the two groups (sorry, I forget which), I found another publisher doing the same thing. For the other of the two groups, the other publisher used four digits properly, but it seemed unlikely any of those titles would ever end up in the ISFDB. So I decided to leave the exceptions in place.
  2. The reason behind not storing valid ISBNs as entered is to support ISBN searching with any semblance of efficiency. So we remove the punctuation on the way in (only if a valid ISBN, with no leading "#"), and then re-format on the way out. With the recent "inferred" ISBNs cropping up, it seems a better way to handle this would be to have the existing field capture the value as entered and have another field capture the "normalized" number....
Anyway, that's the history. I have no problem removing the exceptions and allowing Tor's numbers to be formatted according to the rules. --MartyD 01:38, 5 June 2010 (UTC)
How about this then. Have the ISBN be displayed correctly and have the note field pre-filled in with how Tor displays it. Of course, the down side of this is the Tor number being accidentally deleted. I guess the other way it could be done is to use a double ISBN display only for those "incorrect" numbers, so it would display something like this: 0-8125-####-# (0-812-5####-#). This might be the better way than pre-filling the note field. So forget my first suggestion and go with this one. --Astromath 02:59, 5 June 2010 (UTC)

Adding Publication to short story

In some instances, novella length, but originally novel (short novel, the novella of today standards) in publication, have been changed to Short Story. Unfortunately the short story format does not have "ADD PUBLICATION TO THIS TITLE". Therefore, I have to request a change in category to NOVEL to add publications. I think that with 'chapterbook' and this problem we need to have the 'add pub' link inserted in the short story format choices. Note, I agree that the books, sic, concerned are now of novella length, but this problem is compounded by the 'enlarged type' publications for many Fearn titles. Worse, yet, Philip Harbottle, who bought the literary estate of Fearn, had access to unprinted manuscript forms of many titles and it is very hard, unless you have the several formats, to separate. This is my observation, and I am not 'vested' or 'determined' to have my way, but I thought it best to bring the problem, which is increasing, to your attention. Thanks, Harry. --Dragoondelight 12:37, 17 June 2010 (UTC)

The original reason why "Add Publication to This Title" was disabled for Shortfiction titles was that editors were using it to enter collections/magazines/anthologies where the displayed short fiction title appeared. At the time, we didn't have full-fledged support for Chapterbooks, so we just disabled the option. I suppose now that Chapterbooks are better supported, we could re-enable the link, default the added publication to Chapterbook and add a new Chapterbook title record automatically. Would that work for everyone's purposes? Ahasuerus 03:00, 18 June 2010 (UTC)
I know I have wanted it from time-to-time. --MartyD 10:24, 18 June 2010 (UTC)
I see that we already have a similar (although more ambitious) feature request. I think the "Add Chapterbook edition of this title"option as discussed above is a more straightforward approach. Ahasuerus 15:16, 19 June 2010 (UTC)
Even if every type other than Chapterbook is disabled, some new editors will attempt to use the function to add publication of the story in a container pub. They will try to work around the default, leaving the cleanup or rejection/explanation to moderators. I see no problem with adding a new Chapterbook record and merging the resulting Shortfiction title records. Mhhutchins 17:30, 19 June 2010 (UTC)
I've only ever really wanted such to add a non-genre Magazine first publication. (Another comparatively new policy change, not yet really supported by our software.) But I think "Add Publication to This Title" for shortfiction should be allowed, although I think there should be NO assumption of what type of container title it is in. Nor Editor/Author, or title. Or date. Add new Collection is possibly the best default, with the first shortfiction entry filled in. But Novel should still be allowed - I can't let the lengthists win this for juveniles and older novels. BLongley 00:10, 20 June 2010 (UTC)

Help Pages conflict

New editor [Chavey] makes a good point: In correcting my edit of Tiptree's Tor Doubles, you referred to the Help:Screen:NewPub page which describes the use of Omnibus. I had based my edit on the ISFDB_FAQ, where (under section 3.2) it says of Tor Doubles "You enter these in ISFDB as omnibuses." As such, it seems one of these pages has the wrong policy listed. It certainly seems odd that James Tiptree, Jr. is in three Tor Doubles: one of which is listed as an omnibus, and two of which are listed as anthologies. And that's especially odd since one of the "anthologies" contains a novella by Tiptree (Houston, Houston) that was published as an independent book. So, the policy seems somewhat confusing to me.

Now, I don't want to get back into how we treat Ace, TOR, or Belmont doubles [all three often mis-represented the contents as novels when clearly many/most are not], but obviously the Help pages are offering different guidelines. With the advent of many smaller presses issuing more Chapbooks, the previous standard of 'individually published' has been stretched somewhat. Thus the Help isn't quite broad enough. Since Chapbooks can be anything from a stapled pamphlet to a high-end leather-bound slip-cased hardcover, from $0.25 to $250, the 'individually published' boundaries are quite large. I'm not advocating any preference between Anthology or Omnibus for the Doubles, just pointing out, as Chavey has, that our Help on the subject needs some tweaking. Thanks! --~ Bill, Bluesman 14:56, 19 June 2010 (UTC)

The Help pages should be made more clear, i.e. different situations require different methods. A single standard for Ace and Tor doesn't work. Ace's "novels" were original publications, while Tor's doubles reprinted many previously published works. We can't retroactively assign a type to a title based on the fact that it was reprinted as a "novel". If the first publication of the title is typed as SHORTFICTION, it should remain that type for all printings, even standalone publications (as "Houston, Houston..."). A publication of two SHORTFICTION pieces is an anthology, that of one or more NOVELs is an omnibus. So some of Tor Doubles will be anthologies (like here), and some will be omnibuses (like here). I'll look over the Help pages to see how this "double" standard (!) is handled. Thanks. Mhhutchins 17:42, 19 June 2010 (UTC)
I've made some modifications of the ISFDB FAQ regarding Ace/Tor Doubles. Any concerns or other suggestions about my changes? Mhhutchins 18:21, 19 June 2010 (UTC)
Looks good. Thanks, Mike! --~ Bill, Bluesman 20:39, 19 June 2010 (UTC)
BTW, keep in mind is that once Publication Series are implemented, "Ace Doubles", "Tor Doubles" and "Belmont Doubles" will be converted to Pub. Series. It should have no impact on their type, but hopefully it will make life a little easier by eliminating one of the exceptions to the rules. Ahasuerus 21:06, 19 June 2010 (UTC)
So when a story less than novel length is published in a Tor Double the magazine version is considered to be the original publication? But if a novella is originally published in a magazine and then as a part of an Ace Double the Ace Double is considered to be the original publication? Isn't that still inconsistent? Shouldn't the logic for the inconsistency also be a part of the FAQ? The only logical reasons I can think of are that the work is labeled by the publisher as a "novel" and that is where a user might possible expect to find the title. And make the novella a variant of the "novel" with an adequate explanation .--swfritter 22:22, 19 June 2010 (UTC)
If there's an implication that a previously published novella reprinted in an Ace Double is an original publication, it wasn't intended. My purpose in adjusting the FAQ was to point out the inconsistency with which we handle the two publishers. Yes, it's a double standard. But it's the way Ace Doubles were entered into the database. (And I personally entered very few of them.) I was attempting to explain the reasons for the inconsistency, but perhaps I didn't do it very well. If anyone can do better, please proceed.
Are there any Ace or Tor Doubles that don't already have records in the db that will change the way all of the others have been entered? The new editor who brought up the question was puzzled by how different novellas by James Tiptree, Jr. were handled when published in the Tor Double series. I think my changes explained why this is an omnibus and this is an anthology. I also corrected the statement that "The author is set to uncredited, as there is no overall author of the book as a whole." I don't understand how something like that remained in the Help pages until now. Mhhutchins 14:28, 20 June 2010 (UTC)
Since I think there should not be a double standard and the Ace 2b novellas should be merged with the magazine versions I am probably not the right person to update the FAQ. But someone should definitely make sure that the justification for the current practice is documented.--swfritter 15:01, 20 June 2010 (UTC)
I would further suggest that if nobody is willing to update the FAQ then we should do away with the double standard. No change to the documentation is necessary to justify merging the Ace 2b novellas with their magazine versions.--swfritter 16:03, 20 June 2010 (UTC)
I agree that if the Ace versions are identical to the magazine versions they should be merged. But I'm not in a position to read and compare both versions (nor am I willing). This falls into the same debate we've been having for years: the "lengthists" vs. the "bookists". (Just as are Harry's comments below.) You and I appear to be in the first camp, but I'm willing to accept that there will be exceptions to some standards. My changes to the FAQ were based solely on how the doubles currently appear in the database. I think your last statement that the current documentation justifies the merging of the Ace "novels" with their magazine versions would find much opposition from the "bookists". Would you not agree that the average db user would look for an Ace Double title under novels rather than shortfiction? Entering them as novels should not be considered a "double standard", it's the "Ace Double standard". Just as we've created a standard for entering SFBC editions that often contradict the standards for entering trade editions. Again, exceptions have to be made under certain circumstances. I'll look over the changes I made in the FAQ and see if I can iron out any inconsistencies. I don't believe that there has to be justification for the current standard, just consistency. Thanks. Mhhutchins 16:37, 20 June 2010 (UTC)
I can live with that. I just can't live with being the one to document it. But I do think that the magazine version should be made a variant of the novel version so the relationship is documented. If we start requiring examination of textual differences we would need to justify every merge by physically comparing stories. If the stories are very close to the same length then then I think we can make the assumption that the differences are probably not significant. Considering the economics I would doubt that the authors/editors of Ace Doubles spent much time modifying the stories.--swfritter 17:44, 20 June 2010 (UTC)
Unfortunately, in this case the economics worked against us since Wollheim had to cut texts to accommodate the format :-( Even when the cover stated "Complete and unabridged", it was often abridged.
Having said that, it wasn't all that much worse than the changes that you will find if you compare a story's magazine serialization and its subsequent book appearance (think the Lensman series!), and we VT those.
To take a step back, it looks like there are two separate issues here and they are somewhat obscured by the "lengthists/bookists" terminology. First, there is a basic principle underlying our design that states that all appearances of substantially identical texts should be linked either because they all use the same Title record or because their Title records are VT'd to each other. I think we all agree with this approach in theory.
The problem is that it's not always clear when two texts are "substantially identical". On one end of the spectrum we have texts which were lightly copy-edited between two editions -- no one is going to argue that they deserve two separate and unrelated Title records. On the other end of the spectrum we have short fiction which was later expanded to novel length -- no one is going to argue that the two works should share the same Title record or be VT'd to each other (we will add support for the concept of works "based on" other works at some point, but that's a different issue.) It's only when the changes are considerable but not drastic that it becomes difficult to agree on what is considered "substantial". Given the inherently subjective nature of the issue, I don't think we can realistically come up with an objective standard.
Second, there is the old "novella vs. novel" question, which is what led to the "bookists vs. lengthists" controversy. Now that we have improved support for Chapterbooks, I don't think it's as serious a problem as it used to be, but we may want to clarify Help re: what counts as a Novel to reflect at least some current practices, e.g. juvenile/YA books are considered novels if the book is over 80 (100?) pages long.
Finally, although the Chapterbook compromise is reasonable for now, I am somewhat concerned that it makes it even harder to convert novels to novellas and back, something that was already time consuming due to our serialization rules. It will probably require software changes down the road, but that's fodder for another discussion. Ahasuerus 18:14, 20 June 2010 (UTC)
I would think the Ace Double novellas are probably the least likely to have been abridged. The novel length works that appeared in magazines and then as Ace Doubles are more likely to have been published in the magazines as serials. The bookists seem to ignore the fact that prior to the 60's the history of science fiction is primarily U.S. magazine centric.--swfritter 17:31, 21 June 2010 (UTC)
I don't think they ignore it completely, but simply relegate it to a quaint corner with gramophones and doilies. :) The first thirty years were predominated by magazines, while books have ruled the market for the past fifty years. What I dread is electronic files replacing books! Mhhutchins 19:03, 21 June 2010 (UTC)

Worlds to Conquer - Contento says novella

Moved from my wiki portion to give the subject greater air and to keep my head from blowing apart! LOL

Also note the subject was actually opened previously only two items up!

In The Locus Index to Science Fiction / Index to Science Fiction Anthologies and Collections. I have the 1984 to 2003 edition which luckily lists the Harbottle anthology and the story is listed as a novella.--swfritter 14:25, 19 June 2010 (UTC)

   That it does. It was also published as a novel first. 

To make the additions I need to 'add publications' so therefore the request to change to get the first entries correct. Now the others may or may NOT be the exact same manuscript story as Harbottle had various editions. IMO, change it so I can properly add the pubs and then you can have a hayday and change almost every publication that Fearn did between 1949-1956 of less than 130 pages into a novella. In fact almost every publication of British SF with less than 130 pages meets the novella standard of the 70's etc. Thanks, Harry. --Dragoondelight 14:34, 19 June 2010 (UTC)

       I suspect we have another Lengthist argument coming on. 

Stick by your guns, Harry! Even the Nebula Awards allow a Novella to be a Novel. And why should the SFWA be arbiters on British publications? And as SFWA was only established in 1965, their relevance to categorisation of a 1952 title should be zero. BLongley 01:10, 20 June 2010 (UTC)

The problem stems in that a story was anthologized under the same title that was used fifty plus years ago. Unfortunately, the original 'novel' publication was not entered previous to the anthology being entered. Since it was entered with the most future information possible it was entered as a 'novella', and to hang the nail a little higher it was entered under the pseudonym also. The moderator does not wish to allow me to change it to novel to properly enter the novel. He thus is advocating the backdating of a story length description upon a previously published title. Side note: We really can not physically be sure that the republished story has not been altered for length or content. Actually, have recently studied the publication era that originally was published in, it could easily be altered from the submitted manuscript form even as originally published. The literary estate was purchased by Philip Harbottle and he has publicly referenced that many of the manuscripts differ and he holds several variants of many titles that John Russell Fear produced. Back to point) The dynamic difference of opinion is whether the 'novella' description subsumes the 'novel' publication. From the simple 'word counter' dynamic the story is a novella in the anthology and using the comparison of several other (not this specifically) 'short novels' of Scion and Russell, as a pseudonym, they are not of the word length to meet the modern 'awards' valuation for novel. In this, I support his view on word length description. Unfortunately and Unfairly the issue is more complex than that. We are cataloging the publication of 'titles' not lengths historically. Thus the definitions have changed in the future, but that does not change the reality of the past. Most simply, most fiction, or other writing, is written for the 'market' with three important factors that determine the word length of the publication. The first is the author, but we readily accept that 'editing' in a multitude of forms often changes his original to reflect the other two factors. The other two factors dominate the issue. The Publisher wants to make more money for the least cost and the Reading Public (RP) determines success by their buying of the product. To be specific to the era of 1949-1956, a prolific period of British SF of a very varigated literary quality, we know that the publishers suffered from 'paper rationing' and other after war effects. Thus the physical printing demand was for short, quick, but complete stories and as the binding reflected the story title they were termed novels, or more precisely in recognition of the many factors, short novels. Understand they suffered from almost every publication malaise possible, especially formulistic plotting and required length. Writers live and die by writing, but there success is only measurable by recording of their work. Writers make their names by their titled products, not by the collection, magazine or anthologization of their work. It is self advertising. In the creation of the 'novel' or 'titled story publication' was chosen to present the writer's product. If, for a lengthist reasoning, we reshape this dynamic, we will change the perception of what happened (change the history, by shaping the visibility determined by initiating the formula used to describe the basic subject matter of the DB). Remember writers mostly work to eat and thus we and the publishers shape their production. We buy 'novels' generally with the idea that they complete in themselves.

In the above case, we essentially lost a titled product to the obscurity of his work's anthologization/collection. We thus lost a portion of the overview of his output and it's 'original publication type'. To my amazement, I recently read that John Russell Fearn is estimated to have had 5,000,000 books published under his various pseudonyms in the period after WWII. In sheer numbers he dominated that market and destroyed his health trying to feed himself, as many others did also. Like a stone dropped in the ocean, he surfaced in the Reader Ocean and was subsumed by that market's collapse. At the same time, some good to possibly great author rose from that morass, but if we 'redefine' his production by his 'belated' republication in modern anthologies using award defined criteria we will obscure the real world history. Fearn is the outstanding case of a writer of SF for his era, but the smallest author of no other known account is also affected by 're-description'. Thus Fearn being republished and redefined would be minamilzed, while John Doe who wrote one novel, which we have minimal data on, stands out because we can not 'Count' his words to categorize/re-categorize his accomplishments. Thus again I say to the moderator, I support the 'named titled publication' of novel, without differentiation for word count (in the past). I also support the categorization of the Award standards models, after their public use, but in no case where used to backdate writing/writers. I am not Revisionist. Thanks, Harry. --Dragoondelight 12:57, 20 June 2010 (UTC)

We should make it clear that all published standards can be ignored. I have a whole backlog of webzines I need to enter.--swfritter 13:53, 20 June 2010 (UTC)
I think we've already covered that, with a suggestion that if a Mod wants to be responsible for it, it's in. I know I'm breaking a lot of magazine rules with Vector - I go by first principles of "we record what's stated first, and sort out the rest later", so they go in by number, not date. (The reviews are a valuable addition for British works however the magazine appears here.) And I will continue to allow Novels that don't qualify for Hugo, Nebula, or any other award if they predate the awards, or are not intended to qualify for such awards (by being targeted at children for instance). Feel free to do the webzines - the number of reprints that first appeared on the web makes me think they deserve at least a stub entry on the same lines as our Non-genre magazine entries. BLongley 23:48, 20 June 2010 (UTC)
Speaking of magazines. How will we ever get/show the impact of the Toronto Star Weekly on Science Fiction? Thanks, Harry. --Dragoondelight 11:00, 21 June 2010 (UTC)
Whenever someone enters the relevant issues, I guess. Do you have a source? If not, Googling site:locusmag.com "toronto star weekly" gives a starting point. BLongley 15:51, 21 June 2010 (UTC)
I saw some of that and of course H&H mentioned a few in the "Star Weekly", but that cause me to think and I found [13] check "The Star Weekly: Magazine Section." Seemed sparse, as advertised, but got to wondering and tumbled this [14]. "The Asteroid Stealers" by Calvin M. Knox became "One of Our Asteroids is Missing". So it apparently was a good source of income for SF writers, but I can not find a listing or as in the first sample, no idea what the type stories they were. The light remark that the Star Weekly could do 3-4 Fearn novels a year, and the general comments that SF writers were using it for 'extra' money makes me think that it was both better known and may have greater influence on SF or ISF. I can find no handle with 'enough' data to start stubs that I am pleased with. Check the cover to the right in this [15]. I guess this is pure snarking but it truly baffles me that in the greatest information age ever we can not see what other saw and read just a few decades past. Thanks, Harry. --Dragoondelight 19:34, 21 June 2010 (UTC)
There's pages of the past, but it requires payment. Not all of us do this bibliography for free. :-/ BLongley 20:03, 21 June 2010 (UTC)
I will give it a try. (When I set aside the time) LOL I really appreciate your finding this. Thanks, Harry. --Dragoondelight 21:56, 21 June 2010 (UTC)

Definition of "shortfiction" [2]

The Glossary contains no definition for the term shortfiction (sic, as one word).

For example, the Nebula Award given by the Science Fiction and Fantasy Writers of America has established award categories:

   * Novel: a work of 40,000 words or more
   * Novella: a work of at least 17,500 words but under 40,000 words
   * Novelette: a work of at least 7,500 words but under 17,500 words
   * Short story: a work of under 7,500 words
   * Script: a script for a movie, TV or radio show, or a play

(There are more rules that are explained on their web site.)

How does isfdb define "shortfiction"? —The preceding unsigned comment was added by Rpfgs (talkcontribs) .

Any work of fiction which is less than novel-length (40,000 words). See this Help page and go down to Entry Type. Mhhutchins 01:15, 22 June 2010 (UTC)

Omni Comix

Hi, I certainly have nothing against graphic novels (some of the best examples of SF works are found in graphic novels, IMHO), but I also want to avoid the imdb becoming a comic book database (some already exist), and I think the only way to do that is to act consistently by limiting inclusions to established threshold exceptions. Which brings me to this. Since the vast majority of this publication is a comic book (so the exception to, e.g., listing the Sunpot series from Galaxy doesn't apply), and I doubt most of the contributors here make the threshold either, at the very least I think the included contents of this should be limited to any SF prose and associated illustrations as is done with the regular Omni magazines or mixed genre literary magazines. Opinions? Jonschaper 04:01, 25 June 2010 (UTC)

Here's a somewhat related question re another comic-related situation that crops up. Jonschaper 02:52, 28 June 2010 (UTC)

Originally, comics were "out", but then they began slowly creeping in, mostly because of various awards that we list (and general "importance to the genre") as well as because some flavors of "graphic novels", manga, etc come close to regular fiction. Personally, I see them as another medium and I am not in favor of including them unless the text can stand on its own (see "light novels"), but there is a wide variety of opinion on the subject. Ahasuerus 03:02, 28 June 2010 (UTC)
If it weren't for the Niven material I wouldn't object to the entire record being deleted. I believe it should be handled like a non-genre magazine with only the speculative fiction content recorded. Mhhutchins 05:31, 28 June 2010 (UTC)
It appears the verifier isn't active so I'll send him a note and delete the non-Niven material Jonschaper 23:54, 28 June 2010 (UTC)
Actually, on closer examination I'm not sure how easy it is to edit. A large number of the entries are illustrations of Niven characters, so I'm uncertain if they should be considered illustrations of the story. It might be better if someone with the publication makes the judgement. Jonschaper 00:06, 29 June 2010 (UTC)
I think we can let one issue stand until it can be double-checked. (There wasn't a second issue was there?) Without a Primary Reference, I can't tell if it's more in than out than, say, The Illustrated Roger Zelazny. Maybe leave a note on the publication that we're not interested in anything not related to works published as pure text elsewhere? BLongley 21:10, 7 July 2010 (UTC)

Books about or sort of about SF books

Hi, I've no strong objection to the inclusion of Harry Potter and Philosophy, but it seems pretty tangential considering that instead of being, say, an SF index, Tolkein Bestiary or Klingon dictionary, it is just using Harry Potter as a springboard to popularise philosophical theories and discussion, so I'm wondering what the exact rules are. Jonschaper 04:46, 7 July 2010 (UTC)

Looking at the essays, they do seem to each relate quite closely to Potter subjects. So I wouldn't delete on sight - but then I wouldn't actively go out to find such. Such titles will probably end up reviewed somewhere in a relevant pub and can be usefully entered to avoid dead-end authors (a pet hate of mine) but I wouldn't go out of my way to enter all the contents either, unless I was really keen on one of the essayists. BLongley 21:16, 7 July 2010 (UTC)

Columns

I was looking at "The Fan-Space" essays for some reason and was thinking there was a variant that needed creating. Or that most of them needed disambiguation by the issue they appeared in. And maybe a series created for them. And then I remembered that David Langford's "Ansible Link" has been merged in the same way that EDITOR records commonly are. And also that "Mutant Popcorn" is occasionally disambiguated not by issue, but by the reviews it contains. I guess if we're not going to add data to any particular edition, we can merge - but if we want to add value to any particular edition we shouldn't. Discuss. BLongley 00:03, 14 July 2010 (UTC)

Hopefully someone will take the time to fix the "Ansible Link" essays.--swfritter 00:32, 14 July 2010 (UTC)
Another one of those "wish we would have done it that way" things. I have always wished we had done "(Complete Novel) Serial" in a different way - a very confusing nomenclature.--swfritter 12:33, 14 July 2010 (UTC)
I agree that merging the "Ansible Link" essays was not a good idea. Ahasuerus 14:15, 14 July 2010 (UTC)
Other issues: Multiple and rotating authors, loss of the flexibility to link from title records to online sources - something that is done by HTML now but (hint! hint!) might be a good idea when title record fields are added. The "Ansible Link" merges make a certain amount of sense but doing them in this manner creates an example which might be used in less appropriate situations. I can sooner imagine ways to to display series more economically with program fixes.--swfritter 15:00, 14 July 2010 (UTC)
Well, merging EDITOR records has given us similar problems like being unable to link reviews of Magazines and Fanzines. "Upon the Rack in Print" is another column that shows up some of the problems - multiple authors, complicated title format, no links to items reviewed. I guess current consensus is that we shouldn't merge columns - but I for one am not particularly willing to unmerge them unless I have something to add to a particular edition of one. I can see "Mutant Popcorn" being expanded in future, for media fans, but I can't recall what "Ansible Link" usually contained. It might usefully be left merged like all the "Twll Ddu" and "Ansible" entries - just a stub entry useful for recording that issues existed and occasionally won awards. BLongley 20:49, 14 July 2010 (UTC)
I'm happy to disambiguate the "Ansible Link", "Mutant Popcorn" and "Laser Fodder" columns in Interzone (by adding the issue details in brackets afterwards) as I edit each issue's entry. There is, incidentally, a Mutant Popcorn collection due out soon, so it might be worth keeping that in mind - after all, its contents will be from the Interzone columns.Iansales 19:50, 29 August 2010 (UTC)
All the more reason for not merging them. Hopefully you will not have to fix up any of the ones that have been merged.--swfritter 21:35, 29 August 2010 (UTC)

Fairy Tales and Ghost Stories

I approved a Fixer submission of The Green Fairy Book, as we already had The Olive Fairy Book. There are many books in that series. Question: Fairy tales with no known author "(?)" are excluded. But a very similar group of stories is ghost stories (see, for example, Lord Halifax's Ghost Book). Are these really different, and should we be excluding fairy tales? The ghost stories likewise have no known author.... Should the Lang anthologies be "out"? Or are they somehow different from what the rules intend to exclude? Thanks. --MartyD 10:55, 18 July 2010 (UTC)

The Lang anthologies are particularly significant works which likely had a strong influence on many important spec fic authors. It is also my impression from looking at the beautifully illustrated Project Gutenberg editions that Lang seems to have had a very strong editorial presence; Lang could almost be considered an adapter - much like the Grimm brothers. Given the authorship issues I don't thank it is of any great value to enter the individual stories but the books, taken as a whole, transcend the fairy tale genre.--swfritter 12:24, 18 July 2010 (UTC)
In total agreement with Swfritter here. Mhhutchins 13:18, 18 July 2010 (UTC)

Adaptations and Retellings

Do we have a policy on how to handle retellings and adaptations? I have encountered a couple lately and am unsure as to the preferred way of handling them. For example, there is this retelling of H. Rider Haggard's She. When I entered this one, I chose to list it under the adapter, though I now think that may not be the best scheme. Tuck lists it under Haggard. OCLC lists both Haggard and Ward. There are also several retellings of Robert Louis Stevenson's Dr. Jekyll and Mr. Hyde: (this one and this other one, that does not yet list the adapter - see OCLC #12128687). I see the options as:

  1. List the original author only, with the adapter reflected in the notes. This would be similar to how we handle translations.
  2. List the adapter with the original author listed in the notes. This makes the relationship between the original and the adaptation hard to find.
  3. List both original author and adapter. This would be similar to how the recent fad of mashups (e.g. [16]) appear to be handled.

If we go with the last method, do we make this a variant of the original work? Do we make it a "soft" variant, i.e. add the adapter to the pub only without a new title record, similar to how we handle the presence or absence of subtitles.

I probably lean towards the last option with a "hard" variant (i.e. two title records). What say others? --Ron ~ RtraceTalk 23:08, 18 July 2010 (UTC)

I've also been conflicted about how we enter such works, especially those juvenile adaptations of classic works (Alice, Oz, Frankenstein, Dracula, etc.). This all goes back to the database's inability to create variants based on text. Until that's been worked out, I see no reason to start a new policy about how adaptations are handled, as it would be subsumed once the "based on" variant feature is implemented. Variants today are based solely on author credit and book title. Or should be. I've seen several attempts to use the current variant relationship for text variations and, in most cases, they turn out rather ugly. Mhhutchins 23:22, 18 July 2010 (UTC)
I too have tried all of these options and none of them quite fit. I have also tried another approach using series, e.g. see how a recent "retelling" of James Malcolm Rymer's "Varney the Vampyre" is currently listed on his Summary page. Ahasuerus 00:06, 19 July 2010 (UTC)
I wonder if there's a practical way to set up a "stew pot" for feature requests. This thread seems best resolved by adding some sort of Adaptation, Retelling, Revision, etc. title linkage much like variant titles. I don't have the time at the moment to do the research/thinking to see if it's practical to recycle the existing variant title related database fields and logic or if a new structure should be developed. One complicated aspect is that the tree may well be more than two deep. For example, a shortfiction work adapted into a novel which is then re-adapted for the young adult market. Foreign language translations could also be thrown into the stew pot. --Marc Kupper|talk 22:47, 26 July 2010 (UTC)

LCCNs in pub notes

The question of entering LCCNs in pub Notes has come up a few times over the last month or so. Here is what I wrote about it earlier:

  • It's important to distinguish between the LCCN printed in the pub itself and the data in the on-line catalog of the Library of Congress. The former is objective immutable data -- either the pub has this information printed on the copyright page or it doesn't. Like other bits of information found on copyright pages, it can be helpful when trying to distinguish between different editions. The latter, however, can change over time.
  • LCCNs are somewhat tricky. They can apply to multiple versions of the same book, e.g. if a book comes out in hardcover, trade paperback, and library binding, all three versions will usually share the same LCCN. However, if a later edition is procured by the Library of Congress, then it generally gets it own LCCN, e.g. the first edition of Dune uses LCCN 65022547, but the 1984 edition of Dune uses LCCN 83016030. Basically, LCCNs do not neatly map onto our publication and title records.

Based on the above, I am thinking that something like the following guidelines may be useful:

  1. If you are recording an LCCN in Notes, indicate whether it's found in the book, e.g. "Stated LCCN 66-1234" or "No LCCN in the book, but the Library of Congress lists this edition as 77-2335".
  2. If the LOC catalog doesn't have a record for the pub that you are entering/editing, but has a record for an unrelated edition of the same title, do not enter it -- this pub record is an example of what I believe we want to avoid
  3. If the LOC catalog doesn't have a record for your pub and you want to record this fact, specify when the information was entered (in case LOC enters it later)

Does it make sense? Ahasuerus 03:47, 24 July 2010 (UTC)

Makes perfect sense to me. I followed those guidelines without them being stated in our rules, except for number 3. If there's no LCCN in the book, I don't look for one on the LoC website. I only check the website when the link for the book's stated LCCN doesn't work. In that case, I remove the link and stated that there is no record on the LoC website. (Or correct the LCCN if the book printed the wrong one, and believe me that happens more often than you'd think!) I'll now start placing a date on those particular records for the reason you cite. It never occurred to me that there may be decades of books that are sitting in (taxpayer-funded) warehouses waiting to be entered into the LoC database! Thanks. Mhhutchins 05:20, 24 July 2010 (UTC)
As far as I know, the big problem with paperbacks is that they have to be rebound before they can be added to the LOC's main collection and that's an expensive and time-consuming proposition. (That's what I was told when I donated a part of my collection to the LOC some years ago.) For now, warehousing paperbacks is a relatively inexpensive way to keep them around until they have a better way. Ahasuerus 03:38, 27 July 2010 (UTC)
I do not mean to disagree with any of the above, but I think WorldCat's OCLC numbers should also be considered when discussing what to do. Except for the fact that OCLC numbers are NEVER presented on the copyright page, they have similar issues. --MartyD 10:33, 24 July 2010 (UTC)
OCLC/WorldCat data can be useful, but it tends to be messier that LCCNs: multiple OCLC numbers per pub, numerous vaporware titles, etc, so it's conceivable that we may come up with different rules for the two types of identifiers. Ahasuerus 03:38, 27 July 2010 (UTC)
I think we should add an LoCC verification. That would record who checked and when, and it would eliminate the need for identifying people or dates in the notes. In the absence of slots for other identifiers, the LCCN(s) and OCLC number(s) still need to be recorded in the notes. For LCCNs, the "standard" notation beyond recording the number would be: (a) Verifier describing discrepancies between printed LCCN and what is found in the LOC and (b) Verifier noting the LCCN is not provided. For both OCLC numbers and LCCNs, I suggest the standard be that these NOT be recorded if they do not apply to the particular pub in hand unless they're printed in the pub (and then this discrepancy would be duly noted by Verifier). The standard "N/A" verification would be available for books having no LCCN printed and not being found on the LOC if someone wants to go to the trouble of looking, and it is dated if someone is inclined to re-verify later. Just as is true for OCLC today. --MartyD 10:33, 24 July 2010 (UTC)


As a point, the example of LoC to avoid in topic 2 has indications that is the exact story reprinted by Carroll & Graf. Apparently the "Carroll & Graf" Masters of Science Fiction Series came through the Forrest J. Ackerman Agency, (mentioned as common denominator) and they are of the original edition through the Author copyright. Another recent example is "The Metal Monster". In both cases I do NOT have a undoubted source, but the connections have been touted by others. It makes sense as this type of cutout would reduce costs for C&G and Authors. Nevertheless, without this caveat, your basic statement should read: Make NO LCCN connections but direct ones in book notes? I have no problem but do think an Editor Title block slot would solve both the problem and allow 'users' to supplement their research. H.--Dragoondelight 12:29, 24 July 2010 (UTC)
Note that we do not concern ourselves with trying to identify the true "first edition" but rather we record as accurately as practical what's stated in the publications and/or what secondary sources say about that publication. When we construct a publication record entirely from secondary sources we note this. Hopefully that note serves as sufficient warning to others that the publication, or some of its purported details, may be wrong.
I'm not sure I understand what you mean by an "Editor Title block slot." Would this be a field at the publication level or are you asking that the edit-pub function also have a window that allows for editing the parent title's notes field? --Marc Kupper|talk 05:15, 26 July 2010 (UTC)
"Editor title block Slot" refers to something for people to research at the title level, not the printed level. My thought being that many/most people concerns themselves with what the title text gives them, not the printed variations we note. Turn it around. Editor title items deal with the title text, while specific printing material deals with one copy. My mind says that even specific printing material data leading to singular LCCN & OCCL records is still most useful to the most number at the editor level, especially as I consider such most useful when they can be 'most easily' compared with what they state. If you have to 'dig out' each reference at printing level, you are very likely to miss something. Granted a direct stated reference from the printing should be listed at the printed level, not necessarily linked. Still, I think comparisons made available at editor title level are the same as 'Tags'. Though I cringe at the work that it implies, I also stated that I think the 'Title editor' record should have two levels for use. The 'closed', which is normal information, the 'open' which should items like 'Tags'. Check John Russell Fearn for the result of the 'tagging effect'. Again, ideas for thought. Thanks, Harry. --Dragoondelight 12:49, 29 July 2010 (UTC)
I think the movement would solve point 3 also as why go to the trouble to plink dates down on a negative. It is redundant, or seems so as the primary verifier is dating the record below. Secondary adds though would make sense, but I doubt any would bother. As before the LCCN is actually best related to the editor title not to specific books. H --Dragoondelight 12:29, 24 July 2010 (UTC)
Harry, I don't understand this part. --Marc Kupper|talk 06:58, 26 July 2010 (UTC)
I believe he's referring to suggestion #3 by Ahasuerus that the verifier who notes that there is no LCCN record in the LoC database should also date the note in case a record is eventually created. Harry thinks the date stamp of the verification is sufficient to record the date of the verifier's note. This is a valid point. But, I think he's mistaken if he believes that the LCCN should be linked to the book's title record instead of the publication record. As I stated elsewhere LCCNs are assigned to publications not titles. Mhhutchins 04:09, 27 July 2010 (UTC)
Reply for thought only. "But, I think he's mistaken if he believes that the LCCN should be linked to the book's title record instead of the publication record. As I stated elsewhere LCCNs are assigned to publications not titles." Actually we are both right and wrong. The problem is not what they intended to do, but what they did. For instance check "The Wizard of OZ" and you find good support for your statement, though I do NOT think it is even near complete, but it sure looks like it. Thus an appearance to agree with your statement. Problem is numerous other titles, were NOT treated in the same manner. Dune has only 5 entries, so think and compare with WOO. I have NO idea how they divide up the work of doing entries. I do know they do more entries per title for new stuff, which when it is old titles, can leave stunning gaps. My contention thus is the 'overall LCCN' records due to difference in handling of titles is more an editor title level for research than directly printing level model. Another example of inconsistency in what they enter is that they only have 5 Gor novel titles. Either they are 'editing for morality' or someone taps the till, before it becomes a record. There caveat on NOT every book is a nice thing to say, but way too many titles missed being entered. Thus my contention versus 'what they say they are doing'. Just a thought response, not a hack & slash argument contention. My ideas are simply keep your mind open and make comparisons and see what you think at some later date. Proving my contentions in a short period is imprecise/impractical, but stating my opinions about this and letting it percolate may engender some kind of compromise about the value of the LCCN's. In short, I am not shooting to win an argument, but to catch your attention with another opinion that possibly can shape the results of how this DB handles this source. Thanks, Harry. --Dragoondelight 12:37, 29 July 2010 (UTC)
I still stand by my belief that LCCNs are assigned to publications. There could be exceptions, but the exceptions are probably errors. The reason there are only five LCCNs for Dune is because only five publishers have chosen to send copies of their editions to the Library of Congress. Many publishers simply choose not to send any copies. I've never seen an LCCN record which covered more than one edition, unless there are simultaneous editions in hc/tp or trade/library binding from the same publisher. Please direct me toward any record that gives different editions from different publishers which all share the same LCCN. Yes, I've seen publishers reprint the LCCN of an earlier edition on their copyright page, but the LCCN record only applies to that previous edition. Trust me, I'm open to ideas and facts that contradict my beliefs. And am able to apologize when I'm wrong. Mhhutchins 16:08, 29 July 2010 (UTC)
Case of the stated LoC in a publications. Since I reformatted my entries, I have examined several score publications for this item. Some books, even with changes, relate the LCCN 20-30 years after (rare, but there). In all cases, though it would still look better at Editor title record level and would cut down on repetitions. Why should the rare examples who repeat the same in 2nd, 3rd, etc printings repeat a hot link? H. --Dragoondelight 12:29, 24 July 2010 (UTC)
The goal of a publication record is to document what's stated in a specific publication. We don't care about the absence or presence of rare examples. Links in multiple records is fine as long as the linked-to item is about that publication. For example, OCLC does not seem to note the cover price or printing number implying that an OCLC record may well "match" multiple ISFDB publication records. --Marc Kupper|talk 06:58, 26 July 2010 (UTC)
Case of the stated LoC being wrong. Worst type of data entry, a bad reference. I had one recently that actually was for another book. So, they do need checking. My preference again, show the entry, state no record can be found for it and refer it to the Editor Title record which would keep them like 'tags' are done. Encourage the recording whenever found. H. --Dragoondelight 12:29, 24 July 2010 (UTC)
How do we know the LCCN in the book was wrong? It's possible the LOC web site is wrong or that some other publication has the wrong LCCN and as it got entered in LOC first that's what they have on record for that LCCN.
We document what's stated and if someone spots that the stated LCCN does not match what the LOC has for that number then we add a note explaining the details so that someone else looking at the record will understand that we copied the number accurately from the publication. --Marc Kupper|talk 06:58, 26 July 2010 (UTC)
The extra LoC records caused by either back entering or future entering. Again Record Title entry with a Tag format would best serve the user. It is disturbing to find LoC breaking it's own regimen, but there you have it. They broke and the deluge of extra entries will ensue. One link at Editor Title level is the easiest to address, find or eventually correct if the links are corrupted. H. --Dragoondelight 12:47, 24 July 2010 (UTC)
For now, you can add the links in the notes field for the title. Be aware that if it's a variant title that you may want to add the data at both the variant and parent title record. --Marc Kupper|talk 06:58, 26 July 2010 (UTC)
OCLC records 'pretend' to offer more, but actually are complicated 'rat's nests' of multiple entries, often replicating the same data. Therefore, the addition of a 'Tag' format at Editor Title level removes the clutter. Suggest, with no knowledge how it is done or complexities, that a hide/expand button for such data could be added. OCLC is going to be a nightmare, probably till it kills itself in excess or digests all the foreign data. Many records are duplicated in English editions by the German entries. Think of it as evolving like the human population. H. --Dragoondelight 12:47, 24 July 2010 (UTC)
Do you see any value to documenting OCLC numbers? Within ISFDB we offer a link in the left menu that'll look up an OCLC record using the ISBN. I'm not sure how you found the four numbers numbers for The Tomb. Do you do it manually or is there something where you can plug in an ISBN and it'll list all matching records? If I use Advanced search for the ISBN it returns 473405510 and 39514012 with the first not being on the list of four you listed for The Tomb. --Marc Kupper|talk 06:58, 26 July 2010 (UTC)
Personal appraisal LoC or OCLC records without hot links are not very enticing and after checking some entered by others, without the book, frequently aberrant to a research purpose. At the same time hot links, entered occasionally, in various printings cause rancor both ways, when they are there and when they are not. Additionally, they can potentially be 'hacked for other's purposes'. I love links, but centralizing their placement would help. In most cases, the deeper you delve into a research record the less it really corresponds to the direct printing it may come from, thus any record at a central site serves the user and the 'maintainers' best. H. --Dragoondelight 12:47, 24 July 2010 (UTC)
I don't think anyone has complained about the links themselves. --Marc Kupper|talk 06:58, 26 July 2010 (UTC)
Actually, The Tomb has a problem: the ISFDB pub record is for the 1998 Tor edition yet the last linked OCLC number is 244831110, which describes the 2006 edition. Based on the note next to the OCLC link ("'Author's definitive ed.'-- P.1 of cover 'TOR fiction.'" 2004, © 2006. This edition may be a rewrite to better sequence "The Adversary Cycle"), this was known at the time the pub record was created. We definitely need to remove this text from the pub record since it has nothing to do with the 1998 version of the book. Ahasuerus 03:51, 27 July 2010 (UTC)
I agree with removal of the line from the publication record and started to look for the "right home" for it. I put some preliminary notes at Author:F. Paul Wilson#The Tomb. A bibliophile's headache and possibly nightmare. Unfortunately, my local library system only seems to have copies of the 1984 edition and I don't have the time available to chase this one into verifiable publications. --Marc Kupper|talk 18:43, 28 July 2010 (UTC)
Since I am pushing a reformat for better usage. It would also be helpful if both 'Biographical written record and links' at author level and 'title synopsis/review/etc data links' at Editor Title record level were added. Both would enhance the quick research and both would enrich those topic areas more than adding at printing publications level. Thanks all, H. --Dragoondelight 12:53, 24 July 2010 (UTC)
It will be easy to add LOC and Worldcat lookup links to the left-menu of author and title record display as they offer a URL syntax that allows for searching by author name and/or title. It's a matter of figuring out the URL syntax. LOC wants author names in "last, first" format meaning code needs to be developed to reliably generate that and or to at least "AND" all of the words in the author name. --Marc Kupper|talk 06:58, 26 July 2010 (UTC)
The one thing I personally become 'most conflicted' about is the bringing together of a recorded LoC record with a hot link record. Works fine when they are right and become confused/ing when they are different. Attempting to combine them detracts from the 'copyist' side of what there and bleeds into perceptions of what is most correct. A re-edit, by a person without the direct material, leads to compromises that may decrease the surety of the personal side of a verification. Works at newspaper level but does not have to make the verifier feel good about it. H. --Dragoondelight 13:00, 24 July 2010 (UTC)
The way I deal with this is to note both numbers. For example, "The copyright page states 'Library of Congress Catalog Card Number 76-23999'. This LCCN is not found on the LOC web site, however LCCN 76023802 appears to be for this publication."
Note that I don't do this to reformat the LCCN. For example at GNRTNRMVDS1977 you will see that I showed the LCCN as stated with the hyphen but the link uses the formatting the LOC web site needs where I remove the hyphen and inserted a zero to pad it out to eight digits. --Marc Kupper|talk 01:10, 25 July 2010 (UTC)
Last part: Never been a real fan of LoC or OCLC records till I discovered that they did have important data, and junk, buried in them. They can help to re-define some older works into more acceptatble models. I felt guilt, after being told I could add links in notes, and did so. Thus, we are her now, but no matter the acrimony that ensued in the past, the clarity is they have value, more at the Title level than the printing record level. Thus, I expect I will no longer have to feel guilty for NOT pursuing them at the printing level anymore. Thanks, Harry. --Dragoondelight 13:06, 24 July 2010 (UTC)
I agree with that and like that the LCCN numbers and links are being added. The general assumption with ISFDB publication records is that the physical publication itself is the source for all of the information we have in the record. When information about that publication comes from a secondary source we cite where it came from. For example, "The price is from the SFBC catalog for November 1977." What's had been causing contention is that notes about LCCNs are being added to publication records but the source of those LCCNs is not given. Someone who does not have the publication would assume the LCCN is stated. Someone with the publication will be puzzled as their publication does not state an LCCN meaning they would clone the record to make one that matches their publication or if they are certain there should not be a separate record would edit the LCCN note to say "prior to verification this record stated 'LCCN: 12345678.' I can't locate this in publication but assume it's accurate." --Marc Kupper|talk 01:11, 25 July 2010 (UTC)
That's exactly the reason the discussion was brought up to begin with: if a LCCN is recorded in the ISFDB record, we have to assume that it's actually printed in the book itself. I take it a step further by saying flat out that the copyright page contains the complete LoC CiP Data including the LCCN. Example. I used to believe that the LCCN recorded titles, but since working with them over the past several years, I'm convinced they are meant to record a single publication. True, a publisher will often record the LCCN of a previous edition, and often without acknowledging the fact. I came upon this one recently where the LCCN was assigned to a previous edition by an entirely different publisher! Still, in all, I think the LCCN applies to a publication not a title. Mhhutchins 04:24, 25 July 2010 (UTC)
Absolute Digression. Yes, the LCCN was designed for every book, BUT 'The People allowed bureaucrats, politicians and employees at it. I knew a professor of entymology and he worked at the Smithsonian and he laughed at the concept of the integrity of gifts and such there. He told me that collections given to 'the nation' were pilfered by the above. Just breaking a collection of anything means that someone gets to 'toss' the unacceptable. Thus even book collections go phewwt! Just keeping 'known and cataloged' items on the shelf or in the display has been tricky in the past, better today, but new stuff is very susceptible to loss. Just imagine, how many 'presentation' gifts must have come out of stock. The worst part, is space and warehousing, after cataloging. The initial concept rapidly becomes high budget, and there are always 'helpers' who reduce the task. Just imagine that many very early tomes, may be more dust than record today. Think of the need to rebind. Priorities must be maintained. I know that the Smithsonian 'conned/coerved/stole' many Native American artifacts from 'tribal museums' and actually 'constructed' new or repaired items for display. The number of 'losses' was huge. To the point, though. LoC has reduced itself to records management more than anything else and the history of that is mostly at one book level for over 90% of the material. So research for the material is what is available. I will note the number, but I will NO longer continue that process, if interest is there, then the user can do it easily enough. I really do NOT like hot links to outside material in the notes. I am the guy who regularly is mugged and has to listen/view such things. I also think embedded outside links are way too vulnerable for obvious/hidden misuse. In extremis a hot link field at Editor Title level could be totally deleted, while the occasional scattered links could be used as little time bombs for exploitation. The more sophisticated things in the digital field become, the more openings there are for the meddler. So, forevermore, I will NOT add hot links of any outside type. Internal links to material in the db yes. Suggestions to do so, will be ignored. I even dislike the NOT linked OCCL done by individuals without the book in hand. OCLC is just not specific enough for it. I will verify that an OCLC record exists in the Verifier box, but I will leave it to others to find and utilize it to their content. As OCLC and LCCN are not written records, I will not/can NOT know that they will change, through numerous means. I will grind my teeth when others add them though to my verifications, but the 'responsibility' is now, forevermore, theirs. I also will hunt out the ones I entered previously and delete. I am, as stated by others truthfully, a 'whole' or 'Nothing' individual in these matters. Thanks, Harry. --Dragoondelight 13:27, 25 July 2010 (UTC)
I do not believe complete elimination is called for but rather to use wording that makes it clear if someone is referring to information stated in a publication or if it's something else. For example, if someone does OCLC verification it would be handy to know which OCLC record they verified against. --Marc Kupper|talk 08:20, 26 July 2010 (UTC)
Agreed for knowing what the cited OCLC record is, but the problem is that often there are many records for a title, which in various ways may or NOT meet the users needs. Truth is OCLC and LoC are hit and mostly miss for individual printings of a title. The Best records of both are when a book is frankly a 'dud' and is NOT frequently re-printed. Neither record attempts to do more than say yes 'it exists' and OCLC very frequently has multiples of one record, as it rarely defines it to a printing and their years are most confusing. As an example if one uses "The Lost Swords" OCLC records to determine what was printed, you have to wonder if their records pad the what/where of what is published. OCLC has records that tout Tor, Guild America and Nelson Doubleday, often it is a matter of who input the data. In any case, I personally unhappy with printing level attribution to 99% of the printing records, no matter if it is cited in the book or NOT. I have had my say, hope it helps, sincerely doubt it, and remark that after just a little bit of conversation in a wiki post, my mind becomes confused as to issues and effects wished/NOT wished. I have come to the conclusion that eDiscourse is most valuable in limited quantity and it's expansion at some point becomes confusing (at least to myself). So, as it started on my page, and I never really understood the issues of why it was bad for me to do and all right for others to do and not inform me off, so it is with this. The rug will have the dust beat out, but the real utility of LoC and OCCL use will be lost to some and missed by others, NO matter their stance. I leave you all to clear the air, and I as I said will simply NOT use them in individual printing records. Last reply as I am still confused. Thanks, Harry. --Dragoondelight 11:55, 26 July 2010 (UTC)
Remember that pretty much all OCLC records are the result of cataloging at soem specific member library, NOT at any central site. If different member libraries enter the data slightly differently, the records may not get merged and multiple records for the same publication may and do result. -DES Talk 22:21, 1 August 2010 (UTC)

(unindent) I do favor for LCCNs indicting when these are derived from any source other than the actual publication. I do favor hotlinks in the notes until/unless this becomes a separate field, particularly for pre-ISBN books, where our existing automated link to OCLC will not work. The proposals for a 'Library identifier" field (with possible multiple entries) where one could specify the type of ID (OCLC/LCCN/BL/etc) and the corresponding identifier, with a hot link where possible, would be good IMO, if implementation is not too hard. -DES Talk 22:21, 1 August 2010 (UTC)

Documenting unnumbered pages

I'd like to update Template:PublicationFields:Pages to allow for documentation of unnumbered pages. Here's the proposed addition.

  • Some publications have unnumbered pages before page 1. You may count these and would enter the total in brackets. For example, "[6]+522" would indicate six unnumbered pages before page 1. Note, you may need to count backwards from the first numbered page to locate page 1 and then would count the pages before this as the "unnumbered pages." You may also count the unnumbered pages at the end of a publication. For example, "[4]+280+[4]" is a publication with four unnumbered pages before page 1, 280 pages in the main body, and four pages after the body. Do not include hardcover endpapers in the page counts. Endpapers can be identified as they are usually slightly stiffer than the standard pages.

--Marc Kupper|talk 22:16, 24 July 2010 (UTC)

Fine with me. My top of the head question though, What about pagination that changes from Roman to Arabic? Thanks, Harry. --Dragoondelight 22:35, 24 July 2010 (UTC)
I would be fine with using +[#arabic number#]+ to deal with that. For example TYBSF1999B can be described using the "page number" bibliographic system for recording data as:
[i-v] vi [vii-xi] xii-lix [lx] [1] 2-602 [603] 604-609 [610-612].
ISFDB uses the "page count" system and we recorded same publication as "lix+609". The second bullet on Template:PublicationFields:Pages says "Pages without numbers that fall between the two types of page numbering can be ignored." That does not say "must be ignored" but perhaps we can soften the "can" to "may" which would allow someone to enter this publication as "lix+[1]+609+[3]" to account for the unnumbered (and blank) page after page lix and for the three pages after page 609 which are [610] (blank), [611] Publisher's order form, [612] (blank). --Marc Kupper|talk 01:56, 25 July 2010 (UTC)
Before you get too carried away, I don't think the page count field can cope with formats like "[i-v] vi [vii-xi] xii-lix [lx] [1] 2-602 [603] 604-609 [610-612]" - as i recently pointed out to Don Erikson, even an entry like "Unnumbered" gets truncated to "Unnumber". Consider necessary software/database field changes before this goes too far. BLongley 22:37, 25 July 2010 (UTC)
Sounds pretty straightforward, but we'll have to check the code first. Ahasuerus 03:26, 27 July 2010 (UTC)
I would not object to adding the statement to the standards. Most of us have been doing this. Perhaps it should be made clear that this field is for the page count, not the pagination (as you describe in the above bibliographic system for recording pages.) Mhhutchins 04:30, 25 July 2010 (UTC)
Obviously, I confused again. My comment concerned a book that did a v thru xiii then NOT numbered pages and then went to Roman at 17. So, Reset: If "you are suggesting we change to simply stating that x NOT numbered pages are before pagination." "Thus ignoring the 'intricate convultions' of partial pre-novel pagination, then I am for that also. Thus [6]+602. Understood, and it makes it easier, but if you + one end then I would be inclined to + the other, including the tricky endpages, which are even worse in POD tps. Thus 6+602+14 works, and for me well. That is a complete count of the pages, including any end papers. That addresses 99% of page count, and avoids the differing numerics, which can still be shown on Contents and noted if necessary or wished. AS for the long count, I actually love it, know that it would confuse the H out of normal quick users. The last one where the book when through several pagination counts was even longer than your example. I think we are limited to maybe 16 places in that spot. Long count is very good when, you have extras kept from old printings, new printings, new & old changes, New & Old revisions, etc, though it does not help when the printer pops in an excerpt that is paginated separately [Insert material, pictures (can NOT remember inserted written)]. Got creamed on that one. NO one understands it easily. So Mike's suggestion to state it [Page Count], is "ONLY" for the actual numbering, should make the process easier on users and data enters. State 'oddities in pagination relegated to note field' and let the enterer/verifier handle that. I do think the endpaper statement will confuse. If it is a page, it is a page. You will have to make it clear that first and last page drops in pagination have to be in the 'novel count pagination which may extend past the novel end'. Sorry, just remembered all those glossaries, very occasionally having a 'excerpt' in between and then an 'afterword'. WOW! Just rambling and all those little/large inconsistencies of publisher pagination roll out. Rethink, tells me that moderators will still be having to 'inform' users/verifiers quite a bit. It's all comes back to 'book in hand' versus 'book as described'. I would add another peeve in display though, the [], for before pagination works okay, but it displays the end items before the novel and I find that disturbing, so I cheat, and usually do NOT bracket end pagination or end actual counts, thus forcing what I think is a 'better' display. For instance a NOT numbered [Afterword] appearing before the novel is a trifle ugly. Your proposal, if you are talking dropping the Roman numerals, etc will help with my continual 'cheat' there. I am stuck in real internal dilemma when the numerals stop, no matter that the rules say 'ignore it' that + works my mind the other way. So up the count, which is usually intermittent anyway, so really I want to tell you the "true count" of the book, not the actual count as started wherever the publisher decides. So, yes to this [Pages Before Novel Presentation, Not including Prologues, Prefaces and material numbered with the Novel body](Do Not Worry about including material useful to the novel in the Novel Body, but follow the 'presentation of pagination of the printing')+Novel Body, including full pagination as presented by Printing (extend count, if not present till page ONE and to include last page if pagination stops, as frequently found in U.K. publications)+[all pages following Novel body as described previously, including any NOT numbered additional material, including 'About Authors', blank pages, advertisement pages, end pages and tear-out pages to cover]. [#]+#+[#]. Forgot about those dratted under 'window pane' art pages! Maintain the publication pagination in the contents section, except when that conflicts, such as when there are several sets of pagination. 1-165, next novel, 1-166. In this type of case conflicts NEED to be noted in Notes and possibly an alternate method used in contents, depending mostly on the number of 'content items' and conflicts in possible display. Use the numbering, after notation, or state the use of a different numerical system used to separate the page count in contents, such as Roman numerals. Is this sensible? much too long, and my apologies for inevitable conflicts created. Forget something, I know. Just a visualization, not a definitive proposal. Sincere apologies to Mark and Mike for trampling or offending! Thanks, Harry. --Dragoondelight 12:54, 25 July 2010 (UTC)
In cases where arabic numbers continue from roman numbers (i.e. page ix is followed by page 10), it would be best to note the odd pagination in the note field and record the last numbered page as the total number of pages in the page count field. IHMO. As for endpages, those suckers always confused me, but I don't believe I've ever seen them counted as part of the total page count. Unlike magazine covers, which are even more confusing. I agree with Bill Longley's recent statement somewhere in the wiki that he wishes we would have used Bill Contento's method for the page count of magazines that record page 1 as the first inside page of the magazine. At this point, I don't know how we could go back on change thousands of records. Who came up with the current method anyway? Mhhutchins 16:41, 25 July 2010 (UTC)
I think you're referring to here - the recommendation for Bill Contento's method is actually from swfritter, not me. I have personally abused my self-moderating privileges to enter my magazines/fanzines in the logical method - i.e. NOTE when the covers aren't included in the pagination. Or when front covers aren't but back covers are. I'm not going to go back and change thousands of records that quite possibly DO start with 1 as the front cover (which I guess is the original reason) but I hope nobody wants to go apply an illogical rule to what I've entered. (I've entered over five decades of magazines and fanzines now, I don't want to quit over a simple issue like this. There's still boxes and boxes of the Stableford Collection to go.) BLongley 22:55, 25 July 2010 (UTC)
Harry - the initial thinking behind this proposal was to deal with the rather common publisher practice (both pb and hc) of manufacturing books with a number of pages before page 1. Some editors are already doing this and so it's more about documenting an established practice. The page count field is limited to 16 characters. Something like "lix+[1]+609+[3]" will fit meaning it's likely most attempts to document the extra pages at the end of a publication as you suggested will suceed.
The "page count" system ISFDB uses tends to have compact and easy to understand page strings. It fails for those publications that jump from page xvi to 17 for example. I'll think about the endpaper comment. The problem is that the [] system would cause people to pay more attention to the very beginning and end of publications. Sometimes when I'm not sure if a publication has endpapers I count backwards carefully to see where page 1 or "i" is and see where that is in relation to what seems like the endpaper. --Marc Kupper|talk 09:22, 26 July 2010 (UTC)

Harry - I added feature request 3034595 to add better support for [#] in the page number field in the publication contents. I'd add requests for the LCCN and Worldcan lookups at the author and title level but first want to research the exact URL format needed to do the lookups. --Marc Kupper|talk 08:46, 26 July 2010 (UTC)

Yup, it can be done. Ahasuerus 03:24, 27 July 2010 (UTC)
Sorry, everyone, I just put my 'thoughts' out there to be pummeled into some kind of rationality. I am NOT vested in them, other than 'this what occurred to me'. Straightening out the system or a re-do is probably NOT practical. The only system that really does the job is the 'long notation' system shown by Marc and used for really 'top notch' descriptions of usually the 'best' or 'oldest' books. I will try to follow any change, but I am inclined to agree with Bill that the best method is often to do what you see and make sure that method is shown in the notes. Probably someone will change it but if all the material is shown, then it is all one can really do. I am somewhat 'jaded' after experiencing the Perry Rhodan page counting and reading hordes of Publisher page counts. The real simple fact is that the 'usual user' really is best served by a count that tells him how many pages the actual story content is. Probably the occasional 'two or three' page editor adds in some collections inhibit the understanding of how many pages a story has. We will never beat the change of font and spacing issues. I am aghast that 110 page short novels of the fifties suddenly become 222 page tp's for the 'sight impaired' or 'easy reading' group at over $ 20.00. It is strange to ruminate on, but every thing we do and everything everyone else does really dices up the issue of 'how much' material there is. I noticed that some ebook reader format swell content of less than 100kb to a 1 meg. Wish we had a measure that told the reader what he needed and did NOT discourage their interest. Do NOT concern yourselves with my commentary, it was provided for the thought process. The [] request seems most reasonable, though implementation may be impossible. I often wonder when the time will come when adjustment of an old computer system will become unreasonable due to incompatibility with newer systems. True unsung heroes those who make it bearable in the computer sense. Thanks, Harry. --Dragoondelight 11:38, 26 July 2010 (UTC)

(unindent) I've updated Template:PublicationFields:Pages. This will show the material that was added. --Marc Kupper|talk 06:30, 10 August 2010 (UTC)

The Amazing Adventures of Flash Gordon with ISBN

Morning! I hope I am right, but this title seems to be a pulp type panel reprint of a comic, but it does have an ISBN. ISBN's do NOT a book make I hope? I entered "Flash Gordon in the Sand World of Mongo" which is a story with illustrations, much like Janissaries. I entered as chapter book as length seemed short. Half of paging is illustrations, but the dialogue is text. Thanks, Harry. --Dragoondelight 12:42, 26 July 2010 (UTC)

It would help if you link to a publication or something so I can tell what this is about. I don't see a title in ISFDB with "Amazing Adventures of Flash Gordon" nor "Flash Gordon in the Sand World". --Marc Kupper|talk 20:43, 26 July 2010 (UTC)
Sorry but the little information on either book is fragmented. "Flash Gordon in the Sand World of Mongo" is a juvenile/ya book, heavy illustrations, but without illustration captioning in or out the images. [17]. 0064650510 is for Harper Row version. I have the King Features. Descriptions match though, so what do I know about who belongs to who. I entered it as a chapterbook. The first with ISBN 0-448-17348-4, The Amazing Adventures of Flash Gordon, volume two is a comic panel printing in pb format with all dialogue contained on panels. Not a story format, but comic, but it is listed in ABE under ISBN. [18]. Listed everywhere as a Tempo book, no one seems to mention the illustrations are without colorization. Both are listed as Comic, but the 'Sand World' looks much like this [19], which has text and large illustrations, so many that it is almost a graphic novel. Thanks, Harry. --Dragoondelight 21:20, 26 July 2010 (UTC)
Sorry - I had not realized that your entry of Flash Gordon in the Sand World of Mongo was in the queue. That's why I could not find it earlier. It's been approved and is at FLSHGRDNNT1976. I tend to leave it to the editor's discretion on if something should get entered, particularly if they have a physical publication in hand. ISFDB:Policy#Definitions part 2 lists things things that are excluded. That list includes comic books, manga, and graphic novels. If Sand World of Mongo is more like Janissaries then I'd say it's in. Maybe the litmus test can be "if I remove all of the pictures and their captions is there still a story left?" With Janissaries the answer is absolutely yes and it's likely to be a 200 page book padded out to 335 pages with the illustrations which do not have captions you see in manga nor the talk bubbles you see in comics. --Marc Kupper|talk 23:20, 26 July 2010 (UTC)
That's exactly the test that I use when making these decisions. Ahasuerus 03:20, 27 July 2010 (UTC)
If you don't have the publication I don't see a problem with someone adding it to ISFDB with notes explaining that this may be a comic/manga but also giving the evidence that weighs in favor of it being included. Some day an editor will come along with a copy of the publication itself and can use that to improve the notes.
BTW, this is a Yahoo thread about "I need a list of novels based on comic book characters, old and new?" where the answer could likely be mined for a list of publications to add to ISFDB. --Marc Kupper|talk 23:20, 26 July 2010 (UTC)
Sand World of Mongo is a short story, hence chapterbook, with illustrations, which I considered because of Juvenile fiction, which is NOT unusual to have. As for lists/listings, that was part of the problem as both were in lists and coming from the comic book angle it would appear to be a slight change. Still for YA it meets I think the standard. The other bothered me, because of the ISBN and everyone, including Wikipedia, call them books, which they are only if you consider the binding. I did have a nefarious thought though, that it would be nice if ISBN's when found to NOT meet this DB standards, if they could be on a list that pops out for the moderator showing they have been discarded as books. I hate it when I get surprised, but these two and most that I have done recently were part of the 'Ackerman' collections and he collected everything. Quality was pristine and cost was only $1.25 apiece so I took handfuls with litter regard for oddities. Still, ABE & Amazon are light in description on what they are. Many kudos to those who have looked through juvenile SF and made choices, I find many books described as juvenile/YA and in my opinion are good fiction, but others have fascinating titles are no different from Dr. Seuss. Most difficult to parse. Noticed that many Mercedes Lackey are YA, but I do NOT think her new books that different. Thanks, Harry. --Dragoondelight 11:49, 27 July 2010 (UTC)
Re: "ISBN's do NOT a book make I hope?" I am not 100% sure I understand the question, but comic books can also have ISBNs, so an ISBN's presence or absence doesn't determine whether a pub should be "in" or "out". Ahasuerus 17:35, 27 July 2010 (UTC)
What I was trying to say is that once you know that than an ISBN group is used for comics, then when that ISBN pops up, you know to be careful of the contents. Thanks, Harry. --Dragoondelight 19:47, 27 July 2010 (UTC)
Oh, I see! Well, ISBN blocks are assigned to publishers and certain publishers are known to be primarily comics-oriented, so it's certainly a useful rule of thumb. (For better or for worse, a number of formerly "exclusively comics" guys have recently branched out into fiction, so it's less cut and dry than it used to be.) Ahasuerus 21:24, 27 July 2010 (UTC)
I was just bring it up for thought, not necessarily for action. Others can parse such better than I. I will never recover from finding the same ISBN being used for Wallmart reprints of different titles of H. G. Wells. Never considered that a possibility and am sure that printers/publishers will do anything to make more money. "British Science Fiction paperbacks and magazine 1949-1956, showed how often they can morph and then re-morph and rise again. The real clue is the money trail. Thanks, Harry. --Dragoondelight 21:43, 27 July 2010 (UTC)
ISBNs may actually define a book, but not in our terms - I first encountered "trade paperback" as collections of comic books - maybe 4-6 comics bound together as a book. The UK definition of "trade paperback" has nothing to do with the original US definition and more to do with physical dimensions. (Look up A-format, B-Format, C-Format.) I don't want us to end up with collected comics as books in our terms, but over-zealous pruning of suspected comics publishers is a bit unwise - e.g. "Hollywood Comics" are flagged up for moderators as "suspected comics publisher" but they actually do a lot of translations of French Science Fiction Novels that I think we want. Unfortunately, Amazon sometimes don't go to the imprint level that might clarify such. :-( BLongley 23:04, 27 July 2010 (UTC)
Probably the extreme case is the Vance Integral Edition 44 volumes, but only two distinct ISBNs. Since they were only sold as sets by subscription ("not available in stores) They used one ISBN for each group of titles/vols released together. -DES Talk 21:59, 1 August 2010 (UTC)

"Writing as" exceptions or changes

Our current help says that when the cover of a book says "A writing as B" we credit A and ignore B. For example the "Lucky Starr" books are (except in some early editions) credited to "Isaac Asimov writing As Paul French" and we index them as by Asimov and ignore French (except for the early editions where Asimov was not credited).

However, the "In Death" series by Nora Roberts/J. D. Robb is begin handled differently and i think this should be an exception, or perhaps there should be an explicit change of rules. In this case, i (and I think some other editors) have been treating the books as if the only credited author was "J. D. Robb" who is known to be a pesud of "Nora Roberts". IMO this is proper for several reasons:

  1. The "writing as" appears on the covers, but only "J. D. Robb" is listed on the title pages IME.
  2. The distinction separates the author's SF work from her non-genre and paranormal romance work.
  3. Those SF fans who read these books quite probably know the author better as Robb.

I'm not sure how or whether to make a general rule out of this exception, maybe we should only plan to loosen the "writing as" rule in the help. -DES Talk 01:34, 4 August 2010 (UTC)

I have posted about this before, last year, but I don't think any real consensus developed. -DES Talk 01:35, 4 August 2010 (UTC)

I don't have any books that were published like this, so I'm not in a position to check if the cover credit is consistent with the title page credit. As per ISFDB standards, the title page should be the source for credit. If it says "A writing as B" on the title page, isn't that giving B credit as the author, while acknowledging that B is actually A? I'll look up some pubs on OCLC and see how the librarians handle the responsibility role. Of course, they may be as divided as we are! Mhhutchins 03:33, 4 August 2010 (UTC)
This one gives the responsibility to Robb while acknowledging the cover credit. In fact, I went through more than 20 OCLC records for titles that were published in their first edition giving the "writing as" credit on the cover. Not one record gave the sole responsibility to Nora Roberts. There were only couple that gave "Nora Roberts writing as J. D. Robb" the responsibility. ("Responsibility" is OCLC's way of recording who the credited author is, not necessarily the actual author.) Mhhutchins 03:45, 4 August 2010 (UTC)
My experience with OCLC is that "Responsibility" in some cases also lists illustrators and editors, and in some cases seems to list a canonical form of the name rather than the form appearing in the publication -- I ran into this a good deal while sorting out the many forms of Sheridan Le Fanu's name largely from OCLC records.
However our Help:Screen:EditPub#Author currently says:
"Pseudonyms. If you know that a particular author's name is a pseudonym, leave it as the pseudonym rather than changing it to the real name. The only exception is when a reprint shows both the original and subsequent names. For example, Isaac Asimov's "Lucky Starr" books were originally published under the pseudonym of Paul French, but were later reprinted with the cover giving both names: "by Isaac Asimov, writing as Paul French". In these cases you can simply give "Isaac Asimov" as the author...." (Emphasis added)
Your suggestions seem to follow from our general principles, but to conflict with the current help. -DES Talk 04:10, 4 August 2010 (UTC)
I have most of the "In Death" series. I think all credit Nora Roberts on the cover, but I don't recall one that had anything other than J. D. Robb on the title page. -DES Talk 04:10, 4 August 2010 (UTC)
So there's no question of credit for those pubs.
BTW, I think you may be looking in the wrong place of the OCLC record to get the responsibility data. In the upper portion the record will list the author(s). This is where the record will state the actual author (when known) if a pseudonym is used. If you look further down in the details, there's a field marked "responsibility" which records title page credits. If an artist is credited on a title page, the OCLC will reflect how it's credited. See the record for the Arkham House edition of Tales of the Quintano Roo. You'll see it gives credit at the author field to "Alice B. Sheldon; Glennray Tutor". In the responsibility field it credits "stories by James Tiptree, Jr. Ill. by Glennray Tutor." which is exactly how the book credits the work. Another example without: the Doubleday edition of Mockingbird, which gives the author as "Walter S. Tevis" and the responsibility to "Walter Tevis" (the book's actual credit). Even though he only used it at the beginning of his career, Tevis's name with the middle initial is considered the canonical name on OCLC (but not on the ISFDB), so that's why it's used in the author field. Being entered by humans, of course, there will be errors, but the goal is to record in the responsibility field the book's actual credits. You'll occasionally find records (especially for early 20th century titles, as your experience with Le Fanu attests) that don't give clear responsibility data, as the standards were tightened as data was accepted in later years. There will also be records for pseudonymous works that were created before the author's true name was known. Mhhutchins 05:28, 4 August 2010 (UTC)
I do know about the "responsibility" field on the OCLC records and that it often differs from the "Author" field. I did not know that OCLC's standard was that this represented the credits as published -- i had encountered enough cases where this was not so that i thought the rule was otherwise. I have entered a good many 19th C horror & ghost books from OCLC records. Thanks for the info. -DES Talk 14:19, 4 August 2010 (UTC)
Yes, those early records leave a lot to be desired. The librarians seem to have gotten the hang of it as the years went by. :) Mhhutchins 15:10, 4 August 2010 (UTC)
See this previous discussion. I think in the "A writing as B" case, the author is B -- and should be recorded that way, and the current help is mistaken and should be changed -- and B is being acknowledged as a pseudonym of A. In my admittedly limited experience, many books credited this way either were originally published under the B byline or are later parts of a series started under a B byline, where now for whatever reason someone wants the works linked with A (most likely for marketing / name-appeal). --MartyD 10:16, 4 August 2010 (UTC)
I tend to agree, or perhaps the rules should be a bit flexible. In the Isaac Asimov/Paul French case (which seems to have defined this rule) I agree that we want to record Asimov. In the Rora Roberts/J. D. Robb case I think we want to record Robb. I'm not sure how the general rule should read, but I think the Robb case should be followed more than the Paul French one. -DES Talk 14:19, 4 August 2010 (UTC)
I go along with the both of you but believe that B should always be credited as the author, even in the case of Asimov/French. Because French has already been designated as a pseudonym of Asimov, all of the pubs will eventually appear on Asimov's summary page once variants are created. So I see no reason to make an exception for him. These are not as uncommon as you might first think. Don't forget the "Dean Koontz writing as Leigh Nichols" or "Anne Rice writing as Anne Rampling" or "...A. N. Roquelaure" pubs. Mhhutchins 15:07, 4 August 2010 (UTC)
Yes, I did mean "always". --MartyD 17:09, 4 August 2010 (UTC)
I'm not too bothered which way this goes, but I think it has to be consistent - we can't have Asimov/French going one way and Rice/Rampling going the other. There are some other complexities - has anyone else seen Disturbed by Her Song? That's Tanith Lee, writing as and with Esther Garber and Judas Garbah. What would the new rule say about that? BLongley 21:37, 4 August 2010 (UTC)

(unindent) DES, welcome back. I'm trying to understand what needs to be changed. The Pseudonyms section Template:TitleFields:Author looks fine to me. It seems to make it clear that the goal is to document the author as stated on the title page. It then mentions an exception where if both the real name and pseudonym are stated to use the real name. Ideally editors would document the actual wording in the notes, particularly of the name entered does not match the name stated on the title page, but I don't see a need to make that a separate "rule."

Was the intent to make this a discussion about how to enter publications from secondary sources such as OCLC? --Marc Kupper|talk 09:18, 5 August 2010 (UTC)

Let me take a quick step back. One of the main reasons to record author names as stated in the book (and then create VTs when appropriate) was to make it easy for users to search the database. If a user owns a book attributed to "Paul French", then he will presumably use "Paul French" to search ISFDB and we want him to find that "Paul French" = "Isaac Asimov". Similarly, the reason why Help states that a book credited to "Isaac Asimov writing as Paul French" should be entered as by Asimov is that our users will be (presumably) searching on "Asimov" rather than on "French".
Having said that, books can disclose pseudonyms in a variety of ways. As DES pointed out, the "real name" may be listed on the cover, but not on the title page. I have also seen books mention the author's canonical name or other pseudonyms on the back cover ("She lives in Oregon and also writes urban fantasy novels as XYZ"), in "About the Author" and on the copyright page. At this time our Help doesn't specify how explicit the pseudonym disclosure should be before we credit the author's real name rather than the pseudonym, although the example it gives uses a book's cover, thus suggesting that the disclosure is not limited to the title page.
Given that we generally privilege the title page, I think we may want to change Help to state that we will only use the real name if it is listed on the title page. Ahasuerus 12:03, 5 August 2010 (UTC)
That would be good, and would solve the Robb case, since Nora Roberts is consistently credited on the cover but not on the title page. However after the discussion above, I am suggesting going further and removing the exception altogether. I am suggesting that if the title page says "Joe Jones writing as Fred Smith" then the publication should be credited to Fred Smith with a VT parent with the author as Joe Jones, just as if the title page said simply "Fred Smith" but it was reliably known that Smith was a pesud of Jones. This will produce a line in the Jones bibliography for "Title [As by Fred Smith]" which is about as close as we can do to reproducing what is stated, and it will mean that a search on Either Jones or Smith will find the book, which is obviously the best outcome for the user, and doesn't require us to guess how a user will search. -DES Talk 14:24, 5 August 2010 (UTC)
Now Hauck has reached Wyndham, I'm reminded of another example - Stowaway to Mars has John Wyndham writing as John Beynon on cover and title page. So we wouldn't have to change those ones. I don't know how many publications would actually be affected by the proposed change - anyone else have more examples? BLongley 19:23, 5 August 2010 (UTC)
To clarify - this example wouldn't need a change if we go for Ahasuerus' help update, but would if we go for DES' suggestion. 19:38, 5 August 2010 (UTC)
Actually, we have a perfect example on our front page right now: Fantasy in Death. The UK pubs show "Nora Roberts writing as J. D. Robb" and the US ones just show "J. D. Robb". I've no idea what the title pages will show, but obviously current help suggests we've mis-classified all the UK editions. And Amazon obviously supports the Pseudonym, but I wouldn't let their practices over-rule ours. Just another data-point - I'm still not too worried about the overall outcome of this discussion, I'll find my relevant books eventually. BLongley 23:28, 5 August 2010 (UTC)
As It happens I have a US hc of Fantasy in Death. The cover shows "Roberts writing as Robb" and the title page just Robb. I don't know what UK versions may show. -DES Talk 23:36, 5 August 2010 (UTC)
The above was an error, it seems my mind filled in what I expected it to see. Both cover and title page of my copy of "Fantasy" show just "J. D. Robb". -DES Talk 23:53, 5 August 2010 (UTC)
While I initially liked what DES wrote about entering the author credit as Fred Smith I started thinking about how pseudonyms and pen names get used and believe we should credit the real name if it's disclosed on the title page as that's the person being credited as the author. The only "rule change" is that previously we allowed an editor to extract the author credit from a "secondary source" (the front cover for example) and here we'd be restricting the source for the "author" field to be the publication's title page. --Marc Kupper|talk 01:50, 6 August 2010 (UTC)
This is not the most vital issue on this site. That said, if we credit the "real" name (Jones in my example) and a user searches for the pseudonym (Smith in my example) the user will not find the work sought (unless there are editions credited solely to Smith). However if we credit to the pseudonym with a VT to the canonical author, searches for either name will find the desired record. What is the advantage of crediting the "real" (canonical) name and ignoring the pseudonym in such cases? -DES Talk 04:59, 6 August 2010 (UTC)
DES' method automatically allows for the VT meaning someone will find the title regardless of which author they use. A downside is we'll need to track down and fix some publications and titles. Ahasuerus' method is in line with current/previous practice. To allow people to use either name we would need to add a placeholder VT. I believe we are already doing this. Another option is to enter the author name as "Joe Jones writing as Fred Smith". I checked and no one has abused ISFDB in that fashion yet. None of these astonish me any more or less than the others. --Marc Kupper|talk 08:45, 6 August 2010 (UTC)

Pros and Cons of "Writing as"

(unindent) Given both searching and display will behave meaningfully and as expected, I prefer DES' proposed approach. Although it's going to mean work to track down and change existing entries, I think it lines up more closely with what the pub says and has the benefit of documenting existence and use of pseudonyms in a "natural" way. It also avoids the inconsistency the current method would cause for "Dying in Death" by J. D. Robb for printings 1-3 and "Dying in Death" by Nora Roberts writing as J. D. Robb for printings 4-6. IMO, these ALL should be by Nora Roberts [as by J. D. Robb]. Other than the legacy entries, what are the drawbacks of doing it this way? --MartyD 10:20, 6 August 2010 (UTC)

Well, the rework is a major issue - several of us have been through our collections several times already as standards evolve. And I hate having to rework my stuff. :-( But I'm failing to see the advantage of DES' way - we'd STILL need a note in each pub to indicate whether it was "Nora Roberts writing as J. D. Robb" or just "J. D. Robb". That distinction seems to be useful in separating British and US publications, not sure about Australian and Canadian. I think I'm leaning towards maintaining the status quo until we can have "writing as" recorded at publication level - because I can see a use for the distinction, and the proposal doesn't support it. Are there many books where there is NO publication only by the pseudonym? As we can still create the pseudonymous entry title anyway. BLongley 20:32, 6 August 2010 (UTC)
Bill L is correct that to fully record what is in the pub we will need a note in any case, as there could be "by Roberts", "by Robb", or "by Roberts writing as Robb", and nothing we do with the author credit will allow the three cases to be clearly distinguished without a note. Come to think of it, here is a fourth possibility: I recall an edition of Police Your Planet credited to "Lester del Rey and Eric Van Lhin", and come to think of it Remember When was in fact credited to "Nora Roberts and J. D. Robb" (half of the book was a contemporary mystery romance, the other half was in the future as part of the "in Death" series, with the 1st half forming backstory).
"Writing as" books are not very common. Aside from the J. D. Robb series, i cant off hand think of any that were initially and always published with a "writing as" credit, and those are in most if not all cases on the cover but not on the title page. Asimov's Luck Starr books I think all has at least one edition on which Asimov was not mentioned.
As to the pros & cons, IF a book is published as "writing as" from the first edition, and if we follow the current rule, no search under the pseudonym would find the book, whereas under the modified rule either search would work. -DES Talk 22:00, 6 August 2010 (UTC)
We can still add the variant title for the "as" author. Then the search would separate the "writing as" editions from the purely pseudonymous publications (if any). Of course, that still doesn't separate canonical-name only editions. BLongley 20:06, 7 August 2010 (UTC)
Having special software support for "writing as" would be nice, but I'm not sure it comes up often enough to be worth the coding effort and complications of interface. -DES Talk 22:00, 6 August 2010 (UTC)
Keep in mind that "writing as" statements may not always neatly match our canonical names/pseudonyms. For example, what should we do if we come across "Collected Boxing Stories of William Fitzgerald Jenkins (originally published as by M. Leinster and Will F. Jenkins)"? Ahasuerus 00:31, 7 August 2010 (UTC)
I wouldn't count that as a "writing as" at the Collection level. The sort of authors I'm concerned about are "Stephen King writing as Richard Bachman", "Dean R. Koontz writing as Leigh Nichols", "Kenneth Bulmer writing as Alan Burt Akers", "Jayne Ann Krentz writing as Jayne Castle", "Robert E. Vardeman writing as Daniel Moran", "C. M. Kornbluth writing as Cecil Corwin", "Robert Jordan writing as Jackson O'Reilly", "Suzy McKee Charnas writing as Rebecca Brand", "Frederik Pohl and Lester del Rey writing as Edson McCann", "Katie MacAlister writing as Kate Marsh", "C. M. Kornbluth and Judith Merril writing as Cyril Judd", "Edmund Cooper writing as Richard Avery", "Will Murray writing as Kenneth Robeson", "Linnea Sinclair writing as Megan Sybil Baker" and "Sherrilyn Kenyon writing as Kinley MacGregor", just to pick the first few examples I've found in notes. The Bulmer/Akers rework alone is enough for me not to want to change. BLongley 20:35, 7 August 2010 (UTC)
I've now seen "Dean Koontz writing as Leigh Nichols". :-( I know people have been using Dean R. Koontz for years, sometimes incorrectly by any standard. I think we need a software change to allow variant authors of variant authors of canonical authors if we want to progress this. :-( BLongley 23:34, 8 August 2010 (UTC)
The idea of entering the author name as "Dean Koontz writing as Leigh Nichols" with a VT back to Dean Koontz is starting to appeal to me more. If at some point we add software support for "writing as" it'll be easy to track down and fix the relevant records. --Marc Kupper|talk 22:27, 9 August 2010 (UTC)
Yes, that's beginning to appeal to me more too. That way we DO get self-documenting pseudonyms without additional pub notes. BLongley 19:02, 12 August 2010 (UTC)
It may help eliminate ambiguity and some of our arcane rules, but would it also confuse our users when we display "also as by Dean Koontz writing as Leigh Nichols" on the Summary page? Ahasuerus 02:12, 14 August 2010 (UTC)
I agree - that one seems strange. I thought about it for a few days and what DES is proposing makes sense. We would enter the publication as by Leigh Nichols and VT that to Dean Koontz. The summary bibliography will show "also as by Leigh Nichols" which seems to be the least astonishment case. If we enter the publication as by Dean Koontz (per the current help) then we would not have a record that he wrote as Leigh Nichols. Also, since it's a VT it will not show up on Leigh Nichols' bibliography which is also correct though may astonish someone with a "Leigh Nichols" publication in hand. --Marc Kupper|talk 09:35, 15 August 2010 (UTC)
With "Collected Boxing Stories of William Fitzgerald Jenkins (originally published as by M. Leinster and Will F. Jenkins)" the publication is making it clear that William Fitzgerald Jenkins is the author and confirming earlier publication under a pseudonym. Had the title page said "Collected Boxing Stories by William Fitzgerald Jenkins writing as M. Leinster and Will F. Jenkins" then I would use "William Fitzgerald Jenkins writing as M. Leinster and Will F. Jenkins" as the author name. That separates the concept of writing in a particular voice (pen name) vs. writing under a pseudonym.
I've been thinking about the data structures and code to implement adaptions, revisions, etc. and so "writing as" can be included as part of that. --Marc Kupper|talk 22:27, 9 August 2010 (UTC)

Letters

This is more of a confession than a request to change rules. "In With Reservations - Individual letters to the editor published in magazines." I'm now including all letters in BSFA publications as it often turns out that they are "well-known sf personalities". Have a search on "Letter (Vector"" for instance. As I don't know whether they are known SF personalities till I see the volume of their output, I'm including them for now (and should probably have included them in the past.) I've probably missed a few in my early days, and I think I've misclsssified a few responses as "Letters" when the response should really have been an Essay. BLongley 23:21, 8 August 2010 (UTC)

A certain amount of uncertainty re: "well-known sf personalities" is probably inevitable. Not everyone will immediateley realize that, e.g., the Christopher Lowder who wrote a letter to "The Science-Fiction Collector 4" in 1977 is better known as Jack Adrian. Ahasuerus 13:54, 10 August 2010 (UTC)
I don't recall a prohibition on letters and so doubt anyone would complain.
I have wondered at times at "why do we have publication records?" In other words how much is too little, just right, or too much information? --Marc Kupper|talk 06:14, 10 August 2010 (UTC)
I argued that letters should be limited to people with indexed fiction already in the db, but the consensus was elsewise. -DES Talk 14:33, 10 August 2010 (UTC)
For the same reasons as BLongley, I argued for all or nothing--it would be an unending task otherwise. I choose to enter nothing.--Rkihara 15:57, 10 August 2010 (UTC)
The studied ambiguity of the final Help text was actually geared to come as close as possible to satisfying Des while allowing editors to make their own judgment calls. No editor should ever be pressured into entering more data than they want to. But what about verified pubs? Should we feel obligated to enter all letters before we verify the pubs? As an inclusionist I would leave the final call up to the isfdb editor and allow the widest possible latitude.--swfritter 17:35, 10 August 2010 (UTC)
DES wrote Template:PubContents:Letters and it seems unambiguous. The third bullet on Help:How to verify data#Primary Verification says "you should enter all contents before marking the pub as verified." How about getting rid of the word "all?" That'll allow the rules defined under edit-pub to apply rather than seeming to add a new rule here of "all" contents.
Terms like "significant" and "well-known" are fairly imprecise and open to interpretation. The wording in the template is based upon something I wrote on the Policy page and I intended it to allow flexibility.--swfritter 00:50, 11 August 2010 (UTC)
Ideally, editors that do not enter all of the letters (either because they did not enter any or only entered the ones from notable writers) would add a note explaining the letters exist on pages # to # but were not entered in the ISFDB contents. --Marc Kupper|talk 21:17, 10 August 2010 (UTC)
The intent of that "All" was that an editor should not verify, say a collection or an anthology without entering all the stories and other major contents. I can see some looseness in what interior art gets entered, but I don't want to imply that it is OK to verify a stub entry. If you want a change, how about:
"you should enter all contents (with the possible exception of interiorart and letters) before marking the pub as verified."
Would that cover things to your satisfaction, Marc? -DES Talk 22:56, 10 August 2010 (UTC)
As Marc stated there should be some mention that interior art and letters have not been entered or have not been comprehensively entered although I don't think that there need be more than a general statement to that effect. Subsequent verifiers should not feel any obligation to provide prior notification to other verifiers when adding or expanding such data although it is good manners to notify after so doing.--swfritter 00:50, 11 August 2010 (UTC)
Yes, the new wording DES suggested looks fine but in thinking about this - I'm not in the habit of looking for and adding excerpts, about the author, glossaries, pronunciation guides, lists of characters, historical time-lines, etc. If I see them I usually add them but I generally don't consider the material "major content" particularly as it's rarely clear as to who the author is. The letters are in the same category. I thought Help:Screen:EditPub said "mostly no" to dedications and acknowledgments but now I see that's at Help:Screen:NewPub#What to include and is a long list of things and if they should or should not be included in the contents. (that list does not mention letters and should also be templated and included in edit-pub). Thus I'm we can use "you should enter all content per Help:Screen:NewPub#What to include" with the exception of interior-art before primary verification." --Marc Kupper|talk 04:33, 11 August 2010 (UTC)
Templating done, but I'm not sure I like the results, take a look at the revised Help:Screen:EditPub, please. -DES Talk 06:13, 11 August 2010 (UTC)
I agree - editpub's help is getting a little long but I do like that it has "what to include" now. A thought is to incorporate "what to include" as new section on ISFDB:Policy and to also separate out "major content" that would be the minimum standard for primary verification. I would classify letters, even from significant people, as "minor content." That allows publications to be verified even if none of the letters are indexed. The only ting I'd ask for is a note saying the letters exist so that someone looking at the contents does not think it's complete. --Marc Kupper|talk 19:00, 11 August 2010 (UTC)
The de facto standard for magazines is that interior art is "major content" although it is the editor's option to enter only a single interiorart title when a story has more than one illustraton; I think nearly all of the major magazine contributors would agree. Others would know more about books.--swfritter 21:11, 11 August 2010 (UTC)


I'm not comfortable with labeling interior art as "major content." I'd much rather see people marking publications as primary-verified even though they only checked the metadata as then we know they have a copy and can ask if we need more detail. I've seen *many* publication records that have clearly been edited against a physical publication but are not verified. --Marc Kupper|talk 09:16, 12 August 2010 (UTC)
What was said above was that illustrations are "major content" for magazines and I agree with that. For most novels, i don't think of interior art as nearly so vital. -DES Talk 12:39, 12 August 2010 (UTC)
Thus I'm fine with someone not including the frontispiece illustration or a map when verifying a novel. Likewise, many people do not enter the starting page number for the story in a novel. --Marc Kupper|talk 09:16, 12 August 2010 (UTC)
I myself never enter the starting page number of a novel unless I am entering other content with page numbers, such as an introducxtion or other essay, or possibly a map. -DES Talk 12:39, 12 August 2010 (UTC)
I used to not do it either but since I started paying attention to the numbering or page count of the front matter or preliminary pages it's come up. If a novel starts on page 7 then it's a hint that pages 1-6 are the preliminaries. If the novel is on page 1 it's a hint the preliminaries exist and are unnumbered. --Marc Kupper|talk 20:05, 12 August 2010 (UTC)

(unindent) It shouldn't be necessary to leave a note when omitting letters, as letter columns are generally easy to identify. A note added to the magazine grid should suffice if needed.--Rkihara 22:05, 11 August 2010 (UTC)

If I don't have a copy of the magazine and/or am unfamiliar with it then I would not know the letters existed. The note I'm thinking of would be worded like "There are letters on pages 23-27 that were not included in the contents list for this publication record." or "Each story is illustrated with two or more drawings though only one interior-art has been entered for each story in this publication record." It serves as a heads up.
A note I tend to use regularly is something along the lines of "Each story in this anthology is prefaced by a one to two page introductory essay. It's clear from the content that the editor wrote each of these." --Marc Kupper|talk 09:16, 12 August 2010 (UTC)
If I were not going to enter the essays separately I might well leave such a note. I don't always mention 1-paragraph entries with no significant content, even in a note. -DES Talk 12:39, 12 August 2010 (UTC)
If the letter column is entered as an essay, indicating that it is a letter column, then a note that there are letters in a letter column seems unneeded. If it is just entered by department name ("Brass Tacks") then a reader not familiar with the magazine might not know that it is a letter column, although if it is in a series that fact would soon be apparent. -DES Talk 12:44, 12 August 2010 (UTC)
The standard is to only list significant letters by well known sf personalities. It would seem to me that we need only to document a deviance from the standard which would mean there only needs to be a notation when all letters have been entered.--swfritter 16:04, 12 August 2010 (UTC)
I enter all letters regardless of the author, but would not expect other editors to do the same. The problem with "significant letters by well known sf personalities" is that it takes years sometimes to establish such status. A letter by Greg Benford in a 1962 fanzine would not be entered if this were 1964. A letter published in the August 2010 issue of NYRSF by the winner of the 2018 Hugo for best short story would not be entered if he were unknown. What ISFDB user in 2018 would know that such a letter existed unless it were entered at the time of publication? Mhhutchins 17:11, 12 August 2010 (UTC)
Obviously a compromise that satisfied nobody. Editors should not be pressured into entering more data than they want to and should not be discouraged from entering as much non-trivial data as they want to. As far as verification: I don't think full entry of letters should be required and, as above, I think there should be a note stating that all letters have been entered where that is the case.--swfritter 18:14, 12 August 2010 (UTC)
Something I realized in reading over the replies is apparently these magazines have a column and that contains the letters. I had been thinking the letters were stuck into the magazine much like individual essays. If they are part of a column then I believe the existing rules are fine, that editors can enter zero, some, or all of the letters, and that there's no need for a note though I believe the note would be a nice courtesy. Editors are always free to enter the full list of letters into the publication's bibliographic comments wiki page and if an author is later notable then the letter will get spotted by people using web search engines and likely added to the publication contents. --Marc Kupper|talk 04:24, 13 August 2010 (UTC)

Recording of obvious misspellings

Bringing this discussion to the forum, I don't see how we can distinguish between "obvious" misspellings and intentional ones. Should there be an exception to the standard? I personally feel that if a story as titled in the book isn't recorded as printed, then it flies against the ISFDB standard. Should there be a subjective decision to change a title? I don't believe it should be changed. Mhhutchins 15:06, 13 August 2010 (UTC)

I think there is enough internal evidence within the pub (copyright and contents page) to avoid the VT process as long as all pubs that share the error also have sufficient internal evidence to indicate the correct title. My editing life definitely got easier when I found the same section in Help that Hauck refers to.--swfritter 16:43, 13 August 2010 (UTC)
Mhhutchins, I'm a little confused. It seems you said you want the stated title to be used literally but also that the rules should not be changed. Template:TitleFields:Title's "Short fiction, essays and poems" section allows some editor discretion with how these titles get entered. --Marc Kupper|talk 22:58, 13 August 2010 (UTC)
Read again. I didn't say the rules should be changed. When I said "I don't believe it should be changed" I was referring to the object in the sentence immediately preceding it: the misspelled title. Mhhutchins 04:04, 14 August 2010 (UTC)
For shortfiction we use the title that's stated at the head of the story. If I see that title (or author name) is misspelled, but that the correct title was used on the page headers for the story, table of contents, copyright page, and the story content, then I would enter the corrected title in ISFDB plus a note. If the bad title/name is used elsewhere in the publication, particularly both at the front of the story and table of contents, then I would enter the title as-stated and add a VT to the correct title. Unfortunately, neither of the publication notes([20][21]) makes it clear how the title is spelled in various parts of the publications.
In this case Google for "Where Hesperus Falls" gets 6,440 hits and "Where Hesperut Falls" has six with one of those being ISFDB's note about the misspelled title, a single quote in a forum thread about "the best opening line ever", and the other four apparently being a bootleg copy of the book meaning I'd go with "Where Hesperus Falls" and a note. --Marc Kupper|talk 22:58, 13 August 2010 (UTC)
What would you do if a story is misspelled in every instance in which it appears in the book - title page, content page, copyright page, running header? What would you do if a story is titled differently in every appearance in a book? (It can happen.) Yes, "Hesperut" is an "obvious" error to those who know the story, but how would anyone who only has this publication know that? There are hundreds, maybe even thousands, of cases in the database in which errors or misspellings are the basis for a variant title record. Think of all those stray apostrophes and misplaced or dropped punctuation in titles. Should I or any other editor undertake the task of deleting these variants, making the titles conform, and placing notes in each publication about the misspelled titles? When I first came to the ISFDB, I thought the standard about recording data "as is" was too rigid, especially when it came to author credits. But over the years, I've come to understand and even respect the intentions behind the standard. So, now I'm being told it's left up to my own discretion about how to enter data? Wow! Why didn't someone tell me this years ago? :) Mhhutchins 04:04, 14 August 2010 (UTC)
Just read that section of Template:TitleFields:Title's "Short fiction, essays and poems" and it appears that it specifically spells out the circumstances in which an exception can be made. It doesn't leave it to the editor's discretion. So in the case we're discussing here, the title should be changed to the canonical title and the misspelling should be noted. Thanks for pointing out that section of the help page that I've overlooked before, and forgive the previous rant. Mhhutchins 04:33, 14 August 2010 (UTC)
When there's language like "if you know that one form of the title is the usual one" then I'm assuming that's editor discretion as "usual" is subjective. I do not think we want every single error and punctuation variant showing up in author summary bibliographies. The error or punctuation variant is something that can be documented at the publication level. I agree this can be tricky at times. Did a publisher or editor "intend" that the title be "Where Hesperut Falls?"
In re-reading that piece of help I'm wondering if it should be changed. It gives an example of "Bell, Book and Candle" vs. "Bell, Book & Candle." In that case I absolutely would use "Bell, Book & Candle" and add a note documenting the discrepancy as I usually do not have enough data to know what the "usual" title is. Other's may have verified a specific title but who knows, maybe they also switched to the "usual" title? --Marc Kupper|talk 10:12, 15 August 2010 (UTC)
This discussion makes me wonder if we need to organize our Help pages better. There is presumably value to having overlapping "How To" pages, but if they provide diverging advice, that's a bad thing for obvious reasons. Ideally, we would list all related instructions on the same page/template, e.g. "What to Enter in the Title Field", "What to Enter in the Page Field", etc. Ahasuerus 02:52, 16 August 2010 (UTC)
I agree but also think the help one gets when someone clicks on the link at the top of an edit-pub or similar page could be vastly simplified and that it would link to a detailed reference. I know Mike Christie did some simplified help but I have not looked at it in ages. --Marc Kupper|talk 08:31, 17 August 2010 (UTC)

Dramatic readings & Dramatizations

"The Mist : In 3-D Sound" has notes that say: "This is not a reading, but a full cast dramatization" Assuming that note to be accurate, I would be inclined to delete this, just as I would a record for a movie. Would anyone object to my deleting this? (I found it as part of the Chapterbook cleanup project.) -DES Talk 20:55, 17 August 2010 (UTC)

It should be deleted. According to the standards, dramatizations are out. Because it changes the original text. But then, why do we allow juvenile adaptations of everything from Star Wars to The Wizard of Oz? Mhhutchins 21:58, 17 August 2010 (UTC)
Because they are still text, even if a different text? Just as we allow original juvenile fantasy, and novel expansions that change the text of short stories. But Dramatizations aren't text themselves, and they aren't a faithful reading of a text. Moreover to properly credit them we would need a cast list, a director, a script-writer, often a music credit -- we just aren't properly set up for them, any more than we are for films, IMO. Or for comics. -DES Talk 22:18, 17 August 2010 (UTC)
I'm agreeing that dramatizations should not be in the database, and one of the principle reasons is because they change the text of the original story, not because they're "acted". Many readers of audiobooks "perform" as they're reading, but stick to the text. I also agree that adaptations should be in, but believe that the similarities between audio dramatizations and printed adaptations are closer on a basic level than many of us want to think. Mhhutchins 22:41, 17 August 2010 (UTC)
Here's the path to database destruction: printed text -> audio readings of the text --> audio dramatizations of the text ---> video dramatizations of the text (films). First two steps are in, the last two steps are out. Consider the similar path: printed text -> printed adaptations of the text --> audio adaptations of the text ---> video adaptations of the text (again, films). Again, the first two steps are in, the last two out. We just have to make sure the lines are drawn sharply so that we don't stray down the path. Consider this path: printed text -> visual printed adaptations of the text (graphic novels) --> animated adaptations of the text (anime). There's a hazy line there between the first two steps and I can see some straying over. I personally don't see much difference between textual adaptations and visual ones, but am in the minority here. I will kill records for graphic novels and manga, but am hesitant about doing so for work by major writers. If we can include Asimov's science books, why not Gaiman's graphic novels? I consider the latter much closer to our raison d'etre than the former. Mhhutchins 22:57, 17 August 2010 (UTC)
I don't really care so much what is "adaptation" and what is origianl, I care about format. I would tend to exclude even a faithful-to-the-text reading with a different reader for each character, in part because of the cast issue. On Graphic Novels, i use the "is there a viable story if the images are all deleted" if not it is OUT, if so it is IN. Thus the original editions of Niven's The Patchwork Girl and Pournelle's Janissaries (both of which illuded about one full-page illo every 3-4 text pages, and both of which have been repritned without the illos, and both of which were promoted as "Graphic Novels") are IN. However Niven's Star Trwek comics would be OUT (even though a scene that later appeared in the early chapters of The Gripping Hand took first form here, IIRC). -DES Talk 00:21, 18 August 2010 (UTC)
Going the other way - we have no trouble with a movie or TV show novelization being "in." BTW, why don't we index comics, anime, manga, etc? Was it an arbitrary "let's only index stories told using words in print" or is there some other reason? --Marc Kupper|talk 04:40, 18 August 2010 (UTC)
I wasn't there for that decision. However, some reasons i think may have applied: The relatively huge number of comics issues; comics have their own community separate from the book community; the bibliographic sources which deal with genre books mostly do not deal with comics; to properly record a comic the names of persons in multiple roles must be captured (penciler, inker, colorist, writer, storyboarder, sometimes editor); the sense that comics were a separate thing from books. Also AIUI the ISFDB grew out of an SF awards database and out of bibliographic standards used on the usenet group rec.arts.sf.written (or its predecessors). Neither source would have included comics as in-scope. -DES Talk 14:05, 18 August 2010 (UTC)
That historical background makes sense but why are graphic novels, and by extension, Dramatic readings & Dramatizations, particularly if they are recorded and thus presumably verifiable, barred today? At present the only reasons I can see are
  • Lack of technical support in ISFDB. For example, we don't have title record types for song much less, dramatization and all of the comic roles. We don't have support for "adapted title" which is how The Mist: In 3-D Sound should get filed unless there's evidence this is a word-for-word reading of The Mist plus sound affects.
  • Few or no moderators who would be able to evaluate the accuracy of submissions. An editor could go wild and add a rats nest of junk from secondary sources and we'd never notice as we don't have copies of the works that'll enable some basic fact checking.
I can't say they those two are reasons for not approving a submission for "The Mist: In 3-D Sound" or deleting it once found. Had I ran across that publication record I'd likely keep it as it's based on an in-print specfict story meaning I know it's specfict. --Marc Kupper|talk 15:21, 18 August 2010 (UTC)

Subset collections as variants -- correct?

I recently noticed the following Stephen King colleoctions:

All of these contian some, but not all, of the stories from Night Shift but all are currently indexed as VTs of Night Shift. Moreover, the third (The Stephen King Collection) appears to be an omnibus of the first two plus Lawnmower Man and Other Stories from Night Shift. (which is not currently marked as a VT).

I think it is a mistake to have books with such different content marked as VTs under our current system. (When and if we add support for "based on" or something similar that will change, I think.) I would like support for removing the VTs on these, and replacing them with appropriate title notes. -DES Talk 14:26, 19 August 2010 (UTC)

I note also the existence of 1159628 and 169348, duplicates or near duplicates of the above that are not marked as VTs. These I will merge appropriately, once the VT issue is settled.

I note also that I recently created 1159656 and 1159628. I ran a "find duplicates" to detect possible merge targets, but did not find any because the "find duplicates" logic ignores VTs. I think this should change, see below. -DES Talk 14:33, 19 August 2010 (UTC)

About the variants: I disagree if the collection or anthology's title specifically mentions the collection from which the stories are drawn (more examples here and here), or if an anthology is abridged by one or two stories, retitled and marketed as a bargain book (example here). These should remain variants despite having different contents. If we do as you suggest the Bantam pb of Adventures of Time and Space would have to have its own title record. Despite having only 8 of the 35 stories of the hardcover, it is currently under the same title record as the original edition. In this case, I would prefer that a variant for the pb be created. Another Bantam pb (here) that pulls stories from the same collection has its own title record, because it is differently titled. I would also prefer that this be made a variant of the original title. Either way we choose to handle it, we should be consistent. Mhhutchins 16:19, 19 August 2010 (UTC)
My view is that VTs should be used for cases where the same content is published with different title and/or author credits, but not for cases where the content is significantly different. I would create a separate title record (with a note and a link) for the abridged collection Adventures in Time and Space. I know I've seen notes like "X (1994) is NOT the same as X (1983) do not merge") A number of "Best Of" collections ahve multiple versions with different contents. Or consider Lunar Activity and Phases two collections by the same author, with in fact the same cover art (mirror image) and about 2/3rds identical content. should these be VTs? DES
No. The second collection doesn't identify itself as being drawn from the first one. Mhhutchins 19:33, 19 August 2010 (UTC)
All right, I understand the distinction you are making. I don't see the point of it, however. We don't mark novel excerpts (which do state what they are drawn from) as VTs of the source novels. -DES Talk 20:11, 19 August 2010 (UTC)
We also don't mark "split" novels, like Cyteen' (one edition of which was published in 3 parts) or The Reality Dysfunction (published as two novels in US editions) as VTs, nor should we IMO. -DES Talk 20:47, 19 August 2010 (UTC)
How about making The Man Who Sold the Moon (the collection, not the story) a VT of The Past through Tomorrow (the collection of all the RAH "Future History" stories, which i see we don't have indexed). DES
No VT here either. I'm talking about anthologies and collections that are pulled from previous published ones. Not subsequently combined like The Past Through Tomorrow. I was specific about which ones I believe should be made into variants. Mhhutchins 19:33, 19 August 2010 (UTC)
Why should the order of publication mater. If we have two collections one a subset of the other, we should record the subset as a VT if published later, but not if published earlier? That makes no sense to me, and we won't always know the original pub date when a record is being entered, should we remove the VT if we later find an earlier edition of the subset, or add a VT if we find an earlier edition of the superset? And when working with secondary sources, we won't always know if the subset explicitly identifies its source. -DES Talk 20:11, 19 August 2010 (UTC)
The "Principle of least astonishment" also seem to me to suggest that when our display shows: "X also appeared as: Varient Title: Y" on an author's summery bibliography, or "Variant Title of: X" on a title bibliography, the user is entitled to assume that the contents are identical (or very close) and that having one of these, there is no need to find the others. I admit that in the Night Shift case the title ("...and other stories from...") gives the matter away, but it doesn't in the Adventures in Time and Space case, and may not in outer subset collection cases. DES 18:30, 19 August 2010 (UTC)
I agree that VT should signify identical text, but it's not always used that way. Until we have the ability to create a VT based on text, this is the best we can do. Once that feature's been implemented these can be easily transferred because they're already identified as having a relationship. Mhhutchins 19:33, 19 August 2010 (UTC)
That VTs are misused in some cases is not an excuse to misuse them in others. Help:How to record a variant title says "It often happens that the ISFDB lists a single work under two different titles, or under the same title but with different versions of the author's name." It says nothing about works with different text. Nor does Help:Screen:AddVariant. -DES Talk 20:11, 19 August 2010 (UTC)
What do others think about this matter? -DES Talk 20:11, 19 August 2010 (UTC)
Sometimes it seems appropriate to abuse the VT mechanism and other times not. We don't have adapted/expanded/abridged/etc. title support. When I use VT to link these I also append something like "(adapted)" to the title so that someone viewing a bibliography knows it's an adapted work and not a variant title. Recently I used a series to link the adaptation on H. G. Wells (for "The Invisible Man" and "The War of the Worlds") but I'm not happy with the results. It seems a VT would be a better fit. --Marc Kupper|talk 01:56, 20 August 2010 (UTC)
I don't like your series solution there, and I wouldn't be too fond of a VT solution either. I would probably incline to a separate title record with a note and link. But that has drawbacks too. At least in that case I can see that the VT captures a linkage not otherwise easily captured. I don't see that as being true in the subset-collection case. Lots of collections and anthologies overlap, the fact that one is "selected" from another seems to me worth mention in title level notes (and in this case the title itself says so) but I don't think it is important enough to mangle the VT mechanism to document, particularly when it seems to me in some cases to create as much confusion as it solves. -DES Talk 02:20, 20 August 2010 (UTC)
A flexible "based on" relationship, that could deal with adapted, expanded, abridged, and revised works, as well as fixups, would be wonderful and solve many problems. But I don't expect it soon. -DES Talk 02:20, 20 August 2010 (UTC)
Immediately after my last post I jumped into the shower and remembered why I did the series approach. Had I used a VT then the adapted titles would not have shown up on Malvina G. Vogel bibliography. She's the one that adapted the Well's work. Thus something to keep in the back of the mind for developing support for "adapted works" that sometimes (all the time?) these will need be visible on other bibliographies. The goal was to show The Invisible Man and its adaptation together. Three alternatives come to mind 1) Remove the series stuff and let this be a title that appears on the Well's bibliography nearly 40 years after his death. 2) Remove Wells as an author and give Vogel sole credit. This is troublesome as the publication mentions Wells' name but not Vogel on the cover and spine. The title page has the adaptation credit in small print. 3) Merge the adapted title into its parent and the publication record will disclose that it is an adaptation. --Marc Kupper|talk 03:04, 20 August 2010 (UTC)
If we do a quite general "based on"/"related to", as I view it, each work would be independent, and would appear on whatever biblios its own author credits place it, but there would be a repeating field in which each work wold be based on or related to one or more other works, with the nature of the relationship specified in each case. Such relations as "fixup", "expanded from", "revised from" "adapted from", "translated from", "incorporates", "subset of", "split from", etc would be supported. Each relation would have a proper reciprocal: "Expanded from/Expanded into", "translated from/translated into", "incorporates/incorporated into", etc. This would be a relation between texts, not titles, and so would be orthogonal to the VT relation. But until this or something of the sort is implemented, we must deal with what we have. -DES Talk 05:39, 20 August 2010 (UTC)
I'm not sure what you mean by "a repeating field." Can you clarify that part? Thanks. --Marc Kupper|talk 07:28, 23 August 2010 (UTC)
I mean a field that, like the Author field on a Title or Publication record, or the Website field on an Author record, can appear more than one time; that one title can be based on or related to many other titles. That is required if the feature is to handle fixups in particular, but would also be wanted in other situations. -DES Talk 14:26, 23 August 2010 (UTC)
See the long discussion at ISFDB:Proposed Design Changes#Large Issues: Roles & "Based on", and the shorter one at ISFDB:Proposed Design Changes#Based on. -DES Talk 16:16, 23 August 2010 (UTC)
Thanks - the repeating field complicates things though I'm not sure if by a lot or a little. --Marc Kupper|talk 18:01, 23 August 2010 (UTC)
Getting back to the original topic - I don't see a clean solution but am not particularly disturbed by the use of VT here. I also would not be disturbed were there no variants at all. These seem to be four separate collections with three of them happening to be subsets of the stories available in Night Shift and, presumably for marketing reasons, borrowing the original collection title. --Marc Kupper|talk 03:04, 20 August 2010 (UTC)
That is almost exactly correct, there are actually five collections: three of them subsets of the first, and the last one being the combination of the three subsets, which still does not quite include all from the original Night Shift. Note also that the subsets seem to have mostly or perhaps exclusively been issued as audio books. -DES Talk 05:39, 20 August 2010 (UTC)

Feature request: Find Dups to check VTs

Currently the "find duplicates" logic ignores VTs. As mentioned in the section above, this can cause probable merge targets to be missed, particularly when the author is prolific and visual inspection can easily miss the similar titles. (for example, "Gray Matter" and Four Other Stories from "Night Shift" is currently a VT of Night Shift) but Gray Matter and Other Stories from Night Shift is not. Find Dupes does not suggest merging these two.)

I suggest changing the "find duplicates" logic so that VT titles are suggested as merge candidates, except that:

  • The logic will not suggest merging a VT with its parent, and
  • The logic will not suggest merging two titles each of which is a VT of the same parent.

I would think this logic could apply all the time, but it could be made optional much as the "similar title mode" is optional.

I will create the FR in SourceForge, but only if others agree that this is a desired change. -DES Talk 14:43, 19 August 2010 (UTC)

I agree it should find VTs, but I don't agree with your second exception. The most important case is a title with no parent should turn up as a merge candidate with an identical VT, as the most likely scenario is someone's adding new titles via contents and then wanting to merge them. I'm not sure why you'd want to exclude two identical titles sharing the same parent -- under what circumstances would those NOT be merge candidates? --MartyD 00:32, 20 August 2010 (UTC)
Two titles records sharing the same parent may have identical titles but different author credits, one major reason for having VTs. The "strict" find dups logic now suggests merging records with different author credits, such as one having a co-author and one not having a co-author. I am not sure just how "identical" the find dups logic insists that two titles be, even in strict mode, and a small difference might be enough for there to be good reason for a VT. This applies much more strongly in "similar" mode. But my main thought was, records don't get marked as VTs by accident or automatically. If two records are both VTs of the same parent, some human thought that was proper. But the details can obviously be worked out. I agree that the most important case is finding a title record with no parent, i.e. a title NOT marked as a VT, and merge it with an otherwise similar title that IS marked as a VT. That is the case I most want handled. -DES Talk 01:01, 20 August 2010 (UTC)
I have run into the same problem, i.e. "a title NOT marked as a VT [...] and [...] an otherwise similar title that IS marked as a VT". Definitely worth a FR. Ahasuerus 03:57, 23 August 2010 (UTC)

"With" in publication titles

Apparently Help has been changed to state that "with" should be in lowercase for story titles. The same change has not been made for pub titles. If no one objects in the next couple of days I will make the same change for pub titles.--swfritter 14:54, 24 August 2010 (UTC)

Please do. I think i recall the discussuon that led to the change, but searching for "with" in the wiki may be fruitless. -DES Talk 15:46, 24 August 2010 (UTC)
True - a search for "preposition" worked though. ;-) See here - and I still prefer "with" myself. BLongley 18:55, 24 August 2010 (UTC)
Feel free to change them. I personally think it's not worth the effort. As long as searches for either come through, what's the point? Mhhutchins 00:55, 25 August 2010 (UTC)
Based upon this discussion and the referenced discussion I have added "with" to the list and hopefully found all the places the change is appropriate. Whether the standard change is made or not, the usage has been so inconsistent that at some time a script will have to be run to make things consistent if so desired. Definitely not a high priority.--swfritter 13:05, 25 August 2010 (UTC)

Preschool books OUT?

"Poof!" is a book, labeled as "preschool" and as "ages 4-8" by amazon and OCLC. I edited it in the course of the chapterbook cleanup project, and found a possible bug, which I reported. In response, Kraang wrote "This looks more like a preschool book that should be deleted."

Currently our "scope rules" exclude:

  • Fairy tales with no known author, and
  • Animal books for very young children, i.e. books for preschoolers which depict simple scenes from animal life featuring anthropomorphized animals

This is neither, it is a book by a known, modern author, showing a children's version of the classic magician's duel by shape transformation. It also seems to echo the ballad "Get Up and Bar the Door" a bit.

We have, of late, added a good deal of "Juvenile Fantasy", much via Fixer submissions. None of the RoAs nor the scope definitions explicitly excludes works aimed at an audience below a particular age, although in practice some are excluded as "graphic novels" for having too little text. Should we change the rules to explicitly exclude such works? or whould we explicitly let them in? A clear cut rule would be nice. -DES Talk 03:50, 25 August 2010 (UTC)

Target age is a difficult criterion, as it's both subjective and often without authoritative source. If they otherwise fit the definition of speculative fiction, and a human being takes the time to enter them, I think we should let them be IN. The spirit of the RoA incorporates an inclusionist bias, which seems to apply here. As a side benefit, we avoid opinion-only arguments about whether something's truly preschool or is post-preschool "juvenile" --MartyD 10:38, 25 August 2010 (UTC)
Both KiddieSpecFic and Graphic Novels should be in if someone takes the time to enter them manually and should not be deleted if they are already in. They especially should be in if they are reviewed in a SpecFic magazine or receive major awards - these may be clues that they are of interest to the SpecFic community that we are supposed to be serving.--swfritter 13:15, 25 August 2010 (UTC)
"PreSchool-Grade 2-A great read-aloud" from Library Journal. This title mosts likely came in with all the 2004/5 preschool picture books. As for there inclusion I'm on the side of keeping them out. A walk around this section of a book store leaves me with the impression the db would double in size to include them. Some of these authors must crank out one a week! :-/Kraang 03:39, 26 August 2010 (UTC)
I won't seek them out but won't turn them down either. Many of these books have grade, age range, or reading levels stated in the books and so I add that data to the title notes. I looks the books over and if the foundation seems to be "make believe" then I don't enter them into ISFDB. I'm still undecided about series such as ThunderCats and Princess of Power both of which are space opera for young children but also skirt the edge of graphic novel territory. --Marc Kupper|talk 08:26, 26 August 2010 (UTC)
There are some challenges associated with books for young children, e.g. many of then are borderline graphic novels or picture books. At the same time, many of them are recognizably SF, e.g. Dav Pilkey's "Captain Underpants" adventures include spaceships, time machines, robots, pterodactyls, etc. They also include crossover material like mini-comics and "world-famous cheesy animation technique, Flip-O-Rama". (And enough potty humor to drive adults up the wall, but that's a different question.)
On balance, I think they can be allowed in our exclusive club. We just need to maintain our current policy of preventing Fixer from submitting them automatically to avoid queue saturation.
Re: Marty's comment that "[t]arget age is a difficult criterion, as it's both subjective and often without authoritative source", there are, admittedly, borderline cases and some overlap, but generally juveniles fall into one of 4 categories:
  • 0-4 (toddlers)
  • 4-8
  • 9-12
  • 13-17 (YA)
This is useful information and is usually provided by library catalogs since it helps parents decide what books to select for their children.
At the moment, we support "jvn" in the Storylen field, which is similar to the "Audience" data used by WorldCat. However, this designation is rarely used (in part because the field is shared with other data elements) and IMHO could be profitably replaced with "Reading Level" or "Target Age". Ahasuerus 23:56, 31 August 2010 (UTC)

Collaborative Pseudonyms

For M. Barnard Eldershaw we apparently have two authors who always collaborated with each other using the one pseudonym (see wikipedia). I suppose this isn't the most critical case since the two apparently only authored the one SF title, so I don't feel strongly on this issue, but without creating a variant it would be impossible to list biographical data for both authors, and it's possible that there'd be other more significant pairings in the same position (although I'm just speculating here). Should a variant be created to allow entry of individual biographical data? Jonschaper 03:58, 31 August 2010 (UTC)

We did so in the case of Robert Randall, but there the two also wrote lots of individual works of SF. Still it is a precedent. -DES Talk 04:17, 31 August 2010 (UTC)
I think the stronger analogy is "Travis Tea", whose variant entry required the creation of several otherwise unlisted authors along with existing ones. But that still doesn't quite fit the Eldershaw situation a variant would mean creating a parent that a) otherwise wouldn't exist, and b) are of less importance than the pseudonym (whereas Garrett and Silverberg are of obvious importance and exist independently of the pseudonym, as are some of the people making up "Travis Tea"). Yes, I'm sort of arguing against my position here, but I want to make sure nobody has objections on those bases.Jonschaper 04:36, 31 August 2010 (UTC)
Actually, I just thought of the perfect precedent: Steve Barlow & Steve Skidmore, AKA "The Two Steves". I'll go ahead and submit a variant on the strength of that. Jonschaper 04:41, 31 August 2010 (UTC)
At this time, collaborative pseudonyms are supposed to be disclosed (when known) and biographical data is entered for individual authors. It's a bit more work, but it follows the rest of our rules. Ahasuerus 23:30, 31 August 2010 (UTC)

Titles "written as by"

Something I don't like about ISFDB are bibliographies like Robert Randall where we don't get a nice summary of the titles "written as by" that name. I'd go ahead with the variants and hopefully at some point we'll be able to view bibliographic summaries for pseudonyms and pen names. --Marc Kupper|talk 07:50, 31 August 2010 (UTC)

Sorry, a tangent from the original point, but does anyone know the history of the current behavior of displaying empty bibliographies for pseudonyms? It seems in direct contrast to the handling of variant titles, i.e., if I go to the VT, I see it is a variant of some parent, but I see all of the publications under just that variant; if I go to the parent, I see all of the publications under any variant. I've wondered why author bibliographies are not set up to work the same way, i.e., if I went to the pseudonym, I would see it's an alternate name for some name(s), but I would see all of the titles written as by that pseudonym; but if I went to a non-pseudonym, I would see all of the titles by that author written under any name. --MartyD 10:08, 31 August 2010 (UTC)
You re right that would make sense. It would, however, make it harder to know which titles credited to the pseudonym do not yet have variants set up. (currently those are the only ones which appear). If all titles which are credited to the pseudonym are displayed, there would need to be a clear indication of which ones have variant parents already set up. As to the history, the current behavior was in place when I started editing here. -DES Talk 12:38, 31 August 2010 (UTC)
"Show all titles" still works for Robert Randall and similar authors, but that isn't as nice as an automatic display. There might be a problem in the fairly rare case where a pseudonym is used by multiple distinct authors (not working together). Not that we couldn't display all titles so credited, but it wouldn't be easily clear to the user who actually wrote each title. -DES Talk 12:43, 31 August 2010 (UTC)
Briefly, the history of the current behavior is as follows:
  • Spend a lot of time/effort making sure that all titles authored by the same person appear on the same page (took a while)
  • Realize that this makes it hard to view "house name bibliographies", e.g. all books written as by "Alexander Blade"
  • Scratch head, sigh and add it to the wish list, where it has remained for the last couple of years
One quick thing that we could do would be to display "Show All Titles" right next to the "Pseudonym. See:" line. Perhaps change it to something more intuitive like "View titles for this pseudonym"? Ahasuerus 23:38, 31 August 2010 (UTC)
Display all titles as the solution is a bit of a hack. I can do an Advanced Search too. I believe the "least astonishment" would be for the normal bibliographic displays to show all of the titles and there would be a special display used to show titles that have not been VTed. Granted, we need to come up with a way to display titles on the pseudonym's page. --Marc Kupper|talk 10:43, 24 September 2010 (UTC)

Title case for non-english works

The ISFDB standard is to normalize the case of titles, that is, to capitalize all words except for a specified few. This is in conformance with standard English-language literary and Bibliographic practice, and was, I think, adopted because of the number of pubs that print titles in all caps, or small caps, or other odd variants for display effect, without intending to create variant titles, and which we don't want to track as variants.

However, we are now starting to include a number of non-English titles, and working to improve support for them. In French, the common standard is to capitalize only the first word of a title as I understand it. Other languages may have yet different rules. Should we normalize them all to follow English language practice, or follow the rules of the language involved? It would be good to have a single consistent practice, recorded in the help. I can see arguments to support either choice. -DES Talk 15:19, 31 August 2010 (UTC)

I lean towards one rule for all languages, if only for practical purposes. I wouldn't like to have to search for Hungarian, Italian, Portugese, Polish etc. rules. The English standard works for me, and I have used it for French, German and Dutch titles. --Willem H. 18:07, 31 August 2010 (UTC)
When this suggestion last came up, one editor pointed out that we would need to create a dictionary of "non-capitalizable" articles and prepositions for each language and moderators would have to check it every time. That's quite a bit of work. Ahasuerus 23:22, 31 August 2010 (UTC)
I don't see why we can't allow each language's standard be used. For an example, see this title by René Barjavel, and this title by the Strugatsky brothers. Mhhutchins 23:16, 31 August 2010 (UTC)
Also, we don't make variants based on capitalization, so I don't see why we should be concerned about whether the word in a title is capitalized or not. Searches don't require capitalization either. I see this as a non-issue. Mhhutchins 00:20, 1 September 2010 (UTC)
As I see it, there are pros and cons of any choice.
  1. Using the English-language rules for all titles simplifies procedures and gives everyone a clear bright-line standard, but might look odd to users accustomed to the standards of other languages, particularly users whose primary language is not English. It would also require listing the "no-cap" words for each supported language.
  2. Using each language's standards in works of that language would require documenting a standard for each language for which we have works, and would require editors who might be entering the occasional work in an non-English language (particularly when it is a translation -- I have entered translations into French, Spanish, and German from OCLC data in the past week, and OCLC is notorious for not preserving capitalization from the actual book, nor using standard bibliographic capitalization) to check these rules, and moderators who approve such subs to check the rules also.
  3. We could declare that no strict normalization standards apply, and "follow whatever the book does" at least for non-English-Language titles. This avoids documenting and learnign additional rules, and does help document the actual publication. Mhhutchins points out that searches are case insensitive, although I'm not sure if this is always true for accented characters. Similarly the "Find Duplicate Titles" appears to be case insensitive (again i didn't test accented characters). The disadvantage is that wwe lose any sense of standardization on this issue, but perhaps that isn't huge. i would dislike droppign the normalization rules for English-language titles, however. (I'm not sure if Mhhutchins was suggesting doing so.) There there is a clear and widely accepted bibliographic standard, and lots of titles entered in accordance with it.
Whatever we choose, I would like a clear rule, recorded in the help, so we don't need to have this discussion again. -DES Talk 03:26, 1 September 2010 (UTC)
As we concluded (I think, see above) for Eastern vs. Western names, I think our "rule" should follow the native custom/standard. We could briefly document these as they are discovered but have no expectation that moderators unfamiliar with the particular language will be able to enforce them. As Mhhutchins points out above, it really doesn't matter all that much. I think it is much friendlier for editors if we don't try to force what is to them unnatural capitalization of words. --MartyD 10:29, 1 September 2010 (UTC)
Would that mean that editors who are not familiar with a given language would be free to enter titles/pubs in that language (translations, probably) with the English-language capitalization rules if they do not know the "native" ones? -DES Talk 13:55, 1 September 2010 (UTC)
As french, I'm a bit hampered by the need to convert titles in my tongue into english capitalization system, especially for "small" words whose grammatical status is not always evident (avec, dans, etc). It sometimes strikes me as very odd to write "Il Mange" but in fact I tend to agree that it's quite a non isssue as said by Mhhutchins particularly for reference purposes. The only potential problem lies in the accented characters as a frenchperson will instinctively type "étoile" instead of "Etoile" (this needs testing but it uncovers the problem that search is not operated on the diverse titles (translated) of the same book, i. e. a search on _"Le Pont Sur les Etoiles" gives 0 matches even if there are two french printing of the title, this can be problematic for searches by non-english users who try th find a english-written book with only the translated title). I'll be happy to try to enter titles in "the english way" (that's how I proceed now) but all I want is a relaxing of the moderation on this parameters (edits refused because they're not to standard) which, if too strict, can easily discourage potential contributors. Hervé Hauck 16:32, 1 September 2010 (UTC)
I don't think the accented characters form a problem in searching. A search on all titles gives the same 40 results for "etoile", "étoile", "Etoile" and Étoile". no problem there. No results for "Le Pont Sur les Etoiles" is because there is no title record (the publication refers to "Star Bridge", solution for this depends on the implementation of foreign language support). If you try the "ISFDB Publication Search Form" in the advanced search option, the two pubs are shown. As for "relaxing of the moderation", I would be very happy with any clear set of rules (even if it was like "these rules are for English-language titles only, the editor decides on other-language entries"). As the rules are now, Benario's addition of this pub should at least have raised some eyebrows (as I understand it, we don't support translations of short fiction yet). --Willem H. 07:12, 2 September 2010 (UTC)
That submission should not have been accepted. Mhhutchins 14:29, 2 September 2010 (UTC)
I just talked to benario about this but I think that it'll be better to have a moderator give him (her) explanations (particurlarly on the merging of titles). For the search on french titles, I've made the same tests (with Etoile), but I think that the use of the advanced search is a bit too complicated for non-english persons to use naturally. Hauck 09:49, 2 September 2010 (UTC)
Off-topic, and this should be moved to Community Portal or somewhere else if someone wants to pursue it, but: Use of advanced searching is a bit too complicated for many people, regardless of English fluency. The problem is the quick search looks at title records, not at publication records, and for reasons that I don't understand, we're setting up the non-English translations as publications without the corresponding non-English title as a variant (in contrast with an English translation of a non-English work, where the original-language title is usually set up as the parent with the English title as a variant). If we always had variants, the quick searching would work as expected. --MartyD 11:28, 2 September 2010 (UTC)
This site was primarily set up to record English language publications (I'll let those involved with the creation of the database give the rationale). Making non-English translations into variants would have created so much clutter on the author's summary pages that it would have been hard to find titles. Until there is full support for foreign language publications, when the user can set their own displays for language, we have to enter foreign titles under the original title. The corresponding English translation of original non-English titles is handled as parent/variant relationships because there's far less of them, and out of respect to the original work, IMHO. Mhhutchins 14:29, 2 September 2010 (UTC)
That was my understanding, and the reason I have been waiting with the Dutch collection (no problem, always enough to do). I hate doing things twice, when all it takes is a little patience. --Willem H. 14:49, 2 September 2010 (UTC)
Another thing we could try to do is rephrase the rules to be more generally applicable across languages. I.e., instead of a list of words not to be capitalized, we could have something along the lines of "determiners and prepositions of fewer than five characters". I also agree that rejecting edits based on capitalization is not very friendly, independent of any rules. --MartyD 11:08, 2 September 2010 (UTC)

Attributions for Collected Letters

A question of how to attribute collected correspondence recently came up regarding the correspondence between Ursula K. Le Guin and James Tiptree Jr. — whether or not to add the correspondent(s) as authors, or to credit only the editor(s). It originally seemed to me that the attribute should go to the editor, as in the letter columns in a magazine, but some thought shows there are three cases depending on the number of writers, one, two, or many. The first could be treated as a collection and attributed to the author, the third like an anthology and attributed to the editor, the second(?). Comments?--Rkihara 15:27, 7 September 2010 (UTC)

It is my view that if all of the letters are from a single person, that person should normally be credited as author (or coauthor with the compiler). It is my view that if it is an exchange of letters between two people, those people should be credited as authors, possibly with the compiler/editor credited as well. It is my view that if there is a collection of letters between one person and many individuals, then the central person should be credited, possibly along with the compiler/editor, possibly along with "et al" or "and others".
It is my experience that most published book-length collections of letters do so credit the author(s) of the letters. And i think that is supported by the majority of the cases (but not all) that the ISFDB has already indexed. Examples follow:
Of course, for a book. who is credited on the title page matters. -DES Talk 00:01, 8 September 2010 (UTC)
If an editor is credited on the title page of a book publishing correspondence, I think we should credit both the editor and the letter writer(s) in the author field, if the letter writer's name is part of the title. Otherwise the credit would not appear on the letter writers' summary page. With that in mind, I'd have to change a pub that I verified! Now, non-book-published correspondence throws a wrench into the works. Individual essays made up of correspondence between two or more writers seem to be more of the editor's work than the letter writers. It was the editor who created the work, not the authors of the letters. I would lean towards giving the editor sole credit in these cases. If the essay were entirely of one author's correspondence then I'd give credit to both the editor and the author. Sorry to be seemingly wishy-washy, but different situations call for different responses.
I'd even carry this further in the case of art books which collect the work of a single artist. Both the editor and the artist should be credited in the author field. Editors of single-author collections is another matter. Until we can make additional roles for such a book, we have to stick with editors only being credited in notes field. (Even though this means I won't be credited in the db for such work.) Mhhutchins 04:19, 8 September 2010 (UTC)

Conflicting rules?

I sure hope this is an exceptional case. User Hauck has entered this pub according to the foreign language rules (entered as a publication under the main title record). No problem there, but the pub has (as documented in the notes) an misspelled author (Colin Rapp vs Colin Kapp), and according to the pseudonym rules: "If a title record exists under the canonical name DO NOT add this pseudonymous publication under that title record". How should we deal with this situation, or, which rule prevails? --Willem H. 21:08, 11 September 2010 (UTC)

I would be inclined to follow the pseudonym rules and create a variant in this case, in line with the general "enter the book as it exists" principle. An argument could be made that this is a mere publication typo, not an intended pseudonym, but in general we document what was actually published, rather than trying to determine intent. I am hoping that the current foreign language rules will be changed when we have better language support anyway, while the pseudonym rules I think should be more enduring. -DES Talk 00:37, 12 September 2010 (UTC)
If you look at "Colin Rapp's" author page you will see this title is now a "Stray Publications:". :-)Kraang 01:51, 12 September 2010 (UTC)
The same problem exists with a couple of Strugatsky brothers books (published in French using "Strougatsky"). See Boris Strougatsky or Arkady Strougatsky. These have the added wrinkle of the main titles' not being English. I am thinking the best approach at the moment is to make variant titles, even though one non-English title would be the variant of another. Just making pseudonyms isn't going to be very helpful. --MartyD 11:32, 12 September 2010 (UTC)

Chapterbooks - merging with fiction titles

It is my current understanding that that there should be both a chapterbook title entry with separate title entries for content; no merging of chapterbook titles with other titles. Pub Help is not explicit about the subject and should be modified to state the current standard.--swfritter 14:20, 22 September 2010 (UTC)

Chapterbook titles should not, in general, be merged with SHORTFICTION or NOVEL titles, as I understand it. If one decides that a chapterbook would be better treated as a collection 9for example when it contains multiple works of fiction), then it might usefully be merged with a COLLECTION title record. In theory the same rule might apply to an OMNIBUS, but I find an example hard to imagine. ---DES Talk 14:29, 22 September 2010 (UTC)
Note also FR 2909853 (Prevent accidental shortfiction/chapterbook merges) which assumes that such merges are usually errors. -DES Talk 16:51, 22 September 2010 (UTC)
Help:Screen:NewPub doesn't include anything specific to the CHAPTERBOOK type (or any of the other types it deals with) and doesn't mention merging at all except in connection with art records (where it advises against merges). Help:How to merge titles does say "If the Types are different, you should probably not merge." but could be more explicit. I will edit that page. -DES Talk 14:36, 22 September 2010 (UTC)
I just made this edit which I hope clarifies things a bit. ---DES Talk 14:53, 22 September 2010 (UTC)
Help:Screen:EditPub#PubType is also relevant. In general "Edit pub" is more through than "New pub", because there are things you might do when editing an existing pub that you can't or are unlikely to do when creating a new pub. ---DES Talk 14:58, 22 September 2010 (UTC)
Thanks! The subject has created so much confusion that perhaps Pub Help for this specific subject matter should have a link from the Chapterbook section to Merge help perhaps with a short sentence like "See Help:How to merge titles"?--swfritter 15:00, 22 September 2010 (UTC)
Perhaps. Where exactly do you think the link should be placed. -DES Talk 15:05, 22 September 2010 (UTC)
Does this edit do the trick? -DES Talk 16:22, 22 September 2010 (UTC)
Good. That will make it easier next time I forget.--swfritter 15:55, 8 October 2010 (UTC)

Last name - shouldn't it be the last name of the canonical author

Author data Help doesn't specify. But since the data is used to build the author directory shouldn't it be the last name of the canonical author? For example, Cyrano de Bergerac has the lastname field as Cyrano which is actually the author's true last name. For sorting purposes it might also make sense to take out spaces; a look at the phone book and other sources suggests that this is the normal method.--swfritter 15:53, 8 October 2010 (UTC)

I cleaned up the help to better explain (hopefully) how the last name field should be used. Something that would help would be if edit-author shows the last name that would be computed. If it's "wrong" because of how the canonical name got parsed then we'd know that the Last Name field should be filled in.
I agree, Cyrano de Bergerac should be filed under Bergerac and not Cyrano. Leaving the Last Name field blank will do this.
The part about the space is addressed indirectly with "The field should contain the last name only." Maybe we should do a cleanup project to search for "Last Name" fields that contain spaces or do not seem related to the canonical name. One way to do this would be for each author where LastName is set to compute the Last Name field and to build a list of the case where they don't match. Some of those will be valid usages of the LastName and others will be errors. --Marc Kupper|talk 02:05, 11 October 2010 (UTC)
I'm not sure I understand the comment about spaces. Proper last names do indeed include spaces. For example, Ursula K. Le Guin's last name is "Le Guin". Currently, the Last Name field is set to "Guin" which is rather unexpected for anyone looking for her in the directory. I can understand setting it to "LeGuin" per swfritter's suggestion, but not "Guin". --JLaTondre 22:15, 11 October 2010 (UTC)
Marc, which Help are you referring to? In the Help I referenced above I am suggesting that the text "The field should contain the last name (surname) only" to be "The field should contain the last name (surname) only of the canonical name". Compressing the spaces would basically be for sorting purposes but it is not critical that we maintain lastname in such a way since a programmer can write a sorting routine to do this on the fly when generating a sorted list. I would think "De Bergerac" is more properly a two word name starting with "de ".--swfritter 14:03, 12 October 2010 (UTC)
Swfritter, the help I referred to is the Last Name bullet of Help:Screen:AuthorData. The text for that bullet is at Template:AuthorFields:LastName. It was linked at the very beginning of this thread and so I had not linked to it again. --Marc Kupper|talk 16:59, 13 October 2010 (UTC)
The real question is do we want the name "Cyrano de Bergerac" to sort under B or D. Usage where surnames include "particles" such as "de", "van", "von", "ben", "al", "di", "d'" etc is inconsistent. Most of these mean "of" and indicate that the name is (or more likely once was) a patronymic or a locative -- that is a father's (or parent's) name, or else the name of a place with which the person who first used the name was in some way associated. Some sources ignore these for purposes of sorting, others do not. In some cases the name is particualerly well knowin in one form or the other. In the case of Cyarano, I would favor sorting under "B", but a case could be made for the other outcome. In The case of UKLG, the usage "Le Guin" is so common in SF that to sort her under "G" would be perverse, but other less well known names starting with "Le" might better sort under something other than "L". For example if the well known military figure of the 100 years war "Gilliam Le Marchal" (known in English as "William Marshal" or "William the Marshal") were an author, he should probably sort under "M". That said, swfritter is right that the phrase "of the canonical name" should be added -- we don't want anyone putting in "Clemens" for the last name of Mark Twain. In fact I will add that phrase now. -DES Talk 23:08, 12 October 2010 (UTC)
Agree about Bergerac. A common sense approach is perhaps to ignore the particle for foreign born authors but use it for American/British born authors. The particles still have a semantic meaning for other languages. We will just have to trust editor judgment.--swfritter 12:43, 13 October 2010 (UTC)
I too am surprised that Ursula K. Le Guin's LastName field is set to "Guin." Stephen, you did not comment on this and so it's not clear if there's agreement among the thread participants that she should be filed under "Le Guin" in the author directory.
I'm not sure if we can make bright-line rules about particles and the country of birth. I believe what will help is a better understanding of how the "Last Name" field is used among the moderators and editors. I'm sure many times it's been set when it does not need to be and other times it's set to something that'll cause astonishment when someone else views the author directory. --Marc Kupper|talk 19:59, 13 October 2010 (UTC)
Every case should be decided individually; I wouldn't waste time trying to make up a set rule. The main reason this subject has not come up for discussion is that almost nobody uses the lastname field for anything but it finally dawned on me that a concatenation of lastname, a delimiter, and canonical name results in a lastname, firstname sort. Although, with the advent of random access searching, even last name sorting is a standard that is disappearing.--swfritter 22:16, 13 October 2010 (UTC)

Spaces and sorting order

This is a sub-thread to focus on swfritter's original comment about "For sorting purposes it might also make sense to take out spaces; a look at the phone book and other sources suggests that this is the normal method." Initially I had misunderstood that when I replied above.

I looked at a phone book and found examples where it's clear they remove both spaces and hyphens and then sort the list. That would be a feature request for the code. It seems reasonable though a bit of a challenge unless there's a mysql option for "order by" that allows us to ignore spaces/hyphens in the fields being sorted. --Marc Kupper|talk 18:43, 13 October 2010 (UTC)

Actually I think this would be a mistake in principle. Yes, phone books often do this, but library catalogs generally do not. Personally i think that "Le Guin" should sort before "Lee" -- the "no-space" rule would sort it after. Since it would also take extra work, I think it should not be done. -DES Talk 21:11, 13 October 2010 (UTC)
I agree it would be a mistake. A more immediate problem is where we've disambiguated authors like "Richard Foster (Born 1949)" - his last name certainly is NOT "1949)"! BLongley 21:30, 13 October 2010 (UTC)
I personally alphabetize accounting for spaces, e.g.: de Camp, de Lint, Dean, DeChancie, del Rey, Delany. Leaving out spaces we'd get Dean, de Camp, DeChancie, Delany, de Lint, del Rey. Looks strange to my eyes. Mhhutchins 21:51, 13 October 2010 (UTC)
Contento squeezes the spaces but it is not necessary to make this a standard for data entry; this is an indexing issue that can be done on the fly for display purposes if someone wishes to do so. It is actually also my own personal inclination to leave the spaces in. It would be kind of nice if the parsing that is used to create the lastname took the "(" into account. Easy enough to write some SQL to look for "(" in existing records.--swfritter 22:06, 13 October 2010 (UTC)
Yes, but there is "[" too. We're up to "Martin [VI]" for some reason. :-( BLongley 22:24, 13 October 2010 (UTC)
I checked my public library and they seem to leave the spaces in but they are removing the comma that's between the last and first name. They also sort "Lê" the same as "Le." A fixup project to deal with ( and [ seems straightforward. --Marc Kupper|talk 08:20, 14 October 2010 (UTC)

Should diacritical marks be the basis for a variant record?

I'm holding a submission which wants to make "The Tune in Dan's Cafe" a variant of "The Tune in Dan's Café." I don't think that diacritical marks should be the basis for a variant record, as they appear to be more the printer's choice, not necessarily the author's. There doesn't seem to be a stated policy concerning this or I was unable to find anything in the help pages about it. Mhhutchins 19:43, 22 October 2010 (UTC)

As the submitter of the variant, I was initially inclined to merge, but decided to err on the side of caution since other small, debatable points of punctuation are often recorded as variants. Personally I agree with Mhhutchins that it's not worth taking note of. BrendanMoody 19:54, 22 October 2010 (UTC)
I did find a discussion about the subject here, but it didn't lead the any changes in policy. This help screen, under Accented characters, is mentioned. I personally think, that if the story in The Mammoth Book of Ghost Stories drops the accent, it should be a variant. --Willem H. 20:26, 22 October 2010 (UTC)
Those links are in regard to author names, not story titles. Mhhutchins 17:21, 23 October 2010 (UTC)
That discussion started with Stableford's Salome vs Salomé and Poe's A Descent into the Maelstrom vs A Descent into the Maelström. Both are now set up as variant titles. --Willem H. 18:59, 23 October 2010 (UTC)
And those decisions were not based on a stated rule or standard in the help pages, so that page should not be used to backup such a decision. A consensus about titles has never been arrived at. Mhhutchins 05:14, 24 October 2010 (UTC)
I lean towards accuracy in reporting what's stated in publications and so would set up a variant title. --Marc Kupper|talk 04:02, 23 October 2010 (UTC)
Yes, but how many titles are in the database without diacritical marks simply because the editor who entered them didn't know how to enter à é î ö and û, or simply thought it just doesn't make that much difference? Mhhutchins 05:14, 24 October 2010 (UTC)
I approach this from the point of view of someone searching. If searching with and without the diacritical produces the same result, IMO it doesn't matter. But if searching with/without the diacritical produces different results, then it does matter, since searching by the one that was merged away could produce empty or incomplete results. Using the two examples above, "Salome" and "Salomé" are equivalent, but "Maelstrom" and "Maelström" are not. I also think it is unreasonable to expect editors (or moderators, for that matter) in general to recognize that the system will treat certain diacriticals differently from others, much less which ones and why. So I think the rules should at least permit/encourage variants based on these differences. We could have an additional clause permitting the combining of titles where diacriticals don't matter -- in such a clause I'd think we should call for recording the title with the diacritical -- if we could figure out how to state that succinctly. The same rules should be applied to author names. --MartyD 11:04, 24 October 2010 (UTC)
I find Marty's comments persuasive. I also think that if the diacritical is in fact commonly used, it should be on the canonical title (or author name) if possible. I would suggest that it be considered desirable but not mandatory to create VTs based on diacriticals, and that a list be created of which diacriticals do not affect searches. (that is, a search for the unaccented form will find the accented form also, and vice versa). Diacriticals on that list need not be variants, and any knowledgeable editor can safely merge them, leaving a note indicating that there is no need to create future variants. Merges should be to the most accurate form (or most common form for a person's name).
This approach slightly reduces accuracy, but not in a way that impairs usability, I think. It assumes that there is no case where two publications can be usefully distinguished mostly or solely by a diacritical. IMO the alternative would be a (future) standard of strict accuracy, with VTs for all diacritical forms actually seen in print, at least. That seems more work for little gain, but I could live with it. -DES Talk 14:56, 24 October 2010 (UTC)
One of my big pre-ISFDB regrets was that I decided that Philip Jose Farmer and Philip José Farmer were the same and so I only used Philip José Farmer in my database. Later, I was looking at a book and saw that it had Philip Jose Farmer. I then realized I had no idea how common Jose was vs. José and that I would need to revisit every single one of my Farmer books to answer the question. At times I have merged things in my database and every time have regretted it.
Thus I'll object strongly to any merge proposal unless it were done in a way that it's possible to instantly recreate the pre-merge lists and that every affected title and publication record contain notes explaining that the title was modified.
Another problem with merging is we loose the ability to easily see which name should be canonical. Right now someone someone can look at the title records for Dan's Cafe and Dan's Café and can instantly see the verified publications to help him or her decide which version should be canonical. Many new editors do not enter notes at all meaning any system that relies on inspecting all of the publication records for notes on stated names/titles is likely to get an incomplete picture.
A software change that seems like it'll work would be automatic merging during display of author bibliographies. For example, let's say we have The Tune in Dan's Cafe and The Tune in Dan's Café. The software could show this as a single line using whichever name is the parent. When you click on the title it'll show the VT link as it does now. The same concept could be used to show Philip José Farmer as a single bibliography even though there may be non-variant titles entered under Philip Jose Farmer. --Marc Kupper|talk 06:13, 25 October 2010 (UTC)

(unindent) Regarding searching. There's a bug in advanced search in that it's not finding "Maelström." The left navbar search works and finds four titles. Ideally, both searches would also match titles containing "Maelstrom." --Marc Kupper|talk 05:27, 25 October 2010 (UTC)


As a side note, the ¨ mark which is the difference between "Malström" and "Malstrom" isn't diacritical (at least in meaning that it "changes the sound value of the letter to which they are added". In Scandinavian languages (and that exaple word comes from Norwegian, I believe) and in Finnish "a" and "ä" are two completedly separate letters. I personally think, that there should be variants if the difference is in the "umlaut" character, but in case "Cafe" and "café" the difference is so slight that the variants aren't necessary. Tpi 07:49, 30 October 2010 (UTC)

Promotional Excerpts

We have never really developed a consensus on "promotional excerpts". By that I mean excerpts, often a first chapter or section, from one book, printed at the back of another, often unrelated (except by publisher) book. Previous discussion, including this old R&S thread, seems to have leaned towards not including them, but practice seems to have been different. A basic search finds 1540 fiction titles containing "(excerpt)", and random sampling of the results of an advanced search suggests that half or more of these are promotional excerpts. At the moment entering or not entering these seems to be completely up to the individual editor, with the only standard being to append "(excerpt)" to the title and to class them as SHORTFICFTION if they are entered.

This was brought to my attention by this recent help desk thread. I think we should have a clearer consenses and a help page entry on this subject.

As I see it, the arguments for recording promotional excerpts are:

  1. It makes the record of the book more complete and more exactly matching the physical volume.
  2. Sometimes different printings and often different editions of a book will have different excerpts included.
  3. The excerpt can help date the book in which it appears (or at least its first printing) as excerpts are generally of recently published works.

And the arguments against recording them are:

  1. They are pure advertising matter, and we generally don't record simple ads for other books included in a book.
  2. Almost no one looks for or wants to find a promotional excerpt, or would care if it was present when buying a used book.
  3. They clutter the bibliographies, particularly of popular authors (whose biblos are already long)
  4. They are almost always of books already indexed in the database.

I am inclined a bit toward excluding them, although i have entered some in the past. But I don't feel strongly about the matter. I do feel that this is an area where greater uniformity of practice (particularly as regards modern MMPBs goes) would benefit us.

What are your views, please? ---DES Talk 16:00, 27 October 2010 (UTC)

I'm more on the for side even if the cluttering part is annoying. As a book collector, such points are useful to me for strict identification purposes. There is also sometimes interesting info in them, like some virtual vts, e.g. books that are excerpted and presented as forthcoming under a title that won't be used by the publisher (IIRC I entered at least one of those). Hervé Hauck 17:51, 27 October 2010 (UTC)
I'd recommend that:
  1. That they be included but entered as "Title (preview)" instead of "Title (excerpt)". This would distinguish them from "real" content excerpts.
  2. That "Title (preview)" records of the same work be merged together regardless if the excerpts are exactly identical.
This, in my opinion, provides the benefits of the above "for" arguments while still addressing the biggest "against" argument (to me at least) which is that they clutter the listing. --JLaTondre 23:10, 27 October 2010 (UTC)
My reaction echoes Hervé's. I can see the record of these previews sometimes providing useful information about titles, author name variations and/or pseudonyms, and perhaps even dates in some cases. If they're not recorded as content, there's pretty much no hope of finding the information by searching. If someone wants to bother recording them, clutter seems a small price to pay. I second the suggestion about merging these things together without worrying about whether the text is identical. --MartyD 23:31, 27 October 2010 (UTC)
If we merge them, won't we lose info on title and name variations, and dates, if these differ between different extracts/previews? Or would we only merge if author and title are identical, ignoring only possible text variants? (Such merges won't do that much to reduce clutter -- if every novel, much less every collection, in a popular author's bio has a matching preview, it could add a lot. Still merges would help. I'd love a separate title type for such things, but that won't happen right away, if at all.) I'm not objecting to this idea, merely making sure that if we adopt it it will serve the purposes intended reasonably well. ---DES Talk 00:03, 28 October 2010 (UTC)
Good detail point! Speaking for myself, I was supporting the notion of merging if title and author are identical without worrying about whether the text is the same. --MartyD 00:34, 28 October 2010 (UTC)
That makes sense, if we are to record such things in general. (and possibly set up VTs otherwise). Do others think such previews are worth recording at all? I really don't think that the merging will do very much to alleviate the biblio clutter they can cause in some cases. -DES Talk 01:22, 28 October 2010 (UTC)
I am generally in favor of including them and agree with what Marty and Hervé proposed above. I don't have a strong opinion re: merging excerpts, but usually err on the side of caution since I think I have seen one or two examples where a promotional excerpt was cut in half in a later printing to make room for another excerpt. It's hard to be sure unless you have both printings on hand.
Another caveat to keep in mind: It's not always easy to tell a "promotional preview" from an excerpt. Sometimes publishers keep reprinting the same pre-publication chapter as a "promotional excerpt" even after the book was published, so the same text serves in both capacities. Given this uncertainty, I think we are better off with "(Excerpt)", which is more generic, rather than "(Preview)". Ahasuerus 03:01, 28 October 2010 (UTC)
I think the desire of change of term was to distinguish not between pre- and post-publication excerpts, but to distinguish promotional excerpts from those cases where an excerpt from longer work is presented for itself, say in an anthology. Contrast such titles as (Excerpt) Mutant Hell, 20,000 Leagues Under the Sea (Excerpt), A Connecticut Yankee in King Arthur's Court (Excerpt), and A Modern Utopia (excerpt) with such entries as A Childe Cycle Concordance (Excerpt), A Company of Stars (excerpt), A Dark and Hungry God Arises (excerpt), and 1633 (excerpt). The promotional excerpt appears at the end of a novel or other book, is of a forthcoming or recently published work, has no connection with the book in which it appears except being by the same author, or perhaps merely the same publisher, and is often unpaginated. it is also not generally listed in a table of contents. It serves much the same function as the trailers that appear before a movie or video -- to advertise other somewhat similar products. The non-promotional or content excerpt appears in an anthology, or less often a collection, as one of the content items. While a promotional excerpt may not be of an unpublished book, it is intended to serve as a preview to a reader who has not (or may not have) read the book of which it is an excerpt, i think. So i think "preview" is a not inappropriate term. -DES Talk 03:28, 28 October 2010 (UTC)
The argument for merging (if I understand correctly) is that the only important thing to record is that there was an excerpt of a work by Title X and author Y in pub Z, and that we would merge even if we knew the texts to be different, so there is no need to check the text. -DES Talk 03:33, 28 October 2010 (UTC)
I like the idea of "preview" to separate them from excerpts though I suspect most of the time we'll be able to detect that something is a promotional preview as it'll be a shortfiction work at the end of a type novel publication. A new title type seems useful as these should not be in the "Short fiction" sections of author bibliographies. I'm fine with merging them. IF someone should discover there are two different previews they can add a title note about this to the merge record or to have two title records.
If we add a new title type I could see excluding them from publication displays and author bibliographies by default. They'd still be visible in edit mode. --Marc Kupper|talk 07:23, 29 October 2010 (UTC)
If there were to be a new title type, whether to display them could be a switch like the "concise" display now is, and the initial setting could be a user preference. The value of a different term would I suspect be mostly in author bibliographies, where the distinction now is not obvious unless you click through to the pub level. -DES Talk 14:40, 29 October 2010 (UTC)

Non-Bibliographic posting

A situation has arisen regarding a long-time internet book/magazine seller. He has decided to 'retire', at least from doing any business on the internet. His inventory, in it's entirety, is being offered at a substantial discount (40% off listed prices). This has nothing to do with this site's purpose, more to do with the probable fact that most of the contributors here are collectors in some fashion, certainly all are readers. This appears to be a one-off deal, not an 'occasional' sale event. I wouldn't want us to become salespersons in any way, shape or form. Nor should this site be used to promote things. Is it inappropriate to list the website address, even for a limited time, just so those that use the ISFDB can at least know of this? Would it be a better idea to have a dedicated wiki-page for such things? Thanks! --~ Bill, Bluesman 15:57, 28 October 2010 (UTC)

Don't keep me in suspenders. At least let me know who it is on my Talk. Thanks.--swfritter 19:22, 28 October 2010 (UTC)
Same here.--Rkihara 20:18, 28 October 2010 (UTC)
The website has been posted on both of the talk pages, as requested. The individuals can delete it at their leisure. Perhaps this is a good way to do this type of thing? Though, I think the Community Portal might be a better "notice board"? --~ Bill, Bluesman 22:45, 28 October 2010 (UTC)
I'm bummed. Jim Gardner is one of those old fashioned sellers that loves books. He is a walking encyclopedia that is particularly strong in cover artist identification. I tried a couple of times to see if he'd become an ISFDB editor. Maybe he'll now have the free time for this. --Marc Kupper|talk 07:10, 29 October 2010 (UTC)
I am thinking that the easiest way to handle this type of information would be to post the details on one's Talk page and them leave a brief message on the Community Portal. (BTW, in my experience 40% off cover price may or may not be a great deal on certain items. When buying in quantity, you can often get 40%+ off many used items.) Ahasuerus 06:58, 31 October 2010 (UTC)
Having the details on the user talk page seems reasonable. That 40% is off the seller's normal list prices and not 40% off the cover. In this case the list prices are rather high and even at 40% off is still higher than what I normally pay. My collection average is 65¢/book for specfict. --Marc Kupper|talk 06:24, 4 November 2010 (UTC)
Where do you buy your books? Over here prices for reasonable looking paperbacks start with €3.00 (around $4.00), if there's anything for sale. My first shipment from Jim Gardner arrived this week, and even with postage ($1.50 per book) I don't feel overcharged. --Willem H. 21:41, 4 November 2010 (UTC)
The curse of living in Europe. Not so the price but the simple avaliability for US editions. Hauck 18:52, 7 November 2010 (UTC)
I enjoy the chase and so get most of my books from thrift stores and friends of the library sales. The money goes to good causes in both cases. When I see a book I want to get I check AbeBooks to see how many copies are available. If there are more than a dozen or so I add it to the "long list" and wait until I spot a copy in a thrift store or library sale. If a book is hard to get then it's on the short list and I periodically check used book stores. Used book store prices are generally $2.75 on up around here. I order on-line if it looks like it'll be near impossible that I'll run across a copy by chance. The main public library for our region has a once-a-year discard. I could load up on thousands of specfict books at these events but restrict myself to collections and anthologies. We get a couple of new books each month too. --Marc Kupper|talk 09:03, 8 November 2010 (UTC)

Publication format: tp vs. ph

In the Help screen for Editing Publications, one of the Publication Format rules is:

tp - Trade paperback. Used for anything larger than a paperback. Thus any book that is at least 7.25 in (or 19 cm) tall, or at least 4.5 in (11.5 cm) wide/deep, and is not a hardback, is a tp.     (remainder of the rule omitted)

However, this statement would seem to incorporate many books that should properly be listed as "pamphlet", under the applicable rule slightly later. Thus I propose that the last phrase of this rule be changed to:

... and is not a hardback or pamphlet, is a tp. ...
Chavey 00:52, 4 November 2010 (UTC)
I agree. That help page should be changed. Perhaps even better: "tp - Trade paperback: Any [perfect-bound] softcover book that is not a pamphlet [saddle-stapled, side-stapled, or saddle-stitched] and is at least...[size limits]...should be entered as 'tp'." Mhhutchins 02:04, 4 November 2010 (UTC)
Chavey, I'm curious about "would seem to incorporate many books that should properly be listed as 'pamphlet'." Can you give some examples?
At present I don't see a need to change the help in this case. For example, neither of the sections for pb and tp mention the binding. The section for ph/pamphlet distinguishes these from both pb and tp in that they are short and by the binding used. I believe the assumption is that someone would read the entire help section and choose the Pub Format that seems to be the most reasonable fit rather than reading top-down and stopping at the first item that seems to fit. For example, some magazines are perfect bound and it's assumed an editor would make a best judgment choice from the magazine sub-section rather than entering "tp."
I'd like to keep the rules as accessible to to a lay person as possible and to not expect that editors need to identify the various forms of binding when in order to know which Pub Format to use. For example, the publishing trade considers "cardboard articles" to be paperback books and yet we classify them in ISFDB as hardcover as that's what the average member of the public would call these. --Marc Kupper|talk 05:43, 4 November 2010 (UTC)
I think it's wrong to assume that any person (lay or not) would continue to read once they've found the answer that satisfies their inquiry. If they read "Thus any [not my emphasis] book that is at least 7.25 in (or 19 cm) tall, or at least 4.5 in (11.5 cm) wide/deep, and is not a hardback, is a tp." they would have to conclude that a 8.5"x11", twelve-page, side-stapled book should be entered as "tp". They would feel no need to continue to read the remainder of the section. I think we should also add that both "tp" and "pb" are perfect bound. After all, the field is called "binding" and not "size".
About your example concerning magazine binding, I believe there are certain perfect bound magazines (that are not digest size) that should be entered as "tp" (think of the non-genre magazine Tin House). But "tp" is not among the choices under magazine bindings. Mhhutchins 17:32, 4 November 2010 (UTC)
Note that the field is called "Pub format" and not "binding" nor "size." Adding "perfect bound" as part of the definition for pb/tp makes me nervous as I have some paperbacks that are not perfect bound. See the decision tree below: --Marc Kupper|talk 08:49, 8 November 2010 (UTC)
You're correct. This must have changed without my noticing it. It was "Binding" for a long time. "Pub format" is more descriptive of the field. Thanks for bringing this to my attention. Also, I've never seen a paperback that is not perfect bound. Can you give me an example? Thanks. Mhhutchins 15:16, 8 November 2010 (UTC)
Since Marc asked, the item that occasioned my question was The Fiction of James Tiptree, Jr. by Gardner Dozois. I was verifying this book, and this field was missing (if I remember right; I don't know how to look back at what my submitted update was -- it was submission 1470396). The booklet is 36 pages, staple-bound. I thought I remembered something about a pamphlet format, so I went to the help pages on publication format. I read this rule, and it certainly seemed to be telling me that I was supposed to be entering "tp" as the publication format. As you suggest, I did continue reading and found the pamphlet entry. But then I sat and cogitated awhile trying to figure out how I was supposed to handle what appeared to be two conflicting rules. Eventually I decided I should enter it as "ph". But even then I wasn't sure if I had entered it right. It seems to me that having two rules that contradict each other is just a bad idea. Especially since it requires very few additional words to clarify the issue, and to send the reader of that rule to the appropriate section to consider. And since Marc asked for examples, the other case I ran across recently is the 12 page chapterbook The Tower, which also fits both the definition of "tp" and of "ph". Chavey 23:16, 4 November 2010 (UTC)
I agree that "ph" conflicts with "tp." I'd been resolving this with a rule of thumb of "is the spine flat?" I should add "does the cover stock wrap around the spine?" and perhaps "is the spine thick enough that there's room for the title?" If not, it's probably a pamphlet.
I'm not sure why we have "Digest" in the first list as that also conflicts with "tp." It appears the decision tree is:
  • Products sold or given away in book stores and conventions
    • Hardcover - this covers all sizes and types of hardcover publications.
    • Non-hardcover
      • Pamphlet - I believe anything that does not have a flat spine goes in here? Spiral bound, comb, VeloBind and ZIP bound? VeloBind is flat on the back and it's not "loose" binding. I have some VeloBind bound work and consider them pamphlets.
      • Non-Pamphlet
        • tp, pb which are size based.
    • dos - This seems to cover both hardcover and non-hardcover dos-a-dos.
    • ebook
    • audio
  • Periodicals sold in news stands or mailed
    • Digest, Bedsheet, Tabloid, A4, and A5 all of which are size based.
    • ebook
    • audio
I believe the "pb" in the magazine section is for those magazines packaged in paperback books with ISBNs. Sometimes they are called bookazines. The Perry Rhodan books from Ace would be an example of this. --Marc Kupper|talk 08:49, 8 November 2010 (UTC)
I don't believe the source of the publication should play a part in the decision. There are many books in my library that have never been on the shelf of a book store. Nor have I ever seen an ebook sold in book stores, convention huckster rooms, nor mailed. Also there have been several periodicals issued in paperbacks that are indistinguishable from paperbacks (think of the mid-60s format used by the British magazine Science Fantasy) Other than these qualms, I like your idea of a decision tree. Thanks. Mhhutchins 15:16, 8 November 2010 (UTC)
Assuming by "decision tree" that you mean something you progress through until you find the correct answer, wouldn't "dos" have to go above "Hardcover"? I.e., if we want someone to read until they find the description of what they're looking at, mightn't a laymen read to "hc" or "pb" and then stop without getting to "dos"? Of course we could put an EXCEPT: just before the "dos" section. Chavey 01:56, 9 November 2010 (UTC)
(unindent) Well, this is one of those threads that went on with lots of discussion, and then suddenly stopped without any apparent resolution. So, at the risk of being presumptuous, I have updated the Help screen for Publication Format to include what I think is the conclusion of the discussion above. I have gone for a minimalist approach, trying to make as few changes to the existing text as possible while still clarifying what the conventional wisdom seems to be. I have added a sentence to each of the "hc", "tp", and "pb" descriptors. At least for the next couple of days, these new sentences will be in italics, so others can identify them easily, change them if I've erred, etc. Chavey 00:29, 29 April 2011 (UTC)
Since it's been a week, and there were no comments, I removed the italic format that I had used to indicate the changes to the Help pages. These changes were made to the PubFormat Template, so they affect all of the Help screens that use that template (NewPub, NewNovel, EditPub, etc.). The specific changes made were:
(a) Under "hc" format, I added the sentence: Unlike other formats, this notation takes preference over "dos" for the case of hardcover dos-a-dos publications, such as Up to the Sky in Ships / In and Out of Quandry.
(b) Under "tp" format, I changed the first sentences to read: Used for anything larger than a paperback, except when the format "dos" or "ph" (pamphlet) applies, in which case those notations take precedence. Except for those cases any book that is at least ...
(c) Under "pb" format, I added the sentence: If a book qualifies as a "dos" or "ph" (pamphlet), those notations take precedence over "pb".
(d) Under "dos" format, I added the sentence: We do not use this format for hardcover dos-a-dos books. Chavey 03:53, 7 May 2011 (UTC)

Cover Photograph

If a cover is a photograph, and the photograph is credited to a photographer, does their name go into the "cover artist" field? The alternative would be to demote it and place it in the Notes field instead.
A Google search for "photograph" or "photographer" on the ISFDB site found nothing. Chavey 05:43, 9 November 2010 (UTC)

I'm not sure if there's anything in the help pages regarding this, so I'll just tell you how I do it. If it's only one photograph and the single photographer is credited I will put his name in the cover artist field. If it's a montage of photographs from different sources, many of which are credited to a company or stills archive, I will credit only the person who is given credit as the cover designer, as I feel that they were the person to actually create the cover. I also do this if no photographer is credited and only the cover designer is credited. Their name then goes into the cover artist field. If both a photographer and a cover designer are credited, I put the photographer's name in the cover artist field and give the designer credit in the note field. Mhhutchins 03:31, 10 November 2010 (UTC)

Complete the Story Contests and Serials

I've encountered a complete the story contest. The first time I ran across one of these, I entered both parts as short stories. However, it seems that perhaps a SERIAL type makes more sense. I don't see an issue with having different serial types by different authors (well, perhaps one, which I'll detail below). The first half of the story that prompted this is here and is currently listed as short fiction. The second half, appears in this fanzine, though I haven't entered it yet, while I'm figuring out how. Assuming that we agree these should be serials:

  1. If the complete story was never published as a whole, should we create a title record for the complete work with both authors' names? This is actually a larger question for serials in general. The help page suggests that serials should only be linked when the full work has been otherwise published. However, it doesn't explicitly state not to do so. As an experiment I made a parent title here where the full story hasn't been published together. It seems to work OK on the various pages. However, I'll happily undo it if this discussion goes that way.
  2. If the first part of the story has been reprinted in an anthology, as it has in this case, and we are certain it is identical (as I am in this case), should we add the serial to the anthology, i.e. have a serial that isn't in one of the periodical containers.
  3. In this case, my source for both fanzines where this appeared is The Best of the Baum Bugle 1965-1966 which I intend to enter as an anthology once I've got the individual issued entered. However, if we have part 1 and part 2 both as variants of the complete title, there will be two variants of the same title that appear in one anthology. I don't know if this will create a problem, since they will be different variants, but I know I've seen warnings when the same title appears twice in one container.

Does this sound like a workable way to enter these titles? Thanks. --Ron ~ RtraceTalk 13:28, 12 November 2010 (UTC)

Connecting linkages in collections

A not-uncommon structure in collections is to have a "linking story", split among several pieces, that is used to connect, or weave together, the various stories of the collection. How such split stories are handled does not seem to be specified in the Rules and Standards, and several different approaches occur in the database. For example, Fever Season (ed. by C. J. Cherryh) has the story Fever Season divided into 6 parts, which occur before and between the six stories of the anthology. These are listed in the Contents of the ISFDB entry as 6 different short stories: "Fever Season", "Fever Season (reprise) (#1)", ... "Fever Season (final reprise)", pretty much the way it is listed in the book's table of contents. On the other hand, consider The Starry Rift by James Tiptree. It has the 10 pp. story In the Great Central Library of Deneb University which is divided into three parts, each with different titles (but with no Table of Contents). We have 6 publications of The Starry Rift, none of which list these three story portions separately in the Contents. In fact: two of the publications (one verified) do not list the story at all; two of them list the story once in the Contents with the page on which the first part begins, and with no note about the other two parts; and two of them list the story once in the Contents and include a Note mentioning the continuation of that first part later in the book. I don't have enough samples to tell if one of these approaches is more common than another. So what's the right way to handle such a split story? Chavey 08:58, 13 November 2010 (UTC)

See Rules_and_standards_discussions#Entering_Braided_Stories for a discussion of this. As with many R&S discussions, this one did not reach a definitive conclusion. The last recommendation was to enter it as a single story and to explain it in the notes. --MartyD 11:33, 13 November 2010 (UTC)
Thanks for the references; I hadn't noticed that the "Braided Stories" discussion was what this topic was about. Since in both cases I mentioned, these "serial sub-stories" each constitute a single story, I'll go with this last recommendation. Chavey 16:55, 13 November 2010 (UTC)

Editors of non-genre magazine

(copied from the Moderator board)

I approved User:Chavey's submissions removing the Editors of Cosmos credits from the publications in http://www.isfdb.org/cgi-bin/pe.cgi?28211 as those just made the pub credits match the title credits. But now I'm having approver's remorse... It looks to me like this is a non-genre magazine, and the credits on the title records are in fact missing the Editors of Cosmos they should have. There's no entry for this magazine in either Wiki page. Can anyone shed light on the removal of the Editors credits from the title records? Thanks. --MartyD 13:05, 14 November 2010 (UTC)

If I remember correctly the "Editors of X" format was created because it was not always possible to identify the Editors of some non-genre magazines: but if the editors can be identified, then I see no reason to add "Editors of X" as well. We don't have "Editors of Omni" for instance. And although we have "Editors of Playboy", we also have some named Editors, e.g. this edition. BLongley 16:55, 14 November 2010 (UTC)
Well, the Help:Entering_non-genre_magazines explicitly states one should always use "Editor of X" and optionally enter any identified editors as co-editors (near the end of the help is also a mention of doing pseudonyms). I don't understand the reasons for (or benefits of) doing it that way for non-genre while not doing it for genre. But that's grist for yet another R&S discussion. I'm mostly trying to figure out whether accepting the removal of the "Editors of" credit from the Cosmos titles was intentional or accidental. I don't want to blindly undo something someone else is trying to do. --MartyD 17:12, 14 November 2010 (UTC)
I think Help_talk:Entering_non-genre_magazines#Editor_names was the last relevant discussion. But since we've started putting Magazine EDITOR records in series, my opinion is now "any of them will do, bar option 4" - the pseudonyms would still be a pain. BLongley 19:47, 14 November 2010 (UTC)
Checking Help, I see that it says, in part:
If the actual name of the editor (or editors) of a non-genre magazine is entered
(instead of using the "Editors of Magazine Name" format), it may be desirable to
make such a name a pseudonym of "Editors of Magazine Name". (see Help:How to record
a pseudonym for the procedure.) However, this cannot currently be undone, and if
it later proves that the person involved ought to have an ISFDB record in his or
her own right, an unfortunate and unfixable situation would occur. So it is probably
best not to create such pseudonyms without first discussing the matter, probably at
the Community Portal.
Alternatively the actual editor may be listed as a second editor (co-editor) along
with "Editors of Magazine Name"
This doesn't look right. First, we can now undo pseudonyms, so that whole section needs to be removed. Second, the pseudonymous approach looks unnecessarily complicated and is currently not used. The main reason behind using "Editors of ..." for non-genre magazines was to avoid creating a lot of "uncredited" records and Author records for people who had nothing to do with SF, e.g. various folks who edited The Saturday Evening Post, Rogue, etc. So if the editor is a known SF personality, we enter him/her explicitly; otherwise we use the "Editors of ..." convention.
I suggest we copy this discussion to R&S and clean up Help. Ahasuerus 20:10, 15 November 2010 (UTC)
P.S. For now, I have removed the comments about pseudonyms being unremovable so that we can concentrate on the applicable stuff. Ahasuerus 20:14, 15 November 2010 (UTC)

(unindent) The section in question currently reads:

Editor Pseudonyms
If the actual name of the editor (or editors) of a non-genre magazine is entered
(instead of using the "Editors of Magazine Name" format), it may be desirable to
make such a name a pseudonym of "Editors of Magazine Name". (see [[Help:How to
record a pseudonym]] for the procedure.)
Alternatively, the actual editor may be listed as a second editor (co-editor) along
with "Editors of Magazine Name".

My proposal would be to change these guidelines to simply state that if the editor was a SF personality -- which typically means that we have a record for him/her -- then enter the name as specified in the magazine (or in the secondary source if all you have is a secondary source.) If the editor is unknown or is not an SF personality, then use the "Editors of ..." convention.

Keep in mind that back in 2008 relatively few EDITOR records were in series. As of 2010-11-15, we have 3,884 EDITOR records linked to a series and only 1,190 EDITOR records that are not. As long as an EDITOR record is a part of a series, it's easy to see how it relates to other magazine Titles in the same series even if one/some of them are attributed to an SF personality and others are attributed to "Editors of...". Ahasuerus 21:14, 15 November 2010 (UTC)

I don't like the "SF personality" criterion. Why not just always use the editor(s)'s name(s) if known, otherwise use "Editors of Magazine Name" if unknown (instead of "uncredited") and do so for any magazine -- genre or non-genre? Nothing remotely subjective about that, and it's easy to follow. I understand the potential pitfall of having names that may conflict with those of SF personalities, but we have ways to cope with that. --MartyD 01:21, 16 November 2010 (UTC)
I agree. If it's known and obvious, the editor's name should be credited. For example: Alice K. Turner should be credited for the years she was Playboy's fiction editor (if I knew the exact dates of her tenure I would have already adjusted the "Editors of Playboy" record.) Mhhutchins 01:52, 16 November 2010 (UTC)
I agree too. Name if known, "Editors of" if not, and put them in series. BLongley 13:46, 16 November 2010 (UTC)
Well, it's workable, but keep in mind that we will eventually end up with a lot of non-genre editors in the database. Thousands of SF stories (especially ghost stories and "weird stories") were published in non-genre magazines in the late 19th-early 20th centuries and it's possible to identify many editors' names if someone wants to spend the time to do the necessary research. Do we really want to list the names of, say, most newspaper editors in London, Paris and San Francisco between 1850 and 1930? Ahasuerus 04:55, 18 November 2010 (UTC)
I'm assuming that we won't really make much effort to trace the editors, and will probably only add them if we have Primary sources. Since we've adopted "Editors of" I've given up trying to find Editors of Colliers and suchlike. BLongley 20:09, 18 November 2010 (UTC)
The help as it stands permits identifying and recording the actual editors even now, and it doesn't seem to have been much of a problem. I'm sure we could craft wording that suggests one go with "Editors of" and not put much effort into identifying editors of non-genre magazines (versus presenting use of "Editors of" as some sort of last-gasp, if-you-really-can't-figure-out-the-editor, at least for the in the non-genre case). --MartyD 22:45, 18 November 2010 (UTC)
In this case "be lazy about research" seems good. :-) BLongley

Magazine Series questions

I've just done a run through "Absolute Magnitude" and noticed that the first two issues were as "Harsh Mistress". They've been put into a different series. However, in 2002/3 the Magazine was actually "Absolute Magnitude & Aboriginal Science Fiction" and that has been put under the "Absolute Magnitude" series. I'm tempted to put "Harsh Mistress" into the same series but appreciate that searching for a "Harsh Mistress" series would then fail. Or we could put both and/or all three under a super-series like we do with "Astounding/Analog". They still get separate grids (if in the wrong order) but the linkage is clearer. Thoughts? BLongley 19:27, 16 November 2010 (UTC)

Also, the "Harsh Mistress" series has combined 1993 and 1994 - we haven't rounded up to more than one-year groups even for Astounding & Analog, so I'm tempted to split those down to the two years, even if there's only one issue in each. Again - thoughts? BLongley 19:27, 16 November 2010 (UTC)

I believe it's a good idea to split the two issues of "Harsh Mistress" into two editor records and place both editor records into the "Absolute Magnitude" magazine series, and As long as they're linkable through the magazine series, it's not necessary to have them combined into one editor record. Mhhutchins 22:14, 16 November 2010 (UTC)
OK, done. Still doesn't look quite right but I guess some Series numbering could fix that. Unless anyone knows Months of publication? BLongley 20:13, 17 November 2010 (UTC)
I fixed the order by placing an arbitrary month into the record for the Fall/Winter issue of Absolute Magnitude. Mhhutchins 20:41, 17 November 2010 (UTC)
Looks good to me, thanks! Of course, even entering accurate dates doesn't always help. See Black Gate where knowing the correct dates of a quarterly messes up the years. Not quite so bad there though. BLongley 00:33, 19 November 2010 (UTC)

E-books without ISBNs revisited

Our Rules of Acquisition currently state:

In ...

  • e-books with ISBNs (note: software support added in mid-2006)

Conversely, e-books without ISBNs are generally "out". It may have been a useful rule of thumb 3-5 years ago when e-publishing was still young, but we are running into more and more e-books without ISBNs, especially Kindle books, that complement otherwise eligible books with ISBNs. For example, I just added volume 1 in Thomas DePrima's "A Galaxy Unknown" series. It was published by lulu.com in paperback, so it's definitely "in". On the other hand, volume 2-4 in the series have only appeared as Kindle books and do no have ISBNs. Based on our current rules, they should be "out", which seems wrong as it will leave the author's bibliography incomplete. I have noticed quite a few similar cases over the last 12-24 months, especially with novelette and novella length works which may not be commercially viable otherwise.

Going forward, we can expect a significant number of e-books to drop ISBNs since they cost money. With Kindle books, we have "ASIN"s to uniquely identify them, so perhaps we could start by changing the Rules of Acquisition to include e-books with ASINs, with other formats to follow later when the dust settles?

Granted, I am not crazy about privileging Amazon this way, but they seem to have not only the lion's share of the market but also the most stable identifier/catalog ID. Do other e-books, e.g. iBooks, have anything similar? Ahasuerus 03:04, 3 December 2010 (UTC)

As Lulu is a vanity press I would not call publication there as evidence of "definitely 'in'." However, I do agree that eBooks with ASINs should be accepted as they seem to be stable. The real challenge is self-publication where Thomas DePrima is publishing via Lulu and Kindle. Apparently he does his own artwork too. --Marc Kupper|talk 09:46, 3 December 2010 (UTC)
Vanity published books were excluded in the 1990s, but the policy was changed to include them years ago -- see RoA -- so I don't think it's a problem any more. Ahasuerus 20:20, 3 December 2010 (UTC)
I don't enter Kindle books as I understand that you don't actually purchase them for certain. Remember when people's Orwell Kindle books disappeared? BLongley 16:59, 3 December 2010 (UTC)
That's one of the trickier things about e-books -- in many cases you buy a "license to read" rather than a physical item and the license can be revoked under certain circumstances. Still, I think that it's important to record the fact of electronic publication even if there is a (hopefully small) chance that the publication will be later voided. It's particularly important for books that haven't appeared otherwise and would be missing from the author's bibliography. (Parenthetically, I will note that in 2009 WSFS passed a constitutional amendment that added the words "or the equivalent in other media" to their eligibility rules.) Ahasuerus 20:20, 3 December 2010 (UTC)
I think "license to read" is not a "publication". I'm leaning towards e-pubs that will resist the failure of Amazon or others, but so far that is not convincing me toward Kindle. PDFs (another can of worms, but a bit more generally available) are a bit more stable, if they can be downloaded and kept beyond the demise of the publisher. BLongley 02:56, 4 December 2010 (UTC)
Well, if we had our druthers, I too would opt for a more permanent medium. However, we don't have much control over how SF is published and how people read it these days. If an author chooses to publish some of his or her works in paperback and some other works via Kindle, then the only two choices that we have are to (A) ignore the Kindle titles or (B) include them. If we ignore them, then, given Kindle's rapid rise, our author bibliographies will be increasingly incomplete and increasingly useless. A randomly picked example from the "Kindle Fantasy Bestsellers" list is J. R. Rain, who has self-published a few books in paperback, but most of his books are available exclusively on Kindle.
No offence meant towards J. R. Rain, but would we really miss those "publications"? BLongley 18:20, 4 December 2010 (UTC)
It's true that for a number of years the "Kindle only" world was mostly self-contained with a bit of overlap with the world of POD/vanity presses. I remember how surprised a certain "Kindle author" was when she suddenly broke into print and her book became moderately popular. Even when e-publications of mainstream authors became common, they were almost always accompanied by regular paper editions, so we could ignore "no-ISBN e-pubs" (if there were any) without creating gaps in Title biblios. However, over the last year I have seen more and more author-maintained bibliographies refer to "Kindle only" publications of novellas, especially in the world of paranormal romance. Of course, now that I need to come up with an example, I can't find one :-(
The other thing that got me thinking about this topic was this Usenet review of recently published SF books. Note that items 3-6 on the list are the "Kindle only" publications that started this discussion and that they are mixed with mainstream SF books. I then checked Amazon's list of "Kindle Fantasy Bestsellers" (mentioned earlier) and noticed that it includes both regular bestsellers (Martin, Tolkien, etc) and "Kindle only" books. Now, I don't know how they compile their bestseller lists (there are always ways to tweak them if you really want to), but it seems to suggest that there is more and more overlap between the two worlds.
Having said all that, if we decide to include e-books without ISBNs, we'll probably need some kind of threshold or criterion. I am thinking that something like "a stable ID" would be a bare minimum, but then who is to say what a "stable ID" is? Kindle uses ASINs, which are generally stable and Gutenberg IDs have been stable for many years, but what about other sources? Perhaps we could change Rules of Acquisitions to say "e-books with ISBNs or the following stable IDs: Gutenberg, Kindle, XYZ", and update the list as more "stable IDs" emerge? (BTW, it looks like Barnes&Noble assigns ISBNs to all (?) of its e-books, so we don't have to worry about them.) Or we can punt and wait another 6-12 months and see where the market goes. Ahasuerus 00:45, 5 December 2010 (UTC)
Now that I have been thinking about it for some time, it reminds me of the vanity press situation as it existed 10 years ago. As long as the world of commercially edited and produced books was separate from the world of vanity publishers, we could exclude the latter with impunity. However, with the rise of the Internet, POD, etc in the late 90s, more and more commercially published authors began exploring self-publishing options, especially if they were having trouble sustaining their careers in the traditional publishing world. By 2005 there was so much cross-pollination that we had to throw in the towel and let the barbarians in. Ahasuerus 04:26, 4 December 2010 (UTC)
I've seen authors offer to email people copies of their books. Usually for proof-reading or review purposes, but electronic distribution is still secondary in my view. Tying it to one particular publisher makes me even less keen. (I note that we still put a referrer ID into our Amazon links, does someone get a kickback from sales? I used to assume that Al funded the site from those.) BLongley 18:20, 4 December 2010 (UTC)
I know we have a referrer ID with Amazon, but I don't know what the extent of the relationship is. Amazon generally provides its data to the public (cover scans, the API, etc) in order to increase sales, so perhaps it's something that we needed to do in order to be able to link to them and still stay on their good side. There may be more to it, but I don't know the details. Ahasuerus 23:55, 4 December 2010 (UTC)
I think the referrer ID can also used for giving a small monetary kickback to the referring party.--swfritter 01:05, 5 December 2010 (UTC)
That's typically the case, but I don't know if we ever set up an arrangement with Amazon. Ahasuerus 01:14, 5 December 2010 (UTC)
Anyway, I'm not pushing for deletion, just pointing out that I won't be adding them. BLongley 18:20, 4 December 2010 (UTC)
I don't think it will be a priority, but it will be nice to be able to complete an author's bibliography without breaking the rules or running the risk of having the data deleted later. Ahasuerus 01:14, 5 December 2010 (UTC)
P.S. A recent example of non-Kindle e-publications being withdrawn is the removal of Poul Anderson's "The Escape" by Project Gutenberg -- see this Locus article, which provides further pointers. Ahasuerus 01:43, 4 December 2010 (UTC)
The Escape already documented. PG has apparently already used the original ID for another title. The copyright status of nearly all of the magazine derived SF at Project Gutenberg is now in question. PG is (was?) the gold standard for copyright verification and as such is the feeder to many other providers such as manybooks and librivox. Even if some of the titles are removed it is still valid to document them as having once been available.--swfritter 14:18, 4 December 2010 (UTC)
I think I once got shouted-down for suggesting we record publications for bad-ISBN books. (Or maybe it was just my suggestion to cross-reference such, I can't recall it exactly.) But I'm in favour of recording such references, whether it's a bad ISBN, ISSN, SFBC ID, BCA Catalog ID, whatever. BLongley 18:20, 4 December 2010 (UTC)
That sounds like Feature Request 3127708, "Add support for external identifiers", which was described a couple of months ago. Ahasuerus 01:22, 5 December 2010 (UTC)
A rule I have ignored for a long time especially if the author already has otherwise valid entries in the database. Henry Kuttner for instance.--swfritter 19:01, 3 December 2010 (UTC)


(unindent) I know it qualifies as "in" and thus moderately related to the main discussion but here's specfict driven down to $0.00 on Amazon - http://www.amazon.com/Tarzan-of-the-Apes-ebook/dp/B002RKT00C --Marc Kupper|talk 22:52, 4 December 2010 (UTC)

Yes, pricing is always an issue with e-books, perhaps more so with ISBN-free e-books. Once you no longer have a physical item in your hands which has a price printed on it, "publication price" or "list price" becomes harder to determine. Ahasuerus 00:53, 5 December 2010 (UTC)
Something I've been doing the past few months is if I'm looking at a publication record I'll check Amazon USA and Canada and use that to add a note that looks like
  • December 6, 2010: List prices reported as $7.99 on Amazon.com and C$9.99 on Amazon.ca.
I use that format for publications that I know state USA/Canadian prices. If I know it also states UK I'd add that to the report. If an item is no longer offered for sale (as a new from Amazon with a list price) on one of the Amazon sites I'll note that. Sometimes I'll note things like "only six copies left." Thus if I was dealing with the publication record for that Tarzan of the Apes kindle edition I'd have:
  • December 6, 2010: List price reported as $1.99 on Amazon.com. Amazon.ca and amazon.co.uk do not have a record for ASIN B002RKT00C.
I never record the sale price. The reason for this is that at some point I'll automate the tracking of the list price and the terms-of-service for the Amazon API disallows storing of the sale price (presumably so someone does not construct a site that does competitive analysis.)
The goal behind these notes is discovering that we need to add a new publication record as the price has changed but it'll allow us to better date bracket later printings. --Marc Kupper|talk 01:59, 7 December 2010 (UTC)
Keep in mind that the Amazon API that you mentioned above has been significantly curtailed over the last few months. There is quite a bit of uncertainty re: its future and the development community is mostly unhappy with what appear to be drastic and inexplicable changes. Caveat developer! Ahasuerus 00:08, 8 December 2010 (UTC)
Time to look at alternative sources maybe? BLongley 00:30, 8 December 2010 (UTC)
There are numerous sources of similar data, including Google Books, LibraryThing, COPAC, xISBN/ThingISBN and OpenLibrary. At some point I hope to teach Fixer how to use them, but it's time consuming and Amazon still (sort of) works for our purposes. I had to make adjustments to salvage Subjects after the last round of changes, though. Ahasuerus 00:56, 8 December 2010 (UTC)
I know when I was learning "Oracle Application Express" development, their Amazon shopping example was out of date within months. (Getting it to work again for my own purposes was a proof to myself that I could do these web-2.0 thingies.) There's lots of other projects out there that we could try and code for - and remember there are people that want to use our APIs too. I will get back into coding again sometime soon, if only to put something on my CV to show that I've been doing more than just manual bibliography since the last job ended. (Actually, I think I'll put my ISFDB experiences down as "user tuition, user interface design, problem-solving, data-cleansing, programming, analysis, facilitation, and all sorts of other good buzz-words. I've probably worked harder on ISFDB than on most of the jobs I used to get paid for.) BLongley 00:30, 8 December 2010 (UTC)

Entries for photographs in the database

I accepted an editor's submissions updating this pub in which many of the INTERIORART records are actually photographs. I'm not sure if there has even been a discussion of this matter, or if any consensus was arrived at. I personally feel that such entries would open a can of worms: what criteria is used to determine which photographs should be entered into the database, how should they be credited, do we merge records of identical photos, etc. Many of the photos are uncredited so having individual records seems pointless. Why not just note their existence in the pub's note field? Thanks. Mhhutchins 18:15, 3 December 2010 (UTC)

They may be uncredited but they are searchable by title. So if I want to find documentation of a picture, obituary, etc. of somebody searching by title is the only method.--swfritter 18:48, 3 December 2010 (UTC)
Hello, since I am the editor concerned, I'd like to add my point of view. I certainly wouldn't want to include all photos included in a publication, but in this case I think they are quite interesting, because they point out some highlights of Brian W. Aldiss' career & history and/or are artistically interesting. Maybe this could be a 'rule of thumb' for further decisions on the matter. Thanks. Stonecreek 15:03, 4 December 2010 (UTC)
If we choose to add photos there should be a set of standards established from the start. Perhaps they should be titled "Photo: <description>" to make searching easier. I personally can't imagine going back to add photo credits to any of the thousands of records that I've verified, but I'm not going to stop another editor from doing so. Mhhutchins 16:01, 4 December 2010 (UTC)
I don't think we need to define a standard. If I'm looking for a photo of Robin Straus I'll do an advanced search for photo AND Straus in the title and that'll find it regardless on if someone used "Photo of Brian W. Aldiss & Robin Straus", "Photo: Brian W. Aldiss & Robin Straus," or "Brian W. Aldiss & Robin Straus (photograph)." Thus, I'm fine with how Stonecreek entered the publication.
At present I see 33 titles with "(photo)" or "(photograph)" appended including two puzzles that are ESSAY rather than INTERIORART. Both of those are both verified by Mhhutchins.[22][23] Are they typos? There are two titles prefaced with "(Photographs)", 30 prefaced with "Photo:" and none with "Photos:", "Photograph:", or "Photographs:". I did not search for titles containing "\bphoto.*" or"^photo.*" --Marc Kupper|talk 02:58, 7 December 2010 (UTC)
Those two records are for photographic essays (by Brown and Clarke). They are not single photos but a series of photographs and titled as shown. Thus the reason for typing them as ESSAY and not INTERIORART. I've removed the appendages and added notes to the record explaining the nature of the essays. Puzzle solved. Thanks. Mhhutchins 03:13, 7 December 2010 (UTC)

When to include "About the Author"?

As best as I can tell, the rules say that I should always include an "About the Author" content item when it exists. But many of these author notes are very short, and have very little bibliographic, biographic, or SF data in them ("When not writing she enjoys cooking and gardening. She lives with two cats in central Ohio."). My impression, from many publications verified by long-term moderators, is that everyone ignores these "Abouts". Is that formal? Is there an informal cut-off for when we include such things as a separate content item? Chavey 15:19, 10 December 2010 (UTC)

I never knew that the rules insist that these bios be included. It's been my experience that it's pretty much up to the editor to determine how much data he wants to add to the record. That being said, I think most experienced editors, even those anal-retentive ones like me, know when to stop. If you personally feel a piece doesn't warrant recording, then don't include it. And don't be surprised if a later editor has a different level of tolerance for minutiae and wants to add more data. That's the nature of the beast when it comes to an open database. Just hope that future editors respect the verification etiquette and lets the primary verifier know they're updating the record. Mhhutchins 15:59, 10 December 2010 (UTC)
Back to the question: what criteria do I use when determining if bios should be included? 1) If written by a bylined third party, always, regardless of the length. 2) If uncredited, only if its length approaches the greater portion of a page or more. 3) If on the dustjacket, never. (Except for one rarity that I'm aware of.) Mhhutchins 16:05, 10 December 2010 (UTC)
Good question! Like Mike, I wasn't aware that such a rule existed. Can you point us to the Help Page that says that? I think it's more of a guideline than a rule. I agree with Mike on "bylined third party, always" but for "uncredited" I'd only include it if it contains something we record in the database for biographical data - birthplace or date, real name as opposed to pseudonym, for instance - or some other bit of trivia that I would consider worth putting on the author's wiki-page. (Husband and Wife relationships, for instance.) BLongley 19:45, 10 December 2010 (UTC)
The place where I read it is at NewPub, Contents, where under "include" it says:
Biographical sketches of authors. Include these when they are separate entries. Don't include them when they are part of a blurb or lead-in.
Those "About the Author" entries are invariable always separate items, so this seems to say (essentially) "always". Chavey 03:34, 11 December 2010 (UTC)

"Monkey: Folk Novel of China". In or Out?

In general, the ISFDB excludes "Fairy tales with no known author". The earliest Chinese novel for which the authorship is officially documented is Journey to the West, one of the Four Great Classical Novels of Chinese literature. There is, however, some substantial dispute over the authorship. Thirty chapters of the 100 chapters of this book have been translated into English as Monkey: Folk Novel of China by Arthur Waley. (An abridged version was also published for YA-age readers by Alison Waley, presumably related to Arthur.) The Wikipedia article on this abridged version describes the plot as follows: "The young monk Tripitaka ... encounters and frees the Monkey King, and he and Monkey thereafter recruit Pigsy and Sandy. They liberate a captive princess and punish her abductor, who has also murdered her father. The father is resurrected and reinstalled as king. They meet several bodhisattvas and fight fierce monsters, before finally arriving at Buddha's palace, where Tripitaka dies and become a spirit. Monkey, Pigsy, and Sandy are rewarded with immortality before they return to China with the scriptures." Pigsy is half pig/half human; Sandy is a semi-intelligent Water Buffalo. Pretty clearly fantasy by our standards. So, with respect to the Arthur Waley & Alison Waley versions of this story: In or Out? Chavey 22:01, 12 December 2010 (UTC)

Answered my own question. I just discovered that Monkey is already in here, under its short name, and listed under the original (presumed) author. Chavey 03:06, 13 December 2010 (UTC)

Dating on later printings

My understanding of the data entry rules is that for later printings of books, when the date is not specified (and not known from other sources), we are supposed to enter the pub date as "0000-00-00". But on several occasions I've run across books where, say, the first printing is Jan. 1987, so the date is entered as "1987-01-00", then the second printing is entered as "1987-01-02", the 5th printing is "1987-01-05", etc. This is generally accompanied with a note that says, more or less "Instead of doing the 0000 thing, I think this is a better way of doing it, 'cuz it gets the later printings in the right order." Well, right now I'm trying to enter a 6th printing of Chanur's Homecoming, and in addition to the 1st edition (with a true date), there are 6 other printings -- 5 of which have this alternate dating approach, while the 7th printing has (apparently) a true date. Not surprisingly, the 7th printing, with a 2001-04-00 date is listed well after an apparent 8th printing (or later) which is listed as 1987-01-10. I don't know who started this way of entering dates, but I know that three of these books are verified by editors with a lot more experience that me (4,000; 20,000; and 29,000 edits, vs. my piddling 832 edits). So should I: (a) ignore the oddity of those other edits, enter the 0000-00-00 date, and hence mess up the nice organization of all those other printings, with my 6th printing being out of sequence with the others; (b) follow their lead on this particular book, with a "1987-01-06" date, and ignore what seems to be the official rules, at least for this one book I'm entering; or (c) change all of those other dates to "0000-00-00", as the rules seem to imply, and risk getting several more senior editors annoyed at me? Chavey 20:24, 20 December 2010 (UTC)

Marc Kupper came up with this method of dating so that pubs would be more ordered in the listing. I think he's the only editor who ever did this. Personally, I don't like it for several obvious reasons. Foremost, because it defies the standards as established to enter data either stated in the publications or from secondary sources. It really doesn't even order that well when intermediate pubs are dated to the actual date of publication. For example if the first and third printings are dated 2000-01-01 and 2000-01-03, respectively, and the second printing is dated 2004-01-00 because there's a reliable secondary source that dates it to the actual date.
My suggestion in this instance is to follow the standards and enter your printing as 0000-00-00. I do not suggest changing another editor's verified records, but if you have the same printing, feel free to notify the editor that the record doesn't follow ISFDB standards. Hopefully, some decision can be made between the two of you. Also, the number of edits made by an editor isn't a measure of quality or reliability. Just take a look at my talk page and see the times I've had entry errors brought to my attention (though I would point out, that percentage-wise, I'm doing OK.) Mhhutchins 21:34, 20 December 2010 (UTC)
As luck would have it, adding printing numbers is on my list of things to do this week, although I need to get 5 minor changes tested and out of the way first. I am still not sure how we could best handle "actual printing numbers", which, as we all know, are not always the same as "stated printing numbers". Especially when certain, ahem, unnamed publishers are involved. Ahasuerus 22:50, 20 December 2010 (UTC)

Audio Book Bindings (again)

The discussion petered out last time but P-Brane seems to be volunteering to fix them. However, it still looks simpler to adjust help to match practice rather than vice versa. From last backup:

pub_ptype	count
audio CD	832
audio (MP3)	581
audio cassette	426
audio		159
audio MP3 CD	135
audio (CD)	49
Audio (CAS)	33

If DES's bracketed suggestion is preferred, can we get a script run to adjust them? I don't fancy moderating 1000+ frankly cosmetic edits - but if we are moving slowly towards "binding" being a drop-down selection-list then we need to look at fixing things like this. (I'm off now to fix the bindings of "8½" caused by the loss of inch details when re-editing publications.) BLongley 18:57, 23 December 2010 (UTC)

I saw P-Brane's submissions changing the form to "audio (CAS)" and was going to reject them until I saw this. I don't remember the discussion that made these the standard, but I can live with them. (I would not have known otherwise that (CAS) meant "cassette".) I agree this is basically cosmetic, but will moderate to accept such changes now that a standard has been established. I'm not much into audio anyway so I'll let those who are make the decisions. Mhhutchins 19:20, 23 December 2010 (UTC)
The only discussion I can recall is the one I linked to above. Looking at the Help Template edit history, it seems DES added the "standard" unilaterally. BLongley 20:49, 23 December 2010 (UTC)
I'm afraid I wouldn't know what "(CAS)" stood for either and I suspect that it wouldn't be a useful abbreviation for most users. Ahasuerus 20:57, 23 December 2010 (UTC)
We've got "Audio (LP)" in help as well, and I'm sure that that is going to take even more explanation to the MP3-only generation than to the CD generation. BLongley 21:18, 23 December 2010 (UTC)
Well, "LP" was a common abbreviation, so anyone who knows what an LP was will recognize the term. "CAS", on the other hand, was not a common abbreviation or at least not one that I recall running into at the time. Perhaps I was just out of the loop, though. Ahasuerus 01:51, 24 December 2010 (UTC)
Re: standardization, how many choices in the drop-down list will we need? About as many as Locus-1 uses plus a few more for different flavors of "audio"? Ahasuerus 20:57, 23 December 2010 (UTC)
I don't think we need more than about 40 bindings even if we separate all audio formats and epub types. I think we have 210 at the moment, including things like "large format tp" - which seems like a tautology by our definition. BLongley 21:18, 23 December 2010 (UTC)
40 is a lot for a drop-down list. How about we have 2 drop-down lists, one for common bindings and another one, next to it, for "additional binding"? The submission validation code that I am working on can check that only one of the drop-down lists was used. Ahasuerus 02:34, 24 December 2010 (UTC)
Can you do linked drop-downs? We really only need hc, tp, pb for most of our entries - add pamphlet, dos, ebook and audio to that list, with sub-options for the ebooks and audio-books in a second list. (Actually I can see a use for sub-options for "dos" too - Wildside doubles are tp rather than the pb for Ace doubles.) BLongley 17:22, 6 January 2011 (UTC)
As the one who accidentally instigated this latest flurry -- I was unaware of the previous discussion -- all I can say is we should either moderate to what the help calls for or change the help. We certainly shouldn't moderate contrary to what the help calls for.... I have no preference; the bracketed approach and the unbracketed/long-form approach both seem ok. I don't like the inconsistency with the handling of "ebook", but I realize that presents a broader multi-format problem than the audio recordings do. --MartyD 11:56, 24 December 2010 (UTC)

Personal tools