ISFDB:Community Portal/Archive/Archive06

This is an archive page for the Community Portal. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from

Archive Quick Links
Archives of old discussions from the Community Portal.

---

1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 · 31 · 32 · 33 · 34 · 35 · 36 · 37 · 38 · 39 · 40 · 41 · 42 · 43 · 44 · 45 · 46 · 47 · 48 · 49 · 50 · 51 · 52 · 53 · 54 · 55

---

Expanded archive listing

Variant Titles Again

Ok, this isn't to do with my Kuttner cock-up, but a more general moan on Collections. For editors, variants are a multi-step nuisance, and I'm now afraid to try some things that need doing. For instance: Robert Sheckley's "Can You Feel Anything When I Do This?", is an obvious variant of "The Same to You Doubled". Same with Varley's "The Persistence of Vision" and "In the Hall of the Martian Kings". "The Barbie Murders and Other Stories" is "Picnic on Nearside" but the publication is under the wrong title. There are (probably) inaccurate Novel versions of some collections. Cloning encourages people to leave misinformation in (e.g. how the author is credited in the actual publication, variant spellings are easier to leave uncorrected, etc). Before I mess anything more up - any suggestions on how to make this a more pleasurable experience? BLongley 15:06, 2 Feb 2007 (CST)

Moving a pub out from under a title it's tied to is the "unmerge" option that you have seen mentioned (and mourned) elsewhere on this page. We're all going to be glad when it comes back. Personally, I don't bother to do these, unless it's within the scope of something I'm working on in detail, such as Le Guin or Infinity Science Fiction. I keep notes on my user page of unmerges I need to do when the capability returns. The only way to do them without the tool is to reenter them as new publications, and then delete the old publications. For collections this means reentering and remerging each content item. Of course, once you have one successfully entered, the others can be cloned; but cloning assumes that the new pub is going to stay under the old title. That suggests a feature request, for "Clone this Pub to a New Title", if you think that's worth submitting. Mike Christie (talk) 22:32, 2 Feb 2007 (CST)

I added a note to "Feature:90097 Unmerge Facility" about a way to re-enable unmerge that should be safe.

Also, while a particular variant title relationship may be "obvious" it would be best if these could get documented in the title notes along with the source for your information. For example, with The Same to You Doubled is there a statement on the copyright page about Can You Feel Anything When I Do This? If so, then please add a note to the appropriate publication and/or title records that explain exactly where this statement was found and then quote the statement itself. Marc Kupper (talk) 02:57, 3 Feb 2007 (CST)

It's actually on the title page of The Same to You Doubled. I've added a note. BLongley 13:34, 3 Feb 2007 (CST)

I've checked Varley's "The Persistence of Vision" and "In the Hall of the Martian Kings" and there's no difference in content. However, neither refers to the other and both were originally published in 1978 so what's the guideline on choosing which is the variant? BLongley 10:23, 4 Feb 2007 (CST)

This area is somewhat subjective, but my general rule is that in the absence of any other reason to pick one title as canonical, the earlier title gets precedence. Per Nicholls that would be "The Persistence of Vision". Mike Christie (talk) 11:49, 4 Feb 2007 (CST)

One thing I HAVE done when there's only minor edits on a cloned collection is to mark the page-number as "NA" and add the corrected Author or Title as new under the correct page number. It's a great help to me when I revisit it to delete the incorrect attribution/title, but it would be far simpler to be able to edit the fields. What is/was the reason behind the edit rules? BLongley 15:06, 2 Feb 2007 (CST)

The problem it was intended to solve was twofold: duplicate data entry of content (where an existing version of the collection existed), and the requirement to merge each item after entry. The clone tool solves both of those problems perfectly: if you have an identical version of a collection already in the ISFDB (which is the most common case, though nowhere near universal), then cloning the pub and entering some page numbers and fixing the pub header can get you a twenty-story collection entered in under a minute, with no need to merge. However, having solved that problem, it revealed further requirements as you outline. So the answer really is that the requirements are still evolving as we understand our own process better. It is still in beta, after all. Mike Christie (talk) 22:32, 2 Feb 2007 (CST)

> What is/was the reason behind the edit rules?

A publication consists of a pub record and one or more pub_content records. The pub record contains the data you see in the upper part of the publication-edit page (from Title to Note) and each pub_content record contains a page-number field and a link to a title that you see in the contents. When you clone a publication, a copy of its pub and pub_content records is made and given to you to edit. You can safely make changes to the main publication data (Title to Note fields) as you are working with your own copy. You can also safely make changes to the page numbers as you are working with your own copy. The thing that is not safe to change is the title data that’s pointed to by the pub_content records.

Let’s say you have a publication that contains three stories. Story1, Story2, and Story3. You clone this and let’s assume the edit restrictions were not there and you change the title of Story2 to “The Story2.” You would discover that you would have also changed the story title in the original publication’s content and for that matter, in all publications that include that title in their contents. Basically, when you clone a publication the title records are already merged for you.

This is one of the biggest “gotchas” in ISFDB. For example, let’s say you see a collection that has Story1, Story2, and Story3. You look and realize that Story3 is entirely wrong and so you change its title and author. What someone may not realize is that now all of the collections/anthologies that used to list Story3 now instead list the new title/author you used.

Thus the rule is – if you need to change the title or author for something in the Contents section you first click on the title and see if it’s being used by any other publications. If not, then it’s safe to just edit the title. If there are any other publications listed you instead do an [Add Title] which gives you a dedicated (but blank) title record that you can fill in and would also do a remove-title to disconnect the original title you wanted to change from your publication. That original title will still be referenced by it’s other publications.

Note that a title’s Date, Type (novel, shortfiction, etc.) and Length fields are also shared between publications and when you change them the change will be visible in all publications that reference that title. Usually this is a non-issue as these fields are not as “mission critical” as the title and author fields.

There has been talk of allowing editors to make changes to the title/author fields (both in cloning and edit-pub). ISFDB would silently make a copy of the title record and apply your changes to that. Thus your change would do an automatic “unmerge” of that title. If you need to make a global change to the title that would be visible to all publications that reference it you’d just click on the title and do edit-title. Marc Kupper (talk) 02:32, 3 Feb 2007 (CST)

Collections problem

Another mystery: I want to add another publication to the already existing one for John Dickson Carr's The Burning Court. It was originally a 1937 hc novel. Right now in the DB there's a 1954 Bantam pb shown as the only entry. Putting in the 1937 hc is no problem. But my own personal copy of the book is in a 1959 hc collection called Three Detective Novels. I don't see how I can list this under the entry for The Burning Court -- right now, it looks as if the only info I can enter is for an entirely different title, ie, Three Detective Novels, which in turn will create a new book entry, not a new publication entry for an existing book. Hayford Peirce 17:49, 2 Feb 2007 (CST)

Add your book as a new-collection called "Three Detective Novels." In the contents you would add "The Burning Court" as one of the stories and once approved you merge "The Burning Court" with the record that's already in ISFDB. At that point the linkage from "The Burning Court" to your collection will work perfectly. Does the book bill itself as a "collection" or "omnibus?" I suspect either term would apply. I have a simliar issue coming up in that I have a book that's an omnibus of two previously published collections and includes 16 new stories. I have not decided yet if I want 18 items in the contents (the two original collections plus 16 new stories) or if I will list all of the short stories directly, or maybe even both, list all of the short stories with page numbers but also include the two collections without page numbers to show that the titles are related. The book clearly bills itself as containing the published collections. Marc Kupper (talk) 19:34, 2 Feb 2007 (CST)

Thanks for the above tips -- it appears to have worked perfectly. As to whether it's a "collection" or an "omnibus" I just don't know. I threw away the dustjacket 48 years ago, like a dope, and I don't know if there was any description there or not anyway. There's nothing inside the book. My own feeling is that if you republish 3 out of maybe 50 novels from a single author that is a "collection" of a small part of his work. I have a couple of Saint "Omnibus"s that, at the time they came out, tried to be more representative of his works, I would say, a mixture of novels, novellas, shorts. But, absent the publisher sticking one word or another on the book, I would say it's probably just a coin flip between the two words in most cases. Hayford Peirce 11:31, 4 Feb 2007 (CST)

Whose signature is this?

This is a bit of an off-the-wall request but I have a copy of of Klein's The Overlords of War (translated by John Brunner) that credits Karel Thole for the cover art but I think that's wrong as Thole did the first DAW printing and I have a 3rd printing with a different cover. TIA

Marc Kupper (talk) 19:42, 2 Feb 2007 (CST)

I pulled out the only Karel Thole cover I know I have, but it doesn't have this signature (or any other). Sorry, I don't recognize it. Mike Christie (talk) 08:18, 3 Feb 2007 (CST)

I wonder if perhaps it is Karel Thole as it does seem to have the letters T-h-o-l-e embedded within it. I know of one Thole cover that I have but like yours could not find a signature. Marc Kupper (talk) 03:46, 4 Feb 2007 (CST)

Subtitle rules

The help text said subtitles were optional, but I think various discussions have concluded that this is a bad idea. I've updated Template:TitleFields:Title and Template:PublicationFields:Title to assert that the subtitle should always be included; please revert those changes or edit them if there is more tweaking needed. Mike Christie (talk) 13:05, 27 Jan 2007 (CST)

I thought subtitles should not be included for titles and should be included for publications. I guess at this point this means that all publications should be matching their title record and we will be adding variant titles for every publisher permutation of the title? Marc Kupper 20:24, 27 Jan 2007 (CST)

I may have missed a recent discussion of this are, but how (and why) could we ensure that "that all publications should be matching their title record"? This is clearly impossible with omnibuses, anthologies and collections, so is the proposed rule only applicable to novels published as standalone novels? I can see that it could be a useful software-performed check that would display a warning when the two titles diverge, but why make it a hard rule? Ahasuerus 22:35, 27 Jan 2007 (CST)

Why is it impossible with omnibuses, anthologies and collections? They have titles, for example, The Faded Sun is the title of an omnibus and it's publications would be called The Faded Sun. Actually, if you look you will see both The Faded Sun and The Faded Sun Trilogy because the book has "Trilogy" on the cover but not the title page. Note that I'm writing about I'll call the "primary titles" which are the titles of publications and not the "content titles" where, yes, those title records will rarely match the names of their publications. Not impossible but rarely. :-)

I was wondering why it now seems to be a "hard rule" that subtitles are now to be part of the title record and also that minor differences in a title, such as adding a comma, also result us needing to add title record and variant titles (see Template:TitleFields:Title Novels). I'm getting concerned that the author bibliographies will have a bunch of variant titles simply because of minor variations in the titles from publisher to publisher (or as stories get exported across pond and people either add or remove the leading "The.") The old rule was we came up with a "canonical title" and did not pay all that much attention to minor variations in publication titles. In fact, the first line of both the title-title and pub-title templates support this. Marc Kupper 23:53, 27 Jan 2007 (CST)

Well, maybe I'm misremembering the discussion. I'm not asserting that the way it's currently worded is the way it has to be; I want to capture a consensus. But I do think what I put in is reasonable. Addressing specific points:

• "impossible with omnibuses, anthologies and collections"; like Marc, I'm not clear why that would be. Can you explain?

Sorry, folks, that was confusing :( I was referring to the impossibility of making Novel titles match Omnibus titles or making individual stories match the title of the Anthologies/Collections that they are published in. It is certainly possible to make Anthology Title_titles match Publication_titles etc. I'll comment further once I have some time to digest the rest of the thread. Ahasuerus 09:18, 29 Jan 2007 (CST)

• Marc, you comment that "the first line of both the title-title and pub-title templates support" the argument that minor variations in pub titles are unimportant; do you mean this sentence: "The title should appear exactly as published, even though this may be different from the canonical title"? If so, what I meant that to say was "If your title is different from the canonical one, don't change it to match -- different titles are recorded as different" which is in line with my changes to the help text.
• trivial differences in titles and forms of author names -- should they always require a separate title record? I think they should. One good reason is that it makes the rule absolutely unambiguous. Another is it encourages title-titles to match pub-titles, which I think is likely to be a plus for unambiguous identification of the primary title record for a pub. The last reason, and to me the main one, is that it accurately represents the data. I don't think the variants are harmful; and who knows, maybe some day a bibliographic researcher will want to look at US vs. UK title habits. Finally, I gave one example in the last discussion of a seemingly trivial name change that was quite significant: Le Guin published "Nine Lives" (I think it was) in Playboy as by "U. K. Le Guin" instead of her usual by-line of "Ursula K. Le Guin". This was at the editor's request, and she later figured out that it was to hide her gender. I think we ought to capture data, and let our users worry about interpretation.

Having said that, I acknowledge that the data on what pub used what title doesn't disappear, so long as we keep the pub records strictly accurate. So there is no loss of completeness; just a different approach to what's represented.

Another thought: I now think I recollect that we agreed that subtitles were to be used in full on pub records; but perhaps we did not agree for titles? Maybe that's the source of the disconnect here -- I was assuming that it was sufficiently valuable to keep the titles matching that it made no sense to introduce the change for pub records and not title records. I know that the fact that we have two fields for title, one on title and one on pub, means we can have a difference.

So maybe it comes down to: "What does the title record represent?" Some possible answers:

1. It represents the text, under any publication. Definitely not true, or we wouldn't use variant titles at all.
2. It represents the text, under exactly one title and author combination. That's the way the help files now read, since my recent edit.
3. It represents the text, under a set of titles that can be grouped as being "essentially the same" by human decision, where those titles are all by a single form of the given author(s) name(s). I think this is the interpretation if we drop subtitles from title records, but not from pubs.

But there are also title records put in as placeholders; for example, a title for a canonical name where the author never published that work under their canonical name. For example, I don't believe Moorcock ever published "Printer's Devil" under his own name, only as "Bill Barclay", but we will have to have a title record for "Moorcock/Printer's Devil" for that to show up correctly in his biblio. So the title field is a way to tie data together; interpretation is the job of display.

It occurred to me at one point that if the title and pub records really did always match, you wouldn't need both fields; just pick up the pub-title from the parent title record. That won't work for translations (under the current approach, anyway), or for magazines. Even if it did, that normalization brings risks with it; editing a pub title would then edit the titles of multiple other pubs. So the separation of the fields in the implemented database doesn't have to imply they have different functions.

That's enough for one post; I hope this is coherent -- I only had a couple of hours of sleep. (Got up at 01:30 to watch the Australian Open final.) Mike Christie (talk) 07:26, 28 Jan 2007 (CST)

One more specific scenario I thought of. If we have subtitles on pubs but not on title records, consider Le Guin's "Tehanu: The Last Book of Earthsea". There's an omnibus that contains it under the shortened title, "Tehanu". I don't have the pub to check if it's really listed this way in the omnibus, but let's assume it's actually "Tehanu: The Last Book of Earthsea" on the interior title page in that omnibus. (If it isn't, this same scenario could occur with other titles, so bear with me). That title is the same title record we use to record other pubs of Tehanu, so we either have a vt title record in this case, using the long form, or we use the shortened form of the title, thus losing the information that this pub does use the long form of the title. I think this is a good argument for using subtitles in both places. Mike Christie (talk) 09:09, 28 Jan 2007 (CST)

Author Bios

Bringing the discussion here as a central place; see Talk:ISFDB FAQ for Kathryn Cramer's original post on this topic.

Al posed some questions over there:

"We should discuss how the ISFDB should link to the author bio's (new link like the current wikipedia link? Replacement of the wikipedia link? Automated link generation as with the current author bibliographic discussion pages?), as well as what changes should be made to the Main Page such that it appears more... encyclopedic, and less as merely a wiki to support the bibliographic efforts."

Some suggestions and more questions:

• A new link for the bio; "Author Bio" instead of "Bibliographic Comments". We need the biblio page for organizing information about given authors.
• Automatic link generation is good -- no reason it can't be linked from the author display. Doesn't need a field in the table; it's going to be a standard name.
• The main page is easy to change; just wait till you've made the relevant changes elsewhere and then it will be easier to see what changes to do there.
• The FAQ will have to change, along with Help:Contents/Purpose.
• What do we do about verifiability, sources, and conflicts? One of the benefits of having people like Kathryn contributing information about authors they've known is that they know stuff that is unverifiable. I'm not concerned about what Kathryn posts, but what if someone else posts negative information about someone -- do we require some kind of verification for that? I seem to recall this happened some time last year, and would presumably recur if we open up to bios. I don't quite know what the right answer is here. Maybe we should just open for business, and hope that interested editors will help manage the articles. Is there any liability for TAMU on libel, if incorrect information is posted, Al? Mike Christie (talk) 16:28, 27 Jan 2007 (CST)

I am sure we can handle technical questions, but I suspect that first we want to make sure we fully understand what we are trying to accomplish and why the current solution is inadequate. It sounds like there is a number of different, although perhaps overlapping, concerns with farming biographical data out to Wikipedia. Let me see if I understand them correctly.

First, we have Kathryn Cramer's original post re: finding a home for original articles, personal accounts, anecdotes, etc -- which Wikipedia can not and will not take -- in the ISFDB. As Mike pointed out in his response, many of us "would like to have a place to read that material". However, there are already Web sites that cover this ground, e.g. http://www.fanac.org/ . We may want to investigate what kind of information they have, what their plans and resources are, etc and then either determine whether we want to do more than they are already doing or whether we want to partner with them.

If we do decide to provide an ISFBD (or perhaps an ISFDB-affiliated) place for original articles, then we will next need to consider whether a Wiki is the best way to make them available. Wikis are great when many people are collaborating on the same project and are constantly interacting with each other and supplementing each other's information. 5 different users may edit the DAW list or the Le Guin Project page in any given month, and there will likely be few problems with that.

On the other hand, if somebody like Teresa Nielsen Hayden were to provide us with a memoir of Roger Elwood, would we really want to make the resulting article editable by any ISFDB user? Or would we want to make it only editable by the original author and the ISFDB administrators so that any other original articles (by other hands) about Elwood would be separate documents/pages, all of them linked from a master "Biography:Roger Elwood" page? I think we would want to go with the latter approach, which should help minimize revert wars and contributor frustration.

Regardless of which approach we choose, publishing original articles/anecdoted would necessitate another departure from Wikipedia standards, namely contributor anonymity. Anonymous contributors work fine as long as the data that they supply is independently verifiable, e.g. the fact that the third edition of book X has N pages. It doesn't work nearly as well for original articles for obvious reasons. (Have I ever mentioned that I once saw science fiction writers N and P sacrifice a small animal to Cthulhu?) And once we do away with anonymity, we will also need to have a way to confirm contributor identity, which is to say a way to prove that User:Hayford_Peirce is the same person as Hayford Peirce the science fiction/mystery writer. I think this would be a fairly strong argument in favor of moving the Bio project to its own subspace with its own user registration scheme. It may still leave the ISFDB open to lawsuits from various corners, from vanity publishers to Scientologists, though. More research would be presumably called for.

Next we have the two possible issues that Marc raised: data duplication vis a vis Wikipedia and diluting genre boundaries. We may be able to find a reasonable compromise in this area, though. For example, Wikipedia's article on Robert Conquest concentrates on his government career and his subsequent work as a historian. A similar ISFDB article, on the other hand, would concentrate on his sf novels/poems and his work as an SF editor in the 1960s. We could cover his views of the New Wave (see Reginald-2) and other SFnal areas that may not be considered worth covering according to Wikipedia standards, but would be of interest to genre readers, fans, and historians.

Which brings us to Al's comment about Wikipedia's standards for inclusion being too restrictive for our purposes. This is certainly a good point and the primary reason why there are detailed Wikis outside of Wikipedia for everything from World of Warcraft to Star Wars.

Finally, Al's other comments about adding "critical evaluations of a work's impact on the field" to the proposed Bio project and covering non-biographical tpics like Terraforming open a whole different can of worms and I'll need to think more about them before I can comment in a meaningful way. One thing that immediately comes to mind, though, is that it would be likely very difficult to support "critical evaluations" in a Wiki. No critic worth his salt will agree with any other critic 100% of the time, so we would probably have to treat "critical evaluations" as I suggested we treat "original articles" above. Ahasuerus 01:18, 28 Jan 2007 (CST)

Some good points in that. I think issues about standards of inclusion and genre boundaries can be addressed via policy. I think the issue of whether a Wiki is the right technical mechanism is also a good one, but is secondary to the question of contributor anonymity, which had occurred to me as an issue too. I hadn't thought through the implications as thoroughly as you have done; I think now that this is the key point. If Kathryn Cramer contributes an article, the main thing to verify is that the author is Kathryn Cramer, not that the article can be verified from independent sources, because the value of the information is derived from whatever authority the writer has on the topic. That in turn implies that the appropriate editing is not Wikipedia-like open editing, but the editorship of a single central editor (or perhaps a small group) -- the editing work would more closely resemble what is done with text in a publishing house than what happens in an open collaborative project.

Open collaboration could still work for information verifiable from third party sources, and our standard for notability would be more inclusive than Wikipedia's. But that's not what Kathryn is asking for, it seems to me.

I'd like to hear from Kathryn, or others who might supply text for bios, how they feel the editing/verification issues mentioned above should be handled. Mike Christie (talk) 07:38, 28 Jan 2007 (CST)

One last remark on this topic: while I don't have a problem with putting biographic material here, we should not bring existing Wikipedia articles here. The Wikipedia material is licensed by the GNU Free Documentation License, while the ISFDB content is licensed by a Creative Commons Attribution License. These licenses are not compatible, and the ISFDB will not move to the GFDL. The GFDL is more restrictive than the Creative Commons license, in that the GFDL does not allow Wikipedia material to be brought into the ISFDB without changing the ISFDB license to the GFDL, while the Creative Commons license allows Wikipedia to use our material without changing their license. The original authors of Wikipedia articles are naturally allowed to bring their material here, as they have licensed the material to Wikipedia, and still retain copyright.

Yeah, but how much material of any Wikipedia article actually *belongs* to any one contributor? For instance, I did all of the original work on fairly long articles about Bill Tilden and Pancho Gonzales. Over the years many edits and changes have been made by various editors, and some new material has been added by others. But let's say that at least 80% of the articles as they exist today is *my* material in its original form. Do I actually have copyright on that? Suppose Mike Christie goes into it and adds a single comma? Then what? Does he have copyright to the comma? A better example might be the John W. Campbell, Jr. article. Both Mike and I have contributed to that and so have a bunch of other people.... This looks like a real can of worms to me. And, of course, as someone notes above, how do you even know that I, Hayford Peirce, am the s.f.-writing Hayford Peirce? I might be an escaped character from a John Barth story.... Hayford Peirce 12:10, 3 Feb 2007 (CST)

You're right, but practically speaking I don't think it'll be a problem. If I wanted to lift the Campbell stuff, I'd just find a diff that showed a version that was essentially all yours and my text, and use that. I don't think Wikipedia is going to get sniffy about that. I think Al was more saying "Don't go and grab all the bios from WP regardless of authorship". For the Campbell case I think we wrote pretty much the whole thing, so it would probably be OK, but for most others we shouldn't take stuff we didn't write. As for you not being Hayford Peirce . . . if we never rely on your identity as an authority for our data, does it matter? If we do need to rely on it, we need some verification, of course. That's one of the points raised above. Mike Christie (talk) 12:25, 3 Feb 2007 (CST)

Ah. This whole Wikipedia thing is very confusing in some ways.... But, oh well, as you say, almost all of this is entirely moot, as is the necessity of verifying my identity. How would that be done, by the way? Is there a PayPal equivalent of verifying identities? Hayford Peirce 13:07, 3 Feb 2007 (CST)

I don't think there's a simple, foolproof solution. Practically speaking, the easiest thing would be if a writer has a website that is clearly maintained by them, we could accept a correspondence with that website (e.g via email, or a note posted on the website) as evidence of identity of an ISFDB editor. Other than that, I don't quite know what we'd do. If people start editing the bios, no doubt the question will come up. Mike Christie (talk) 13:33, 3 Feb 2007 (CST)

I think the key is that the SF field has its own sources of infield expertise and regardless of the process arrived at, I think anyone who is going to edit a bio at ISFDB has more basic common sense about what is and is not important in th SF field than the random officious wikipedia editor.

I am easy about a number of the specifics, but the ISFDB is just a much more sensible forum for bios. One doesn't really want or need for an author bio associated with a bibliographic database to be The Unauthorized Biography (ala the Harlan Ellison Wikipedia bio). What I'm looking for is not perfection, just a larger helping of common sense. --Kathryn Cramer 17:51, 9 Feb 2007 (CST)

If there exist copies of what bios were in the ISFDB before bios were moved to Wikipedia, then that could be used. The current problem I have there is that I did some work on editor bios, which brought the attention of some of WIkipedia's more authoritarian editors who note that editorial bios (which probably came from the ISFDB or a similar source) are "inadequately sourced". Now, material on authors is much more readily available in general than material on editors. While our house technically does contain most of the print materials to do that, we don't, for example, have our issues of Locus all together and in order. The editor bios may get radically cut or dropped from Wikipedia now that they have attracted the attention of the "deletionist" types.

Kathryn

Now we have the bio links in place from the ISFDB, I think bio material will eventually come back. Up to Al if he wants to retrieve old bio information and use that. At the moment, though, I think we have a group of editors who are more bibliographers than biographers, so it may be a while till we see activity in that area. Mike Christie (talk) 14:32, 23 Feb 2007 (CST)

Last I checked, the original author bios were still in the database, just dormant. It should be easy to export them to the Wiki if that's what we want to do.

As far as Mike's comment that "we have a group of editors who are more bibliographers than biographers" goes, well, it depends. At one point I wrote quite a few biographies of various European Social Democrats and Communists for Wikipedia, from relatively short pieces about obscure folks like Vladimir Posse to longer articles about major figures like Arvo Tuominen to a 100Kb biography of Leon Trotsky. I also updated numerous SF articles on Wikipedia (Simak, Laumer, etc).

However, the ISFDB has a long history of concentrating on bibliographies and that's what I have been working on here for years, on and off. One of these days, once the biblios are more or less stable and self-sustained, I may concentrate on biographies, but I suspect it will be a while. On the other hand, we see new editors join all the time and who knows what their priorities may be? Ahasuerus 18:44, 23 Feb 2007 (CST)

Title Unmerge Returns From the Dead

A redesigned title unmerge tool is now online. This time you select which publications you want to unmerge from the merged title. Normally the name of the new title will be the same as the publication that was split off; if the title in question is a work of shortfiction, the name of the new title will be the same as the original title (if we need more exceptions like essays, interior art, etc let me know). The moderator screen will show what the new title will be for each selected work, so there shouldn't be any surprises. Any awards associated with the title stays with the original title. Alvonruff 05:34, 3 Feb 2007 (CST)

Great news; I'm sure everyone will be happy to hear about this. It's going to get stress-tested pretty quickly! Mike Christie (talk) 08:27, 3 Feb 2007 (CST)

Seems to be working fine right now. <and there was much rejoicing!> Ahasuerus 22:56, 4 Feb 2007 (CST)

I've been really happy with being able to unmerge. Thank you! There is one type of unmerge I have not tried yet in that I was on a parent title and the unmerge offers me the publications listed in both that title and in it's the variant titles. Presumably unmerge will still be safe should I choose a publication that's the only one listed for a variant title or to unmerge a publication that does not have a title reference (title-type does not match). Marc Kupper (talk) 18:33, 6 Feb 2007 (CST)

Moderator Help

I've created Help:Screen:Moderator to help moderator consistency. Please edit as necessary to make this more accurate -- both editors and moderators, please, since editors will probably have opinions on how moderators should interact with them over possible errors. I know there are some submission types missing; please add at will, or I will as I get time. Mike Christie (talk) 13:08, 3 Feb 2007 (CST)

Nominating Unapersson for moderatorship

I'd like to nominate Unapersson for moderatorship. We have no defined process for this, but rather than define one up front, I thought it would be best to learn by doing. Below is a short nomination paragraph, with links to contribs, and Unapersson's edit count. After that I've put some notes about consensus.

Nomination statement for Unapersson

I nominate Unapersson (talk • contribs) for moderatorship; he has accepted the nomination. With 911 edits, Unapersson has been the most active of the new editors since the beta went live, but has been so accurate that there have been few issues for review on the talk page. He is based in the UK (see this diff, and his website), and that is an additional advantage as it will give the UK-based editors a better chance of having a moderator awake when they are submitting edits. The backlog of unapproved edits is occasionally getting quite sizeable, and with Al focused on the code, and work needed on the Wiki as well, the other moderators could use more assistance. Mike Christie (talk) 14:10, 3 Feb 2007 (CST)

Process

Some initial thoughts:

• A nominee has to be asked (on their talk page), and accept the nomination, before the nomination can be made.
• Editors and moderators alike, please comment if you have an opinion.
• Start your comment with either Support, Oppose, or Comment. Unless anyone objects, I will regularize comments to this format to make them easier to read.
• To avoid a lot of indentation, please put your comments in consecutive bullets, rather than indenting replies. If you want to reply to a comment, then indent below that bullet.
• The model I am using (just for format) is the Wikipedia Request for Adminship; take a look at this page. We don't want a big hairy process like that; I just mean look at the bullet format of the comments.
• Moderator comments and editor comments carry equal weight.
• This isn't a vote; the bureaucrat who closes the nomination uses their judgement based on the comments, and can go against the consensus if they have a strong enough reason. They must give their reasons. (For those who don't know, a "bureaucrat" is a user who has the ability to make an editor into a moderator. Currently we have four bureaucrats: Al, Ahasuerus, Grendelkhan, and myself.)
• Any bureaucrat can close a nomination.
• The nomination should be closed after five days at most. A bureaucrat can close a nomination early if they feel the outcome is unlikely to change -- either for or against.

All this is straw man, of course, so please suggest changes too; but in order to avoid getting tied up in process discussions I am being bold and assuming this is close enough to a good process to be able to start now.

Discussion

Support

• Support, as nominator. Mike Christie (talk) 14:10, 3 Feb 2007 (CST)
• Support, as nominator. Marc Kupper (talk) 14:50, 3 Feb 2007 (CST)
• Support looks a good choice to me. BLongley 16:26, 3 Feb 2007 (CST)
• Support. I was going to nominate as well, since roughly 1000 contributions seems like more than enough to show both commitment and knowledge of most of the editing types. Alvonruff 18:10, 3 Feb 2007 (CST)
• Support as per the reasons given by the nominators. I am sure new moderators will make their share of technical mistakes -- as we all have and still do -- so one thing I would suggest is being extra careful the first few weeks. You will be likely running into various types of submissions that you have not had a chance to submit as an editor, so if you see anything new or unusual, just put it on hold and ask around :-) Ahasuerus 01:32, 4 Feb 2007 (CST)
• Support Excellent. think it would be v. useful to have someone actively moderating on this side of the pond - as noted above, the time difference can be a problem for you all , I imagine. Seems a knowledgeable individual , with good website.Thomas conneely 17:44, 5 Feb 2007 (CST)

Comments

• I hope candidates are asked first if they were ok with the role and given a chance to look over the moderator help pages, etc. before finding out via the public forum that they are being nominated. Marc Kupper (talk) 14:50, 3 Feb 2007 (CST)

You're right, that's a necessary step. I did ask: see User talk:Unapersson#Nomination for moderatorship, but it needs to be part of the process statement. I've made the change, above. Mike Christie (talk) 15:36, 3 Feb 2007 (CST)

• I see that on the Wikipedia discussion page for Admins. there is also a Neutral vote allowed. Should there be one here also? I suppose that anyone who really is neutral about a proposed moderatorship would probably have a Comment as to why he/she was neutral but I don't really know.... Am I just being a Completist about this? Hayford Peirce 17:10, 3 Feb 2007 (CST)

I made it "Comment" thinking it was a more general category than "Neutral". We can change it to Neutral if there's a need for it, or add a "Neutral" list; I think that might be overkill this early on, though. Mike Christie (talk) 07:35, 4 Feb 2007 (CST)

How about both Neutral and Comments sections? Had there been just a Neutral section I would not have commented and I assume there are people who did not have an opinion one way or another that could have felt invited to partipate had there been a Neutral section. Marc Kupper (talk) 18:28, 6 Feb 2007 (CST)

• I see that Unapersson is not on Special:Listadmins yet. Is there a specific time-frame and/or conclusion for these nominations? Marc Kupper (talk) 18:28, 6 Feb 2007 (CST)

Well, it's pretty clear that Unapersson is going to be a sysadmin, but I was going to wait the five days just for form's sake. However, now you mention it, it does seem unnecessary to wait. -*ZAP*- Unapersson is now a moderator. I'll leave a note for him in a bit; right now I have to run. Mike Christie (talk) 18:59, 6 Feb 2007 (CST)

I have left Unapersson a note informing him. I'll create a process page based on this to define moderator nomination process when I next get time. Mike Christie (talk) 22:06, 6 Feb 2007 (CST)

• This is just a comment on the process, not on Unapersson: but I see that "number of edits" has been mentioned as a positive point more than once. I seem to have overtaken Unapersson on that score, and do NOT feel in any way more qualified than him, so can we reduce the importance of that? Or make it clearer what type of edits are being done - someone that can edit collections, omnibuses, magazines and translations, decipher foreign languages, find cover-art images, et cetera, seems to be more rounded overall and in better shape to moderate than others (like me) that just pick up their books and go for "Add Pub" and "Verify"? (OK, I'm trying to learn all the other skills, but the number of questions and "Help!" posts here show I'm several months away from even considering such a role, unless we get specialist Moderator levels that can 1: approve the obvious things and 2: deal with the unobvious things. BLongley 17:44, 16 Feb 2007 (CST)

Link problem in database?

A search on title for Artemis, brings up among

Artemis, Spring 2002 	EDITOR 	2002 	Ian Randal Strock
Artemis, Summer 2001 	EDITOR 	2001 	Ian Randal Strock
Artemis, Summer 2002 	EDITOR 	2002 	Ian Randal Strock
Cover: Artemis, Autumn 2000 	COVERART 	2000 	Bob Eggleton
Cover: Artemis, Spring 2000 	COVERART 	2000 	Alan Bean
Cover: Artemis, Spring 2002 	COVERART 	2002 	David Menehan
Cover: Artemis, Summer 2000 	COVERART 	2000 	Randy Asplund-Faith
Cover: Artemis, Summer 2000 	COVERART 	2000 	Randy Asplund
Cover: Artemis, Summer 2001 	COVERART 	2001 	Bob Eggleton
Cover: Artemis, Summer 2002 	COVERART 	2002 	H. Ed Cox

Clicking on Cover: Artemis, Summer 2000 brings up the entry that cover, clicking on the Publications entry for that magazine brings it up. Clicking on Editors: Ian Randal Strock in that entry brings up his Summary Bibliography, which doesn't list Artemis Summer 2000.

This may be related to the other problem just above but in case it isn't.

—The preceding unsigned comment was added by Dcarson (talk • contribs)}.

I fixed it from the publication by doing an edit-pub and adding an EDITOR title entry. Did you by chance initially create this publication as something other than a New-Magazine and then convert it to type magazine? Marc Kupper (talk) 04:02, 4 Feb 2007 (CST)

In ISFDB1 there was a separate table (file) for EDITOR data. This means that it was often forgotten (which is why it is now autogenerated). When ISFDB1 was converted to ISFDB2, the entries that existed were translated into titles of type EDITOR. There are then numerous magazines without EDITOR titles. Alvonruff 04:33, 4 Feb 2007 (CST)

That one and a couple of others were already in ISFDB. I'm entering the rest and adding things like page numbers to the existing ones. The others seem to have the same problem. How do I add an EDITOR title entry? When I look at the content type menu I don't see that as a choice. Dana Carson 06:24, 4 Feb 2007 (CST)

When you enter a magazine, the EDITOR record is created behind the scenes from the name you put into the "Editor" field. You shouldn't need to enter it yourself. Mike Christie (talk) 08:56, 4 Feb 2007 (CST)

Rfom what Alvonruff these are probably from the first ISFDB and are missing those records. The ones I entered are fine. I'd like to fix the messed up ones. Dana Carson 14:54, 4 Feb 2007 (CST)

You fix the messed up ones by doing edit-publication on the magazine, click [Add Title], and adding a new title record to the contents where

• Page: Leave it blank
• Title: Match the name of the publication/magazine
• Date: Leave it blank (the date will be copied from the magazine's date)
• Type: EDITOR
• Storylen: Leave it blank
• Editor: Credit the magazine's editor

I was in a hurry to get out the door when I wrote my first response but another way the linkage can break break is if you change the tag field value. I just ran a test with this though and it did not break my test magazine but recall doing this with a novel and things broke. In that case the novel had been entered with a completely wrong title (it was one of those books where the series name was displayed in a larger font than the title and someone assumed the series name was the title). I tried to correct the tag to match the corrected title but discovered that broke things and so I cloned the novel to generate a record with a new-correct tag and deleted the original publication. Marc Kupper (talk) 15:27, 4 Feb 2007 (CST)

OK, that looks like it should do it. Problem was that on existing titles there is no choice of EDITOR so I didn't know where to make an EDITOR record. —The preceding unsigned comment was added by Dcarson (talk • contribs).

You are right - there's already a feature-request/discussion about that for existing titles EDITOR is not an option while it is available if you click [Add Title]. I believe the plan is that EDITOR will be available as a choice for both when it's a magazine and will be removed as a choice when the publication is not a magazine. It does mean that if you change the type to/from magazine that you would first need to get that change submitted/approved before the EDITOR choice became available or was removed. Marc Kupper (talk) 13:58, 5 Feb 2007 (CST)

Pagination question

An editor just submitted a NewPub for this pub. Take a look at the note and the pagination; he entered the page range. Do we want to do this for omnibuses? I think probably not, but information about length certainly disappears if we don't. I think perhaps it would be better to have the Pages field at the pub level say "xi+361+398+281+xxvi" and then give page 1 for each content item, but I thought I'd post the question here before editing it. I'll update the help text with the answer, too. Mike Christie (talk) 11:58, 4 Feb 2007 (CST)

"xi+361+398+281+xxvi" is fairly standard in the biblio world, so I think we can adopt it as well. As far as the data that we want entered in the "Page Number" field, I guess the first question is whether Al can think of anything that this would break? Ahasuerus 22:46, 4 Feb 2007 (CST)

Agreed - I would have used "xi+361+398+281+xxvi" rather than "xi+1040+xxvi" as it more clearly documents how the book was paginated. The "1-" in the pagination in the contents confused me as the source of the "1-" was not explained nor apparent until I thought about it for a moment and realized the editor was trying to convey "page 1 to 361." I suspect that as our standard usage of the content-page field is the starting page # that the record should have "1" for each of the titles and with "xi+361+398+281+xxvi" someone should understand that the first story goes from page 1 to 361 though that then creates the problem of that ISFDB will not know about what order to sort the titles in. As it is, the pub's note probably should explain in detail the exact content order and pagination rather than trying to tweak field values so that the contents are reliably listed in the correct order. FWIW, This morning I entered this publication to deal with two collections in a dos-a-dos double. Marc Kupper (talk) 13:45, 5 Feb 2007 (CST)

Do you want me to add in more detailed information about each individual element in the omnibus? I confess it was one of the few pb omnibus editions I've seen where each volume retained individual pagination, as well as adding new content ( the appendix ) not originally in either of the original hc publications Thomas conneely 17:38, 5 Feb 2007 (CST)

I've updated it here; the field length is three characters too short for all the pagination so I left of the last page number ("xxvi") and put it in the notes. I'll add a feature request for the extra length, if there isn't one already. Thomas, I don't think more detail is needed; the pagination makes it clear what's going on. Mike Christie (talk) 07:08, 6 Feb 2007 (CST)

Witchworld: The Turning

OK, I've just looked at my copies of the "Witchworld: The Turning" series and based on the contents I think these are all anthologies rather than novels, as they each seem to contain two novels by different authors set in Andre Norton's world. (Sort of like "Merovingen Nights" series.) The linking text is uncredited, but presumably belongs to one or more of the stated authors. However, there's questions about some of those too: I think we can do an author merge on "P. M. Griffin", "Pauline Griffin", "Pauline M. Griffin" and on "Mary H. Schaub", "Mary Schaub". However, I DON'T think we want an author merge on "Patricia Mathews", "Patricia Matthews", "Patricia Shaw Matthews", "Patricia Shaw Mathews", "Patricia Shaw-Mathews" as there's a Patricia Anne out there and I think some titles are currently mis-credited. Can I ask for comments before I do anything drastic? BLongley 15:51, 4 Feb 2007 (CST)

Another possibility would be OMNIBUS -- novels by different authors under one cover seems like an omnibus to me. However, I think anthology works OK too.

For the Griffin author merge, I'd say it's pretty safe to merge "Pauline Griffin" with "Pauline M. Griffin"; there's only one title under the former, and it's listed as by "Pauline M. Griffin" in at least one source. Without a verified source saying otherwise, I think you're good to go on that one. You could also just change the title record on Oath-Bound, and that would also get rid of "Pauline Griffin"; authors disappear when their last record vanishes. However, I think "P. M. Griffin" should stay; this page of Locus1 certainly quotes that form of her name for the last entry, "Firehand", and the "P(auline) M(argaret)" format generally means she used her initials.

Mary Schaub seems to have generally used her middle initial everywhere; there are only two titles using just "Mary Schaub". I don't think it's impossible that she didn't use it on a couple of stories, though; this Amazon page, for example, includes a review listing all the authors, including middle initials, but she doesn't give Schaub a middle initial. Since we don't have a verified publication under either of the "Mary Schaub" titles, it's not that big a deal to change it, but in a situation like this I tend to leave the records as they are until I can come up with evidence they're wrong.

Haven't looked at the Patricia Mathews ones, since you say you're not going to merge them. Mike Christie (talk) 19:52, 4 Feb 2007 (CST)

New Worlds Quarterly

I've checked my copies and based on the three editions I have it seems clear that it was certainly a magazine by issue 3. Issue 2 was arguable as it had content from "New Worlds" and one from a "Sword and Sorcery" magazine that never appeared, but issues 3 and 4 are all new content. All three editions are labelled as magazine in some way (2 as "Britain's foremost", 3 as "Britain's finest", 4 as "Britain's best"). I'd like to make them "Magazine" and leave the "Anthology" to the "Best... New Worlds" books with proper SBNs and such. I know this is the definitive example of how to categorise something as Magazine in the Help Text, so that'd need updating too if you allow this change. (Oh, and can anyone can verify "Weihnachtsabend" as opposed to the "Weihnachtabend" (no "s") I can see? I have no German skills.) BLongley 18:16, 4 Feb 2007 (CST)

I think it's OK to make these all magazines. I have certainly seen them treated that way by bibliographers in the past, so it's not a departure to do so. Re "Weihnacht(s)abend": Contento gives both spellings, so that's a genuine variant title. Mike Christie (talk) 20:02, 4 Feb 2007 (CST)

There are a few anthology series that can be legitimately called magazines. Some were advertised as magazines and some were not, but I don't think it's that big a deal either way. Magazines are a little more painful to maintain in the ISFDB, but on the other hand they get their own Wiki pages, which may help some users. There will always be borderline cases and we'll just have to toss a virtual coin. One word of caution, though. Original content vs. reprint content is not always determinative when deciding whether something is an anthology or a magazine. There have been many reprint magazines and many original anthologies. Ahasuerus 22:38, 4 Feb 2007 (CST)

Times Without Number

Here's one that came up in a discussion with BLongley; we're debating whether John Brunner's "Times Without Number" is a novel or a collection. It was originally three short stories. You can see all three listed here, in the Ace Double version. These had "numerous minor alterations" from the original stories, per Brunner's notes in the Ace 1969 edition. That edition, according to the same notes, says "Shorter and substantially different versions of the three sections of this novel were . . ." and it gives publication details of the Ace and the earlier magazine printings.

The Ace Double says "Turn this book over for a second complete novel"; they would say "second complete book" when they regarded it as a collection or anthology, so that's (rather circumstantial) evidence for it being a novel.

However, it is not a stretch to regard this as a collection, either. The three story titles are retained; they're prefixed with "Part One", and so on, but the novel is in three section that correspond to those stories, with no additional linking material. I would say the later version is definitely a novel, and there is already a Ballantine pub of that version that records it as a novel. The Ace Double is recorded as a collection, though.

Any opinions about which it is? Currently we have a confusing situation as there are two Brunner "Times Without Number" titles in his biblio; one under Collections and one under Novels. Comments? Mike Christie (talk) 20:16, 4 Feb 2007 (CST)

For what it's worth, Contento lists it as a collection, but then Contento tends to take an inclusionist view since collections and anthologies is what he concentrates on.

I have run into this problem with numerous fixups and was never able to come up with a cut and dry rule. At one point I almost filed a feature request to have "FIXUP" added as a new title type, but then decided that it would be too silly. But then again, maybe not? Ahasuerus 22:43, 4 Feb 2007 (CST)

I would go with "collection" for Times Without Number and to make the Ace double an omnibus of the collection and Destiny's Orbit. My thinking is that when someone views the title in either a bibliography or the Ace double's contents list they will see "coll" in the title list and know that they should drill deeper to see the stories contained. If Times Without Number is filed as a novel it's not apparent that it contains three short stories. You can always add a note explaining the thing is sometimes packaged as though it's a novel and that the stories did not seem to get published outside of this collection. Marc Kupper (talk) 15:26, 6 Feb 2007 (CST)

OK, I've read it now, and have to say that they still read like three short stories to me. 1 is set in York (or Jorque), 2 in London (or Londres), 3 in California. They share a common protagonist, Don Miguel Navarro, and there's enough references back in the two later stories to put them in order - but that doesn't make it a novel to me, each episode tied up perfectly neatly. I'll add some notes and submit my opinion again. BLongley 14:41, 14 Feb 2007 (CST)

Amazon Look Inside as a Source

I was sure that Writing Science Fiction & Fantasy was wrong. Looked at Amazon, the copyright page and table of content are readable online. Different edition, reprint of the paperback. Is is legit to use that to get the data closer to correct?

What the cover says is by the Editors of Analog and Asimov's Science Fiction. The copyright page just has the LOC catalog info, Gardner Dozois . . . et. al. The TOC lists all the articles and that includes many people who aren't editior anywhere. —The preceding unsigned comment was added by Dcarson (talk • contribs) .

The Locus Index and Worldcat.org both list Dozois as the only editor, which is probably what we want to do for now. I have a copy somewhere in my collection, which I may be able to check in late February when I am briefly reunited with it. Ahasuerus 06:07, 5 Feb 2007 (CST)

I have a project in my personal queue to verify the DAW book's using Amazon's "Look Inside" as for many of them I'll be able to verify nearly all of the information we collect about a publication other than the page count. Instead of flagging these as "p" (physical or primary verification) I would use "q" (I'm lazy (aka, efficient) and just used the letter after "p") for these. I've also been wondering just how Amazon generates the Look Inside scans as they are nearly always well clean and well registered. I assume they use an industrial paper cutter to strip off the spine and then feed the loose pages through a scanner.

There are specialized flatbed scanners that will create high quality scans of book pages. Low end products, e.g. Plustek's Bookreader are relatively inexpensive. Ahasuerus 14:13, 5 Feb 2007 (CST)

Back to the magazine issue - Look Inside may be less than reliable as you really want the physical magazine to check what the interior says rather than going by the cover, table of contents, or copyright page and also to check that none of these conflict (and to document if and where they do conflict). When I made additions/changes to ISFDB based on partial data such as a Look Inside or a cover scan I try to always state the source and to indicate that I'm not able to double check the data against other places in the publication. Marc Kupper (talk) 13:17, 5 Feb 2007 (CST)

Backwards Variant Titles

Look at who's back! Variant Titles! :-) I was documenting a Terry_Carr collection which included the following statement about the story The Chaser on the copyright page “‘The Chaser,’ copyright © 1959 by Terry Carr. Originally published under the name of Carl Brandon.”

I see that a title record does not exist for The Chaser by Carl Brandon though the pseudonym had already been set up. From the Terry Carr version of The Chaser I did “Make this a variant title” and had it be a new record by Carl Brandon.

I believe this is accurate in that the story first appeared as-by Carl Brandon in 1959 and was reprinted as-by Terry Carr in 1986. The problem is that ISFDB does not show variant titles in the main part of bibliographies meaning that while this story is visible on Carl Brandon (the pseudonym)’s page it’s not showing on Terry Carr’s page as the Terry Carr version is the “variant.”

Excluding the feature request that variant titles do get displayed on the pseudonym’s page I decided to deal with this by redoing the variant title linkage so that it was “backwards” and made the original Carl Brandon version the variant. The story disappeared from Carl Brandon’s page and now displays on Terry’s Carr’s page with an “as by Carl Brandon.”

Interesting. To me that is the "forward" case, and precisely how I've been entering them. No matter what, Terry Carr wrote the work, so his version should be the invariant one, and the Carl Brandon version is the variant title. (The fact that Brandon is a pseudonym makes it a variant title). For authors, "variantness" is whether the authorship varies, not how the title originally appeared. Maybe this is why we have confusion on the topic. Alvonruff 19:17, 5 Feb 2007 (CST)

My question is – should these “backwards variants” be an ISFDB standard or is there a better way to deal with this situation? Maybe instead of “variant title” these could be documented as being used to set up “as by” relationships where the canonical name is always the parent and you “as by” variant titles. Of course, with ISFDB you do these backwards in that you go to the variant and give the record # or name of the parent version though I’ve been thinking if a feature request to allow VTs to be set up from the parent would add or remove confusion about this subject. Marc Kupper (talk) 13:52, 5 Feb 2007 (CST)

I have been making the "usual name" the parent of the variant tree for the ones I've been doing. For example, I'd say that any story by Murray Leinster as "Will F. Jenkins" has a canonical form as by "Murray Leinster", even if it was never actually published that way, and even though that's actually a pseudonym. I had thought this was the standard, but maybe it's not documented. I agree that when there is doubt, the earlier version should be the parent, but this usually applies to titles as I assume that the canonical form would be whatever the canonical form of the author was. Mike Christie (talk) 19:29, 5 Feb 2007 (CST)

In thinking about what Al and Mike wrote I suspect what’s happening is that the two types of situations where the VT mechanism are used are not separated out well in my mind and I suspect in the minds of others.

• When a story is released under one title and later under another title then the first title is the parent and the second is the variant.
• When a story is released under a pseudonym then it’s reversed in that the initial story is the variant and the title record under author's canonical name is the parent.

Today I had a pseudonym but mentally dealt with it like a title variant. It was a case of tunnel vision in that I saw that The Chaser by Carl Brandon was the first title and therefore anything after that would be a “variant.”

I believe we tend to think in terms of “this title is the variant” and “that name is the pseudonym” meaning there’s some shifting of gears (and a little dizzy in the head) in with ISFDB you go to the pseudonym (author or title record) and say “make This the pseudonym/variant.”

At the author level it may work to have a “manage pseudonyms” link that would show the list of pseudonyms for the canonical name and allow people to add/remove them. If someone did it from an author that’s already flagged as a pseudonyms then you’d switch to the canonical name and to it from there or if there is more than one canonical name, as with house names, you could have a page that says “this name is used by a pseudonym by … and allow people to add/remove from the list of canonical names.

The variant/pseudonymous title version is more complicated and I need do some screen layouts to see if I can come up with something seems to make sense. Marc Kupper (talk) 23:18, 5 Feb 2007 (CST)

I think I see where you're going. If you do work on screen layouts, don't forget to look at the odd cases like Leinster, who has a usual name that's a pseudonym; John Wyndham is another. The house names are going to be particularly nasty; where the authorship is unknown, I think the house name becomes the canonical name. Mike Christie (talk) 23:24, 5 Feb 2007 (CST)

How about this as alternative wording for mkvariant.cgi - I did not change any of the functionality. < is the existing wording and > is the replacement.

< Making the following title a variant title of the specified parent:

> Use this screen to mark this title record as a variant of another

> title or to mark this title record as a pseudonymous title.

Title: Test Novel

Author: Marc Testing

Year: 0000-00-00

Type: NOVEL

-------------

< If the parent title already exists, enter the record number of the title below.

> If the parent (canonical) title record already exists, enter the record number of the title below.

< Parent #:

> Parent/canonical record #:

--------------

< If the parent title does not exist, enter the title information below to create it.

> If the parent or canonical title does not exist then enter the title information below to create it.

> Note that if you are changing the story title (variant title) then you will be creating a record for the original (canonical) story name below and Test Novel will be a variant title of the new title record you are creating.

> If you are changing the author name (pseudonym or a variant spelling of the name) then you should enter the canonical author name below and Marc Testing will become a pseudonym or alternate name for the canonical name. The canonical author name is the name under which a particular person's bibliography is organized by and is usually the name the author is most commonly known by.

> If neither of these seems to fit what you want to do then please check Help:Screen:MakeVariant. (I suspect the most likely reason for clicking the help at this point is someone who wants to make that name or record the variant/pseudonym rather than this name/record being the variant/pseudonym and so the help should address this. Another common reason is what do to if someone wants to change both the name and title and they are unsure of the this-n-that.)

Marc Kupper (talk) 01:17, 6 Feb 2007 (CST)

Your text seems reasonable; you might turn this into a feature request. I am hesitant to add too much text directly to the screens; we did have a feature request a while back for rollover help, though, and that might be a compromise. But your changes are not extensive so would probably be OK. Mike Christie (talk) 06:58, 6 Feb 2007 (CST)

I'll wait to see if there is any other feedback here and then copy/paste this thread to feature requests. In the mean time I'll be doing some editing of Help:Screen:MakeVariant. Marc Kupper (talk) 12:05, 6 Feb 2007 (CST)

Other Days, Other Eyes by Bob Shaw

Contendo, and as a result ISFDB, list this publication as a collection containing

• Light of Other Days • Bob Shaw • ss Analog Aug ’66
• Burden of Proof • Bob Shaw • ss Analog May ’67
• A Dome of Many-Colored Glass • Bob Shaw • ss Fantastic Apr ’72
• Other Days, Other Eyes • Bob Shaw • na Amazing May ’72 (+1)

I'm assuming the novel is a fixup of magazine stories but could not find any evidence of this in the book nor on the Internet. Does someone know of a source that I could cite for this being a fixup or should I convert the publication into a novel and remove the short story links?

As the note for "Other Days, Other Eyes" shortfiction says it's an Expansion of "Light of Other Days", it wouldn't make sense to list both as parts of the same collection, IMO. I think that's a fix-up novel. BLongley 14:50, 9 Feb 2007 (CST)

And his Wikipedia entry agrees, see paragraph 3. BLongley 13:55, 15 Feb 2007 (CST)

Other Days, Other Eyes was serialized (Amazing, May and July 1972) and so assuming it's a fixup that's why Contendo lists all of these as part of the "collection" Other Days, Other Eyes. I think I'll need to chase down copies of Burden of Proof and A Dome of Many-Colored Glass to see if this is a fixup. It's not something that seems to be mentioned in any edition of Other Days, Other Eyes. Marc Kupper (talk) 21:51, 22 Feb 2007 (CST)

I've got a copy of Other Days, Other Eyes on order, I think that goes top of my re-reading list when it arrives. (Sadly, I don't yet possess a copy of every good book I ever read, although my house is becoming a replica of the Colchester Public Library's SF section circa 1975...) BillLongley 16:00, 26 Feb 2007 (CST)

It's arrived. I think I will go enjoy reading rather than editing for a bit. BLongley 16:41, 1 Mar 2007 (CST)

Also, here's the signature. Does anyone recognize this as H. J. Bruck?

Here's a 300-dpi scan of the entire cover if that'll help. Marc Kupper (talk) 02:53, 8 Feb 2007 (CST)

Links to other sf related databases

I was chasing down an attribution and found that someone has built what looks like a complete index for the British Science Fiction Association (B.S.F.A.) magazines. I’m thinking that we should have a page that’s right off the ISFDB main page that could list other fan databases that overlap with ISFDB. Potential sites for inclusion are

Science Fiction indexes/databases

• British Science Fiction Association magazines (index/database maintained by a fan and not the B.S.F.A.)
• Contento's Index to Science Fiction Anthologies and Collections
• The Locus Magazine Index to Science Fiction

Publisher indexes/databases

• The Ace Image Library
• DAW Books (index maintained by Steven Silver)

What I have not thought up yet, and why I’m not making the big bucks as an editor :-), is a good title for the page.

Maybe what I'm thinking of is an expansion and/or renaming of Reference:General_Overview.

Marc Kupper (talk) 14:04, 8 Feb 2007 (CST)

Would it be any different than the Sources of Bibliographic Information page, which is linked from the Main Wiki page? Ahasuerus 14:21, 8 Feb 2007 (CST)

Perfect - I had igored the page as it's under "Current Projects" and also I had thought "Sources of Bibliographic Information" meant projects related to importing data into ISFDB. Would it work for you if I moved "Sources of Bibliographic Information" to the Resources section on the left and gave it a byline of "Lists on-line and print resources that can be used to verify publication and title data?" I suspect then we could combine "Current Projects" and "ISFDB Operations" - probably under the banner "Current Projects." Marc Kupper (talk) 21:45, 8 Feb 2007 (CST)

Moving it out of the Project area certainly sounds reasonable as I have occasionally wondered if our users could easily find the "Sources" link in that section. It would appear that the answer is a resounding "No" :) The proposed byline, however, sounds rather long for the Main Page. Something a bit more compact, perhaps? Ahasuerus 00:06, 9 Feb 2007 (CST)

I went ahead wth the move and also trimmed down the wording. The other Resources bylines averaged 13.1 words and 81 characters. Here are various editions of the wording followed by the (word count, character count).

• Sources of Bibliographic Information – Lists on-line and print resources that can be used to verify publication and title data. (15, 88)
• Sources of Bibliographic Information – Lists resources that can be used to verify publication and title data. (12, 70)
• Sources of Bibliographic Information – Resources that can be used to verify publication and title data. (11, 64)
• Sources of Bibliographic Information – Resources used to verify publication and title data. (8, 52)
• Sources of Bibliographic Information – Resources used to verify data. (5, 32)
• Sources of Bibliographic Information – Resources used to verify bibliographic data. (6, 44)

Marc Kupper (talk) 11:55, 9 Feb 2007 (CST)

Strange Seas and Shores

I just added a pub to Avram Davidson's collection Strange Seas and Shores and noticed something slightly fishy about the contents of the first listed edition. The Doubleday ed. lists both a preface and introduction by Avram Davidson, which would be a bit odd, but is possible. My copy (Ace) has a preface by Davidson, but an introduction by Bradbury, and I wonder if the introduction in the Doubleday ed. should have been credited to Bradbury as well. Contento only lists the Doubleday edition, and doesn't credit anybody with either the intro or preface, implying that both are by Davidson. Can anybody check this from their collection or other refs?Jefe 23:20, 8 Feb 2007 (CST)

I have the hardcover edition in my collection and should be able to check it ca. 2007-02-17 when I get back to it. Ahasuerus 12:22, 9 Feb 2007 (CST)

P.S. If you run into any errors or omissions in Contento's biblios, keep in mind that Bill welcomes corrections (contento[at]best.com). Ahasuerus 12:26, 9 Feb 2007 (CST)

Take a look at the copyright page in your edition and see if there is a later copyright for the "new" or "Bradbury" introduction. Ace tends to be bad about including details on their copyright pages but you might get lucky and find a statement such as "New introduction copyright (c) 19xx by Ray Bradbury." Marc Kupper (talk) 12:34, 9 Feb 2007 (CST)

I checked, and my edition has copyright info for all the stories, but nothing whatsoever about either the preface or intro. The cover says "Introduction by Ray Bradbury" but doesn't suggest whether it's a *NEW! EXCITING!* introduction, or just the old one.Jefe 15:40, 9 Feb 2007 (CST)

The lack of attribution may be a clue that the introduction was added but for now we'll wait for someone that has the hardcover in hand. Marc Kupper (talk) 02:14, 12 Feb 2007 (CST)

Stardance and Stardance II

I've finally dug out the Analog magazine containing the conclusion of Stardance II and verified that it's part of the novel Stardance, not a separate novel. I don't see how to represent this clearly though: I believe the Serials only appear under a Novel if they have the same title? So I could make the Novella Stardance appear under the Novel Stardance in the Stardance Series by making it a serial (Part 1 of 1), but I can't see how to add Stardance II (already nicely serialised but under a Novel that doesn't exist) as well. Any suggestions on two serials in one novel? BLongley 06:42, 9 Feb 2007 (CST)

Play with a variant title called "Stardance (Part 2 of 2)". I'm not sure if it'll work with "Stardance (Part 2 of 2)" being the variant or if that has to be the canonical title and "Stardance II" would be the "variant." Marc Kupper (talk) 12:22, 9 Feb 2007 (CST)

I can't see that that would remove the non-existent novel... Anyone got any definite advice rather than suggestions of what to "play with"? (My mistakes will get noticed longer than a Mod's will. :-( ) BLongley 16:48, 9 Feb 2007 (CST)

What I did was

1. Renamed Stardance (Part 1 of 3) to Stardance (Part 1 of 4).
2. From Stardance II (Part 1 of 3) I did a "make this a variant title" and created a new parent title Stardance (Part 2 of 4).
3. From Stardance II (Part 2 of 3) I did a "make this a variant title" and created a new parent title Stardance (Part 3 of 4).
4. From Stardance II (Part 3 of 3) I did a "make this a variant title" and created a new parent title Stardance (Part 4 of 4).

I left your old placeholder, Stardance II in there for now as one problem with doing it this way is that the shortfiction titles Stardance II (Part 1 of 3) to 3 of 3 are no longer displayed in the bibliography as they are variant titles and ISFDB does not have logic to check for and display these when showing the serials. I'll see if anyone else has a better idea and/or if a feature request to show the variant titles is needed. Marc Kupper (talk) 03:05, 10 Feb 2007 (CST)

ps: Brian, is the Stardance shortfiction at the bottom of the page is the same as Stardance (Part 1 of 4)?

Yes, that's the one I wanted as the first "serial". (And I'm "Bill", Not "Brian".) BLongley 14:50, 11 Feb 2007 (CST)

I remember I had a situation that was very similar this and I dealt with it the way you did where there was a dummy title/ But, I was able to then snug this title right up against the real title by adding a dummy series. Unfortunately, that trick won't work in this case as the Stardance novel itself is a numbered item in a series. While it would look a little messy you could do it by creating three dummy series

• 1 Stardance
• 2 Starseed
• 3 Starmind

The parent for all of these is the main Stardance series and under "1 Stardance" you would put the Stardance and Stardance II titles. This will cluster things right but looks a little messy. Marc Kupper (talk) 03:26, 10 Feb 2007 (CST)

Thanks for trying - I don't think I want to mess with it anymore till we sort out presentation over accuracy. BLongley 14:50, 11 Feb 2007 (CST)

Getting to isfdb from the wiki

I'm not sure where the mediawiki skins are stored but believe it would be useful if the "isfdb" folder linked to http://www.isfdb.org/ rather than Main_Page. There have been numerous occasions when I wanted to get to www.isfdb.org from the wiki and it seems my only option is "main page" and then to click on the word "ISFDB" in the greeting. Another option would be to add an ISFDB link to the "navigation" box in the skin. Marc Kupper (talk) 12:40, 9 Feb 2007 (CST)

I second this. I'm forever clicking the "isfdb" image in the wiki and being surprised that I'm not on the other site.Jefe 15:43, 9 Feb 2007 (CST)

Sounds like a good plan! Ahasuerus 15:56, 9 Feb 2007 (CST)

Thirded. I have resorted to having 3 "Favourites" entries to get me from place to place. BLongley 16:33, 9 Feb 2007 (CST)

Three favorites? Which pages are they as we might as well combine the requests to update the navbar. Apparently the skin that contains the navbar is in a CSS file that may or may not be called monobook.css or monobook/main.css. I was unable to figure out a path to the thing but assume Al knows where it is meaning we can put in a feature request. Marc Kupper (talk) 02:36, 10 Feb 2007 (CST)

The third is my own talk page. I tend to work with 3 or 4 tabs: 1 for editing, 1 for searching, 1 or 2 for Community portal and/or my own talk page where everybody's advice ends up. BLongley 07:03, 10 Feb 2007 (CST)

E.E. 'Doc' Smith

I was surprised to find that's not listed as even a pseudonym. If I go by these rules, he has to go in under that variant: Author - The name of the author of the publication. The name should be entered exactly as it actually appeared in the publication. This includes pseudonyms, abbreviated names ("I. Asimov" instead of "Isaac Asimov", "Robert Heinlein" instead of "Robert A. Heinlein"), etc. As with the title, take the name from the title page in preference to the cover or spine of the book. But with over two dozen publications to enter, a quick scan suggests that they're almost ALL likely to be E.E. 'Doc' Smith, unless there's a special rule for him to take the author from the copyright page instead? ("Edward E. Smith, Ph. D." - and even then it wouldn't come out as a standard isfdb name every time.) I'm fine with creating the pseudonym, but it would suggest there's few to no British publications of his here yet, and that seems so unlikely I though I'd ask around here first. BLongley 14:45, 9 Feb 2007 (CST)

Ok, having chosen Alan Dean Foster to do in the meantime, it seems we ARE lacking British Publications a lot more than I thought? BLongley 16:43, 9 Feb 2007 (CST)

It depends. We have much better coverage for collections originally published in the UK (Contento etc) and for recent UK editions (Amazon.uk.co etc) than for non-recent UK books in general and non-recent novels in particular. Sometimes it's just a question of what past ISFDB editors had access to or were interested in, so certain authors are much more thoroughly covered than others. Ahasuerus 18:41, 9 Feb 2007 (CST)

Well, I added a lot of E. E. 'Doc' Smith works. As usual, whenever I try anything new, it causes problems. :-(

It seems there's two of them now, one with a lot of orphaned works. But if I do an advanced search on "E. E. 'Doc' Smith" under "Pseudonymns" [sic] it causes a Python error. Another Apostrophe-related bug, or Unicode, or something? Or me trying to go beyond my abilities? BLongley 16:42, 10 Feb 2007 (CST)

The search function is buggy and many things get python errors. For example, I usually enter the title in the first field and the author name in the second field. Out of habit I entered "Smith" in the second field and hit [Submit]. Oops, I guess it does not like the first field being blank. I enter "Doc" in the first field and search for authors in both fields. There are just two titles

• Masters of the Vortex
• The Bloodstar Conspiracy

I see what happened. At some point someone "cleaned" up the E. E. 'Doc' Smith bibliography by changing the author name from E. E. 'Doc' Smith to Edward E. Smith in the title records. They should have set up variant titles. It's moderately time consuming but the fix is:

• from E. E. 'Doc' Smith click on a stray publication. I'll use First Lensman
• Down near the end of the list of fields is "# Title Reference: First Lensman " - click on "First Lensman"
• Click "Edit Title Data" and change the author name from Edward E. Smith to E. E. 'Doc' Smith
• Go back to the title record and click "Make This Title a Variant Title or Pseudonymous Work"
• In the lower part of the page where it says "Author1:" change the name to "Edward E. Smith". If the title update has not been approved yet this will already be set up for you.
• Once the title update and make-variant are approved then First Lensman will appear on Edward E. Smith's page with an "as by E. E. 'Doc' Smith"
• You don't need to wait for approval though but can go through the remaining stray publications. Once that is done then from E. E. 'Doc' Smith page you'd do "Make This Author a Pseudonym."

Note that there are two very similar pages

• http://www.isfdb.org/cgi-bin/ea.cgi?E.%20E.%20'Doc'%20Smith
• http://www.isfdb.org/cgi-bin/ea.cgi?E.%20E.%20'Doc'%20%20Smith (this one has two spaces and someone already made it a pseudonym meaning if there are stray publications they are invisible and you need to use the advanced search to locate them.)

Marc Kupper (talk) 20:37, 10 Feb 2007 (CST)

ps: I did did a publication search for "E. E. 'Doc' Smith" (with two spaces before Smith) and found three publications. I changed the author names on all three publications to "E. E. 'Doc' Smith" (with one space before Smith). This deleted the extraneous "E. E. 'Doc'(space)(space)Smith" author and now they will show as strays on http://www.isfdb.org/cgi-bin/ea.cgi?E.%20E.%20'Doc'%20Smith page. Marc Kupper (talk) 20:48, 10 Feb 2007 (CST)

ps2: I was thinking about what I had written and believe there’s a more reliable way to deal with this though it’s more work. The first two steps are the same as before

1. from E. E. 'Doc' Smith click on a stray publication. I'll use First Lensman
2. Down near the end of the list of fields is "# Title Reference: First Lensman " - click on "First Lensman"
3. Click on "Unmerge Titles" that's at the bottom of "Editing Tools:" in the navbar.
4. Make a note about the title’s record # from the URL
5. Check/select all of the titles by "E. E. 'Doc' Smith" and then click [Submit Unmerge]
6. Repeat steps 1 to 5 for all of the E. E. 'Doc' Smith stray publications.
7. Wait for moderator approval of the title unmerges.

1. from E. E. 'Doc' Smith click on a stray publication. I'll use First Lensman
2. Down near the end of the list of fields is "# Title Reference: First Lensman " - click on "First Lensman"
3. Click "Edit Title Data" and change the author name from Edward E. Smith to E. E. 'Doc' Smith
4. Go back to the title record and click "Make This Title a Variant Title or Pseudonymous Work"
5. In the upper part of the make variant enter the record # from step 4 and click [Submit].
6. Once the title update and make-variant are approved then First Lensman will appear on Edward E. Smith's page with an "as by E. E. 'Doc' Smith"
7. You don't need to wait for approval though but can go through the remaining stray publications.

1. Once that is done and E. E. 'Doc' Smith’s page is no longer showing stray publications you can do "Make This Author a Pseudonym." Marc Kupper (talk) 22:45, 10 Feb 2007 (CST)

ps3: I started looking at the strays plus titles and realized the "plan" would vary a little from title to title and so I went ahead with setting up all of the variant titles. The good news is Edward E. Smith's page now shows the as by E.E.'Doc' stuff, the bad news is after the last variant went through E. E. ’Doc’ Smith's is showing stray publications but I can't see why they are strays. I'll ask Al about this. I don't think it's related to the "'Doc'" as I took a pub & title and removed the quotes making "E. E. Doc Smith" and that author's page showed a stray instead of a title. Marc Kupper (talk) 23:50, 10 Feb 2007 (CST)

ps4: All handled - the new bibliographic display logic just gets a little confused because it knows it does not want to display variant titles and when it then did the check for stray publications all of the publications are counted as "strays" because a title record for them is not being displayed. Marc Kupper (talk) 00:17, 11 Feb 2007 (CST)

Thanks Marc! I had to fix the Skylark series and a couple of variant titles, but it looks much better now. (I'll leave the "Lord Tedric" mess to someone that knows where Eklund came into it.) BLongley 14:32, 11 Feb 2007 (CST)

It turns out I had an Eklund to add as well (my cleaner just filed it alphabetically and I hadn't noticed it till now.) "Lord Tedric" is definitely open to other editors now though. BLongley 16:18, 1 Mar 2007 (CST)

Omnibus or New Collection

I'm trying to sort out the tangle over on the Howard Waldrop page and have a question: how does one denote a new book that reprints the content of an extant collection and an extant novel (or possibly, novella) or the contents of two extant collections? My first thought is it's an omnibus, since it takes two separate existing things and brings them together. If so, are there rules for the creation or linkage of the parts of an omnibus. I'm looking specifically at Strange Things Close Up, which combines Howard Who? and All About Strange Monsters of the Recent Past. (Scott Latham 19:43, 10 Feb 2007 (CST))

Nope - there are no special rules and it's handled the same as a collection or anthology. I'd call it an omnibus as it includes at least one previously published publication title. You'd add your omnibus publication and hit [Add Title] a couple of times to add the titles of the stories it contains. Once approved you merge the story titles with the existing title records. Something I do when doing this is I append "(merge)" to the title as I'm entering them as a reminder that they need to be merged.

I don’t know Howard Waldrop’s works but let’s pretend a book called The Best of Howard Waldrop gets released that contains the following:

• Them Bones (a novel)
• Going Home Again (a collection)
• Green Brother (a shortfiction)

You would do

• New Omnibus
• Enter the title The Best of Howard Waldrop etc.
• Hit [Add Title] three time. Hint. Hit it once with the mouse and then tap the space bar twice. It saves on aim-click at a moving target.
• Enter your three titles. All you need is the starting page numbers, titles and author names. Don’t worry about the date and length values as you will be merging them but you should set the type so that just in case you don't get around to the merge right away people won't cat confused about things showing up in the wrong category. As noted above, when I do it I enter the titles as:
• Them Bones (merge)
• Going Home Again (merge)
• Green Brother (merge)

Submit the publication and once approved then using either advanced search or the titles link from Howard Waldrop’s page you locate the titles and merge them. Note that if use you titles that you check two titles, hit [Merge Selected Records], confirm what you want to do, go back a couple of pages in the browser, uncheck the two titles just merged, check the next two, merge those, etc. Once the merges are approved your omnibus should be set up. Marc Kupper (talk) 20:15, 10 Feb 2007 (CST)

Translation question

I've held a couple of updates in the sub queue that relate to translations. I think what they do is correct, but I wanted to check here first as it does look a bit odd. They change this pub to have the German title. I believe the last discussion we had on this concluded that the title should be unified under the first version, rather than having variant titles for the different languages. Here the first version is the 1995 German version. The result is that the title will be listed as "Die Haarteppichknüpfer", the German version, and "The Carpet Makers" will only show up in pub titles. As I said, I think this is consistent with earlier discussions, but does anyone else think the result will look a bit odd to an English reader? And does that matter? Mike Christie (talk) 09:57, 11 Feb 2007 (CST)

It looks consistent, that seems to be what was done with Harry Potter for instance. Yes, I think it will look a bit odd to English monoglots, but I'm one and don't think it matters. BLongley 12:19, 11 Feb 2007 (CST)

I don't worry about it too much either and this morning when I saw that an English title record had translated publications I just shrugged as some day ISFDB will have better language support and we'll deal with it then. I like how Pierre_Barbet was done where the English titles are variants of the orginal French works. Technically that's not correct translations are regarded as "new works" but it serves to show the linkage. The current Andreas_Eschbach page is a little confusing in that aspect and I would VT the titles as with Barbet, particularly as there are both English and French translations of his story. Marc Kupper (talk) 02:11, 12 Feb 2007 (CST)

At this time English language translations are privileged in that they get not only a Publication record under the master Title record, but also a separate Title record. Thus, the master Title record for a Jules_Verne novel will be in French, but there will also be one or more English language Titles for the book. On the other hand, if a book was originally written (emphasis on "written" since sometimes English language books are only published in translation) in English, its foreign language translations do not get a separate Title record, only a Publication record under the English language Title record. Unfortunately, this approach doesn't work well for short fiction pieces since they don't have "Publication records" aside from the anthology/collection/magazine Publications in which they appeared.

As we discussed a few weeks ago, we may want to change this approach and make the database layout language-agnostic if and when we allow our users to specify language preferences. Until then, creating Variant Titles for every translation would make Summary Bibliography pages for popular writers like Heinlein a navigational nightmare. Once I have a little bit of free time, I'll try to find the writeup that I did at the time and convert it into a feature request. Ahasuerus 11:58, 12 Feb 2007 (CST)

I've been thinking about this recently and think the kind of thing that might be ideal is if there was a language field against a title, as well as a translation field that points at the original title it's a translation of. So like with variant titles you can mark a title as a translation of another title.

Then show the English language of a bibliography by default with links in the header to the views in different languages where there are translations/original works. The behaviour of a entry when a user specifies a different language by choice would be to show that language for each entry, but fall back to english if titles have not been translated into that language. You could also use the accept headers a browser sends to choose the default language. IANA keep a list of language codes but that's a far bigger list that we'd ever need.

Of course it would be a while before it would work complete as there will be a backlog of titles without the correct language set, but something like this will need to done at some point so that's always going to be an issue. I'm happy to download a copy of the database, set it up and do some experiments. Is it worth me doing that or is someone already looking at this? --Unapersson 17:27, 16 Feb 2007 (CST)

That's really a question for Al, and he's indicated recently he's about to take an extended break. However, I suspect that getting a copy of the database and working with it is probably a good thing to do anyway, since you'll get more familiar with it. I haven't done it, but I hope to find the time. Mike Christie (talk) 17:34, 18 Feb 2007 (CST)

Why I'll Never Be a Moderator -- Charles McCarry

I've been doing a little editing on the Charles McCarry page. One obvious object was his novel "The Secret Lovers". It was originally published in 1977 or so. But the McCarry page listed only a 2006 new edition. I then added the earlier, 1977, edition to the "publications" for that entry. This has been cleared by the moderators. And yet, under "Christopher Family Chronicles", the date shown for "The Secret Lovers" remains stubbornly "2006". I've been clicking on various things but don't see anything that would let me correct this. I'm leery of "merging" things, since I'm afraid of messing too many things up. Could someone more gifted than I fix this up? Thanks? Hayford Peirce 21:29, 11 Feb 2007 (CST)

The date shown is in the title record. Go to the title, hit "edit title data" and you will see the date. Ideally ISFDB should be computing this from the entered publications but for now we manually update the title record once we become aware of earlier publications. Marc Kupper (talk) 01:56, 12 Feb 2007 (CST)

Magazine Search

I'm not sure if I'm doing something wrong or if there's a bug. I can never remember the full format of something like "Analog Science Fiction/Science Fact, October 12, 1981" so I use Advanced Search and put "Analog", "October" and "1981" into the three boxes. Which finds the COVER, but not the title. OK, no biggy as it's only two clicks to the publication (well, if there IS a cover entry, I've seen one where there was a cover photograph instead and this trick didn't work). Even after I've found it a search on the full title won't work - I thought it might be down to field-size but the cover's title is longer than the pub one. Am I missing something obvious? (Really, it's not a problem, I have no more magazines to enter at the moment.) BLongley 15:29, 13 Feb 2007 (CST)

That's one that catches everyone the first time; took me a while to understand what was going on there. What happens is that the title records for magazines are of type "EDITOR". These are created automatically by the "New Magazine" version of the newpub form. They have the editor as the owner of the biblio, so if you go to Campbell's biblio, for example, you would see all those records there -- and at one record per issue it was a huge, and rather unreadable, list. So what is done is to merge all the title records for a given year into one title record, and give it a title of the magazine, plus the year, with no month. This gives one record on the biblio but means you can't search for the title of a single issue.

To see what's going on, go to Advanced Search and put "Analog" in for title, then tab down to the second field in the Title section and put "1966" in, also for the title. That will return 13 records -- 12 cover art records and the year title record. Then go back to Advanced Search and enter the same things for the publication: "Analog" and "1966" (separately) and you will get 12 records -- the 12 individual issues. Does that make sense? Mike Christie (talk) 16:18, 13 Feb 2007 (CST)

Much clearer now, thanks. Of course, I WOULD have to have been looking for the June 1965 edition - no coverart. :-/ (It had a photograph instead.) BLongley 13:58, 14 Feb 2007 (CST)

Lord Darcy

There are several publications under the title Lord Darcy. The SFBC in 1983 and Baen in 2002 tp, 2004 pb. While it is the same title is isn't in some ways the same book. The SFBC looks like it had all the Darcy stories at the time, the Baen one has all of them except the 2 novels written after Garretts death by Kurland. Should it be two separate titles? Dana Carson 21:08, 13 Feb 2007 (CST)

There has been debate about how to handle this in the past. The main thing is that the publications accurately reflect the contents. My preference would be to have two titles "Lord Darcy", and give each one a title note that said something like "1983 SFBC version including Kurland novels" and "2002 Baen version without Kurland novels". That way anyone who tried to merge them would see the incompatible notes. I don't think we have discussed whether or not one could be made a variant of the other, or whether they should be listed as separate versions. Again, speaking for myself, I would not make them variants -- the contents are substantially different, not just a minor difference. A reader browsing the bibliography would want to know about both, separately, and would not regard one as a publication of the other under a different title.

I have seen titles where a version description is included in parentheses after the title. For example, see this title and this one, where someone took that approach to a short story. I don't think this is a very good way to do it. Mike Christie (talk) 21:40, 13 Feb 2007 (CST)

Some moderators have done this to my edits: e.g. Melancholy Elephants (revised US 1985). It does make it clear the editions are different, and shouldn't affect searches. It doesn't feel quite right as a solution, but I haven't tried merging titles with different notes yet to see if there's sufficient warning given. BLongley 15:17, 14 Feb 2007 (CST)

The SFBC version doesn't include the Kurlands if its publication record is accurate. So the difference is 2 new stories in the Baen book. It seemed to me that different publishers, and different editors should mean it is a different book however.

I'm also thinking about that sort of thing for Dream Makers by Platt where he has a 1980 version, a 1983 version that looks like it was 2 books, and a 1987 that apparently has all the interviews from the earlier ones and a few new ones. I have the 1987 US version, looks like the 1987 UK is the same. The 1980 has a contents listing under one title, are the others variants, it looks like it. No contents listed for the 2 1983 books. The 1987 lists when the interviews were done so you can make a good guess. Dana Carson 00:22, 14 Feb 2007 (CST)

There's a few instances where collections have been made variants of previous shorter collections too, even under different titles. When they have a different title, I think I'd like them kept separate: when they're the same title but different contents, I'd prefer them to keep their proper titles but it should be clear there's a difference: and when they're different titles for the same collection and contents I'd like variant titles. Just my tuppenny-worth. BLongley 15:17, 14 Feb 2007 (CST)

I've been thinking of asking for a revised story mechanism that would work like variant titles but am not sure if it should also be able to handle fixups where two or more stories get sewn together and if an things like excerpts are "revised stories." At the moment it seems we have available are (parenthical notes). Variant titles don't work well as they remove the story(ies) marked as variantes from the main bibliography.

FWIW - I believe the publication titles can always be "as stated" without the (parenthical notes) and that the (parenthical notes) are appended to the title-record titles. Marc Kupper (talk) 19:05, 14 Feb 2007 (CST)

Debated issues

I've created a subpage off my user page, User:Mike Christie/Debates, to capture some of the debates we have on this page that eventually scroll off into the archives. I will try to start work on expanding those bullets into sections that summarize what the various viewpoints have been, and what the current consensus is, if there is one. I'll also try to make sure the help files are in sync.

If anyone would like to add another issue to that list, please do; or edit/create explanations of the issues. Eventually I want to dump most of the content into the help, one way or another -- most likely as part of the FAQ. I will be working on this as a background task, so it'll probably take a while; all suggestions and help appreciated. Mike Christie (talk) 18:16, 13 Feb 2007 (CST)

I've added some opinions on your "To Do" list. Of course, feel free to over-ride them if they've been discussed to death before and a consensus was reached before I arrived. BLongley 15:41, 14 Feb 2007 (CST)

Different printing, same ISBN and cover--different publication?

Publication BKTG12380 (The Descent of Anansi) has been reprinted with the same ISBN (0-812-51292-8) at least in 1991 and 2000, and probably more times than that. These are labeled "third printing" and "fourth printing", but have the same ISBN. Do they get their own publication entries? grendel|khan 08:59, 14 Feb 2007 (CST)

I notice we still only have one pub under that ISBN - haven't we convinced you? BLongley 14:46, 19 Feb 2007 (CST)

Oh, yes! As per the Help:Screen:NewNovel, "we are interested in recording each different reprint of a publication". See Ursula_K._Le_Guin's bibliography, e.g. A Wizard of Earthsea for a recent example. Ahasuerus 13:17, 14 Feb 2007 (CST)

A more accurate answer might be "Some of us are interested in recording each different reprint of a publication". I do record them, but it's not for my benefit: I'm more concerned about NOT buying yet another copy of the same work under a different title or a different cover. (I think I've bought "Sirens of Titan" 3 times, fooled by different covers, and bought John Varley Collections twice because they were published under a different short-story name for the title.) But, yes, record them all - and then sort them out bibliographically for MY benefit! ;-)

I think I know which shelf I should attack next. BLongley 15:57, 14 Feb 2007 (CST)

I've been caught a few times by cover changes and own duplicates as a result. I've personally gone back and forth on if I should record every single printing in ISFDB. For a while I was just noting the different printings for a publication that was otherwise identical in a single publication record. That turned out to cause a problem in that I never came up with a way to clearly communicate to viewers and/or editors that “this pub record is for the 2nd, 4th, and 5th printing. There have been verified sightings of the 2nd and 5th printing, the 4th printing has been spotted on Abebooks, and it’s assumed a 3rd printing exists…” Another hassle was that many publishers will have “First pubname printing: some date / number line” which is clear enough but should the record for the first printing be shared with the second or later printings where the pub-date is unknown? Finally, there are often gaps in the known information. For example, we may know that the 3rd printing was $1.75 and the 5th was $1.99. We don’t know if the 4th printing was at $1.75, $1.99, or even an intermediate value.

Going to one record per printing vastly simplified things as then it’s easy to tell which printings have been spotted in bookseller catalogues, which have been physically verified, and for things like the 4th printing we can add a placeholder record with a blank price and a note explaining that we know about the 3rd and 5th and are waiting for someone to show up with the 4th. One record per printing also simplifies that those publishers that print in the USA and Canada often overlap the printing numbers and, just to mess up bibliographers, may have different prices…

The hassle with one record per printing is that it’s a lot of records plus that at present there is no way to organize the records, particularly as many later printings do not state a date meaning they are all stuffed up at 0000-00-00.

FWIW - I tend to record a lot of detail when I enter publications. My thinking is to be able to eventually use the notes to construct help pages for each publisher that explain how to tell one printing from the next, what the codes mean, etc. Books printed in the last ten or so years tend to be better (probably because many publishers got swallowed up by the “publishing groups” and now all use the same systems) but as nearly all my collection is older books I’m still dealing with trying to tease usable information out of things like how the spine or copyright page is formatted. Marc Kupper (talk) 18:35, 16 Feb 2007 (CST)

Submissions held in review

I understand that during approval of submissions, a reviewer may flag a submission he is working on, to let other reviewers that it's being looked into. The submitter only sees that it is pending. If the submitter could also see the name of the reviewer, he would be able to create an entry on his (or the reviewer's) talk page, explaining what was intended. I think this could expedite proper review and ease the burden on reviewers. What does anyone else think? (Scott Latham 18:26, 14 Feb 2007 (CST))

Sure, sounds reasonable. Anything that helps us communitate better is a plus in my book. While we are at it, might as well request an update to the Moderator screen so that any submissions by users with moderatorial privileges would display differently (color? checkmark?). That way other moderators -- and there will be more as time goes by -- won't try to cross-approve submissions. Ahasuerus 18:53, 14 Feb 2007 (CST)

We get to see the reviewer in Recent and Rejected Edits, shouldn't be too hard for Pending, I hope. While we're at it, could we have the "Subject" as a Hyperlink too? During multi-stage edits it would be great to have a quick way to go back to something that's approved for the next phase. Or to see where we went wrong in a rejection. BLongley 14:49, 16 Feb 2007 (CST)

Second the hyperlink request. Lots of submission require 2 steps and being able to go the Recent and clicking on the link to start part 2 would be useful. Dana Carson 04:42, 17 Feb 2007 (CST)

Something that may help is that with Reject a reviewer/moderator can enter a note but notes are not an option for Hold. If Hold supported notes I'd usually jot down what I plan to look into. I started to write down what I look for when reviewing submissions but got distracted and need to get back to that as I believe that would also help the editing process as then people would know what sort of things hold up their submissions and could add notes explaining why the submission looks strange thus heading off the need for the moderator to hold and research or ask about the item. Marc Kupper (talk) 18:12, 16 Feb 2007 (CST)

Larry Niven

I'm tempted to tackle some of the Larry_Niven titles this weekend. A few questions:

1) There's some bibliographies in the publications that don't match our current classifications. E.g. we have "Svetz" but not "Leshy Circuit" - we call that "State". Any particular reason to stick to our current series titles?

OK, first warning - "N-Space" has several unreliable Magazine dates around 1969/1970, and lists at least one publication inaccurately ("The Words in Science Fiction" is NOT in "Antigrav".) BLongley 14:16, 17 Feb 2007 (CST)

2) Some ShortFiction has been assigned to particular series: this would actually help me in the short term, as some collections have been split into the "Known Space", "Gil the ARM", and "Other" categories for later collections. I could at least separate the collections that are definitely NOT in the series they're attributed to more easily with that info, but is it something we'd like to keep around afterwards? There'll still be several books that are part-in, part-out.

Assuming my edits go through, series up to 1976 will match the bibliography in "Tales of Known Space". I'm already tempted to make "Gil Hamilton" a sub-series of its own based on later data. (And have to add some more books on my Amazon wish-list...) BLongley 17:16, 17 Feb 2007 (CST)

3) "Magic Goes Away" - we've got a Series with one book in and an Anthology Series with the other two. I think they'd look better together. Any suggestions?

You have some time to answer, I see "N-Space" will need a lot of work anyway... (Who lets people produce monstrous problems like that? I think it would be easier to find Heinlein's "Requiem" and do that one... :-/ ) BLongley 16:53, 16 Feb 2007 (CST)

I'd say take your best stab at the series. This is probably the most contentious series in the ISFDB, and was the location of numerous revert wars in the 90's. Go for it. Alvonruff 14:22, 17 Feb 2007 (CST)

OK, will do. Leave it with me for a day or two, then we can see what it looks like. BLongley 16:48, 17 Feb 2007 (CST)

Question 4: What do we do about stories like these? Some are already included here, but others don't have the requisite publications. Should I create placeholders just so we can include the missing stories and essays? I suspect not as we don't seem to have many "Playboy" publications... a few empty ones, although this one looks like it's there for the same reason I'm suggesting... BLongley 09:46, 18 Feb 2007 (CST)

If they have dead-tree versions, then we can include them -- I think you can just create the title records if you wish, though there seems no reason not to also create as much of the pub record as you can based on the info quoted there. Mike Christie (talk) 17:27, 18 Feb 2007 (CST)

OK, will do, but probably next weekend. In the meantime, if anyone can do the contents for this pub it would save some work. I'm on to "Dune" at the moment. BLongley 16:36, 19 Feb 2007 (CST)

Got that publication ordered now, so don't worry about it. BillLongley 16:05, 26 Feb 2007 (CST)

It arrived, I added it... remind me to stop volunteering. :-/ BLongley 16:15, 1 Mar 2007 (CST)

Russell T Davies and regularizing author names

While looking over the Dr. Who edits I saw Russell_T_Davies and realized that

1. Help:Screen:AuthorData is silent on the issue on if the middle initial should use a period.
2. Template:TitleFields:Author is also silent on this issue and also does not mention author name kerning though that’s mentioned in the author level help screens.
3. Template:PublicationFields:Author is the same as TitleFields.

Both the title and publication help pages state “The name should be entered exactly as it actually appeared in the publication” though down at the very bottom of the title/publication pages it mentions regularizing names though does not bring up a middle initial as an example. Should this get added?

FWIW – this author is in Wikipedia as “Russell T. Davies” though many of the references, particularly for external sites, use “Russell T. Davies.” I tried to link from the wikipedia back to ISFDB using {{isfdb name|Russell T Davies}} but wikipedia regularlized that into "Russell T. Davies" which gets an Author-Not-Found error. Marc Kupper (talk) 02:49, 17 Feb 2007 (CST)

I think a period for middle initials is a reasonable convention. I've seen debate about whether "Forrest J. Ackerman" gets a period, since his middle name is just "J", but I don't think it hurts to add the period to avoid just the kind of link problems you mention with Wikipedia. I suspect we'll run into occasional title pages that don't use a period for middle initials, and I'm pretty sure we don't want to make those into variant names -- so that means we regularize. Mike Christie (talk) 17:25, 18 Feb 2007 (CST)

I figured I would make the wording/definitions for canonical names, title authors, and publication consistent and did a diff of the title and publication descriptions. One surprise was that for titles you enter the author names alphabetically. Maybe I read it before but had forgotten about this one. I ran a test where I added Marc Testing1, Marc Testing2, Marc Testing3, and Marc Testing4 as authors to my test novel’s title record and ISFDB came back with “Marc Testing and Marc Testing4 and Marc Testing3 and Marc Testing2 and Marc Testing1” meaning it’s also flipping names around at the title level.

Thus I’ll drop the alphabetical wording but add a note about that ISFDB does not maintain the author order. Marc Kupper (talk) 11:54, 19 Feb 2007 (CST)

I agree with Mike that the period (or lack thereof) after the middle initial is just a typographical convention like kerning or using commas between the last name and the suffix ("Jr.", "Ph.D.", etc). Although we are certainly interested in substantive differences like "C. M. Kornbluth" vs. "Cyril M. Konrbluth", I don't think we want to record minor differences in spacing and punctuation. If we did, we would end up with a lot of pseudonyms, variant titles and complex relationships that would, IMO, overall serve to obfuscate rather than clarify things.

If we adopt this proposed approach, then we may want to clarify the section(s) of the Help files where we state that we want to capture information exactly as it appears on the title page. Ahasuerus 15:54, 19 Feb 2007 (CST)

Unnamed "about the author" sections

We generally catalogue in the Contents section "About the Author", "Editor Bio", ect. essays. I looked at an anthology tonight that has a paragraph or two at the end of each story about the author. These don't have a title but are separated from the story itself in that they are written in italics. Should I add "About the Author (storyname)" for each of these in ISFDB? I'm assuming the reason for cataloguing these is help people that are trying to locate biographic data on an author/editor/artist. Marc Kupper (talk) 03:53, 17 Feb 2007 (CST)

I agree these would be helpful, but I don't think they're worth cataloguing. I think if they have a title they can be included, otherwise not. At some point maybe we could enhance the publication record to make comments about intros, afterwords, and so forth, that don't merit their own entry as contents. Mike Christie (talk) 17:03, 18 Feb 2007 (CST)

Mike, you wrote "At some point maybe we could enhance the publication record ..." Do you mean to add new title types besides Essay? Locus has a rather long list of codes some of which I see infrequently enough that I need to look it up.

ai - author's introduction to story, ar - article, aw - afterword, bg - biography, bi - bibliographic material, br - book review, cl - column, cn - contest, cr - criticism, cs - comic strip, ct - cartoon, ed - editorial, ex - extract, fa - facetious article, fp - frontispiece, fw - foreword, gp - group of related stories, gr - game review, ia - illustrated article, ibc - inside back cover, ifc - inside front cover, il - illustration, in - introduction, ins - insert, is - introduction to story, iv - interview, ix - index, lk - linking material, lt - letter, mr - movie review, ms - miscellaneous, pm - poem, pp - prose poem, pr - preface, pt - photography, pz - puzzle, qz - quiz, rd - radio review, rr - round-robin, sf - special feature, sg - song, si - section introduction, sl - serial segment, sp - speech, sr - story review, sy - symposium, te - true experience, th - theater review, and ts - true story.

Marc Kupper (talk) 21:37, 18 Feb 2007 (CST)

When doing the Spider Robinson edits, I could have used "Song" usefully. I used "Poem" instead, but some entries actually had musical arrangements, in which case it might be a useful distinguishing category. The Merovingen Nights series also has a lot of Filk in there, and the difference might be useful to musicians. I'm not one - I sing in the key of "Z" and it sounds like "L". ;-) Which hasn't stopped me getting four filks published and winning an award for one... but I don't recall any of the ones we have recorded winning anything yet? If anyone's interested, I can go track down the British Filk awards. Or add some of the song-books. BLongley 15:18, 19 Feb 2007 (CST)

Vandalism again?

See This page. I've been a bit late getting back to it, but I'm pretty sure I didn't put a rant about Child Sex Tourism in the notes. BLongley 16:48, 18 Feb 2007 (CST)

Since (at the moment) edits are not stored historically in a way that lets you see who did what, there's no way to tell who did that. When you see something like that, just go ahead and clean it up. With the moderator system in place, I don't think we'll see much recurrence. Mike Christie (talk) 16:55, 18 Feb 2007 (CST)

The vandalism, or erroneously inserted text, was done a time ago as I found the text in the version 1 database implying it got inserted before we had the reviewer/moderator system. I went ahead with editing the title tecord to remove the text. As it's not spamming a product/site here's the original title note for those wondering exactly what was there. It looks like a random chunk of text got appended to the note:

Note: This book was published after the original BBC Television series was brodcast, but before the motion picture was made (US Title of film: Five Million Years to Earth). discovers and fights "child sex tourism"--wherein pedophiles travel to Thailand to sexually molest children outside the reach of United States law enforcement.

Per Google, the ISFDB record seemed to be the only place this text existed. I searched ISFDB for any similar text (looking for each of the more unique words, one by one, in the text) and did not find any further evidence of text similar to this.

Also - I left it in place but the storylen is nvz which is a code I don't recognize. Marc Kupper (talk) 18:04, 18 Feb 2007 (CST)

nvz stands for novelization. It's yet another example of this poor overworked field being shameless overloaded. Al has mentioned that he was planning to move nvz/jvn information elsewhere, but it will probably take a while. As far as the original issue goes, it sounds like a cut-and-paste gone wrong. The source was possibly TV Guide or something functionally equivalent. Ahasuerus 21:58, 18 Feb 2007 (CST)

Thank you Ahasuerus - I added nvz to Template:TitleFields:Length. Marc Kupper (talk) 12:17, 19 Feb 2007 (CST)

Publication Prices

In a conversation with an editor about an SFBC publication I realized that Template:PublicationFields:Price is a little more ambiguous than I thought it would be. I’m bringing this up here as it might be a misunderstanding on my part. I had believed that

• We are entering the cover price as stated on the publication.
• If a price is not stated we would enter “None”.
• If your copy had a price but it was clipped or erased then enter “Clipped” plus a comment should be added to the publication notes explaining why the price is stated as “Clipped” and hopefully someone else with a copy of this publication can fill in the data and remove the comment. Strictly speaking, the comment is overkill we we try to make it clear that it's not the word "Clipped" that's stated in the publication.
• If you get the price from a catalogue or some other secondary source then add a comment to the Notes field describing that the price was not stated in your publication (or was clipped from your copy) and give the secondary source you used for the price. Please also note the publication date of the secondary source if it's a web site then use today's date. The goal being that people reading the comment in the future will understand fully where the price and at what times the publication would have been sold at that price.

Marc Kupper (talk) 17:33, 18 Feb 2007 (CST)

I've been leaving it blank or adding a note if I can't discern a price - "Clipped" seems too long. And inaccurate, as most of my pubs without a price have either been scratched out, blacked-out, or have had a sticker stuck over it so long that removing it would take the price off anyway. "Clipped" is usually reserved for dust-jackets where you can just take off the bottom right corner of the inside front without doing a lot of desecration, and I don't have many of those. (If I owned hardbacks rather than paperbacks, I'd need a larger house.) I'd like "blank" as the default for faster data-entry, but I agree any non-primary price should have an explanation. I do make especial effort to find a price if I can't verify the date either. (Anyone got the newbie guide to reading printing history notes from a number line sorted yet?)

One thing that might be worth including in the help - British publications might still have "Shilling" prices as late as 1973. My March 1973 edition of "The Dragon in the Sea" is listed as "30p (6/-)" for instance. (Although they tried to obscure the "30p".) BLongley 15:07, 19 Feb 2007 (CST)

I agree that "Clipped" usually comes from the hardcover book world. I'll think about blank vs. some other word, perhaps N/A. The reason I was thinking of a word rather than blank is that it makes it clear that someone looked for the price and could not find it. Marc Kupper (talk) 18:14, 19 Feb 2007 (CST)

OK, I'm already using "NA" as a place marker for titles that need to be removed from publications on my next edit - and it's shorter than "N/A". Perhaps we need a "proposed conventions" discussion somewhere? (Feel free to take that example from my talk page.) But as I'm still just editing, I'm looking at it from my "data-entry" point of view. The faster the better. BLongley 16:11, 20 Feb 2007 (CST)

No SFBC book comes with a price on the cover. Prices are available from several sources, the SFBC website if it is recent, Locus if it is over 2 years old, and receipts if you buy it new from SFBC (your first buy, you get several books at $1 each which is not a valid price). All the prices for SFBC books in the ISFDB came from this kind of source and those I looked at did not identify a source. Also please look at the first two entries on the "major contributors" page. About 100,000 entries were made from other than the primary source (the books themselves). For SFBC books, I believe that the SFBC Web Page is the most reliable source of a price, Locus is second best, and if neither have a price, no entry might be the best choice. rbh 21:44, 19 Feb 2007 (CST)

That's fine - I'm trying to figure out if I've misunderstood entering publication data into ISFDB or if the help pages need to be updated. I had been operating on the assumption that any time I enter information that is not stated in the publication, or is not immediately clear (estimating the publication date of a reprint from advertising for example), then I was adding notes explaining what was not stated in the publication and then listing the exact source of the data I entered.) The existing help pages are ambiguous about this.

My thinking is that if someone has a publication in hand and looks at ISFDB they should be able to tell if their publication matches a record in ISFDB. If someone enters a price for an SFBC publication but does not include a note explaining the source of that price then someone looking at the record will see that their copy does not have a price and conclude they have a different edition and there must be one out there that states the price.

I agree with you that there's around 100,000 records in ISFDB where there's no source data. I believe the goal of primary verification is we can hopefully start accumulating a set of records where we have a better confidence in that the data is valid and has been confirmed against the primary source (the publication) and when secondary sources were used then each one was documented. Marc Kupper (talk) 00:29, 20 Feb 2007 (CST)

Printing # Number Line

Anyone got the newbie guide to reading printing history notes from a number line sorted yet? BLongley 15:07, 19 Feb 2007 (CST)

Here’s an off the cuff response and I’ll see if I can edit it into a real “how to guide” down the road.

Most number lines are pretty straightforward and are one of:

• A sequence counting up – 1 2 3 4 5 6 7 8 9
• A sequence counting down – 9 8 7 6 5 4 3 2 1
• An alternating odd/even sequence - 1 3 5 7 9 10 8 6 4 2

With these the lowest number is usually the printing number. One exception is that sometimes you will see a sequence that counts up like "1 2 3 4 5 6 7 8 9 0" (or down from 0 9 8 ... 3 2 1) and in this case the zero is short for ten. Presumably that means the tenth printing would have just a "0" for the number line.

Things that make a bibliographer’s life interesting

• Some number lines also include a line of numbers that indicate the year of the reprinting. Usually these are easy to figure out as the years are stated as two digits but sometimes it’s not that apparent. For example I just grabbed a Microsoft Press non-fiction book and see “2 3 4 5 6 7 8 9   8 7 6 5 4 3.” The first part looks like the printing number and as the copyright is 1993 I figured the last part must be the last digit of the year meaning this is a 2nd printing done in 1993. An Aqueduct Press book used “14 13 12 11 10 09 08 07 06   1 2 3 4 5” which is a first printing printed in 2006.
• Some publishers, Random House comes to mind, have a number line that starts with “2” for the first printing and you then need to look for the words “First Edition.” Some people say that for the second printing they just erase the “First Edition” and others say that they will erase both “First Edition” and the “2” meaning the printing # is one less than what’s on the number line.
• Other publishers don’t use a number line but instead you will see a line of years. For example, “1985 1987 1987 1990.” In this case it’s a 4th printing done in 1990. There were two printings in 1987.

It’s my understanding that the reason the number line came about because it was (and maybe still is) easy to erase something from a printing plate in that you rub it out with a blue pencil (the plate is a negative image meaning white text on a dark background). Thus the original plate would have “1 2 3 4 5 6 7 8 9” and to do the second printing someone would just erase the “1” and they were set. To add a word to a plate meant redoing the entire thing. With modern desktop publishing that’s less of an issue but before 1986 when Ventura Publisher came out any additions of text to a book meant serious work.

Cheers, that makes sense. And the odd/even sequences means you can keep the number line nicely centred. (Even/Odd appears too, e.g. My "House Corrino" has a line 2 4 6 8 10 9 7 5 3 1, and I'm pretty sure that's first printing as I was waiting for it to be published in pb since I saw the tp.) BLongley 14:24, 20 Feb 2007 (CST)

Something I have not figured out yet is why publishers like Bantam would carefully note the dates of the first through nth printing where “n” is almost always a number that’s a couple before the printing of the book you are holding. For example, a Bantam 16th printing would list the dates of the 1st to 14th printing. I suspect that from time to time someone will make up a set of copyright page plates and store them until they are needed to run a printing. It means that to get the date of the 16th printing someone needs to find the18th or higher printing. Marc Kupper (talk) 18:14, 19 Feb 2007 (CST)

Another warning that might be useful - is 0 zero or is it 10? I just grabbed a couple of Bantam Spectra George Alec Effinger novels and one is "OPM 0 9 8 7 6 5 4 3 2 1" - which makes it look like it's high. The other from the same series, same publisher has "KR 0 9 8 7 6 5 4" which seems to confirm it - I guess a 0 in the centre of an alternating sequence would be a 10 and one on the edge might be a 0 though. (For review copies maybe?) But any ideas about "KR" and "OPM"? BLongley 14:24, 20 Feb 2007 (CST)

Just found an example of '0' on the middle - Piper's Federation has '2 4 6 8 0 9 7 5 3 1'. BillLongley 19:30, 24 Feb 2007 (CST)

The other thing that makes me nervous about printing histories is the number of imprints (not publishers) it might go through. I've seen British books go through Peacock/Puffin/Penguin editions that from page-count would probably all be from the same plates with different title pages, and they're all the same publisher. But do they restart the counting? OK, British publications tend to list the edition more clearly (Well, apart from the fact "First Publication in Great Britain" will often be printed on ALL the reprints too) and list a few previous editions instead, but in this age of international publishing I can't be sure the US isn't having the same problem now. BLongley 14:24, 20 Feb 2007 (CST)

Oh, and this is definitely getting off topic: but "NEL Open Market Edition 1985. First NEL Paperback Edition March 1986. Reprinted before publication 1986." WHAT got reprinted and what edition do I have? BLongley 14:40, 20 Feb 2007 (CST)

I'd guess the Open Market Edition is a trade paperback and "Paperback" is a mass market paperback. "Reprinted before publication 1986" means that the publisher did a print run but the market demand was high and so they did a second print run even before, or perhaps while, they were shipping the first printing to bookstores. You probably have the second printing of the MM Paperback. Marc Kupper (talk) 10:24, 21 Feb 2007 (CST)

What I tend to do is to look around for the most recent date. I've seen books where they will have "First hard cover: date, First paperback: date, First Canadian printing: date, First Mass Market: date, etc." I generally use the most recent date which is not always the last line in the list.

Still off-topic: "First NEL edition July 1968. New edition November 1969. Reprinted November 1970. New edition April 1972. Reprinted August 1972. Reprinted September 1973. New edition December 1974. New edition August 1975. New edition May 1976. New edition October 1977. Reprinted January 1978. This new edition July 1978." I wish it HAD just been a number line. :-( BLongley 14:56, 20 Feb 2007 (CST)

I had not seen that style recently. I agree it would be nice had the numbered the items 1 to 12. I'm wondering if the "new editions" are reprints with just a new price or if they are changing more things such as the cover art. Marc Kupper (talk) 10:24, 21 Feb 2007 (CST)

Anyway, thanks for starting this off Marc, hopefully we'll get a useful guide out of it. BLongley 14:24, 20 Feb 2007 (CST)

• I revised the text a little to add a note about the lines that start or end in zero.
• I believe OPM is Offset Paperback Manufacturers, http://www.opm.com, and would be the company that physically printed or manufactured the book.
• When I first saw "KR" I assumed it was a code for the year of publication but since have decided it's like OPM and designates a manufacturer or bindery. I could not figure out who KR is though I found some sites that lists the codes or the books that help explain the codes. Most of these are about identifying the first printing but presumably you can use the same ideas to figure out which printing you have. It’s my hope to some day to be able to build page(s) on ISFDB that are similar to these but focuses on the publishers that produce speculative fiction.
  • http://www.lazylionbooks.com/identifyingfirsts.htm
  • http://www.litterascripta.com/bibliomania/ident.shtml
  • http://www.backcreekbooks.com/columns/03012000.php

Just received the third and final book in the series, and that's got 'RAD' at the beginning of the number line. Nice to see some consistency from Bantam... :-/ BLongley 14:33, 28 Feb 2007 (CST)

• As for when publishers restart the counts - I've wondered at that too. With DAW Books I discovered that they usually restart the count for each binding. The hardcover will start at 1 and the next year the paperback also has 1 but they also usually will say "First Paperback Printing: date" When it comes to imprints it seems that most of the time they don't restart the numbering and really all that changes is whatever's at the bottom of the title page (Ballantine to Del Rey / Ballantine for example) and on the copyright page there will be a note that the book is was published by an imprint of a division of a company of a publishing group.

A problem I've seen with all of this is that it's hard to tell what information someone should be collecting. You open up a book there's a whole smorgasbord of codes and it’s hard to tell which are relevant. I recently started tracking the entire number line for DAW books just to see if there’s a pattern. They nearly always count up 1 2 3 4 … up to 9 or 10 but I’ve seen sequences ending in 11, 12, 13, 16, 19, and 20 plus one book that counted down with “20 19 18.” I just started doing this and after a while will see if there’s a pattern. Marc Kupper (talk) 01:43, 21 Feb 2007 (CST)

Moderator nomination process

I created some notes on moderator nomination process at Moderator Qualifications. I also edited the quals a little, de-emphasizing edit count a little per BLongley's recent comment about that. Please edit as needed to improve the process.

Are we yet at the point of needing another moderator? I will be limited in activity myself from around Feb 26 through Mar 17; I'll be on a business trip to Singapore and Australia, and then on vacation without net access for a week. Mike Christie (talk) 05:55, 20 Feb 2007 (CST)

Moderator Qualifications looks very good other than down at the very end it introduces the term "bureaucrat" without definine who or what these are. Recently there have not been times when I checked the New Submissions queue and found a large backlog. But, I personally would not mind more reviwers/moderators but it would also help if the suggestion about color coding moderator submissions gets color coded gets implemented so that we don't accidentially approve submissions a moderator may have chosen to leave in the queue while they work on things. Marc Kupper (talk) 15:43, 20 Feb 2007 (CST)

"Bureaucrat" does jar a little, but it's the way things are. Define it and have done with it, I'd suggest. I have seen a large backlog on my submissions at times, but I'm odd anyway: I just leave them for the next day. (OK, if the "pile of things to do" gets put back on the shelf for later due to lack of floor-space around my PC, I may miss an intended update, but I try not to do any damaging intermediary edits so me missing one shouldn't harm. I'll do a full pass when I want to update MY lists from the ISFDB rather than vice versa.) If people are really missing the guidance and/or approvals, then the case for a Limited Mod may come into place: I don't think we'll get it sorted as an absolute capability anytime soon, but there may be a few people that can be trusted with mod capabilities that will only use the ones they are sure of. BLongley 16:44, 20 Feb 2007 (CST)

The queues are still pretty long by late afternoon (Central US time) and on weekends, so it wouldn't hurt to have one or two more. Alvonruff 16:49, 20 Feb 2007 (CST)

They can also get quite long (dozens of submissions) around 2a-4am EST when, not surprisingly, there is a shortage of awake moderators in the US.

As far as the issue of "Limited Mods" goes, I wonder if simple restraint would work? Basically, if a new moderator sees a submission that looks complex or unusual, he can simply let it be and concentrate on his own submissions with occasional (and obviously optional) forays into approving straightforward submissions by other editors. The main reason why we have moderation in place is to prevent clueless or damaging behavior and once it has been established that a particular editor is both clueful and well-meaning (and also knows his limits), I think we can trust him to stay out of the areas that he isn't familiar with (yet). And if there are any questions or uncertainties, well, there is always this forum and other Wiki venues. The advantage would be that a number of active editors could approve/correct/verify their own submissions much faster, which would make their lives much easier.

That's often my mode of operation, but I always wonder if some think I'm leaving a burden on them. Alvonruff 19:09, 20 Feb 2007 (CST)

While we are on this topic, I should mention that "we have been found" (ObSF: D.F. Jones). At first the Wiki was only spammed by commercial spammers (casinos, porn, etc), but I have now seen some minor amateur vandals pop up. They are easy to deal with if you check "Recent Changes" a few times a day, but painful Wikipedia experience (I made many thousands of edits and patroled 600+ articles back when I was active) suggests that eventually we may be subject to much more sophisticated attacks by very devious and advanced vandals. The lengths that some people will go to in order to establish a "cover" and then subvert a community project are almost unbelievable, but it happens. There is not much we can do about it aside from being extra careful while reviewing submissions by new editors. Ahasuerus 18:18, 20 Feb 2007 (CST)

Sounds like the plot of an SF story. Any links or references that describes these kinds of attacks, or should we just sit back and enjoy them like a clever novel? (first post from a UMPC, using a stylus. Now where are those tildes?) Alvonruff 19:09, 20 Feb 2007 (CST)

One example of a devious Wikipedia vandal would be Wumbo. See, for example, his clever edit of the wikipedia:Propaganda of the Deed article. Note the self-referential reference ([[Propaganda_of_the deed#List_of_assassinated important_figures and_other_propaganda_by_the_deed_acts|Lev Bronstein]]) inserted in an otherwise innocuous multi-section edit, which he later used to subert the Leon Trotsky article. He was eventually exposed as a vandal, but not until after hundreds of his edits had contaminated Wikipedia. He apparently came back a few months later and resumed his activities as wikipedia:User:Skrooball.

I don't know whether the ISFDB will ever attract this type of sophisticated vandal from outside the SF world, but there are people in the SF community with a rather, er, peculiar sense of humor, so I figure that it's better to be safe than sorry. Ahasuerus 16:47, 21 Feb 2007 (CST)

In reading over the process, I see that you are seeking input from editors on specific nominations. We have little to work with to make a judgement, we don't see enough of another editor's work (mostly comments on this page) to make an informed judgement. I've noticed that the most recent nomination only has responses from existing moderators which makes sense as they have seen the nominee's body of work and can make an informed judgement. rbh 06:57, 24 Feb 2007 (CST)

I agree to some extent, though any given moderator may have seen only a small fraction of an editor's edits. There are a couple of things that other editors do get to see that might be relevant, though. The main one that comes to mind is communications skills. If you've seen an editor get short-tempered in an exchange, that might be grounds for an oppose or a neutral comment on the nomination.

I've nominated two editors so far. My first criterion is whether the edits I've seen demonstrate an understanding of the data structures -- not necessarily complete, since my own understanding is incomplete, but thorough enough to be safe. I think most editors reach that point within a hundred or two edits, though there are some gotchas that tend to take longer to emerge, such as editing titles of content items. My second criterion is whether I have seen enough of their wiki communication to be confident that they would be both accurate and patient in working through any changes with editors whose edits they moderate. This assessment is something any editor can make equally well.

Oddly, this means that editors whose work is exceptionally accurate and rarely raises questions will take longer to establish their communications abilities, as they won't be called on to respond to wiki questions very often. For example, A.kesrith has almost as many edits as Scott, but has only made two edits to the wiki so far, so it's harder to judge communications skills. Mike Christie (talk) 08:54, 24 Feb 2007 (CST)

I think it often takes more effort for an editor to review another editor than a Mod would take, but it doesn't mean it can't be done. I'm not doing my edits in a particularly planned fashion, and will often go off on a tangent - one of which is often looking at recent updates and saying, "Hey, I have some of those..." and adding my pubs to a hot topic. I'm quite willing to question a verification or a title-merge or somesuch when I'm sitting here with a Primary Source. Or I'll go off and find another pub with more data: I've done that for as little as a misplaced apostrophe in a title. So long as you don't get into a mind-set of "Mods know all" you can question, argue, or frankly rebel against someone else. But how you do it, and how they respond, will be an example of communication skills. If a Mod ever pulled a "I'm the Mod, you aren't" card I'd demand a Mod removal process. Thankfully, I've not seen a need for one yet. BillLongley 16:29, 26 Feb 2007 (CST)

Ballantine editions of Sarban

I thought I'd spend time cleaning up the Sarban page, but ran into a snag: the Ballantine editions of Ringstones and The Doll Maker are not collections, but short novels, and they omit the other short stories in the originals. How does one handle this? Luckily, the titles are different - should I just unlink them from their predecessors and recharacterize them as novels? (Scott Latham 20:16, 20 Feb 2007 (CST))

Did you already fix this? Sarban's page looks ok in that each of the publications seem to be matching the titles they are filed under. Marc Kupper (talk) 00:25, 21 Feb 2007 (CST)

No, Marc, I haven't done anything yet. The point I'm trying to make is that the original Peter Davies Ringstones and Dollmaker are collections, and the Ballantines with similar titles are not: they reprint only the longest piece from the Davies editions, not the full content. (Scott Latham 16:51, 21 Feb 2007 (CST))

Sorry - you had said the titles were different and when I saw that all of the titles were the same I figured it had been taken care of. If the titles are different, The Doll Maker for example, then just rename the publications using edit-pub and unmerge them using "merge titles." As there are two Ballantine publications you will end up with two separate title records for The Doll Maker which you would then merge using the "Title" option from the author display.

If they have the same title then I suspect we are still ok as you could still unmerge a publication and then change the type for both the title and publication to NOVEL or COLLECTION. As they have the same title I'd add notes explaining that both standalone novels and collections exist with the same title.

The first thing though is to get the publication names corrected using edit-pub. Will you be working on Sarban in general? I see that Ballantine published a standalone copy of Ringstone too. Marc Kupper (talk) 19:50, 21 Feb 2007 (CST)

ebooks, Gutenberg, CDs, etc.

Web only publications are out for good reasons (although there is a part of me that hates ignoring data). The list of what is cataloged lists ebooks however. So should I add the Baen ebooks. Also Baen has done a dozen CDs in various books that each have 25 or so (very overlapping) books on them. Then there are the Gutenberg versions of out of copyright SF including things like the copyrights on H. Beam Piper weren't renewed. Add in writers like Doctrow and Stoss that put copies up of their current books. I'd think that if there is a physical publication listing those as publications makes sense. Dana Carson 03:35, 21 Feb 2007 (CST)

It seems that if the eBook is a "reprint" of a physical publication then it should be included in ISFDB. Indexing the Gutenberg titles has some appeal to me as then someone looking would at least be aware that the story can be read on line for free rather than chasing down a dead-tree version but let's see what the concensus is. Marc Kupper (talk) 19:54, 21 Feb 2007 (CST)

I've added links to online versions of some stories, but only in the notes of a dead-tree publication. If there is a VERY stable site I don't mind adding that sort of information, but I want to be careful about not pointing at some of the pirate sites. (I know that my own physical collection is dwarfed by the number of SF publication available on some Russian websites, for instance.) BLongley 12:29, 27 Feb 2007 (CST)

Is it safe to merge cover-art records?

If I see multiple title records for what I know is the same artwork is it safe to merge them or is this likely to break something in ISFDB's internal handling of the covert art records? Marc Kupper (talk) 19:24, 22 Feb 2007 (CST)

I have merged cover art records in the past and it didn't break anything, so it's presumably OK to do from the technical point of view. The only reason I have been hesitant to merge them more often is that it's hard to tell whether the art is really identical unless you have both editions/printing at your side. Most of the time it will be, but there are enough exceptions to make me nervous about it. Ahasuerus 17:58, 23 Feb 2007 (CST)

Canonical Title vs. Variant Title: Which is which?

Our wiki glossary defines Canonical title as "the title under which all publications of a particular work are listed. For stories or novels that appeared under more than one title, the canonical title is usually the first title for that work, but may be a later title if that title is much better known." I was surprised when I went to enter my Dell paperback copy of Clifford Simak's Time and Again to find that its variant title First He Died was being used as the canonical title. The true first edition (not yet entered) was titled Time and Again, and all the other publications listed as of yesterday use that title. Does anyone object to my changing the canonical title to Time and Again and entering my Dell copy as the variant title? (Scott Latham 11:11, 23 Feb 2007 (CST))

Oh, sure, this is certainly a mistake and needs to be corrected. There is a similar problem with "Time Is the Simplest Thing" aka "The Fisherman". Also please note that the display logic for Serial-Novel associations is based on simple lexical matching. It works when a Serial title matches the main Novel title, but it doesn't work when it matches a Variant Title. Thus, once you flip flop "Time Is the Simplest Thing" and "The Fisherman", the Serial publication will become an orphan and will be displayed under "Serials". This is not a data issue, but rather a known display bug and will be fixed in due course of time; nothing to worry about :) Ahasuerus 13:37, 23 Feb 2007 (CST)

Nominating Scott Latham for moderatorship

Ref: Moderator Qualifications#Becoming a moderator for the nomination process.

Nomination statement

I nominate Scott Latham (talk • contribs) for moderatorship; he has accepted the nomination. Scott has 516 edits to date, which is fewer than some other active editors. However, Scott has been thorough and detail-oriented in the work I have seen. I would like to point to this long exchange with Marc Kupper, which despite a complex situation and some slight misunderstandings led to a good resolution of the problem in a very good-natured way. I believe Scott would be a patient and thorough moderator, and would be a valuable addition to the moderator ranks. Mike Christie (talk) 14:48, 23 Feb 2007 (CST)

Support

• Support, as nominator. Mike Christie (talk) 14:48, 23 Feb 2007 (CST)
• Support. Scott has been a careful and cautious editor in my experience, so he should make a good moderator. Ahasuerus 18:00, 23 Feb 2007 (CST)
• Support I'll second that, plus I learned about City of Truth which I just finished and enjoyed. Marc Kupper (talk) 18:00, 23 Feb 2007 (CST)
• Support. Sounds good to me, the edits I've seen have all looked good. --Unapersson 08:02, 24 Feb 2007 (CST)
• Support. Alvonruff 13:07, 24 Feb 2007 (CST)
• Support. I've had disagreements (well, two) over edits in the past, but in the end it all worked out for the better, and we both learnt something. I trust his intentions, and feel sure he's not going to go "Yay! I'm a Mod!" and go crazy. ;-) BillLongley 17:53, 25 Feb 2007 (CST)

Oppose

Comments/Neutral

Nomination closed. I have promoted Scott to moderator per the consensus. Mike Christie (talk) 18:29, 1 Mar 2007 (CST)

Bob Shaw

I have finished physical verification of my Bob Shaw books, including a collection of his fanzine writings and a few other editions that were not in the database or had not been verified. The only issue that I ran into was a previously Verified edition that had an extraneous "The" prepended, but otherwise it looked pretty good. We are getting somewhere :)

I may be over-paranoid, but was the "extraneous 'The'" one of mine? During the downtime (or my personal downtime, as you all seem to have had two days of edits without me) I've been reconsidering some of my posting styles, e.g. adding page numbers to already verified works. (I find it makes me spot added/deleted content far more easily if I do that, but it also leaves me with far more variants to sort out afterwards.) Oh, and the downtime also allowed me to order another 4 or 5 pubs which I will have to get round to. BillLongley 18:06, 25 Feb 2007 (CST)

Nope, it was Shadow of Heaven (1969), which was revised for Gollancz in 1991 and published as The Shadow of Heaven. The Avon first edition was erroneously listed with a prepended "The" and marked as Verified by Mark. I corrected the error based on my copy and then Unmerged the Gollancz editions and turned them into a variant Title. I am sure verification mistakes will happen from time to time since other, more established sources like Locus/Contento occasionally drop articles or even more consequential words. Heck, I made a Verification mistake just a few weeks ago!

As an aside, although we have often discussed this fact here, it's worth repeating for those who weren't around at the time: US/UK editions should be entered as published including added/dropped articles (a transatlantic specialty!) and variant British/American spelling. We can then create Variant Titles as appropriate. Ahasuerus 01:43, 26 Feb 2007 (CST)

OK, glad to hear it. That's what I've been doing: it does get a bit complex when you're adding variant author names as well as variant titles (I normally put a book aside to verify it after the update, but sometimes forget I changed the author AND the spelling AND the date, e.g. Gray/Grey Lensman by Somebody-or-other Smith). BillLongley 16:40, 26 Feb 2007 (CST)

Unfortunately, my collection uses a highly efficient but non-author based storage algorithm, so doing author-based projects can be rather time consuming since it requires retrieving books from fdifferent areas. Once I have more permanent access to my collection, I could go from #1 to #20,000 linearly, but who knows when that may happen... Ahasuerus 19:20, 23 Feb 2007 (CST)

BLongley

Just an FYI for anyone trying to contact Bill Longley: he is running into trouble connecting to the ISFDB -- it may be that his IP number is being blocked by TAMU for some reason. He is working on trying to fix it. If anyone wants to contact him, let me know and I'll get his email address to you in some non-spam-harvestable way. Mike Christie (talk) 08:25, 24 Feb 2007 (CST)

No luck getting a new IP address, my ISP keeps giving me the old one back. I'm using IPhide.com to get here at the moment, but as I don't fully trust proxies I'm using a separate account. Hopefully we'll soon figure out why I can't get here direct and I can go back to the old one. BillLongley 12:44, 24 Feb 2007 (CST)

Yay! I'm back! Did anyone find out what the problem was? And is there any way to do an Editor merge? (Not that I'm overly worried about my edit counts, of course...) BLongley 12:15, 27 Feb 2007 (CST)

Awards Data

There's a Terry Carr anthology that has an award listed against the title "Best SF Novellas of the Year 1" whereas the actual book has the title "The Best Science Fiction Novellas of the Year #1".

If the title records are merged will the award data remain as it is? --Unapersson 18:06, 25 Feb 2007 (CST)

This was fixed a couple of months ago and merging Title records should result in all related Award records getting merged. There is no such thing as a 100% guarantee in this business, but it seems to be working fine at the moment (fingers crossed). Ahasuerus 17:08, 26 Feb 2007 (CST)

Are the awards still lexically matched or how are we supposed to deal with cases where the official awards list uses what we'd call a "variant title?" Marc Kupper (talk) 18:14, 26 Feb 2007 (CST)

Last I checked, Al was putting a basic Award Editing application together, so he probably has the latest scoop :) Ahasuerus 19:01, 26 Feb 2007 (CST)

Publication showing up twice

Skylife: Space Habitats in Story and Science has 2 publication entries. They both lead to pub #30757. Is this a display bug or am I missing something? Dana Carson 18:43, 26 Feb 2007 (CST)

The problem is underlying data corruption which is masked by the display logic. There is only one publication record in this case, http://www.isfdb.org/cgi-bin/pl.cgi?SKYLIFE2000 , but it contains two pointers to the Title record, which confuses the application that is responsible for displaying Titles. You can't see the duplicate pointers in the Publication record since the Publication display application won't display ANTHOLOGY Titles for ANTHOLOGY Publications. (The idea is that our users want to see the stories and essays contained in Anthology/Omnibus Publication records, not the Anthology/Omnibus Tiles that they point to.) You can confirm this by converting the Publication record to a, say, NOVEL and then pulling it up via Edit Publication -- you will see two entries for the Title.

I have seen this problem a number of times, but I am yet to determine what causes it. I think it has been reported in the Open Bugs area, although perhaps incompletely. I am afraid the only way to fix a Pub record with two identical Title pointers (that I know of) is to delete and re-enter it :( Ahasuerus 19:00, 26 Feb 2007 (CST)

I screwed up the other day and succeeded in making this happen. If you look through the Bob Shaw books you will find one publication that was deleted and re-added…

With that Bob Shaw story there were two title records with the second having something like “(UK 1991)” appended to its title. There was a publication under one title but it referenced the other one in the Contents. I was researching this and realized I should merge the titles and then do an unmerge as the USA and UK editions were not all filed under the correct title. The merge did give me a big fat warning about the two titles being referenced by the same publication which I ignored. I then went to the publication and changed its title type to something else to expose the normally hidden parent title link. When I did the delete-title to get rid of the extra title link ISFDB deleted both of them! (Marc has shot himself in both feet now after poking himself in the eye). I now had a very lonesome orphan of a publication record that had zero title references. Fortunately I knew there was a good chance I would shoot myself in the feet and had been keeping notes on the URLs involved and so I created a new title record using the data from the original and deleted the orphan. Marc Kupper (talk) 19:27, 26 Feb 2007 (CST)

OK. I'll delete and reenter it, it needs contents entered anyway which is most of the typing. Dana Carson 20:15, 26 Feb 2007 (CST)

As an experiment I cloned it and the clone only has one title record meaning I had not needed to delete/reenter the publication before. You can see the clone as I just appended "(clone)" to the title. Marc Kupper (talk) 21:51, 26 Feb 2007 (CST)

Verification

I've read the help again, and it's lacking. WHAT are we verifying? I see people verify Collections without bothering with the contents. Which is fine, we at least know the pub exists... but an unverified pub may have more detail of contents, that presumably someone with a primary source COULD have verified. Or I add an Artist. Or an introduction, or page numbers... I hate adding stuff to an already-verified publication, it makes me feel I'm criticising the verifier. So what is the level we're verifying? BillLongley 16:56, 26 Feb 2007 (CST)

Bill, as you know, I'm not an expert, but what I intend when I do a primary verification of a publication is that I've had a copy in my hands and I've compared all the data fields with the actual book. The one area where I tend to defer to another editor is on attribution of uncredited cover artists: I assume whoever made the entry knew more than I about the artist, or had other research materials. All that said, I know I've made mistakes in some verified entries, and appreciate it when someone fixes them. And I've made small changes (e.g., corrected page counts) to some verified entries in the same spirit. (Scott Latham 18:00, 26 Feb 2007 (CST))

I will compare/correct all data fields, the contents, and link to a cover image. If there are blanks in a verified record then I fill them in. For example, I added an interior artist to a publication yesterday. If a verified record states something that does not match my copy then I ask the original verifier to check their copy as there's always the chance the publisher made a mistake.

Scott wrote: "The one area where I tend to defer to another editor is on attribution of uncredited cover artists: I assume whoever made the entry knew more than I about the artist, or had other research materials." If you see something in a publication that is not apparent to you then you should query the original verifier as they should have added a note explaining the cover artist or interior artist was not credited and then listed the source of the data. For example, a couple of days ago I had a book where I found that ISFDB already had a record though not verified. This record stated a cover artist but I could not find any evidence in the book itself as to who did the artwork. In that case I deleted the cover artist's name but also added a note explaining that the original ISFDB record had credited so and so but I could find no evidence in the publication itself that he had done the cover.

I will compare/correct all data fields, the contents, and link to a cover image. If there are blanks in a verified record then I fill them in. For example, Saturday I added an interior artist to a publication. If a verified record states something that does not match my copy then I ask the original verifier to check their copy as there's always the chance the publisher made a mistake. Hopefully people do the same with me as I want people to be looking over my shoulder.

Scott wrote: "The one area where I tend to defer to another editor is on attribution of uncredited cover artists: I assume whoever made the entry knew more than I about the artist, or had other research materials." If you see something in a publication that is not apparent to you then you should query the original verifier as they should have added a note explaining the cover artist or interior artist was not credited and then listed the source of the data. For example, a couple of days ago I had a book where I found that ISFDB already had a record though not verified. This record stated a cover artist but I could not find any evidence in the book itself as to who did the artwork. In that case I deleted the cover artist's name but also added a note explaining that the original ISFDB record had credited so and so but I could find no evidence in the publication itself that he had done the cover.

I believe the original question of “WHAT are we verifying?” is a very good one as I’ve often wondered this myself. Speaking for myself, both as an editor and moderator, is that I am trying to leave behind a trail of reliably sourced data. When I enter something into ISFDB I also try to indicate exactly where I’m getting the information from. With a publication the implication is that whatever is stated is a verbatim copy of what’s in the publication. If I need to regularize a title or author name then I add a note explaining what was stated in the publication and that I entered regularized the data. I’m also entering printing number information in the notes field as we have one ISFDB record per printing but at the moment ISFDB does not have the separate data fields to help distinguish the printings. Any time I spot something that I could see as being a source of confusion such as the title, author(s), ISBNs, etc. being slightly different then I will document exactly what’s stated where. If I find that something about a publication is frequently stated wrong elsewhere on the Internet I’ll add a note such as “Other sites often list the title as “abc” but it’s stated as “xyz” on the title page, front cover, spine, etc.

The goals of what I enter are:

1. That others will be able to look at a publication and feel very comfortable about if their own copy of the publication matches the record or not. Or, if they don’t have a copy of the publication people should be comfortable with if the data matches or does not match other sources.
2. I link as much as possible including lesser roles such as book designers. For example, this morning I entered Traitor's Sun with a pseudo-title to deal with a co-author that was only credited on the copyright page.
3. Related to linking is that I also add contents for secondary items such as “About the Author” pages so that if later someone is trying to locate an official bio they should be able to look under “Essays” for that person’s bibliography. Marc Kupper (talk) 19:01, 26 Feb 2007 (CST)

Let me try this again: I'm not asking about what else people would add (although it might be interesting to know what level of detail each editor is willing to go to), it's about what minimum level of information it is useful to verify, and how people react to a "Verified" publication rather than an unverified one. For instance, if I've verified a publication of unknown date and price I'm not really saying anything except that there is a physical copy of the book with that name, author and publisher: I'm not able to add anything that would help distinguish it from any other edition. Leaving it unverified may make someone else go and try and find out more from other sources: verifying it might make someone prod me for more information that simply isn't on my publication. I can see a case for NOT verifying something if somebody else might be able to add the information from their copy - e.g. if my copy has the price scratched out, leaving it for someone with an un-mutilated version might be better. If we all know what "Verification" actually means, then I might save some verifications, or other editors might know that it doesn't mean the Verifier has necessarily gone to the same level of detail as they would. BLongley 12:09, 27 Feb 2007 (CST)

Good issue - I don't think we have addressed partial verification.

My personal practice is that if for whatever reason I have a partial verification then I will go ahead with adding/verifying as much information as I know, I will mark the publication as verified, but I will also add notes explaining the circumstances and which fields have been checked and which may be in error. If I used something like Amazon’s Look Inside and all I’m missing is the page count and I’m comfortable with marking this as “verified” though I’ll definitely explain that this is based on an Amazon Look Inside and that the page count may be wrong. Amazon’s Look Inside is the only time I mark publications as “primary-verified” when I did not have a physical copy of the book in hand and even with that I only verify the publications where the copyright page is included.

When I see "verified" what does it mean to me? Well, it tells me it's likely the person has a physical copy of the book (or knows where to get it again) and that their copy matches all of the data stated in the ISFDB record. if I have a question about that data that I can ask them. Ideally, it also means that all of the metadata (the upper part of the publication listing) has been entered and double checked. If there are contents entered I'd also expect that they have also been double checked. If I see a verified publication that is missing its contents I may send a note to the verifier asking them to enter and double check the contents. If they no longer have the publication then a note should be added explaining that the metadata was verified but that the publication itself is no longer available.

That could get tricky after a while. Contributors may wade in and out over time (some of us are even mortal); collections may be sold, lost or otherwise change hands; certain physically verified items may no longer be readily available if they were verified at a library or a bookstore. Ahasuerus 19:49, 27 Feb 2007 (CST)

I know that I'm taking this editing burst as a sign that I really SHOULD clear out some books I have to admit I have no actual intention of reading. It's not working yet - for instance, nine books have arrived this week so far and they'll take up more shelf-space than the seven I intend to discard, having entered and verified those seven: but I'm sure my cleaner will eventually nag me into verifying some of the crap that prior girlfriends left here. At that point, you have no more than a week to get me to confirm the details of my David Eddings or Piers Anthony before I turn them into cash, fertiliser or ashes. BLongley 15:22, 1 Mar 2007 (CST)

I do double check verified publications as I already have the book in hand and so it’s quick scan.

Related to this is that, ideally, if there are any changes to a verified publication, then the person doing the changes should also send a note to the original verifier explaining what is being changed so that the original copy can be re-checked. It might be a worthwhile feature request to make this automatic if a person other than the original verifier makes any edits then the original verifier gets notified. The goal here is to prevent edit-wars where it turns out there are two copies of a publication that are identical except for one thing. In that case the editors would be able to add notes explaining how to distinguish the publications. Marc Kupper (talk) 14:28, 27 Feb 2007 (CST)

Thanks Marc, but AGAIN we've wandered off topic into "what I would edit" rather than "what would I verify". I think we've established that you will not take verification as an absolute end to a publication, there might always be an improvement, and so will check it anyway. For you, "verification" is almost meaningless. For others, it might be a sign that "enough has been checked", and so they'll go carry on elsewhere that they feel might be more productive. It's that level I'm looking for: verification may be a BAD thing if there's not enough info, but verifying it puts people off further edits. Remember I'm complaining about the Help Text: we're asking people to choose yes or no about whether a publication record is "good enough", without enough guidance. BLongley 16:46, 27 Feb 2007 (CST)

Well, the original problem that we were trying to resolve by intoducing the Verification flag was that we had thousands of records whose origins were uncertain. We had vaporware titles, multiple records for the same Publication, records that were plain wrong or hopelessly confused, etc. The Verification flag gave us at least some degree of assurance that the book/magazine actually exists and is not a figment of some Webbot's imagination. It is not a guarantee that every Verified record in the database will be thoroughly documented, but we should probably edit the Help pages to ask editors to add as much information as they can before marking a Publication as Verified.

As far as "verifying it puts people off further edits" goes, well, it depends. If I were going through my collection sequentially and had plenty of time, I can see myself spending the extra 30-60 seconds that it takes to confirm that a previously Verified record was entered correctly. But if I were in a bookstore or a library and time was at a premium, I would probably go for the low hanging fruit first. Ahasuerus 19:49, 27 Feb 2007 (CST)

BLongley, how about if you propose how you would like the help text to be edited. There seem to be two pages unless you want to get into what the goals of ISFDB are, etc.

• Help:Screen:Verify

OK, I've had a stab at the first. I'm trying to balance the fear of editing versus the potential laziness of verifying. BLongley

• Help:How to verify data

For example, should we encourage people to double check already verified books and to start a discussion should their data not match what’s in ISFDB? That is my own operating procedure but I had not thought about making it “the rule” because, as you noted, others have different views and priorities.

Is there is a need for a "preliminary verification" flag that could be used by someone that will be checking minimal data or doing verification off sources such as bookseller listings? Would these data records would be of higher quality than the dissembler-pruned-by-Ahasuerus method we have used for years? Marc Kupper (talk) 03:32, 1 Mar 2007 (CST)

Just to make sure that we are all on the same page, here is the approximate sequence of events. Dissembler goes out into the WWW wilderness, grabs whatever looks like SF biblio records that are not already in the ISFDB and creates regular ISFDB submissions. Al reviews the submissions, identifies any weaknesses in the acquisition algorithm and either rejects or approves them. As he recently put it when Dissembler grabbed a book about lesbian bikers, we would only want to include it if it was about alien lesbian bikers. Next I pull up all Publications published during the current month, e.g. http://www.isfdb.org/cgi-bin/fc.cgi?12+2006 (yes, I am way behind at the moment), and review them one at a time, merging duplicate Titles, creating Variant Titles and Series, verifying Publication records against online sources, etc. I then review the rest of each affected Author's bibliograpy and try to clean up their Titles. I wish I had time to clean up all Publications for the Titles that I review, but there are only 24 hours in any given day and I have other responsibilities :( Ahasuerus 15:54, 1 Mar 2007 (CST)

ps: I woke up this morning thinking about "levels of verification" where "Level zero" would apply to the existing dissembler generated data, level 1 means a human has made “some” effort to see that the record corresponds with reality, and we'd then have succeeding tiers of levels up to a “gold standard” where the parent title record would also be verified and filled in (synopsis, tags, etc.). I don’t have time today to detail the levels I had in mind as it includes voting numbers (multiple people need to physically check the publication before it could get promoted to “gold standard”. Or – we could use the same terminology as the wikipedia to stamp that authors, series, titles, and/or publications have been brought up to a certain quality standard. Marc Kupper (talk) 14:14, 1 Mar 2007 (CST)

It might be useful once things are stable to add some canned searches that let people do things like pull up collections/anthologies without content entries, books without cover info, etc. so they can fill in the missing data. Dana Carson 20:54, 1 Mar 2007 (CST)

Author pseudonyms

Sticking to "As in the publication" looks fine to start with, just needs us to add variants. But adding my John Wyndham pubs tonight gave me the problem of Title Page admitting that it was "John Wyndham writing as...". Sometimes the cover and spine mention variants too. I've stuck with First-named for all titles, but I could understand doubling up the Author names. Trebling up the names looks overkill... but should it be done? BillLongley 17:11, 26 Feb 2007 (CST)

The thinking is that a person or title has a canonical name or title and that anything else is a variant. Thus when Kenneth_Bulmer is writing as Alan_Burt_Akers who in turn is writing as Dray_Prescot then Kenneth Bulmer is the canonical name and Alan Burt Akers plus Dray Prescot are both pseudonyms. We don't try to go three deep.

That's what I thought... there's an awful lot of late pubs by 'John Beynon Harris' that I frankly don't believe had that name on though. BillLongley 09:37, 27 Feb 2007 (CST)

We generally enter what's on the title page into ISFDB. If your title page says "John Wyndham writing as..." then the credited author is John Wyndham but you should add a note explaining what's on the title page, cover, and spine. This is also a great opportunity to add a note to the author's bibliographic notes giving the stated source for a pseudonym. Marc Kupper (talk) 19:12, 26 Feb 2007 (CST)

How I would improve the Summary Bibliography

What I'd really like to see, perhaps more than anything else, is an additional "Notes" box in the "Author Data" imput on the "Summary Bibliography" page for every author. For Donald Hamilton, for instance, I would like to be able to add "Hamilton has written 22 straightforward espionage thrillers about the U.S. government assassin and counteragent Matt Helm. Two of these novels, however, contain minor elements that are speculative fiction in nature, and thereby merit inclusion in this database." I would like to be able to add similar statements for a number of other authors or various other Notes. This would, I think, be a lot more useful for the casual user of this DB -- you click on an author's name and there is the prime info about him, not buried two or three layers deeper in links that you have to work your way through. Sure, keep some of the same info in the Notes for each individual novel the way it's done now. But put some overall info in the Summary Bibliography. Hayford Peirce 17:20, 26 Feb 2007 (CST)

This comes up from time to time. Al has a reasonable sounding reason on why the author fields don't have notes but I've forgotten what it is and can't remember a good set of keywords to help search for the old conversation. Marc Kupper (talk) 19:41, 26 Feb 2007 (CST)

It wouldn't hurt for series to have a notes field also. Dana Carson 14:15, 27 Feb 2007 (CST)

US vs. CDN printings and entering the price

(this was originally on Kraang and Marc’s talk pages but I’ve shifting it to the community portal to give everyone a heads up). Marc Kupper (talk) 18:07, 26 Feb 2007 (CST)

Marc-From 1978-ca.1990 before dual pricing del put out a CDN & US printing. The CDN are always higher to reflect the exchange rate. With multiple entries would it be helpful to put the letter "C" before price(ie. C$2.25) on CDN printings to help identify faster. Other publishers had similar pricing policies, but most only printed in US. You end up with two identical books(ie. printings) but priced $3.95 & $4.95. See Chalker's The Demons at Rainbow Bridge (ACE) for example.Kraang 20:15, 23 Feb 2007 (CST)

See ISFDB:Community Portal/Archive/Archive03#Codes for Foreign currency - I'm updating Template:PublicationFields:Price to reflect this. BTW - how can you tell it's a USA vs. Canadian price in the Del books? Marc Kupper (talk) 00:25, 24 Feb 2007 (CST)

Marc- One of my collecting interests was Chalker 1st printings.From 1978 to present the exchange rate varied from 10% to 60%. IF you compare my VERIFIED books with similar dated books you will see the "Printed in Canada"(on copyright page) are always higher in price. A perfect example is Chalkers "Exiles at the Well of Souls" both 1st printings $1.95(Manufactured in US) & $2.25(Printed in Canada). Being in the US you would not see the alternate priced books except as used copies(most of these would be in CDN) Most of the books on the ISFDB are sourced from US. Dual pricing made this issue disappear. If its agreeable i will go and add the letter "C" to the "Printed in CDN" on the ones i have personally seen. It may be helpful to put a note somewhere for editors about the DEL books to include the printing info in NOTES. On single priced books the printing info(on copyright page) is significant others may be unaware of this.Kraang 20:21, 25 Feb 2007 (CST)

Yes, I agree that it seems that if a book is single priced, and it states “Printed in Canada” that it should be entered into ISFDB as C$2.25 but also to add a note explaining what the book states. You have brought up an interesting issue in that it’s not apparent from casual inspection that someone would be looking at a price stated in Canadian dollars versus U.S. dollars.

I believe we should add a note to the publication price template about if the publication states a single price in dollars to check the country of printing and if it’s not the USA to both code the price (C$ for Canada for example) and to add a note that the stated price is just $xxxx and that it was entered in ISFDB as a Canadian price because the publication states it was printed in Canada.

It turns out I have two copies of Marune: Alastor 933 by Jack Vance.

• The covers are identical except one states $2.25 and the other $2.50.
• The cover layout of where DAW positioned things is consistent with the books being a first printing.
• The spines are identical except one copy has 225 after the ISBN and other has 250. This is the price code and agrees with the front cover prices.
• The back covers, front-end-paper, frontispiece, the story, and advertising, including book codes and prices, are identical.
• On the title page of the $2.50 copy the text "DAW BOOKS, INC. ... New York, NY 10019" has been shifted up 7/16th of an inch and below this, in a different font and not quite aligned (indicating it was pasted in after the fact) is "Published by / The New American Library / Of Canada Limited".
• The copyright pages are identical ("First DAW Printing, January 1981 / 1 2 3 4 5 6 7 8 9") except down at the bottom where "Printed in U.S.A" in the $2.25 copy was replaced by "Printed in Canada / Cover Printed in U.S.A."
• An irony is that my USA copy has a bookstore stamp from "Book Rack" in Madison, WI which you’d think would be more likely to have a Canadian edition on its shelves.

BTW - something I'm curious about is that the advertising in the Canadian edition must be stating USA prices. If a bookstore carried a "Printed in USA" title did it sell the thing for the cover price or was a sticker applied to factor in the exchange rate? Marc Kupper (talk) 18:07, 26 Feb 2007 (CST)

Pass-on

Could another moderator take a look at the three subs I've held? I think two may be duplicate pub pointers to the same title, and the third is an odd-looking MakeVariant that needs looking at. I'm off to Singapore in a couple of days, then on vacation, and may not be around much; and I don't think I'll get to look at these between now and then. Oh, and someone else should probably close Scott's nomination when it gets past five days, which I think is tomorrow. Thanks -- Mike Christie (talk) 22:27, 26 Feb 2007 (CST)

I took care of the subs.

• One of them was a publication that had two title links back to the parent and so I cloned it first creating a record with just one title link back to the parent and approved the original pub-delete. That clone-pub can fix this is really handy as it was a collection with about a thousand stories(well - not really a thousand but it seemed about a full screen-full.)

Interestingly enough, I own a couple of books with 1,000+ Titles (reviews) each. Al swears that it shouldn't break anything in the application. I can't say that I am looking forward to entering them, but it will need to be done at some point... Ahasuerus 10:14, 27 Feb 2007 (CST)

• I could not see anything wrong with the next pub-delete. It had a note asking the moderator to also delete the title but there are still publications left.
• The make variant was a strange one and resulted in my writing a vignette about... nothing. (as it was not supposed to be a variant title at all).

Have a good trip. Marc Kupper (talk) 01:56, 27 Feb 2007 (CST)

Enjoy the trip and grab some of those famous copyright violations if you get a chance :) Ahasuerus 10:14, 27 Feb 2007 (CST)

Arbor House Treasury of Great Science Fiction Short Novels

If anybody happens to have a copy of Arbor House Treasury of Great Science Fiction Short Novels handy, could you please check it to see whether it contains The Dying Man or Dio by Damon Knight? We have two Publications on file which differ on this point and even though Contento says it's Dio, it would be nice to have independent confirmation before we blast the other one into the ether. Thanks! Ahasuerus 23:47, 27 Feb 2007 (CST)

I have both of the Anthony Boucher anthologies with a similar title [1] [2] but not the Silverberg edition you are looking for [3]. Marc Kupper (talk) 02:46, 1 Mar 2007 (CST)

I have approved the submission and will check my copy when I have access to my collection ca. march 30. Ahasuerus 12:01, 5 Mar 2007 (CST)

In the book club edition of Arbor House Treasury of Great Science Fiction Short Novels (of which I have an actual copy) the story by Knight is Dio. Mhhutchins 17:26, 6 Mar 2007 (CST)

Unexpected spam censorship on the wiki

When trying to clean up / refactorize a bit the Open Display Bugs, I got this warning after pressing "Save":

Spam protection filter

From ISFDB

The page you wanted to save was blocked by the spam filter. This is probably caused by a link to an external site.

The following text is what triggered our spam filter: <pre style="overflow:auto">

Return to Main Page.

(yes, I was trying to prevent PRE-formatted dumps of error messages from making the page too wide, in a way which I've learned on Wikipedia).

I have some basic user knowledge of MediaWiki, but haven't been able to find where this installation would have its custom spam filter. If Al or anybody else knows (it might be hidden too deep in the defaults; it's strange that other uses of styling are OK), and considering it's quite harmless and allowed even on Wikipedia, would there be a way to enable it here too? --JVjr 09:51, 1 Mar 2007 (CST)

Please be aware that ISFDB is running a very old version of MediaWiki. I don't have time at the moment to dig into the details of the error message you saw and to comment on them. Marc Kupper (talk) 13:33, 1 Mar 2007 (CST)

Beta recruitment

The main ISFDB page links to ISFDB:Beta recruitment, which we're not really using. Should it link to ISFDB:Beta instead? Mike Christie (talk) 18:33, 1 Mar 2007 (CST)

I use ISFDB:Beta recruitment (as well as the New Submissions and related pages) to leave welcome messages on new editors' Talk pages, but I agree that it's not a particularly informative page. Perhaps we could add a pointer to ISFDB:Beta without removing the current one? Ahasuerus 19:12, 1 Mar 2007 (CST)

Are Review records ever merged?

I held up a publication update for a bit as there were many changes to the story titles plus an author name for the Reviews section as I was concerned about merged records and if the changes would impact other publications. After chansing down each review/link I found that none of them had been merged and approved the update. My question is - while it's technically possible to merge two title records for reviews would this ever happen? I had updated the help for publication edits to warn about the consequences of making changes to the story titles/authors in the contents section but am thinking if the help should be updated saying it’s ok to make changes to Reviews, Interviews, etc. and that editors do not need to do the add-title/remove title thing to implement changes. Marc Kupper (talk) 22:27, 1 Mar 2007 (CST)

Most reviews (although not all) appear only once in a magazine. Exception would be someone like John Clute who has published collections of reviews. In that case, they should be merged as they are the same text (magazine version vs book version)- but my guess is that no one has ever tried it, and something undesired is likely to happen. You should let me experiment with merging both reviews and interviews over the weekend. Alvonruff 05:59, 2 Mar 2007 (CST)

Editing Awards?

I've been cleaning up the Theodore Sturgeon biblio, and found that the shortfiction title for "A Saucer of Loneliness" [4] has the award nomination for the collection by the same name, but I can't find any way to fix this through the UI. Any thoughts? Jefe 15:11, 2 Mar 2007 (CST)

I think Award info is fixable by mods only, but not being one I'm not sure of that. If so, just bug the Mod that looks active most recently to fix it. ;-) BLongley 16:30, 2 Mar 2007 (CST)

Al is working on Award editing at the moment -- see the first note in the ISFDB Feature List#Editing ISFDB Feature List. Ahasuerus 17:18, 4 Mar 2007 (CST)

Identifying duplicate titles

Something else that occurred to me was that it would be very useful to have a tool that searched an author's bibliography and identified duplicate titles. There were a slew of short stories that needed to be merged; I think I found them all, but he wrote so many it's almost impossible to be sure just by scanning the page. Obviously, there are lots of times when a duplicate title is perfectly legit, so we wouldn't want this to be an automated merge tool or anything, just a quick way for someone to identify potential duplicates. Jefe 15:11, 2 Mar 2007 (CST)

I could do with something similar, but it might be partly my fault. For instance, today I took the Larry Niven page, copied it into Notepad (to lose the HTML), copied it again into Microsoft Word, did some "search and replace" to get rid of leading formatting characters, sorted it, and came up with a large set of titles that needed merging. And most were due to me adding a volume of "Draco Tavern" stories: as I couldn't add the series to every title as I added the publication, they almost ALL needed merging (apart from the three previously uncollected ones). Now that was a pain for me to do 20 merges, and presumably a pain for the moderator approving them all too. It'll also be a pain for everybody else that adds any of those titles from any new collection. Still, I don't think there's anything in the help that says I have to classify short stories into series, however useful that may be to those that like (for instance) Draco Tavern Stories but hate Known Space stories. I'm not even sure where I got the idea from, I just saw a page that had the Short Fiction arranged into a series and found it useful, before I found out how much of a pain it is to maintain. If we could guarantee all series eventually got published in suitable collections there'd be no need for us to do it (I had 23 Draco Tavern Stories nicely categorized but adding 26 in one collection looked a very good idea) but I don't think we can ever guarantee that it wouldn't be better to do the categorizations early, if people other than me want them.

I do a lot of Short Fiction work here (and MAJOR apologies to the Mod that had to look at my "100 Great Science Fiction Short Short Stories", but it's not my fault only 94 passed muster) and would LOVE some tools that allowed me to replace an entry with the same story under a pseudonym already recorded, or to delete while cloning, or even clone one pub into two for the "republished in X volumes for the Y market".

Anyway, trying to get back on topic: 1) Yes, I'd like that tool. A lot of Searches are broken though, not sure if we'll get that one soon. 2) Is adding a series to Short Fiction of appeal to anyone but me? 3) I'll review my Sturgeon too. BLongley 16:30, 2 Mar 2007 (CST)

Such a tool might be useful, but I don't think it's realistic to hope for it in the situation when Al barely manages to correct the worst bugs, let alone add more important features (I for one would like if "Diff publications" also compared publication data and eventual Notes, and found duplicate titles in contents).

However, to find duplicates, I just use Advanced Search - ISFDB Title Search form for given Author (if zie is too copious, add limitation to Ttype = SHORTFICTION etc.). As the results are sorted alphabetically, it's quite easy to spot a repetition. This works well enough that I don't find it such a bother as other areas of ISFDB work. (It would be nice if "AND NOT" worked, though, so I could exclude COVERART from results for simple title search - I usually do just that with search term wide enough to find a few other duplicates than the one I originally noticed.)

But creating some way that would prevent creation of duplicate titles when contents of a new coll/anth are entered would IMAO be time better-spent. --JVjr 05:52, 3 Mar 2007 (CST)

P. S. Sturgeon's short fiction seems OK now, but there are some duplicates elsewhere. BTW, does anybody know why A Saucer of Loneliness is displayed as the last story even though with date of 1953? That mistakenly assigned award for Collected Stories volume of the same name, mentioned above? Very weird bug... --JVjr 05:52, 3 Mar 2007 (CST)

New Tools for Finding Duplicate Titles

Now that people are updating collections, anthologies, and magazines, there is a lot more time being spent tracking down the duplicate titles that are introduced through that activity. To help find these titles, I've put a new tool on the author page to help find them.

If you look in the navbar, you'll see a new item called "Dup Candidates". If you click on this, the tool will show and group titles with the same exact title REGARDLESS OF TITLE TYPE. Since collections are often named the same as one of the short stories it contains (and short stories are often expanded into novel form), you can't assume that everything in a grouping needs to be merged. This identifies candidates for merging.

There is a more aggressive version of the tool which you can invoke by clicking on the link titled "Similar Title Mode". This will find titles which are similar, but not necessarily identical. This will find more false positives than Exact Title Mode, but will find titles followed by parenthetical information, have differing capitalization, and other small changes. Alvonruff 08:29, 4 Mar 2007 (CST)

Perfect! - the more tools the better. Marc Kupper (talk) 11:20, 4 Mar 2007 (CST)

It's really a good idea! Rudam 11:47, 4 Mar 2007 (CST)

Looks enormously useful! :) Ahasuerus 17:28, 4 Mar 2007 (CST)

This is terrific -- thanks!Jefe 11:54, 5 Mar 2007 (CST)

It has been an enormous help the past few days. Thank you. Here's an idea for a similar application which is to add find-dup to the publication view. It would build a list of unique author names from the title records and run the dupe logic on each of them. After you add an anthology or omnibus for example you could run this to check to see if any of the new titles you added need to be merged. You could limit it to just the titles in the publication or you can check the entire author bibliographies for duplicates. Marc Kupper (talk) 17:55, 7 Mar 2007 (CST)

Project Gutenberg links

I've been adding notes to titles that are available on Project Gutenberg, since Greg Weeks over there has been clearing a lot of pre-1964 titles that never had their copyright renewed. See, for instance, 3043 or 52200. I've been using the text

Available on Project Gutenberg as <a href="http://www.gutenberg.org/etext/18800">etext 18800</a>.

or whatever the etext number is. The SF over there goes up to 1964, so I think we should be able to get a fair number of links in. grendel|khan 10:43, 3 Mar 2007 (CST)

Serials are confusing to me, so I didn't touch Null-ABC (113021 and 113022); there's a combined version on PG at etext 18346. PG authors to look at include H. Beam Piper and Andre Norton. grendel|khan 11:39, 3 Mar 2007 (CST)

I'd started doing the same, we should probably coordinate. I'm entering them as a publication however. See The Black Star Passes for instance. Dana Carson 02:46, 4 Mar 2007 (CST)

I like the note approach more (especially as Gutenberg texts are always copies of one particular paper edition), but I guess this is for the more avid bibliographers to decide. However, it's strange that the publication entry for the Gutenberg The Black Star Passes doesn't have a Title Reference to the story, even though its Title entry does: does anybody know the reason, and fthe way to fix it? Also, note that the Gutenberg e-text is of [the book collection containing 3 novellas], not merely the title one it's assigned to. --JVjr 05:20, 4 Mar 2007 (CST)

We definitely need a consistent policy on these things; it's madness to do it helter-skelter. I kind of like the idea of entering them as publications (though I'd name the publisher "Project Gutenberg" instead of just "Gutenberg"); if we enter them as publications, they can easily contain more than one title, which certainly solves that problem. Also, when bug 20002 gets fixed, it'll be nice to see a summary of PG's SF works, which you can't get from their catalog. There is, of course, the question of other etexts like 73523's (and what's up with that %%-markup?); I'd be very wary of adding actual publications for anything other than PG etexts, since PG really does act like a publisher. grendel|khan 11:20, 4 Mar 2007 (CST)

(Also, I appear to have proposed this some time ago, but it never got acted upon.) grendel|khan 11:22, 4 Mar 2007 (CST)

Good point on the publisher name, will correct them to Project Gutenberg. You can see the SF Bookshelf for Gutenberg. I'm also adding etect publications for theings like Baen that puts out an etext of all its books. Also intend to add publications for the CDs of etexts that they do. What do we do for things like Accelerando where the author has put it up under "Accelerando" is © Charles Stross, 2005. The downloadable version on this website is licensed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs license . I'd like to track things like that. Dana Carson 14:28, 4 Mar 2007 (CST)

Is there a hold light on this thing?

I just ran into a particularly messy series of submissions and put them on hold to research. I got curious on if I could put a note on my user page giving a general policy/reason on why I put things on hold and so did a random update to a title, held it, and am trying to figure out how I, as an editor, can see my held submissions (I wanted to make sure there’s a link to my user page). I would think it would be under “My Pending Edits” but that’s an empty list. How do editors know which submissions are on hold? Marc Kupper (talk) 11:33, 4 Mar 2007 (CST)

Marc, the New Submissions page shows all the holds, including the ones you put on my Kuttner updates. (Scott Latham 17:23, 4 Mar 2007 (CST))

Would those be mine? If so, I think it's pretty sure that you can blame me or Peter Haining. BLongley 18:12, 4 Mar 2007 (CST)

Thank you - that's good to know as I had assumed that editors could not see the New Submissions page.

In summary the thing that tends to hold up submissions are changes to the story title and/or author for title records. This can be done indirectly via the Contents section of a publication, directly via edit-title, or by merging titles. Thus the first thing I look at when titles are being changed is how many publications are impacted. If it's zero (empty title) or one, then I approve.

At the moment every single one of the items I have on hold other than Scott's title-updates are being held because of title-updates like this. Unfortunately, the moderator review pages don't link to the publications meaning the first step is to manually copy/paste the record numbers into URLs. For example, item 374581 in the queue is a pub-update that looks great except deep in the contents the story Summer's Lease gets renamed to Truth to Tell. I copy paste the author's name "Joe Haldeman" into a URL and on his page I see Summer's Lease (1974), Variant Title: Truth to Tell (1974). I have two choices

1. Reject this submission and have the editor re-enter a couple dozen things.
2. Approve it and then change the title record that used to be Summer's Lease back into Summer's Lease, add a new title for Truth to Tell (merge) to the publication, remove-title Summer's Lease from the publication, approve, and then search for and merge Truth to Tell

I usually do the latter route as it seems like less work overall though it pushed a bunch of work onto me meaning they sit in the queue until I have time. The other item on hold has the same issue where if I approved it then Fritz Leiber's Lean Times in Lankhmar shortfiction gets blown away and replaced by Swords Against Death collection.

The remaining items I had on hold were less challenging in that I realized they will not break much even if the update is wrong meaning I approved them. Marc Kupper (talk) 19:41, 4 Mar 2007 (CST)

Serial showing up under a book with the same title

For The Lost City by Blaine it shows a serial by the same name by Sloane. How do I tell it that they are not related? Unmerge didn't seem like the right thing, it just lists the book publication as a unmerge possibility. Dana Carson 14:12, 5 Mar 2007 (CST)

You can't, until Open Display Bugs # 20046 is fixed. The connection is currently purely by title. --JVjr 09:18, 5 Mar 2007 (CST)

I was afraid of that. OK, will just ignore it for now. Dana Carson 14:12, 5 Mar 2007 (CST)

Ahasuerus downtime

Ahasuerus e-mailed and said there are problems with his Internet connection. He apologizes and says he will not be able to be as responsive as usual to any pending moderator issues. I'll try to check in on things though my available time is limited at the moment. Marc Kupper (talk) 19:42, 3 Mar 2007 (CST)

Ahasuerus Downtime Part 2

Unfortunately, rumors of my resurrection have been greatly exaggerated. My connection to the ISFDB server (and only to the ISFDB server, interestingly enough) is still flaky and goes up and down seemingly at will. I am not sure what's going on, but with any luck my primary Internet link should be back up some time tonight, so hopefully things will be better. Ahasuerus 11:26, 5 Mar 2007 (CST)

That might be similar to my own previous problems, being ONLY to the ISFDB server. Have you tried a proxy? BLongley 15:56, 5 Mar 2007 (CST)

Is the connection down or just very slow? What I see regularly with ISFDB is I’ll be working away, click on something innocuous, and all of ISFDB will hang. I can submit other things (lookups or updates) to ISFDB on other windows and they all hang and in fact, that’s the state it’s in this instant and while I've clicked on "edit" to submit this message I'm typing this message itself off line and will paste it into the wiki once ISFDB comes back. After about three minutes ISFDB gets unstuck and all the pages then continue on with whatever they were supposed to do. Maybe there are deadly embraces in the code and it’s MySql attempting to resolve them. Marc Kupper (talk) 17:38, 5 Mar 2007 (CST)

I have seen some brief "freezes" (5-20 seconds?) as well as occasional "Failed to connect to the SQL database" errors, which usually go away in a few seconds. However, when my secondary connection goes down, it goes down and stays down for hours, which sounds similar to what Bill experienced a few days ago. Going through a proxy is not an option with the secondary connection. Thankfully, my primary connection is back up, so things should be better now. I have a tertiary connection option as well, but I am fresh out of peanuts to feed the squirrels that drive it :( Ahasuerus 11:22, 6 Mar 2007 (CST)

Artist pseudonyms?

Any advice on the best way to merge an artist ( Les Edwards, his real name) and his pseudonym, Edward Miller? Both appear on the IFSDB. While it may not be of any great use to have this information on every title he dis artwork for, it should be noted. Where and or how is my question? Thomas conneely 18:42, 5 Mar 2007 (CST)

It's handled the same as an author pseudonym. From Edward_Miller's page look at the top for something about "name used by ..." - it's not there meaning you would first do a "Make This Author a Pseudonym" and enter "Les Edwards" (the real or canonical name) in the "Parent Name" field. This just sets up a general pointer from one name to the next but does not do anything with the titles listed in the bibliography.

For each title listed in the pseudonym's bibliography do the following.

• Click on the title
• Click "Make This Title a Variant Title or Pseudonymous Work"
• In the lower part of the page replace the Author1 name (Edward Miller) with the real or canonical name (Les Edwards)
• Click [Submit Data]

If you are using FireFox, IE7, and perhaps other browsers that support tabs, this is rather easy in that you can use CTRL-click from the main bibliography to open up dozens of sub-pages and from each of those you click "Make This Title a Variant Title or Pseudonymous Work" and without moving the mouse you do CTRL-TAB to switch to the next window. It's still a manual process but overall a fast one.

Once all of the make-variants are approved the pseudonym's summary bibliography will then show in bold "Pseudonym for Les Edwards" and it'll link to Les_Edwards where you will now see covers with "as by Edward_Miller".

Any time a new title (or cover art) is added for Edward_Miller you will need to repeat the part about assigning variant titles. Ideally, ISFDB would do this automatically but I believe there are technical issues involving that the pseudonym stuff is also used for house names and two or more authors writing under a shared pseudonym. Marc Kupper (talk) 19:26, 5 Mar 2007 (CST)

Just a quick note to point out that Artist pseudonym handling is currently not quite as robust as Author pseudonym handling. I will need to experiment some more and post a summary when I have a little bit of free time. Ahasuerus 11:29, 6 Mar 2007 (CST)

A Saucer of Loneliness and maybe other stuff is broken

Jvr wrote in the #Identifying duplicate titles thread

P. S. Sturgeon's short fiction seems OK now, but there are some duplicates elsewhere. BTW, does anybody know why A Saucer of Loneliness is displayed as the last story even though with date of 1953? That mistakenly assigned award for Collected Stories volume of the same name, mentioned above? Very weird bug... --JVjr 05:52, 3 Mar 2007 (CST)

This is very strange. Al, could you take a look at Theodore_Sturgeon's page? A Saucer of Loneliness is stuck at the bottom of the shortfiction list. I cloned the title record (using make-variant) and the clone sorts correctly. I also edited the date and it's still stuck. Another strangeness is when I enter a note in this record it's in the XML blob, approval screen, but does not get integrated. As a test I then changed the note and synopsis. The blob looks ok but the approval shows "Synopsis" for both the Synopsis and Note fields and integration only updated the note. Hmm, I can't seem to change title notes on the title that's just above this one though it works on my test-titles. Marc Kupper (talk) 20:09, 5 Mar 2007 (CST)

The order was determined by MySQL. The SQL query does a sort by series number first and year second. In this case the series number was taking precendence, as it was not NULL (it was zero). My guess is that the title was previously part of a series, and then later removed from that series, and the seriesnum field was set to zero instead of NULL. It was interesting that phpMyAdmin could not fix the problem, and I had to directly issue an SQL query to fix it. There are other titles with a seriesnum of zero, but they are either a part of a series or a variant title in a series, so this is the only title in the database with that behavior. I'll look at the title/notes stuff later. Alvonruff 06:24, 6 Mar 2007 (CST)

Well I just did a test of a title, adding it to a series and the removing it, and verified that the field was set to NULL. Some more complex chain of events occuring... Alvonruff 06:27, 6 Mar 2007 (CST)

Thank you Al - FWIW - When processing editor updates I saw one where someone added a note and to I paid extra attention to the integration. In that case I saw "update titles set note_id='52081' where title_id=45202" but what's not logged is the update to the notes table. A synopsis was added in the same transaction and in that case the insert of the note and then the setting of title_synopsis were logged as expected. In this case the note update worked as expected other than it did not log the updates of notes. Marc Kupper (talk) 11:43, 6 Mar 2007 (CST)