Rules and standards discussions

From ISFDB

For old discussions see Rules and standards discussions/Archive.

This page is for discussions about the rules and standards, such as whether certain kinds of publications belong in the ISFDB, or whether the help text defining capitalization should be modified. It also includes questions about interpretation, such as whether a SERIAL type can be used for sequences of short stories subsequently republished as a novel.

To add a new section at the bottom, click here.

Contents 1 Cloning Magazines 2 CoverArt Dates 3 Variant title dates 4 Fixups vs Collections 5 Cartoon entries 6 Summary Mini-reviews 7 Verification Definition 8 Heros in Training by Jim C. Hines and Martin H. Greenberg 9 Change to Help - Title Fields - Entry Type - Interiorart - Page Numbers 10 Perry Rhodan - Book or Magazine? 11 Spoilers in Tags 12 Pseudonyms With Unknown Real Names 13 Help Update - Book Review Dates 14 Consensus on "Dateless" Printings 15 Foreign Language Magazines 16 Minor variations in a publication 17 Annotating Excerpts 18 Amazon Images 19 One for the "EXACT title, including all punctuation and special characters" crowd 20 Cover art that gets reworked 21 Volume Number, Issue Number, Whole Number 22 "Real" Publisher problems 23 Ellipsis 24 Assigning Dates to Serials in Magazines 25 Merging Artwork 26 Series in Fantasy and Science Fiction 27 Website links for deceased authors 28 Modifying references to The Editor attributions with the mag name 29 Magazine Bibliographies and Replicating Other Sites' Data 30 Complicated data entry rules scaring away new editors? 31 Binding Types 32 The Definition of Verification 33 Assigning Pseudonyms 34 Plagiarized Story 35 Ace Doubles 36 Identifying magazine editor essays by initials 37 Non-canonical names in titles of books being reviewed 38 Pub Editor help regarding book reviews 39 Listing Fanzine Reviews? 40 Letters OK in fanzines but not in prozines? 41 Advance Reading Copies 42 Fanzine Reviews Again 43 Cain to the power of N 44 Those Pesky Nuns 45 Magazine data - level of detail 46 Certainty of data for deletion? 47 Canonical name for Artists 48 Interior artwork attributions 49 Prologues and Epilogues 50 Manga and Takahashi 51 Full artist signature over last name editorial attribution 52 Unpaginated pages in page count 53 Shortfiction vs Essay? 54 Tracking relationship of magazine cover art to interior story 55 Linking Cover illustrations to the stories they illustrate 56 "In Memoriam" 57 Current Help on foreign language books 58 Canonical Publisher names 59 Entering Reviews of Ace Doubles 60 Linking to Author images 61 Audio Bindings 62 Amazon Blogs 63 Two Variants of something we don't record 64 Published Bibliographies 65 Publishers 66 Doc Savage and house name and numbering 67 European graphic novels 68 subsequent multi-volume "serialisation" 69 just fictionally speculating . . . 70 Magazines reprinted as Anthologies 71 Publisher names again 72 Uncredited vs. Anonymous vs. N/A plus Reviews 73 Collections in Omnibuses 74 Merging Audio Dramatizations with Stories? 75 OCLC 76 Killdozer by Theodore Sturgeon 77 "Digest" sized books 78 Novels that were later split into multiple volumes 79 Variant titles: for identical texts only?

Cloning Magazines

It seems that Magazines as a PUB TYPE are currently not cloneable. The way we put magazines however, under one EDITOR TYPE for a single year, ie. "The Dragon Magazine - 1980 - Issues # 35-??" will contain all of the Dragons for 1980, it would make entry quicker and easier if we could just enter say Issue #35, and then just clone the rest of the issues for the year under the "The Dragon Magazine - 1980 - Issues # 35-??" EDITOR TYPE, and then go back and edit the few changes and then add in the differences aka the spec-fic, the book reviews, any appropriate Interior Art and Cover Art, etc. Is there a reason that this is not doable at the current time? It doesn't seem to me like it would be any different then putting in different publications of the same COLLECTION for instance. If nothing else, it would save a bunch of typing or cut-and-pasting. Of course if it would take too much work to do the programming, don't even worry about it. CoachPaul 13:26, 28 May 2007 (CDT)

Would being able to clone magazines just like a normal publication be enough or are you looking for something extra? Adding the ability to clone magazines seems doable but I believe you are looking for "something extra" and can't quite understand what it is. Let's sat I'm at The Magazine of Fantasy & Science Fiction, May 2007 and now want to add the June, 2007 issue. I hit "clone" and for a magazine it looks like you only want the metadata (the upper part of the page) but nothing from the Contents section. Is that correct or are you also thinking about things like regular columns? For example clone-magazine could go down the contents list and if it sees a title that looks like

Xyzzy (The Magazine of Fantasy & Science Fiction, May 2007)

then it would create a new title record in the clone's contents called something like

Xyzzy (%title%)

and when the editor fills in the new issue's title, The Magazine of Fantasy & Science Fiction, June 2007 fpr example, the “%title%” would get replaced so that the column is called

Xyzzy (The Magazine of Fantasy & Science Fiction, June 2007)

In fact - now that I think about it, something like %title% may be handy for regular publications because I frequently do things like "Introduction (Story Title)" where the title is copy pasted. If I could enter this as "Introduction (%title%)" then ISFDB could fill in the title when the publication is submitted. Likewise, %author% would also fill in and maybe we could shorten these to %a and %t. Marc Kupper (talk) %%~

I was just thinking of the Meta Data. Say for Magazine SF you have entered the January, 1999 issue, and change it's editor record to SF - 1999. Then when you hit clone pub it creates a second issue of Jan1999 without the Contents from the bottom of the page. Then you just change the title from SF, Jan 1999 to SF, Feb 1999, change the month to 1999-02-00, change the page numbers and price if need be, and put in the new Cover Artist if known. This would make it less then a minutes worth of work, instead of several minutes of cut and paste. The potential savings over decades of magazine entry would be in the hours, but then again, I don't know how many magazines still need to be entered in this way. I know that I have over 20 years of Dragons to do , but if that is going to be the only savings, three or four hours of programming and bug fixing isn't worth the three or four hours it would save me in not having to cut and paste. CoachPaul 06:59, 6 Jun 2007 (CDT)

I want cloning of the ENTIRE data, to allow easy entry of British versions that will be months later but mostly similar. Of course, these might need to be made an entirely separate series so maybe we need to separate the title's parents as well. And some US magazines like "Destinies" DID go through multiple printings that we only cope with by calling them Anthologies. I don't know if the lack of British versions of magazines is because it's so hard to clone, or that we have few British reprint SF magazine collectors here. I DO know that it puts ME off entering the few I have, there seems to be a "Help!" post somewhere for every one I do in detail (fixing the series in general is OK, fixing CONTENTS is so much of a pain they go right to the bottom of the list of things to do.) BLongley 13:52, 27 Aug 2007 (CDT)

CoverArt Dates

The way Coverart seems to work, if I have a book published in 1985, and then a second publication of the same book in 1986 with the same cover, it creates two COVERART records one for each year, yet both books have the same covers. Is it OK to merge these two entries into the same COVERART record with the earlier date? CoachPaul 00:50, 2 Jun 2007 (CDT)

Coverart records don't have any relationship to the Coverart Image URLs on a pub, so yes, it seems safe to do so at the moment. If the database links ever change, then it might become more important: e.g. Josh Kirby's Colour of Magic for the Compact Discworld edition is different from Josh Kirby's Colour of Magic paperback edition. I don't merge Coverart records for now - nobody seems to really care that much about them. But in my A.E. Van Vogt explorations, I've seen and confirmed the same coverart used on different publications, e.g. The Changeling and The Weapon Makers, which may be of interest to someone, sometime. BLongley 16:15, 3 Jun 2007 (CDT)

I tend to be leery of merging Cover Art records until I have seen them both. Publishers have been known to resize or even replace covers in between printings. I am sure the only reason they do that is to make our lives more difficult. Ahasuerus 23:35, 19 Jun 2007 (CDT)

Of course! We are the enemy of the publishers, we tell people that they might be able to buy a cheap second-hand copy instead of their latest deals... they're bound to hate us with a passion! ;-) Still, as we don't link the Coverart URLs or any other interesting stuff to the coverart record itself, it seems to do no harm for now. One thing I did notice, after tackling Dissembler's submissions in particular, is that cleaning up a pub and title doesn't clean up the Coverart record, which is a bit of a nuisance: so merging several of those at once might be a useful trick if several of these are done at once. BLongley 13:56, 20 Jun 2007 (CDT)

Variant title dates

At present Template:TitleFields:Date says "For works that have had variant titles, the date to enter is the first under any title and any pseudonym; variant titles do not have their own dates." I've always thought it was an odd rule and sometimes have set the variant's date to be the first time the variant title was used. Apparently others are doing this too and I started thinking about updating the help to say "For works that have had variant titles, the date to enter is the first time the variant title was published." The downside to this approach is that if someone's looking at a title in the Contents section (magazine, collection, anthology) they may be mislead on how old a particular story is. For example, if an anthology reprints with a new title then someone looking at the Contents may think this was a new story written just for the collection. One fix would be for the existing display code to look at the dates and when doing the "as by" stuff to display the original date of it's different. For example, right now publication ALINMND1991 Alien Minds shows

• 131 • The Exterminator • (1971) • shortstory by Keith Laumer (aka A Bad Day for Vermin) [as by Keith Laumer ]

but as A Bad Day for Vermin was in 1964 it should show

• 131 • The Exterminator • (1971) • shortstory by Keith Laumer (aka A Bad Day for Vermin (1964)) [as by Keith Laumer ]

What are your thoughts on this? Marc Kupper (talk) 23:56, 4 Jun 2007 (CDT)

The as by with both dates seems like the best solution. Programmer time is of course the problem. Dana Carson 00:26, 5 Jun 2007 (CDT)

I agree that a programmatic solution would be preferred. Using the same dates for variant titles seems like a way to lose valuable data for no good reason, but if you use different dates, the way it gets displayed right now in anthologies and collections is misleading. Ahasuerus 23:44, 19 Jun 2007 (CDT)

I'm leaning towards preserving dates of variations, although the display problems are a bit of a nuisance. I was thinking that we could separate them programmatically if ever needed, but as we often don't have the very first edition of a title recorded here except in notes I think we're possibly destroying data. Looking at publications from Non-Genre magazines has convinced me that unless we start creating a lot more stubs for first printings (which is difficult considering some of the mandatory information required for magazines especially) we could NOT recover the information. There's a lot of (presumably secondary-sourced) data which leads to titles like this that I don't like - I'd prefer the REAL title, REAL date, and the REAL publications under it. But there's a lot of work to do on those still: maybe I should concentrate on those while the postal strikes are delaying more physical books arriving here. BLongley 13:58, 11 Oct 2007 (CDT)

Fixups vs Collections

I entered a book the other day The Adventures of Terra Tarkington as a novel and it has been changed to a collection. This was done I assume because I listed several shortfiction entries with it. The entries are based off of the copyright page having

Portions appeared in a somewhat different form in IASFM

   * "Hitch on the Bull Run", IASM 1979
   * "Itch on the Bull Run", IASM 1979
   * "Switch on the Bull Run", IASM 1980
   * "Twitch on the Bull Run", IASM 1981
   * "Bitch on the Bull Run", IASM 1981 

however the book itself is a novel. No TOC, no page heading for stories etc. If we use collection for fixups many of the classic novels become collections. I think the main entry for a book should reflect what is visible when you look at the book. So should I change it back? Dana Carson 01:33, 13 Jun 2007 (CDT)

That's right, when I saw the added stories, I changed the "title type" from Novel to Collection. Typically, if a fixup is classified as a Novel, we enter the stories that the novel was based on in the Notes field, but we don't list the stories in the Contents area. I think that's a reasonable approach since listing the original stories in the Contents are may confuse our users. However, I see that we don't have this spelled out in the Help pages:

Fixups. Sometimes an author will assemble material published separately into a novel. This will generally be classified as a novel. Some fixups are less coherent, consisting of largely independent stories, formed into a whole by the addition of linking material between the stories. In this case it is acceptable to call the book a collection or a novel; the decision should be discussed on the author's project page if there is any doubt.

We way want to expand this paragraph to reflect our current practices. Also, one thing that I have found useful is to create a Series record that includes the fixup and the stories that it is based on. That way the relationship is immediately clear when a user is browsing the author's Summary page. Ahasuerus 11:07, 13 Jun 2007 (CDT)

Kraang was recently asking me about a Van Vogt title which seems to have had the contents gathered into a series, and now has some verified pubs, some with notes and some with contents. In that case, I agree it's a Novel whereas this publication seems more like a collection - I could put page numbers to each short story, whereas most of the van Vogt fix-ups couldn't be separated that way. Now some fix-ups can be separated out IF you know the stories, despite no contents list: but I doubt whether it's worth the effort if 'Notes' is the preferred way forward. But I would like to see the notes on both the short-fiction and the fix-up, people do rearrange series quite a bit and I think merely grouping as a series leaves this information a bit vulnerable. BLongley 13:44, 13 Jun 2007 (CDT)

We also seem to have quite a few 'Classic novels' that on further examination are probably collections in ISFDB terms: e.g. Foundation and again various editors are being inconsistent about whether contents are needed in such a well-known title.

As we do seem to have editors proposing opposite changes, is it time we figured out how to conduct a poll? BLongley 13:44, 13 Jun 2007 (CDT)

And I see the first title discussed is back to "Novel" now, edit approved by the Moderator that changed it in the first place... is that a shift in policy or a "this isn't THAT important an issue, must clear the submission queue" thing? As Kraang's hasn't changed yet... BLongley 15:02, 20 Jun 2007 (CDT)

I approved the last submission since I thought that Dana was going to remove individual Contents Titles next, but that hasn't happened yet. I am keeping an eye on this Publication, but I don't think it's a huge problem: once we agree on a standard, we can run a search for all Novel Publications that include Shortfiction Titles and fix them as according to that standard. Ahasuerus 15:19, 20 Jun 2007 (CDT)

Hang on - that only works if we agree on your standard - "Novels shouldn't contain Shortfiction contents". If the agreement is "FIXUP Novels can OCCASIONALLY contain contents", there's no way we can search for the titles that need changing. BLongley 16:18, 20 Jun 2007 (CDT)

My reasoning for putting content entries was that if you're looking for where a story was published it only shows as a publication if it is entered there. For many fixups they don't give a nice list the way the copyright page for that one did so it may be confusing to do it for the ones that do. If the decision is to not include them I'll go back and remove them. Dana Carson 15:34, 20 Jun 2007 (CDT)

I lean toward the "CAN have contents", IF, and ONLY if, the short fiction is still separable in the fix-up, e.g. by being an identifiable series of consecutive chapters. If I want to find a short story in particular, I'd settle for finding it in chapters 2-5 of a fix-up: but I wouldn't accept a "most of it was rewritten into novel X, along with stories Y and Z" - that sort of mess needs lots of notes both ways. Of course, "Can Have" means no data-quality clean-up searches like Ahasuerus suggests above. Wow, I think I've managed to sit on both fences at once tonight... :-/ Time for me to go unpack the latest arrivals and do some normal editing/verifying again, I'm not used to so much discussion mid-week! BLongley 16:18, 20 Jun 2007 (CDT)

A.E. van Vogt's Rogue Ship was in the unfinished and things to do pile. I've changed it to a novel and put it under the series title Centaurus with the short stories. Have a look at the Summary Bibliograph [1]. Let me know what you think.Kraang 20:16, 20 Jun 2007 (CDT)

That one looks fine - you can't separate the short stories in "Rogue Ship" the novel. (Well, I can't from my edition alone, anyway.) I've just come across a title I gave up on in my early editing days though, and finished it. Now there's seven different short stories in that published elsewhere, a couple under variant titles, and three new stories that could be lifted out and published as stand-alones at some point. The interstitial material probably couldn't be, but I entered it anyway as it all has definite titles. (Quite amusing ones too, IMO.) We have that pub's title verified as a novel, AND as a collection. The Plus point of going for the "Use notes about prior publication" is that we can lose the Shortfiction entries for the interstitial material. The Minus point is that if the short stories not already published elsewhere DO get published elsewhere, there's no obvious link to the first publication and people will probably have difficulty hunting it down. In other situations like this, we may lose the variant titles entirely if the "Chapter Heading" is different from the short-story title. I'll leave this as an example - I'm not so bothered that I'm going to hunt and kill novels with Shortfiction contents, nor am I going to fix this title with the Novel/Collection mismatches. But I DO wish we could agree on a way forward: and I'm actually now STRONGLY in favour of "identifiable short fiction titles in other publications" being recognised, and my dithering is now over what to do with interstitial stuff that has proper titles. I think I'd be happy if THAT moved to "Notes". BLongley 15:27, 22 Jun 2007 (CDT)

Cartoon entries

I have just made changes to the documentation for entering cartoons which should clarify matters somewhat. The changes appear in the newpub and editpub help under the discussion on interiorart. I have made it clearer that the cartoon entries must follow the same rules as other interior art which is not connected to a story - the cartoon should either be signed or significant (i.e. full page)--swfritter 11:59, 16 Jun 2007 (CDT).

Although the amended rule makes sense for digests, keep in mind that there were not that many full page cartoons in the pulp era. Would you say that we want to catalog half-page cartoons published in bedsheets and pulps? Ahasuerus 23:32, 19 Jun 2007 (CDT)

Didn't notice this comment on my watchlist at first. Have changed help to include 1/3 page cartoons on pulps and bedsheets.--swfritter 10:31, 11 Aug 2007 (CDT)

I'd make it "e.g." rather than "i.e." and leave it to editor's discretion. Although nowadays I'm not sure that anyone under 40 understands the difference between the two. :-/ BLongley 15:06, 20 Jun 2007 (CDT)

Summary Mini-reviews

Magazines sometimes do summary mini-reviews. They may publish a full review when a book is first released and later summarize this for a paperback reprint. It's almost like a book announcement. Should these be entered and/or noted in ISFDB? One option would be to add an "announcement" title type that would be handled much like reviews though one sticky point would be that it would be nice to then be able to enter what was being announced. For example, the ISBN, etc. that was stated in the announcement. The title records for book reviews are normally managed in the background meaning people don't see their notes. Anyway, I'm throwing this up to see what people's thoughts are about the summary mini-reviews and how to handle them. Marc Kupper (talk) 18:44, 17 Jun 2007 (CDT)

I'm happy with the standards as they are now. Reviews should have at least one element of criticism, even if it's only a thumbs up or a thumbs down. I don't want to rummage through my collection looking for a book review only to find that it is merely an announcement. The only exception might be the case where such an announcement documents an edition of a title that is not in the database. I definitely don't think this issue is important enough to make any programming changes.--swfritter 11:16, 18 Jun 2007 (CDT)

No change needed, I'd say. Otherwise we might end up with "Also Received" notes or suchlike, where there is no criticism at all. If it shows there was a publication we don't know about, create the stub pub with notes on the source, but forget about adding anything to the pub you found it in. For instance, when I eventually run out of books I may go look at, say, the "Selection of other Humorous titles from Corgi Books" entries in some of my pubs to pick out missing editions of stuff, but I wouldn't want to update the pub I got them from. BLongley 14:31, 18 Jun 2007 (CDT)

However, you do bring up one good point: we're sometimes not recording information that was contained in the review. In the few I've added, often the Publisher, Price, ISBN, format and cover-artist were included in the review. In some cases the lack of this was so bad that the review doesn't even make it clear if it's reviewing the US or UK edition, when the pubs actually had almost completely different contents! (Do a title search for "The Best of A. E. van Vogt" - the dates are what made me split the two up that way, but I hate changing titles and we'd have had to change BOTH, for the pubs and the reviews, if we ever get a review of the other.) BLongley 14:31, 18 Jun 2007 (CDT)

One thing that we had to decide upon back in the Dark Ages (1995-1996) was just how granular we wanted our classification scheme to be. We had Contento's example with dozens of subtypes drilling all the way down to "facetious article" (!). If memory serves, we consciously decided to take a different approach and have relatively few types to avoid confusion and debates about sub-sub-types. The idea was that anything unusual or questionable would be documented in the Notes field. This approach seems to have worked reasonably well and I think as long as we stick to it we should try not to create additional types, although sometimes it's very tempting to create a new feature request :) Ahasuerus 23:56, 19 Jun 2007 (CDT)

I personally like Contento's "Straight To Remainder" type. Marc Kupper (talk) 02:24, 20 Jun 2007 (CDT)

Verification Definition

I have a huge problem with Verifications. Basically, what the heck is it? I think that we need some basic general guidelines that we can all agree on. In one talk page discussion with a mod, I get asked if I verified the fact that there was no CoverArt credit given in a book that I Verified, there was, and I changed it, and in another Talk Page discussion, a different Mod says that it's ok to Verify a Pub without filling in all the metadata information that is in the pub being verified by Primary, and also that it's ok to Verify an Anthology that has only one Publication for the Title Record, and none of the contents has been entered. When Mods differ so greatly in their interpretation of a basic principle here at ISFDB, I think that it's time for some sort of consensus. Just what does Verification mean? I don't want to be just part of the problem and not the solution, so I will offer some suggestions. This refers to Verification by Primary only.

To mark a pub as Verified by Primary means:

1. All Metadata has been checked, double checked, and matches, any Metadata that is present in the Primary, but is not in the db has been added. The Title has been checked against the Title Page, and any differences in the Title between the Title Page, Front Cover, and Spine have been noted in the Notes. Any information that you can't verify, such as your copy has the price inked out, or part of the cover is missing and you can't read the ISBN should be noted in the notes section also.

2. For Anthologies, Collections, and Magazines, all contents should be listed, including page numbers, story names, and pertainant artwork/cartoons. The discussion two above this one deals with the matter of artwork/cartoons. Titles and Authors here should be checked as correct as on the first page of the story, and not from the ToC. Any differences between the Title Page of the Story, and the ToC should also be noted in the Notes Section of the Pub. All Book Reviews and Essays should be added also.

My suggestions, (please feel free to change them, add to them, subtract from them, or totally ignore them, in any way that you see fit), set up some guidelines for Verification that may make it a little less confusing for us all. If nothing else, hopefully at least it will get the much needed discussion on this topic started. CoachPaul 23:20, 19 Jun 2007 (CDT)

Sounds like a good start! I will comment further when I fully recover from this bug and regain a semblance of coherence. Ahasuerus 23:59, 19 Jun 2007 (CDT)

I personally am relaxed in what I accept and so if someone just wants to verify that they have a publication that's fine with me though I do ask they people check that the information that is stated in ISFDB matches their publication. The verification help seems to support this. The other thing is if someone marks a publication as verified it implies they have a copy or can get it pretty quickly meaning you also know who to ask. I do wish there was a way for multiple people to mark a pub as verified as increases the confidence in the data and increases the pool of people you can ask when there is a question.

For item 1 in your list I do agree with many of the things you listed and they should get added to the help. Probably the only thing I'd not include is "any Metadata that is present in the Primary, but is not in the db has been added." I'd make that a suggestion but not a requirement.

I personally would like to see a tighter standard for the cover artist. At moment Template:PublicationFields:Artist says "Enter the artist for the cover art if known. If not known, leave blank." I'd like to be able to flag for verified publications

• Cover artist not verified
• Publication inspected and cover artist is not credited
• Cover artist credited from secondary source (cite exact source)
• Cover artist credited from signature
• Cover artist credited in the publication

The first two are confusing. "Cover artist not verified" is it unknown or not checked? Does the first mean not checked and the second mean unknown? Also remember don't use unknown. Same questions can be applied to classifying unknown versus anonymous authors. The solution also introduces complexity that might no be understood by the casual editor. Ray 12:13, 20 Jun 2007 (CDT)

The reason I brought up this field and not others is "blank" is one of the allowed values but it's also ambiguous. Marc Kupper (talk) 02:48, 20 Jun 2007 (CDT)

I am glad this question came up. My definition of of what verification should be is that all the data for the publication has been entered according to the standards. That requires that the verifier duplicate every act that they should be expected to perform when the data is first entered. That means doing dup checks, checking for pseudonyms, checking artist names, checking spelling, etc. The publication should not be verified unless it is 100% correct according to the editor's best understanding of the standards. If the editor has to make changes to a pub they should not verify the pub. You verify it - you own it. Queries about any pub you have verified should receive a prompt reply. A high standard and well beyond the requirements in Help I will admit.--swfritter 11:24, 20 Jun 2007 (CDT)

Unfortunately it's "I own it - I verify it" for me, not vice versa. I don't guarantee to keep the book afterwards, though keeping it for a week or two should be no bother if people are checking recent verifications. (I see some editors seem to, either that or there's a lot of coincidental submission overlaps.) I know I left a lot of my early edits unverified - sometimes due to delays in approvals, sometimes because I hadn't learnt all the policies so didn't really know if I was done yet. (I was a very iterative editor, still am in some ways - so long as an edit improves the data I'm happy to leave some of the less interesting stuff for later, or for someone else to do - I wouldn't verify in those cases though.) BLongley 13:04, 20 Jun 2007 (CDT)

I can see a case for verifying a contentless collection if it's one of those fix-ups that can be considered a novel by some people - we don't seem to have sorted that argument out and we shouldn't need to for verification purposes, if the verifier firmly believes it's a novel. And I'm not that interested in interiorart so I'd prefer an option to verify the text and just note that some interiorart exists - I'm fine if people add it later. I'm not that interested in reviews either, but people seem to have made a special effort to support them so I'm OK with demanding that they're all entered before verification is allowed. I rarely do magazines so it's no biggie for me anyway. BLongley 13:04, 20 Jun 2007 (CDT)

I'm also OK with verifying partial contents so long as all the SF is entered. For instance I'm not ashamed of leaving the Detection out of Detection, Mystery and Horror. And I don't demand that people add Coverart URLs even when the Amazon link takes them to the exactly right cover anyway - if I add it I just ask the verifier if it's right, and I don't do it from anything less than a primary source or a scan with specific publication notes anyway. So far I don't think I've got one wrong. (Touch wood.) 13:04, 20 Jun 2007 (CDT)

(Unindent) One thing to bear in mind if we DON'T all agree, but fall into only two or three separate camps, is that it WOULDN'T require any programming effort to create a new type of verification. If it turns out that, say, most of us have been verifying the text data only we could change the meaning of Primary Verification to reflect that, and add another entry in "Bibliographic References" for "Primary, INCLUDING all Artwork checks". However there shouldn't be too many of these categories and it really shouldn't be used to DECREASE the amount of effort people should make before verification. So no "I think I saw that in a shop once" level please! BLongley 13:18, 20 Jun 2007 (CDT)

I may repeat myself here, and I apologize if I do, but I need to let everyone know where I'm coming from. To me, SpecFic is Short Stories first, everything else is secondary, and my edits show that. Well over 95% of my edits have had to do with Short Stories, including Anthologies, Collections, and Magazines. If I have recently verified a Pub, it means that I have done my best to make sure that everything is correct and everything is entered to the best of my knowledge and ability. In other words, it can be considered fairly complete, and I have left nothing out. If there is data that needs to be entered, I will generally not verify the pub. If I have just added a month to the date, I may go ahead and verify too. This is different then when I first started doing edits and verifications, and I have been removing my verifications from pubs I verified early in my stay here if I am not happy with what I had done. My plan is to go back and verify any pubs I own that aren't verified later after I've finished adding content. While I feel that verification is important, I would now, never verify a pub that had missing content. People who use the db don't come here to see if a pub is verified, they come here because they want to find out what is in pubs. I belong to a Short Story reading group, and the members are constantly coming here to find out what books the stories are in, it's how I got involved here in the first place. If we list Anthologies, Collections, or Magazines as verified, but don't list the contents, what good does it do for anybody? While it is important to know that Roger Zelazny's "Unicorn Variations" Collection was published in 1983 as a HB and reprinted in 1987 as a pb with a cover by James Warhola, isn't it more important and useful to know what the stories in it were? Verification should be used to give people confidence that the information is correct and as complete as it can be to the verifiers ability. An incomplete pub that has been verified doesn't inspire any confidence in me. If we're just using verification to say, "I have this pub and will answer questions about it", then maybe we need another marker for that information. CoachPaul 13:54, 20 Jun 2007 (CDT)

I'm all for the "Do all SF text content" minimum level, and like short stories most myself. Unfortunately we lack some tools that would make it easy to, say, copy contents from the very good content-listed, artwork-credited ideal verified pub to others, and I think some people think "if one edition is right, I can just verify this other one with less detail - they'll search for the good one, right?". Which is wrong - I don't care too much about variant titles in punctuation alone, but copyright issues often mean US and UK collections DON'T have the same content, and I want people to work to correct that. But again, forbidding verification of contentless Collections and Anthologies is a programming change and we're not going to get any complicated changes like that for a while I believe, given Al's recent news. So all we can do for now is agree on what level of verification there should be: or, as I suspect, we won't agree, but can create another level (maybe more, but let's try and keep them to a minimum) so we can have Text-verifiers and Text-and-Art-Verifiers. OK, having mentioned this possibility of different levels of verification with little effort, some people might want to LOWER the bar, I don't want to. But if we do go that way, we have to recognise that we're questioning ALL past verifications. I know I'm questioning some of my past ones. If the active verifiers are still around then it will be a bit more work for them to upgrade their verifications - but given a little programmer time we could downgrade the bad verifications, if we agree that certain verifications are not worthwhile (e.g. no ISBN on a 1990s pub, no contents in an anthology) and it only takes a little effort to pull off lists of "what you've verified that you might want to upgrade" lists for anyone that wants to IMPROVE. For some of us those might be long lists, but if it's in the interest of improved data, I'd go for it. BLongley 14:30, 20 Jun 2007 (CDT)

Oh, and "I have this pub and will answer questions about it" verification as opposed to "I owned this pub, verified it, and don't want to ever speak of it again" is easy, if that's the way people want to go. I think that would affect me most - say 200 pubs - but I already know I should move 50 or so from my OTHER userid anyway, I'm resigned to changes already. And if anyone else that hasn't downloaded the ISFDB database locally needs a reminder of what they verified, just ask me for the list. It'll only be up-to-date as of the last backup, but for those regretting their early verifications it's a good starting point. BLongley 14:36, 20 Jun 2007 (CDT)

(Unindent) What is the protocol for modifying a verified pub? My own inclination is to implement the required modifications and reset the verified flag.--swfritter 08:14, 21 Jun 2007 (CDT)

If you are adding new information then add the data and send a heads up to the verifier. If you need to modify verified information they send a message to the verifier and explain what's up. If it's apparent a verifier made a mistake. Let's say they misspell an author's name a little then I usually make the correction but also send a heads up to the verifier just in case the author name was spelled that way in the publication. In that case the verifier can undo my edit and add a publication note indicating the name is misspelled in the publication.

I guess the #1 rule is you always notify the verifier if you make any changes. Marc Kupper (talk) 03:35, 22 Jun 2007 (CDT)

Thanks. Good. Pre-notification is too cumbersome.--swfritter 07:20, 22 Jun 2007 (CDT)

After a few "I added the cover-art URL to your pub, as I have the same edition, is that OK?" messages, or some such-like question, they may give you carte blanche permission for that type of edit anyway, in which case you can probably stop asking THAT person about THAT type of edit. Remember it's just that one permission from that one editor though. Any Mod that wants to add to/improve my interiorart entries on stuff I've verified is free to do so without asking me, for instance: if they're competent to Mod then they can deal with the bits I'm not really interested in. However, stomp on my cover-art URL entries (or lack of them) at your peril - I do upload a lot of covers from my own edition when I can't find one online already (usually from Jim Gardner), and leave it blank when there isn't one available. So in those cases I do like to be asked if it's OK (after the event is fine, I'm an active editor and WILL revert it if you cocked-up).

A lot of it is just plain "Respect", or lack of it. I'd ask Al or Ahasuerus to check a detail in their verified pubs rather than change it myself and ask afterwards. But I'd probably go fix J. T. Newby's first verification of his first entered publication and leave him a polite note about what we really try to do here, and hope he reads it. It's not a "Level" thing - e.g. I'll challenge Bureaucrats over some things, and make way for far-more-knowledgeable non-mod Editors that know their specialities. I guess MY guideline is "Ask if the verifier is more experienced than you: Change it and notify if you think you know better: and if you don't know better, ask why it's like that and prepare to learn something. " BLongley 16:14, 22 Jun 2007 (CDT)

Heros in Training by Jim C. Hines and Martin H. Greenberg

Jim Hines is an internet aquaintance of mine and I have been in contact with him about his bibliography page, and I will be adding some to it in the next few weeks when I have the time, and he has given me the list of Short Stories for this book. (They were also published on Amazon.com back in February here.) I'd like to add them to the pub, but wanted to get an ok from a mod or two first. He has also given us permission to link to any images on his website, but I can't figure out how to do it. CoachPaul 18:06, 20 Jun 2007 (CDT)

I don't want to encourage TOO much respect for Mods - we're still mere mortals! But if it helps I think it's fine to add the contents, just don't verify it till the book is out. And permission to link is good, but we're still lacking a central area to record all the permissions granted (I've acquired three myself, but all for one author): so just record it somewhere close to the publication for now and we'll blame you if we ever get sued, OK? ;-) Or get Jim to come here and do it himself - it's not THAT difficult to edit here is it? BLongley 16:57, 22 Jun 2007 (CDT)

Personally, I'd rather that he wrote and edited new books as opposed to spending time here at ISFDB. As for the Mods, I don't hold you guys on a pedistal, but generally you guys are the ones that have been around longest, and tend to know more then us plain old editors. Not that there aren't times in which I've explained to a Mod how to do something or other that I had done many times and they hadn't done at all. The one constant to db work on computers is that something new that somebody hadn't seen or done before inevitably shows up sooner or later. And while editing here isn't all that difficult, editing completly and correctly sometimes is. CoachPaul 17:29, 22 Jun 2007 (CDT)

Change to Help - Title Fields - Entry Type - Interiorart - Page Numbers

I have modified the help to indicate that editors have the option to assign page numbers to interior art to either the first page of the story illustrated or the first page where art appears. In addition I have indicated that multiple pieces of artwork for a story can be entered although it is not required to do so. Help previously indicated that the page number for artwork should be the first page of the story but the second option was an older informal standard that many editors adhered to.--swfritter 08:43, 21 Jun 2007 (CDT)

OK. Fine by me. Someone used big words like "lapidary" and "laconic". Might be worth rereading those comments. Might not be worthwhile. That BLongley bloke talks too much at times. Happy medium possible? BLongley 17:20, 22 Jun 2007 (CDT)

The edits look ok - the very last sentence which was part of the original text caught my attention:

Illustrations not attached to a story are given a title of "Untitled".

I've been making these the title of the magazine/publication so that someone looking at the artist bibliography will know what the interior-art appeared in. Search finds just 14 "untitled" interior art titles plus five untitled cartoons. Anyone have opinions about updating the help and fixing up the 14 titles that used that title? Marc Kupper (talk) 01:06, 24 Jun 2007 (CDT)

This is actually part of a bigger issue. There are other titles like "editorial", book review columns, etc. which have multiple non-unique instances. Some editors have started entering pub information (like magazine title and date) in parenthesis as part of the title to make the titles unique. I am hesitant do to so myself because such data is not part of the actual title. It's probably OK, because most such titles will not be reprinted so there is little chance of needing to merge them. I think brackets might actually be better for editor additions because parenthesis are occasionally a part of titles while brackets rarely are.--swfritter 08:07, 24 Jun 2007 (CDT)

The "some editors" may well be me at times because I when I use dupe-candidates on someone and see a bunch of titles all the same, a few minutes ago it was, "Acknowledgments" for Dozois in iirc, then I sweep down the list and add the publication title to each one. You are correct in that if the essay is ever reprinted in a package that has a different title then the (pub-title) method would be sticky. Perhaps (pub-title & pub-title)? Marc Kupper (talk) 02:39, 25 Jun 2007 (CDT)

I think a reprint would most likely have a different title. For instance, several of the essays in Gold appear to be introductions to various books, unfortunately I can attribute very few as they only give copyright date not original publication name. Still, in the identifiable cases, making the Introduction a variant title of the titled essay might help separate things. BLongley 13:45, 25 Jun 2007 (CDT)

I find the most annoying additions of dates to titles are the ones like "Faust Aleph-Null (US 1990)" - there's no need to mess up the title like that, it's a COMPLETELY different name to the original: and you can keep the date of the change in the variant title record, and the countries that it varied in should be clear when some publications are added. (E.g. did it vary for Canada too?) BLongley 13:45, 25 Jun 2007 (CDT)

I actually want to fix this in code so that editors can enter the title exactly as stated and the display software will see that a bunch of essays or whatever have the same titles and it will then look up the publications and display the title as "Introduction (pub's title)" (or we can use [] as you suggested). Marc Kupper (talk) 02:39, 25 Jun 2007 (CDT)

I think Marc has gotten to the crux of the matter. No publication data appears on the Author pages, just titles. If the first publication data appeared next to or under the title the problem would be solved in most of the cases were discussing. The Locus databases actually indent under the title and show all publication dates but that could get yucky with the ISFDB because there is so much more data. [] might be out because they are also used for such things as series. Perhaps the earliest publication for every title, not just the ones we are discussing, could be listed. I am sure that idea is full of pitfalls too. There would have to be some way of informing the user that there are multiple entries for the title. And I am sure there are a thousand other issues that implementation would give rise to.--swfritter 15:01, 25 Jun 2007 (CDT)

A closer look at the data for for non-unique essay entries ("Introduction", "Editorial", etc.) convinces me that so many them have already been modified with the name of the publication in which they appear that it may be too late to do anything.--swfritter 17:22, 26 Jun 2007 (CDT)

Perry Rhodan - Book or Magazine?

Ready to start on this project as I have time and copies. rbh 21:14, 23 Jun 2007 (CDT)

If I recall correctly, some issues simply reprinted a single novel, but many others had a smallish section in the back that contained serialized reprints of 1920s-1930s SF plus letters to the editor and Forry's musings, so I would think that the whole Ace series was definitely a magazine.

Also, keep in mind that Ace reprinted PR novels out of order and then went back and filled in some gaps. For magazine numbering purposes, the Ace order (1-118 or so) is fine, but for Series order purposes we presumably want to use the German order since it is roughly chronological.

Also, there are numerous subseries in the PR universe, e.g. the Lemur series is about to be published in the US -- see this recent discussion on Usenet. Ahasuerus 21:30, 23 Jun 2007 (CDT)

Spoilers in Tags

Dave Tate has posted the following comment on Usenet (http://groups.google.com/group/rec.arts.sf.written/msg/ecb63be507fd0dd0):

Apologies for posting this here, but I am apparently too old and stupid to figure out the ISFDB Wiki instructions.

Question: does ISFDB have a policy on spoilers?

Specifically, I note that authors and individual books may now have associated *tags*. For example, if you go to Mercedes Lackey's author page, you see the associated tags 'fantasy', 'occult thriller', and 'dragons'. If you click on 'dragons', you see a (currently very short) list of individual titles by various authors.

That's great, and I can see how it would eventually be very useful. However, some of the specific tags represent Big Honking Spoilers (tm) for the books they tag. The tag "lost colony" seems like a particularly likely suspect in this regard. I can think of several books -- including some of the ones already tagged -- for which it would be a major spoiler to know in advance that this is a "lost colony" story.

I imagine there are other likely tags with similar spoiler potential. Is this a concern for the ISFDB team?

I responded as follows:

That's a very good question, thanks for bringing it up. I wrote the current Tag help page (http://isfdb.tamu.edu/wiki/index.php/Help:Screen:AddTags) in response to a few editors' questions some time after Tags went live, but it didn't occur to me to mention our spoiler policy. The Edit Title help page (http://isfdb.tamu.edu/wiki/index.php/Help:Screen:EditTitle) has a comment about "non-spoiler" synopsis data, so it's safe to assume that we want Tags to follow the same policy. I will update the Tags help page and post a note on the Rules and Standards board (http://isfdb.tamu.edu/wiki/index.php/Rules_and_standards_discussions). Feel free to stop by and comment -- all you have to do is create an account by clicking on the "Create an account" link at the top of the page :)

I have added a line to the Tag help page: "Please do NOT add tags that spoil the plot." Would you say this is sufficient or should we dig deeper, e.g. cover situations when a variant title already spoils the plot, e.g. Flight into Yesterday? Ahasuerus 19:24, 24 Jun 2007 (CDT)

That's a tough one because almost by definition tags spoil some element of the plot. Sometimes even organizing the book into a series could be considered "spoiling the plot." Maybe a feature request to disable display of the tags would work as I'd want the freedom to be able to tag titles with plot elements (which is what I thought the point of tagging was). I personally am not fond of tags as most tagging systems do not have a mechanism to add definitions to the tags (quite a few of ISFDB's tags are mysterious to me – Jack Vance pastiche anyone?) while sometimes there are multiple tags to describe the same or similar things (asteroid, asteroid belt, asteroids, faery, fairytale, humor, humorous, sub-light speed, sub-light travel, etc.)

While looking at the tags list for overlaps I saw "space pirates" and realized that in the book I just finished (and tagged) much of the condition the characters were in was caused by space pirates, they thought and talked about space pirates a lot, but it's not until the very end of the book that anyone shows up that may be a real space pirate though even there they could have been classified as religious, political, and/or business zealots. Thus, even though the entire book was about space pirates there were none physically in the story except for one scene. Does the tag apply and would it spoil things for someone? Marc Kupper (talk) 04:10, 25 Jun 2007 (CDT)

The spoiler issue is an important one for me. I see the ISFDB as a research tool and the synopsis data as critical information for academic research. The site is, after all, hosted by a university. Restricting synopsis information reduces the academic value. My primary interest is the SF magazines of the 50's and 60's which are generally unavailable to most people. I prefer the Wikipedia method of putting a spoiler alert before the acutal critical discussion. I can understand a policy of not allowing spoilers for recent works but I think they should be allowed for works older than 10 years as long as there is an alert. As for tags, they should not be spoilerish because the viewer will not have the opportunity to avoid them the way then can if there is a spoiler alert at the beginning of a synopsis.--swfritter 12:08, 25 Jun 2007 (CDT)

Pseudonyms With Unknown Real Names

Is there a way to indicate that a name is a pseudonym, even if the author's true name is unknown? "Arthur Pendennis" is clearly a pseudonym: he is the title character of a novel by Thackeray, and the listed author of another novel of Thackeray's. Somebody wrote a story under his name (The Prince of Snobs) in the Sept 1983 Amazing; the editorial blurb describes it as a lost work of Pendennis', and further describes Thackeray's novel as a biography, so readers were clearly expected to get the joke. I can't find any references online to who the author might be, however. Is there a way to show that this is a pseudonym? ("B. T. H. Xerxes" is a similar case.) Jefe 14:19, 2 Jul 2007 (CDT)

The real author is Arthur Jean Cox according to the Locus Science Fiction, Fantasy and Weird Fiction Magazine Index on CD. Someone still may want to address your original question. One place to make such a notation is in the bibliographic data for the author.--swfritter 15:10, 2 Jul 2007 (CDT)

Hmmm.... I had this problem recently. Ah, found it: Amazon say:

About the Author
Everett Owens is a pseudonym for a well-known children’s book author.

Anyone got any idea who it is? BLongley 16:50, 2 Jul 2007 (CDT)

Xerxes could be either Tom Boardman or Brian W. Aldiss according to Robinson in "Who's Hugh". His sources are Contento for Boardman and informal information from John Clute and others for Aldiss. I would probably but a note on the story and in the bibliographic information for the author.--swfritter 18:29, 2 Jul 2007 (CDT)

ISFDB doesn't do a very good job of supporting undisclosed pseudonyms and house names at this time. If you want to know what stories were published as by "Alexander Blade" or Victor Appleton, you will get what amounts to a disambiguation page instead. This page will point you to Summary pages for all known individual authors (as well as "unknown") who wrote as "Blade" and "Appleton". This works reasonably well when we know who wrote what (although we lack an integrated view of all stories published under that pseudonym), but it doesn't work too well when a significant percentage of the works is attributed to "unknown". At one point Al wanted to create a list of requirements for "pseudonym bibliography" logic, but I don't think he has made much progress in this area so far. Ahasuerus 18:51, 2 Jul 2007 (CDT)

Help Update - Book Review Dates

Unless there is any feedback I will be changing Help to indicate that the Book Review date should be the date the book review was first published (usually the date of the pub being edited) rather than the release date of the edition of the book reviewed. I will also make a notation that current data need not be modifed because future enhancements could make the current standards obsolete.--swfritter 12:42, 7 Jul 2007 (CDT) Changed to - The date the review first appeared. Normally the date of the publication in which the review appears unless it is a reprint. (Note: if you leave this field blank it will default to the date of the publication being added or modified).

I just made a followup change to the pub-editor help about this, & another (hopefully minor) change in the same section. This is for Reviews. I changed "date of the work being reviewed" to say "date of the review".
I also changed "There is room for three reviews in the initial screen display" to say "There is room for one review in the initial screen display" since this is, AFAICS, the way the pub editor works. I hope that's OK. Thanks. -- Dave davecat 13:24, 5 Nov 2007 (CST)

Ack! I then almost immediately observed that I needed to change "The screen will redisplay with a fourth review record visible" to say "The screen will redisplay with a second review record visible". So I did that, too. -- Dave davecat 13:29, 5 Nov 2007 (CST)

Looks good, thanks for the fix! Also for mentioning that you made the fix: some edits do seem minor but eventually people realise they could cause problems not thought about. See this discussion for instance. (And if there's no further input on that, I think I'm going to propose a clarifying edit back: "Artwork should not be merged" is too absolute a rule for my liking. I waffled us off-topic there though, I think. :-/ ) Still, so long as we're open about proposed, or even past, updates to Help pages it can be sorted fast enough. BLongley 16:17, 5 Nov 2007 (CST)

Oh, and I recently discovered that it's not just help we have to update at times, for instance my last proposed update affects the FAQ too. Not one you have to worry about in this case, but it was one I had to worry about. And that last entry is a nightmare... :-/ BLongley 16:26, 5 Nov 2007 (CST)

Consensus on "Dateless" Printings

Can we come to sort of consensus or compromise on dateless printings? I'm with the crowd that would see some sort of date be shown. IMHO having a "Thirty-second Printing of the 1964 Berkeley Medallion Edition" printed sometime in the late 70's or early 80's appear before the original "Galaxy Magazine Printing" from 1958 just seems wrong to me. One of the other Mods had mentioned using the print date of the first printing of the edition and adding the "Print Number" to the Day Field of the Date Field, and I can live with this if there is nothing in the programming to foul this up. Another option I can live with would be to use the last known good date of the Edition, and adding the Print Number to that Date Field to come up with a new date field. CoachPaul 20:36, 14 Jul 2007 (CDT)

I think it a good idea for the moment and Marc has used this idea on some DAW books. If we go forward with this approach there would have to be a VERY CLEAR statement in the notes about whats been done. As a minimum the printing history in full would have to be in the notes.Kraang 21:43, 14 Jul 2007 (CDT)

I've been gently pushing Al to do a couple of things so that I can get a working copy of ISFDB as this is one of the first things I want to fix. I really don't like what I'm doing with the dates though agree it works. Marc Kupper (talk) 19:45, 15 Jul 2007 (CDT)

Whatever we decide to do, we probably want to ensure that it won't be a problem for the eventual software fix that may be put in place. Ideally, it would facilitate the latter. For example, if we decide to add a "Printing number" field to Publication records, we may be able to write a database conversion that will populate this field based on the data in the Date field. Ahasuerus 00:31, 16 Jul 2007 (CDT)

The risk with such a script is that REAL days (or Amazon Days) in the date field could get converted into printing numbers. If we want to automate such, then we need something to indicate where the day in the date was real or used for printing number. And there's not many fields to use for that, it does seem to come down to notes. However, We COULD come up with a convention on what to put in notes to indicate the Day has been used - something that wouldn't naturally be entered as a note for any other reason though, so it would have to be something obscure like "QZXJZ" if we want it short or "DAY=PRINTNUM" if we want it vaguely readable. BLongley 14:03, 16 Jul 2007 (CDT)

The other worry is what the final "Printing Number" solution could be. In the recent example we can already date 1st, 3rd and 5th printings to some extent. Two solutions might be:

1 Leave them as all dated with the one date in the actual pub, (i.e. First Printing Date) and use Printing Numbers 1-5. 
2 Add calculated dates and use Printing Numbers 1-5. 
  Printing 2 would be left with same date as 1, 4 with same date as 3, but 1, 3 and 5 would have a reasonably sure Year and Month.

I'm not sure where we might EVER get dates for the reprints with no reference number/price/ISBN change so I quite like the second option, but WHEN you try and work on dates for the whole set is up for debate. The other worry is printing numbers across imprints: how many editors would spot that a 6th printing in Grafton might have a 5th printing in Granada and a 4th printing in Panther?

Overall, this is why I've left my controversial ones with the verifiable date of a prior printing and the actual printing number in notes: when it comes to manually fixing them there should be small "clumps" of pubs that appear identical until you read the note, then one is clearly a different printing number from the other. (Most of my notes seem to be about printing history.) I don't mind Marc's method either for now, he leaves copious enough notes. I think the compromise for now is to accept either. I'd love a consensus on a way to make it fixable with a bit of SQL when the way forward IS decided, so DO consider a keyword in the notes and I'll move to Marc's method. It might be an idea to mention such a consensus or convention in the Help, and drop a note to all active editors about the change as well. I'm not rejecting "0000-00-00" new entries but I am querying changes from something identifiable TO "0000-00-00" as that seems to be destruction of data to me, and the current guidelines seem to be encouraging it. :-/ BLongley 14:03, 16 Jul 2007 (CDT)

I've hosed myself too often with scripts to globally search/change things and don't do them any more. They seem like a good idea at the time but you generally loose the audit trail. I'm all for scripts to identify data for the humans but even there I get concerned if the human's task turns into an unthinking "approve the change." I'm not too concerned about the existing publications that are already using the dates set. We can find them with scripts if needed and over time as people edit the publications the new data will get filled in. We could agree on some code word or phrase to put in the notes to make the scripted scan easy.

The first task for me is to get a working ISFDB system. Right now I have a working database, working MediaWiki 1.10, but am waiting to hear back from Al on if he's made progress in converting the Python code to know about the new MediaWiki table structure. I could set up MediaWiki 1.4.5 but the code for this is not archived and so would need to get that from Al. Marc Kupper (talk) 20:04, 16 Jul 2007 (CDT)

Foreign Language Magazines

There are a few English language magazines published outside of the Anglophone market, e.g. Orphia, and most genre bibliographers (including us) lump them together with regular magazines. Not a problem so far, but the next step (regardless of whether you are working on magazines, novels or anthologies) is to identify the original foreign language Titles and make them our primary Titles -- see Jules_Verne's bibliography for an example.

Unfortunately, this is not always a trivial proposition even if you find the author's bibliography in his native language. After all, editors love to make seemingly inexplicable changes like Non-Stop -> Starship, and it only gets worse whenever there is another language involved. For example, the Strugatsky brothers' Waves Extinguish the Wind was published in English as Time Wanderers even though the book had nothing to do with time travel. (As an extra bonus, the publisher failed to advertise that it was book 3 in a trilogy).

However, let's suppose for the moment that we have found the original foreign language Title and created a VT for the translated/derived Title(s). If the Title first appeared in a book, then it should be easy to create a Publication record for it and populate the record with whatever data you can find in OCLC and other online sources. But what should we do if the Title first appeared in a foreign language magazine? I suppose we can create a Publication stub for the latter, but should we list it in our Magazine directory? Do we want to have a "foreign language" section for them, with (perhaps) individual subsection for French, German, etc magazines? And how can we tell that the magazine in question is/was a genre magazine and worth listing in the directory as opposed to a New Yorker clone that publishes spec fiction once a year or three? Decisions, decisions... Ahasuerus 16:02, 16 Jul 2007 (CDT)

I speak only as a monoglot myself, but frankly I think trying to cover "foreign" languages here is going to put editors off, maybe some mods too. There's enough problems with English, American, Canadian, Australian, New Zealander, South African, etc, differences already that I'm quite happy to let the polyglots[*] encourage sites like this in their own languages. I've already stumbled across sites like the DSFDB that might be on the same lines as us, but I can't tell as I don't speak the language. (And the site seems short of bandwidth too.) I've found French sites that are useful, talked to Danish and Icelandic and Japanese webmasters to get permissions for coverart etc, but frankly I'm unqualified to moderate any foreign-language submissions here.

I've also found myself looking at Wikipedia sites that AREN'T "en.wikipedia.org", there's OTHERS out there too it seems! And sometime they're quite good, too! So I'm actually in favour of us dropping the "THE Internet Speculative Fiction Database" title and becoming "The English-Speaking Internet Speculative Fiction Database" - but I'm definitely for encouraging links to the other SFDBs out there. Maybe we can donate our Manga entries that are due for zapping under current rules back to the Japanese SFDB or somesuch? BLongley 17:17, 16 Jul 2007 (CDT)

[*] or even the "biglots", but that sounds more "Nac Mac Feegle" than English of any kind. :-/ BLongley 17:17, 16 Jul 2007 (CDT)

There are quite a few categories here as we currently state in our Rules of Acquisition:

1. Foreign language translations of speculative fiction works originally published in English or written in English and published in another language (e.g. Bulmer or Dibell)
2. English language translations of works of speculative fiction originally published in foreign languages. In these cases, we will also provide information about the original foreign language work.
3. Works of speculative fiction published in a foreign language that haven't been translated into English, but whose author's other works have been translated into English. This is done to make it easier for people who are interested in, e.g., Lem or Barbet to see as full a picture of the author's work as possible.
4. Works of speculative fiction published in a foreign language that haven't been translated into English and whose author's other works have not been translated into English. Arguments for exclusion: avoid duplicating the efforts of foreign language bibliographers in a field where we can't realistically compete with them. (True? False? Revisit if/when we have foreign language editors with extensive expertise in the field who would be willing to merge their biblios into the ISFDB?)

The main reason to include (1) is to provide comprehensive coverage of foreign language translations of English language titles. This is typically done by single author biblio sites (e.g. Wilson Tucker, Philip Jose Farmer, Jack Vance) as well as by comprehensive encyclopedias, e.g. Tuck. It is also valued by working SF authors who like to see their complete bibliographies presented in one place and not split across multiple sites, especially if they are in multiple languages. Given our goal of being a comprehensive speculative fiction bibliographic site and our history of support for foreign language translations, I doubt this will change. At some point we will need to beef up our support for foreign language translations of short stories, which is pretty much non-existent at this point.

The first sentence in (2) presumably needs no justification :-) The second sentence ("provide information about the original foreign language work") is trickier and was the reason that I started this discussion. Do we want to list original foreign language Titles only? If we don't list the associated Publication records, then how can we Verify our data? And if we do add Publication level data, do we want to drill to the same level of detail that we support on the English language side (see the Magazine discussion above)?

(3) is a potentially largish can of worms. When the bulk of a writer's work has been published in translation -- e.g. Verne, Lem or the Strugatskys -- it's pretty much a no-brainer. But what should we do with someone like Sergei_Lukyanenko? The first three books in his recent series have been translated, so clearly we want to list the fourth one as well. On the other hand, he has written dozens of other genre works, none of them available in English. Should we list all of them as well?

I suspect that (4) is where our inclusionism will be really tested. Unless we find more editors willing and able to maintain complete bibliographies of additional hundreds (if not thousands) of writers and/or unless we form alliances with other sites, we will probably be in no position to do this area justice. Ahasuerus 18:59, 16 Jul 2007 (CDT)

Well, it's been a week or so now and I don't see any volunteers to help with (4). Nor (3). Or even (2), second sentence. Or (1). We might need to add (5) for English-writing authors that have had some works ONLY published in foreign languages so far, e.g. A. E. van Vogt. ( I think John Brunner may have had similar troubles with some Esperanto editions.) Basically, I have no problems with people adding non-English data here, I just feel I (for one example) am not qualified to deal with such entries. I'd prefer to leave it to the experts - and if we don't get such experts here, then let's just link to the expert's sites. BLongley 18:55, 26 Jul 2007 (CDT)

Interesting points, Bill, but at the moment I am trying to use my precious "collection time" to Verify as many publications as I can while juggling 8 other things. I'll be back on the road by late Tuesday/early Wednesday, at which point I will be able to comment further. Ahasuerus 15:49, 28 Jul 2007 (CDT)

Minor variations in a publication

I have a copy of Cordelia's Honor that has a special dustjacket. I just made a note of this in the notes instead of a new pub. Does that seem reasonable to people? It's already confusing enough with the limited ed hc having the same ISBN as the trade paperback. Dana Carson 12:24, 1 Aug 2007 (CDT)

That's a tough call. Most of the time I add a note much like yours and other times I clone the publication. For example, I had a publication where on the copyright page they detailed how many of various types of limited edition (hardcover, slip cased, gold leaf binding, etc.) works they had made and for that I just had a single publication record with the note that included what was stated on the copyright page.

Part of the problem is ISFDB does not have many fields to help show that a publication is distinct when you are viewing them. I've sometime done it by adding a (parenthetical note) to the publisher name field.

Dana - your note was mildly confusing in that it's not quite clear if all of the hardcover copies were this "limited edition of 500". Something I believe would help if if you would indicate exactly what is stated in the publication or the source of this information. Marc Kupper (talk) 00:00, 2 Aug 2007 (CDT)

From the front flap of the dustjacket. The Stars our Destination, Chicago's premier science fiction bookstore since 1988, is proud to present this special dustjacket for Baen Books' limited hardcover edition of Cordelia's Honor. The fron of the dustjacket has "A special hardcover edition containing the complete texts of Shards of Honor and Barrayar."

It doesn't say how many copies of either the hardcover or the special dustjacket there are.

The copyright page has the ISBN of the trade paperback and cover artist for that. And the LOC data block has P.p. verso and ISBN 0-671-87749-6 (pbk.)

I don't remember where I bought it. Not at a regular store but if it was just one of the dealers at a con that carries speial editions or at a con where Bujold was GOH no idea. Can't find anything about 500, maybe I thinkoed from Dreamweaver's Dilemma which is also a limited hardback. But my note seems too solid for that and the other was hb and slipcover. Why didn't is source where I got that from. OK, google to rescue, http://www.locusmag.com/index/b75.html matches the 500, says no dustjacket so the special dustjacket was probably just for a con where she was GOH or just for ones sold through that store? Dana Carson 16:26, 2 Aug 2007 (CDT)

Annotating Excerpts

As a relative newbie, I've noticed that some ISFDB entries contain excerpts from pending novels as a content entry. I'd really like to know why. This is advertising material and is no different from the list of forthcoming books or one-page catalogs that used to appear to fill out the blank pages of a print run. It seems like a waste of time. Inquiring minds want to know . . .

--Dsorgen 22:14, 31 Jul 2007 (CDT)

If you're referring to the few pages in the back of an otherwise unassociated work, I agree. But when novels are excerpted in periodicals, it's a different matter entirely. (Someone could argue that they're only promoting forthcoming works, as well.) You do have to admit that there's no comparison between a list of books from a publisher and excerpts from the book itself. The thing is, it's hard to make a value judgement when determining what fiction or non-fiction (and in whatever form) to include in the database. It is a good question though and one worthy of discussion... Mhhutchins 23:26, 31 Jul 2007 (CDT)

The stuff that bothers me appears ONLY at the end of books, is almost always unpaginated and normally promotes another work by the same author. One of the troubles is that there is NO guarantee that the work will ever appear. They seem to serve roughly the same purpose as movie trailers and may (or may not) have as much to do with the final work. I heartily agree that discussion is warranted; since, the instructions in the help system say to include such material.--Dsorgen 21:43, 1 Aug 2007 (CDT)

I'd vote for making these optional until ISFDB is updated to have some sort of generation flag for records. In other words, we'd record the current records as "1st generation" and if there's a rules change to include these advertising excerpts then any records verified after that point would be "2nd generation." I see including the advertising excerpts as useful if someone wants to do research on book advertising. For a while I was entering the advertising lists from the back of DAW books into my database. I was trying to capture price points for the books and their catalog #s. What I started to find is these lists had many errors plus the data entry was tedious and so I stopped. Marc Kupper (talk) 00:13, 2 Aug 2007 (CDT)

I recall one publication where the "excerpt" turned out to be a complete short story from a collection. That's rare though. It might inform us about a title-change between planned publication title and actual, but that's rare too. Definitely worth a debate. BLongley 14:37, 2 Aug 2007 (CDT)

OK, new example: this pub contains one chapter of a "Serial Novel", "Starfleet: Year One" apparently published amongst the monthly Pocket Books Star Trek releases from August 1999 on. There's supposed to be twelve parts but unless we check the "excerpts" for each month's releases we'll never know how to construct the complete novel, which I think appeared in Complete form elsewhere but we haven't got that. OK, I suppose most of you don't actually care, but it's one reason FOR recording excerpts, so we can put them back to proper serialisations where relevant. I'll save that till we have some verification of which EDITIONS have this "extra", and who wrote them, as the current cover-art at Amazon suggests the idea may have been dropped, or that some editions no longer have the excerpts. (In which case the entire novels with excerpts become variants of themselves or something...) BLongley 13:17, 27 Aug 2007 (CDT)

Oh, and rereading this thread, the "NO guarantee that the work will ever appear" makes me thing it might be MORE important to record the excerpts, in some cases. If the final work never appears, then this is the ONLY record of the existence of parts of it. Not every author has a Christopher Tolkien or Brian Herbert to finish it... BLongley 13:17, 27 Aug 2007 (CDT)

(unindent) I'd like to reopen this discussion. It seems to me that, at least, unpaginated excerpts of novels that have, in fact, already been published and are already in the database add nothing, and merely clutter the author's bibliography, and perhaps series listings. Excerpts that amount to a serialization, such as are mentioned above, seem worth recording. Excerpts from forthcoming books, that probably will be published, but might not, may be worth recording if there is a chance that the book never comes out, is long delayed, or comes out is very different form. But if the book does come out and there is no significant textual difference (not counting minor copy-editing as significant) then we can probably safely delete the excerpt, IMO. I'm not sure on other cases. But what ever the decision, i think we should come to a consensus, at least on the clearcut cases, and update the help pages accordingly. -DES ^Talk 13:46, 8 Feb 2008 (CST)

I'm leaning towards only including excerpts of books not already present: I'm currently NOT including excerpts if we have the forthcoming book already, unless on numbered pages. That keeps my workload down as I don't tend to buy new books with this problem. I'm never particularly keen to delete contents already here though - it might be a useful pointer to somebody wondering whether to buy a book to be reminded they own an excerpt already. BLongley 14:11, 8 Feb 2008 (CST)

Another discussion I'd like to reopen is whether it's acceptable to move a section like this to the bottom of the page if it becomes active again after some time - given that we tend to archive in order of creation rather than latest activity, it would save losing sight of active discussions. BLongley 14:11, 8 Feb 2008 (CST)

Amazon Images

OK, it seems clear that Amazon aren't going to keep their '01.LZZZZZZZ.jpg' URLs stable. Consequently, I'm not going to use them any more, and would encourage everyone else NOT to Verify such pub entries: but I HOPE that the g-ec1/2/3 URLs are stable enough to continue with. Anyone else have evidence about THOSE URLs stability? I don't mind so much if they very occasionally disappear, but I don't want them changing to something wrong. BLongley 15:02, 5 Aug 2007 (CDT)
I'm now also very much in favour of moving hosts, or at least getting permission from our current hosts to upload our own cover-scans. I've got over 400 uploaded to the UK Amazon site and 170 to the US one, but if they're not going to be reliably useful then I'd rather the ISFDB has them stored instead. It's only about 170MB of images so far, and even if I did my full collection of books that's under a Gigabyte. Make it two or three gigabytes if I do the magazines and fanzines currently in my possession (if the scanner doesn't wear out!). BLongley 15:02, 5 Aug 2007 (CDT)
I guess what we do in the long-term is up for debate depending on the answer to the first question, but for now I suggest all Editors and Moderators try and switch '.LZZZZZZZ.jpg' URLs for g-ec1/2/3 URLs instead, ESPECIALLY the ones that are currently linking to nothing. (Who knows what could appear there in the end?) BLongley 15:02, 5 Aug 2007 (CDT)

I certainly agree that we are currently at the mercy of Amazon.com as well as other third party hosting sites and few things are more irritating than seeing one's work destroyed or at least compromised. Hosting our own images would solve this problem, but it would also raise a few other issues. First, we would be exposed to a variety of legal issues that we are currently (mostly) insulated from. Second, it would raise disk space and especially bandwidth requirements tremendously, which would not only make our backup files much larger, but would also make it more difficult to choose another reasonably priced hosting solution if we were to decide to move. I suppose the backup question could be solved by creating two backup files, one for pictures and one for text data, but the disk size/bandwidth question may prove to be problematic and we will probably want to run some calculations first. Also, IIRC, Al did a fair amount of research in the ever murky copyright area at some point and hopefully he will be able to weigh in shortly.

P.S. I am sorry I have been mostly inactive over the last week, too many other things going on :( Ahasuerus 21:43, 6 Aug 2007 (CDT)

Well, I'm not talking about pinching Amazon's images and storing them locally, just keeping control of our own images. I could host all mine on one of my own sites and link to that but that's not necessarily stable (I have switched ISPs before when they've changed terms and conditions unfavourably) and I'm not acquiring a server of my own just for that purpose if we can all club together. There's no bulk updates available here if I had to move them, either: of course Al could just update the prefix in SQL if necessary but he could also do that if we had to do an emergency take-down if legal problems ever did arise. How many of us are we talking about anyway? I've only noticed my and Marc Kupper's image libraries so far, are any others of us uploading? It might be quite a small problem in the end. BLongley 12:10, 7 Aug 2007 (CDT)

(Unindent) It seems we might have to decide sooner rather than later. More and more Amazon pictures are becoming "Search Inside" only, and when that happens we only seem to know how to create the '.LZZZZZZZ.jpg' URL. Which is not stable for anything currently in print. I've worked around a few - if it's a US pub that went "search-inside" only, the UK site might have a g-ec1/2/3 image, and vice versa. But this might be time to admit that all Amazon '.LZZZZZZZ.jpg' images are only ever going to be reliable for CURRENT editions, in which case it might be time to move them to "Title" level. Except we'd have to show the latest hc/tp/pb editions for each country to be sure of even that. :-( BLongley 19:18, 1 Sep 2007 (CDT)

I'm still uploading new images to Amazon, especially with my out-of-print editions, but really want a better solution as to where we can store OUR scans, if we're not going to be able to rely on Amazon. Is it time to talk to WikiMedia or suchlike if we're still too scared to store them ourselves? BLongley 19:18, 1 Sep 2007 (CDT)

One for the "EXACT title, including all punctuation and special characters" crowd

I've verified this pub as it DOES have the "®" (registered trademark character, (R) if it doesn't come through on your browser.) If it didn't have the "The author gratefully acknowledges the kind permission granted by Technicolor Corporation of America for the use of its trade name and registered trademark in the title of this book" I'd probably complain about the lack of a "U" for the British publication instead, but for once I'm going to be particularly anal about this and see if people REALLY want to be absolutely correct about things. ;-) BLongley 16:57, 5 Aug 2007 (CDT)
We've already decided (informally) that apostrophes don't matter, how about a character that could get us sued if we don't use it in the right places? Maybe a "©" (copyright character like (C) or somesuch) or "™" (trademark character (TM) or suchlike) is important at times? Do those even appear correctly to you? If we need to use them, do we need a standard way of representing them? If we don't, can we agree when we don't need to be so anal about when it's an Em-dash or an En-dash, double-quote or single-quote, whether we use a colon or a dash between title and sub-title, etc? I'm leaning towards simplicity, or at least to making sure searches find ALL the English variations in one go, but I'm open to suggestions. BLongley 16:57, 5 Aug 2007 (CDT)

I don't see an ® on the front cover. Is it on the title page? There is a comment that the cover and spine don't include the leading "The" in the title but the comments are silent regarding ®. You included this quote "The author gratefully acknowledges the kind permission granted by Technicolor Corporation of America for the use of its trade name and registered trademark in the title of this book". I assume this is something stated in the publication? If so, it seems like a useful thing to add to the publication notes.Marc Kupper (talk) 02:50, 14 Aug 2007 (CDT)

Yes, and it will be: however, I'd like to gather some views on whether I should remove the "®" or change it to (R) or leave it as is before I try and explain WHAT the differences are. BLongley 14:08, 14 Aug 2007 (CDT)

As for your general questions. I suspect publications should stick with plain-text ASCII text as much as possible with an exception made for foreign language characters. We already diverge from EXACT duplication of the title in that we often enter "title: sub-title" where the publication often indicates the sub-title by changing the font or with a line break. Sometimes it's tough to tell what the exact title is and in that case I add publication notes.

I believe the goal to keep in mind is that we are trying to document publications well enough that someone else can determine if they have the same publication or should create a new record.

A secondary goal is I try to document things that can cause confusion. Your pub is a perfect example of this where people may state the title as

• The Technicolor® Time Machine
• Technicolor® Time Machine
• The Technicolor Time Machine
• Technicolor Time Machine
• The Technicolor (r) Time Machine
• Technicolor (r) Time Machine
• I won't get into the variants of the form "Technicolor® Time Machine, The"

Thus it's great that you already noted that the title on the title page is different than the cover/spine as it will explain to someone researching this over the Internet on why there's so little consistency with the title in book lists and seller listings. Marc Kupper (talk) 02:50, 14 Aug 2007 (CDT)

I think the worst for this pub is that some Queen's-English-speaking people will look for "Technicolour" :-(

But yes, it's a good example pub which is why I used it here. I also find searching for titles with "(r)" or "(tm)" in leads to titles that need to be reworked to strip out series names, which some people seem to like to leave in at the publication level. I don't, but sometimes am forced to as the series title IS the title of the first book,