Rules and standards discussions/Archive/Archive05

Canonical name for Artists

Authors are pretty simple - most commonly used or recognized. The standard that seems to be developing for artists (now that we have significant data in the system) is the fullest name with credited artwork (Frank Kelly Freas instead of Kelly Freas/Ed Emshwiller instead of Emsh). Does that sound right? If so, Help should be updated.--swfritter 18:26, 14 Jan 2008 (CST)

One reason to go with the longest used form of the name is to avoid disambiguation problems down the road when another, say, "Knight" shows up. On the other hand, there is a fair number of artists who have used an abbreviated form of their name almost exclusively (as far as we know), e.g. Blair. Using a very infrequent form of an artist's name as his/her canonical name just because it's the longest one may lead to a lot of variant titles that could be otherwise avoided. In other words, I am firmly undecided on this one. Ahasuerus 18:58, 14 Jan 2008 (CST)

Perhaps a threshhold? A percentage or "significant number of credits" but not necessarily the most. That would probably mean that Ed Emshwiller is out (Emsh in - my preference) even though he is categorized as the canonical author for four signatures. There are a number of covers assigned to Ed Emshwiller but I think that is because they were entered from secondary sources which is a factor that has influenced perceptions of canonical artist name selection. Frank Kelly Freas, on the other hand, probably is more commonly credited as Kelly Freas, and I think by a wide margin, but early in his career was many times credited as Frank Kelly Freas. Enough times? Sometimes it will just have to be a judgment call. But once someone makes the decision it is hard to go back.--swfritter 20:21, 14 Jan 2008 (CST)

Interior artwork attributions

Artists are generally credited in as many as three places in a magazine - the table of contents, title page of the story being illustrated, and the author's signature on the work. I generally use the fullest name used to credit the author. For instance, Virgil Finlay might be credited as V. Finlay on the table of contents, Finlay on the title page, and sign his name Virgil Finlay. I use Virgil Finlay. If an author is credited on the table of contents with a full name and signs his last name or initials that clearly correspond to that name I use the name in the table of contents. If the fullest credit is a last name I use only the last name. Initial are tricky and require a judgment call. Most editors seem to be using the artist's name for commonly known artists rather than entering initials. In any case, it seems like a good idea to always make a note in the notes of the magazine when signatures and initials are used to determine artist attributions. How close is that to a policy we can agree on? Any additions?--swfritter 18:45, 14 Jan 2008 (CST)

Generally I'm inclined to agree.
An additional issue I've run into is how far credits can be applied. Two different questions come to mind.

• In Analog for quite a few years, normal practice was to list a credit (normally after the author's name) for illustrations as "Illustrated by Kelly Freas", with no other credits. At other times normal practice was to simply put a credit (typically in italic type) Kelly Freas by the first illustration in the story; sometimes this was done even in those years when an "Illustrated by" credit was normal. Consistently, where I could verify (by signature/initials or by clear style), the illustrations have all been by the artist credited on the first one. I've thus been taking this as a credit for the whole story, without bothering to add notes to a pretty large number of issues for a large number of stories. Is that OK? (I can imagine that in a different magazine this might be done differently, though it would strike me as kind of perverse.)
• In a very few cases, even where there's an "illustrated by" credit, there are special or unusual illustrations which aren't (to me, anyway) clearly covered by the credit. The two crude diagrams of time travel trips in "The Time-Machined Saga" (serial in Analog), & a couple of maps in other stories, come to mind. I don't have the book The Technicolor® Time Machine to check, so I'm only guessing that those diagrams were Harrison's & appear in the book. I'm not sure I've been at all consistent with this kind of thing, except that I've tried to remember to add a pub note whatever I did.

I own it, and the 'art' is there on pages 57 (diagram), 109 (map), 148 (diagram). All uncredited, so I'd assume Harrison: there's no other artwork so I presume the Schoenherr work is omitted. I'm not keen on recording such minutiae for books (I'll record major interior-art illustrators like Pauline Baynes, but not every picture) but if it helps with a magazine I'm happy to check books for you. BLongley 13:08, 15 Jan 2008 (CST)

Ooh, cool! I'll go look at what I entered in those issues, & at least clear up the pub notes. Thank you!! -- Dave (davecat) 13:47, 15 Jan 2008 (CST)

I'm not trying to quibble, only to toss in some additional complexities before anything is carved in stone. If we were to decide (say) that any illustration in a story with a general credit is covered, barring contrary credit or signature/initials (or outside evidence), I'd not complain. Ditto if we decided that to call it uncredited but leave a note if in any real doubt. Probably ditto for most anything you're likely to suggest, I think. Thanks. -- Dave (davecat) 09:45, 15 Jan 2008 (CST)

Prologues and Epilogues

There's a discussion going on here about how to handle prologues and epilogues. If you have any thoughts or comments that you'd like to add, please check it out. Mhhutchins 17:14, 19 Jan 2008 (CST)

Manga and Takahashi

I've accepted some submissions from Dcarson that delete pubs, which we agree are manga, thus not in the Rules of Acquisition. But someone has gone to a hell of lot to create this author's summary page. Removing all of this info seems a downright shame, seems an awful waste (to quote Sondheim). But I fear we have no choice. Is the person who created this lovely page still around, and can he defend keeping it? Mhhutchins 18:30, 20 Jan 2008 (CST)

We have a number of well researched and lovingly compiled bibliographies that are only tangentially relevant, e.g. William W. Johnstone's page, which consists mostly of westerns. I wouldn't mind having them listed here if we had support for Non-genre Series, but as it is, it is impossible to distinguish bona fide SF titles from other genres :( Similarly, if we had support for "Graphic Novels", we could list them as such and it wouldn't be as misleading as it is now, but significant software changes don't seem to be in the cards at the moment.

Anyway, given the software limitations that we have, perhaps the safe thing to do would be to list all of these "beautiful but alien" bibliographies on a separate Wiki page and revisit them in, say, another 6 months. By then either:

• the software improves, or
• we find a loving home for these biblios, or
• we finally give up and zap them en masse. Ahasuerus 19:17, 20 Jan 2008 (CST)

Another example: Jonathan Gash's Lovejoy books. I know I verified some, can provide cover images, sorted the titles, created some variants, added some pubs: I can't justify keeping them here though. It's useful to me as I still have to find one or two titles, but I have no idea where a "loving home" will eventually be. Again, a candidate for us reaching out to other sites and asking if it's of interest. I'm not going to delete a single one of the pubs, let alone the titles, as they're based around the town I grew up in (for major variations in the value of "grew up"). If we send them away, I want to know where they went. BLongley 20:21, 20 Jan 2008 (CST)

Wow! That is the most impressive Manga-organisation I've seen here, I don't know why I've not come across it before! (I used to practice my editing/moderating/serialisation skills on unverified Japanese authors, it was always a good way to clear up some translators and artists from the Author directory at least.) I agree, it's a waste of effort if we delete it: but if the creator wants the data still we can probably provide a backup to extract it from. BLongley 19:21, 20 Jan 2008 (CST)

I'm not in principle AGAINST Manga being included here: I think I've only actively deleted Non-SF Manga. (E.g. I've obliterated some Basketball Manga.) But it often falls into the "The Moderators are unlikely to be qualified to deal with such new submissions" category that I feel are unwanted here: same as I feel about most foreign titles really, but whereas I'll have a go at including a French or German title, and make an effort over accents and suchlike, I cannot cope with Japanese, Chinese, Russian, or even Polish titles. If there's an English version, I don't mind if it's pictorial so long as it's SF - we could delete, for instance, a lot of Alan Moore and Neil Gaiman works if we enforce the "Graphic Novels are out" rule, only to have to put them back when it's turned into a film and then novelised. I've repeatedly said that we ought to concentrate on what we all do well (basically, works published in English, of all varieties) and forge links to other sites that can do our non-specialities better - so if isfdb.org.jp or isfdb.org.pl or isfsb.org.de want some of our data, let them have it and we can go back to doing what we do best. (Arguing about whether certain titles are missing a "u" in the word "colour" or not, maybe - but at least we don't have to install a new font to argue over those!) BLongley 19:21, 20 Jan 2008 (CST)

If a non-genre author's bibliography is small and looks like an accidental addition to ISFDB then I delete the titles. For example, I recently deleted Kazuo Koike as only a couple of his titles were in ISFDB. If it looks like a human being put a lot of effort into organizing an author's page then I'll let it be. In the past we have talked about making backups and letting people know they are available but nothing has come of it. It does mean ISFDB could creep into comic books, romance, and macro economics in the long run but I suspect it'll also break down into groups of editors that specialize in a genre or author meaning the only impact is server resources. Marc Kupper (talk) 22:31, 22 Jan 2008 (CST)

I'm not against manga being in the ISFDB. But that's what the current policy is and when I saw it going through the author list I figured I'd clean it up. Not a big manga or anime fan but I have enjoyed what I've seen of her work. Dana Carson 20:02, 25 Jan 2008 (CST)

Full artist signature over last name editorial attribution

One of my recent posts included a number of topics. One of them concerned situations where an artist is editorially credited only by last name but signs a fuller version of his name in a legible manner. Virgil Finlay for instance almost always signs his full name but is often credited as Finlay - same with Freas/Kelly Freas. It makes sense to me to use the more fully qualified signature. Perhaps we can come to some kind of agreement on this one issue.--swfritter 22:08, 22 Jan 2008 (CST)

Entering the full name is fine with me as a citation can be added for the reliable "secondary" source. If I was entering the publication I would use Virgil Finlay and add the publication note "The copyright page credits 'Finlay' as the cover artist but the artist' full name, 'Virgil Finlay', is visible on the front cover and so that was used for this ISFDB record."

In other words, I generally only credit exactly what's stated as the artist credit. If I enter something else for the credit then I'll explain the source.

One exception is if an ISFDB record already credits someone and I'm unable to verify that (either in the pub or via signature) then I'll leave the credit in place but add a note "The cover artist is not credited nor is a signature visible. However, prior to verification this ISFDB record credited /artist-name/. The source of this credit is unknown but is assumed to be accurate." This goes on the assumption that the data already in ISFDB tends to be accurate. If I have a reason to feel the artist credit is invalid then I will remove it but add a note explaining that prior to verification that the ISFDB record used to credit /artist-name/ but it was removed because of /insert-reason/. Marc Kupper (talk) 02:54, 23 Jan 2008 (CST)

What about cases where the credit is for a last name only (say, "Summers") but initials (say, "LRS") are signed? -- Dave (davecat) 10:05, 23 Jan 2008 (CST)

I would go with Summers. In either case you would have to do the variant author thing. As far as the data being accurate. Much of the original data was entered from secondary sources and those secondary sources had a tendency to list legal names rather than the actual credited names. That is probably why so many covers are credited to Ed Emshwiller despite the fact that he was rarely credited by that name. Same with Kelly Freas/Frank Kelly Freas. The interior art, although entered somewhat inconsistently while we are developing standards, is much more accurate. I have been going through some of the coverart records to correct artist attribution errors and also to check that the magazine titles are correct. I might note when you update the cover artist's name in a publication the artist's name for the coverart record is updated but if you update the title of a publication the title of the corresponding coverart record is not updated. Not as bad as Editor records. If you update the name of an editor in a publication nothing else anywhere else is updated. More projects!--swfritter 14:53, 23 Jan 2008 (CST)

I handle it like this

• I enter the Cover Artist as "Summers"
• Add a publication note that explains the credit is "Summers" on the copyright page, back cover, etc (I always say where I found it), that it's signed "LRS" on the front cover, and that this is probably Leo Summers (the canonical name in ISFDB). The only time I don't enter a publication note is if the artist is credited on either the copyright page or back cover with one of their well known names but maybe I should start getting into the habit of always entering a note so that someone later looking at the record will know exactly what was stated and won't wonder if I perhaps translated Kelly Freas into Frank Kelly Freas (I'd never do that but as Swfritter just noted, many people do translate an abbreviation or short name into a fuller or more well known version of the artist name without also noting citations).
• Once approved I then make the cover-art title record for Summers a variant title of Leo Summers (the canonical name). This part is optional and there are many artist records that have never been linked up to the canonical name for that artist. I also do this if I'm fairly certain the pseudonym is in fact the same person as the canonical name.

Considering Summers I see

• Leo_Summers - Canonical name
• Summers - Alternate Name for Leo Summers
• L._R._Summers - Was not marked as an Alternate Name for Leo Summers. I fixed this.
• Leo_R._Summers - Alternate Name for Leo Summers
• Leo_Ramon_Summers - Alternate Name for Leo Summers

Marc Kupper (talk) 15:58, 23 Jan 2008 (CST)

<sigh> I'll have to start adding a lot more pub notes, I guess; I've just been using the credited name. (For Analog/Astounding it's (in my experience always) either an "illustrated by" credit at the beginning of the story, or a credit (in italic type) with the first illustration in the story. Since swfritter basically told me to not assume that canonical names are established for illustrators (except (he grumbles, I can almost hear his teeth gnashing) for Kelly Freas), I've been doing aliases only for Freas (to Frank Kelly Freas).
Somewhere along the line I started quietly changing Analog cover art credits (mostly ones that say "Frank Kelly Freas") to what's actually in the magazine; but I haven't been thorough or systematic about it. At least one more pass through on that, I guess.
-- Dave (davecat) 16:38, 23 Jan 2008 (CST)

Good. I think we are pretty much on the same wavelength. Even if the canonical name is not a perfect choice people are still going to be able to find the data. Artist data is important but it is probably more critical that we get the author data correct. Frank Kelly Freas is fine as canonical with me - at least it was a name he was actually credited with. I think I have been remiss in double-checking the cover artist names - I am going the route of checking the actual coverart records so I can concentrate on one thing.--swfritter 16:44, 23 Jan 2008 (CST)

There are probably more interior art permutations that we could realistically document in the Help pages. For example, suppose Finlay is explicitly credited as "Finlay" at the beginning of a story, but a subsequent illustration, although not formally credited, is signed "Virgil Finlay". Should we use "Finlay" or "Virgil Finlay" in this case? My rule of thumb is that whenever I run into something unusual, I just use the name that seems to make the most sense, set up a VT and then document the details in the Notes field. Ahasuerus 23:26, 23 Jan 2008 (CST)

As far as Cover Art and Editor records go, it's a known minefield. I have also noticed that some folks use the "Make This Title a Variant Title or Pseudonymous Work" option for existing Variant Titles apparently without realizing that it will create a new parent title, but will not delete the old parent title, which will then become an orphan. Another script that I need to write, I suppose... Which reminds me that I really need to resume writing scripts now that I seem to be running on all cylinders again, but the Wiki has been so active lately that I can barely keep up :( Ahasuerus 23:26, 23 Jan 2008 (CST)

Unpaginated pages in page count

When a book has 6 unpaginated pages of "Author Notes" and "Acknowledgements" prior to page 1, ad then has 317 paginated pages, should the page count be entered as "317" or "6+317". I know what would be done with lc roman page numbers, but what if there are none at all? -DES ^Talk 15:38, 25 Jan 2008 (CST)

You just reminded me that I have not been paying attention to this. Even with the roman and then decimal publications I have not been entering x+100. Fortunately, I have list of which books I've verified...

Your question is a good one and your suggestion of 6+317 seems reasonable. Some downsides are figuring out what exactly is a "page" and now we'd be spending time counting them. With hardcover books there's a sheet of paper glued to the cover that forms a page. I forget what the technical term is for this sheet but don't think we want to count it. What if a publication has an accordion folded thing? Is that one "page" or do you count each fold? Sometimes a paperback will have a glossy page just inside the cover with a painting. Does that count?

Thus, I'd say the unnumbered pages are too much of a minefield and would get us even further off ISFDB's mission of indexing specfict. Marc Kupper (talk) 03:17, 26 Jan 2008 (CST)

One of the major benefit of entering the page count is to identify editions. Entering the roman page numbers may help in that matter. I don't believe I have seen any other bibliographic data elsewhere that attempts to document the unpaginated pages.--swfritter 13:12, 27 Jan 2008 (CST)

When there is content on unpaginated pages which I will add to the pub's record, I use the same method as OCLC. Place the number of unpaginated pages in brackets, followed by the paginated pages, e.g. [7]+245 (or 245+[7] if the material follows the paginated pages.) If the content is not going to be individually credited (such as "Acknowledgements", which I never enter) I don't include the extra pages in the page count. IMHO. Also, many artbooks (for example) are unpaginated, but OCLC gives a page count which is usually bracketed. Mhhutchins 13:26, 27 Jan 2008 (CST)

The OCLC system seems reasonable. The common pagination patterns seem to be

1. Pages seem to be counted starting at 1 at the very beginning of the publication but are unnumbered until the start of story meaning the story often starts on page 7 or 9.
2. Pages seem to be counted starting at i at the very beginning of the publication but are roman-numbered, or unnumbered, until the start of story meaning the story often starts on page 7 or 9.
3. The first few pages are unnumbered and the story starts on page 1. This is what triggered this thread and you are suggesting [#]### which is the OCLC format.
4. The first few pages are roman-numbered, or unnumbered, and the story starts on page 1. We have been using x+### for this.

We we need special notations for #1 and 2 above though or is noting ### ok? Marc Kupper (talk) 13:47, 27 Jan 2008 (CST)

When I rechecked my Bored of the Rings I found it was an example of Roman-numbered continuing through the Arabic-numbered, thus story started on page 23. When there's THAT much pre-main-content material it might be useful to note the start page of the main work even if none of the other contents are worth recording: a distortion of 7-9 pages might be OK, 22 pages might make people start thinking they've got an expanded or abridged edition. BLongley 14:00, 27 Jan 2008 (CST)

And I get teased because of the level of detail used when entering magazines. As long as the second part of the page count is the last numbered page that's fine by me - otherwise we've got a lot of work to do.--swfritter 14:21, 27 Jan 2008 (CST)

If I had my way you would - those magazines that continue a story on a completely different set of pages you don't record are misleading to anyone trying to establish a page count for the story. You don't get that problem in most books! :-/ BLongley 14:29, 27 Jan 2008 (CST)

Guess we are going to have to start including ads and font sizes, too.--swfritter 14:55, 27 Jan 2008 (CST)

Don't you dare! Our guesstimates are close enough for me, mostly, but in cases like THIS, where a story is considered to be under 40 pages by one of the most experienced people here, and turns out to be pages 8-45 PLUS pages 102-167, I have to wonder whether we're recording enough about magazine entries of the Speculative Fiction we're supposedly all interested in. I'm used to messy continuations of some columns in British magazines, but a typical Interzone or such just has a couple of pages of "continued from page [x]" for a lot of paragraphs that didn't quite fit. Losing length details of two-thirds of the story seems a bigger issue to me than me not recording an uncredited artwork somewhere. I know I can't win any sort of argument: I've recorded a piece of art that's just fancy text really, in a BOOK, and ADMITTED to it here: I blame Marc Kupper for that though. ;-) BLongley 15:47, 27 Jan 2008 (CST)

Unfortunately, chopping up a story that way was a common practice in the pulp era, which, as Bill points out, makes it difficult to tell whether it was an ss, nt, nv or even a "(complete novel)". However, keep in mind that there are other gotchas when trying to derive this information from page numbers alone. For example, anything printed closer to the end of the magazine is automatically suspect because it likely shared pages with ads, so what may look as a "nt" at fist glance is likely to be an "ss". Also, some editors had a habit of putting half page essays in the middle of novelettes, so even though they appeared contiguously, you couldn't tell whether it was an "ss" or an "nt" from the table of contents. Ahasuerus 16:59, 27 Jan 2008 (CST)

I suppose the bottom line is that deriving story length information from page numbers sight unseen is inherently risky. Perhaps adding a sentence to the Help pages to encourage our editors to document potentially confusing split stories in the Notes field would be the easiest-to-implement compromise? Ahasuerus 16:59, 27 Jan 2008 (CST)

If the total page count becomes important, then we'll probably need a separate field for data entry. Amazing Stories split not only novels, but novelettes, and sometimes a story could have as many as three, full-page illustrations. Not trivial, when a bedsheet could contain >900 words/page, and pulps could run >600 words/page. It probably would hurt to have a field for typical word count for each publication.--Rkihara 17:40, 27 Jan 2008 (CST)

Or we could trust the judgment of the editors, those who enter the original data and those who double check the data and those who find discrepancies in other ways. Wish all the mags could be like Fanastic Universe. No artwork, no continued stories, the same font size used throughout the run of the magazine. If it's 16 pages it's a short story, if it's 18 pages it's a novelette, if it's 17 pages you toss a coin.--swfritter 17:59, 27 Jan 2008 (CST)

Anyway, back to annoying Book Editors ;-) :Another example: here I included the unnumbered pages as there is a Novel "Starfleet: Year One" spread across several publications, chapter by chapter, that I want to find. (When I've found them all, then "Serials in Novels" is a discussion to have.) But it's always on un-numbered pages. Here I just noted that there were Short Story Competition Rules (presumably removed from later publications once the competition had closed) and an excerpt from the next book in the series. I'm a bit inconsistent on "excerpts": when it's just a few pages from the obvious next title, I don't record it. When it's from a surprising work, I probably do - e.g. an extract from a non-genre work by the same authors. When it's a complete short-story as an example of a forthcoming collection, I do. One title I entered this weekend has an "excerpt" I didn't enter, plus a chapter of the Serial that I did. I just can't recall WHICH though. (I picked up a LARGE box of free Star Trek books yesterday and have been processing in a fairly random manner). I can't recall an example of TWO sets of record-worthy details on un-numbered pages, but I'm sure we have some somewhere. BLongley 14:29, 27 Jan 2008 (CST)

I believe something that would help is if we knew why ISFDB asks for “Pages” and why it has a section about the roman numeral pages. It seems it’s not to get the number of physical pages otherwise we’d be counting the advertising, etc. after the last numbered page. FWIW - Publishers often upload the number of physical pages to Amazon. Another alternative is that ISFDB wants to know how many pages long the story itself is. I usually enter the starting page number for the story though for novels this gets concealed in the publication display as novels don’t show the title record for novels. The bit about the roman number pages is a curve ball. Template:PublicationFields:Pages says “…where viii is the highest numbered page with a Roman numeral page number. Pages without numbers that fall between the two types of page numbering can be ignored.” Meaning that if there are unnumbered roman pages between the last highest roman numbered that we would not note in ISFDB, and the start of the story, which may be a page numbered 1 or a page numbered 23… Thus I’m confused as to what we are supposed to enter into ISFDB and why. Knowing “why” seems more important as then we will be able to deal with the weird numbering cases more consistently. Marc Kupper (talk) 16:15, 27 Jan 2008 (CST)

WHAT we enter seems mostly up to us: I typically enter less than you about a book, but if you found it important and I'm cloning it I'll continue it. Same with magazine entries - if it's there, I won't delete it. All I'm really interested in is where we can find the SF. Which does seem to mean recording reviews (so we can find the SF being reviewed if we don't have it already), small bits of fiction added to some publications but not necessarily others, etc. I'm actually quite keen to support some artists too, but can't say they're essential to the ISFDB as I first saw it. (If we deleted ALL artists, how much damage would we do? Some ambiguity between editions for the book editors, wailings of anguish from the magazine editors, I suspect.) BLongley 17:08, 27 Jan 2008 (CST)

I agree - WHAT we enter seems mostly up to us. But, If we all are operating under different rules/assumptions when we do enter things then it'll turn into a mess. Thus I'm trying to understand how others are interpreting the Pages field. Marc Kupper (talk) 23:34, 27 Jan 2008 (CST)

Shortfiction vs Essay?

This item ("Professional Dilemma") isn't the only of these I've encountered, but it may be the first one I've dealt with on ISFB. It's listed in the table of contents as a "Special Feature". In this case, the pub was verified (by Alibrarian, who I'm told is MIA), & he entered (or at least verified) it as an Essay. In form it's fiction, centered around an office of patent attorneys. There are others, at least this one ("The Professional Touch"), also entered as an essay; I think I vaguely remember some more. Arguably these are essays on the complexities of patent law, in fictionalized form; but I'd just call them shortfiction if it were up to me. (And I'm not good at classifying lengths, so (again if it were up to me) I'd just leave the length as sf until someone more knowledgeable weighed in.)

At any rate, my question is what to do about these. I'm guessing that all of these are also in the same series: Improbable Profession, The Professional Look, The Magnificent Profession, The Curious Profession, The Lagging Profession, & The Professional Approach, all entered as Essay. (The last is the only other one in an issue I actually own.) Essay or fiction? Thanks for your thoughts.

(I realize that I'm guessing - on the basis of title & their being Essays - that they're all the same series. I'd not be inclined to change any particular one without some verification of that from someone; I can only be sure of two or maybe three.) -- Dave (davecat) 15:34, 3 Feb 2008 (CST)

I'd LOVE a "Fictional Essay" category! E.g. here you can't distinguish between the real essays about the creation of the publication from the essays about the fictional world described. Of course, I'd also like "Fictional Recipe" and "Real Recipe" so I can deal with books like this and this. I don't want to be held responsible for someone trying a "Bloody Stupid Johnson" recipe on my say-so though. ;-/ BLongley 16:12, 3 Feb 2008 (CST)

Um. I think what you're talking about is different. (I'm not familiar with the pub you cite.) These aren't essays about something fictional; they're fiction apparently included for didactic purpose. Which Campbell was quite capable of doing anyway, of course; AFAICS the only reason these aren't just called short stories (or novelettes if long enough) is that they're listed in the table of contents as "Special Feature" not "Short Story" or whatever. I could be wrong, of course. Dave (davecat) 16:26, 3 Feb 2008 (CST)

Try finding the actual stories in this issue of Amazing Stories.--swfritter 16:52, 3 Feb 2008 (CST)

I tried, but kept getting drawn back to the cover-art. :-( Can we find some female editors that won't get as distracted as I do? BLongley 17:19, 3 Feb 2008 (CST)

• Consider Giant Meteor Impact. This opens with a short piece of fiction, hardly more than a sketch. For the rest of its length, it is a standard essay about meteor impacts and craters.
• Consider The Pocket Song. This is a fictionalized "Introduction" to the verse/song lyric of the same title, which follows immediately in _Takeoff Too!_. The "intro" purports to explain how the song arose within the "Darkover" world, and how it was translated into English. But it has an in-universe PoV -- i.e. it is written as if the Darkover universe is real.
• Consider Uncleftish Beholding, a straightforward science fact essay -- from an alternate universe where English never absorbed any significant amount of vocabulary derived from Latin or the romance languages. (The title "translates" roughly as "Atomic Science")
• Consider the "Interludes" in the original book version of Hammer's Slammers, such as "The Church of the Lord's Universe". They are basically in-universe PoV info-dumps, containing at least some fictional content.
• For the matter of that, consider the appendices to The Lord of the Rings.

We may well want to consider indicating such "essays" as a separate type, but it would require judgment calls in each case, made by someone who had actually read the work in question -- it is not something that can be reliably determined from a ToC, in my view. -DES ^Talk 16:54, 3 Feb 2008 (CST)

We have discussed adding more detailed Title Types like "fictional essay" on a number of occasions. The problem with that appears to be twofold. First, it's hard to tell where we will (or should) stop since there are always further sub-categories that could be (arguably) usefully added. Second, it's much easier to add additional distinct types to a reference like the Locus Index, which is displayed as a hyper-linked Web page. If we were to add additional types to the application that displays the ISFDB data, it would require a significant amount of code juggling and even design changes whenever a new type was added. Do we display "fictional essays" together with "essays" or with "short fiction"? Or do we create a new section of the Summary page for them?

These are not insurmountable issues, but given how precious Al's time is, I am afraid that starting down that path would very quickly bog him down. Ahasuerus 22:15, 3 Feb 2008 (CST)

Fair enough. Perhaps we should agree on a more or less standardized notes entry, or perhaps a tag. That could be captured by a script later, when and if such logic is eventually added, and be useful in the meantime. At the least, we should encourge some indication in the notes field, for future reference. -DES ^Talk 22:57, 3 Feb 2008 (CST)

We could perhaps, also come up with some agreed upon standards as to when to use the "essay" type and when to use the "shortfiction" type in such borderline cases, as guidelines for future entry. -DES ^Talk 22:59, 3 Feb 2008 (CST)

There are lots of 'fictional essays' (reviews and introductions) in Lem's bibliography and I enter them as SHORTFICTION (well, they are fictional and they are short). But I created a series for them and they were all published in omnibus-like collections (so it is clear which texts belong to the series). There is also some grey area here, like an introduction to a not-yet-written book (it is more an outline of the book, as it nowhere assumes the book already exists) which I would be happy to enter as an ESSAY but mixing essays with fiction in a series doesn't work well. --Roglo 03:15, 4 Feb 2008 (CST)

(unindent) If I'm understanding, most of these (I think all except the Amazing Stories issue swfritter introduced) are about the opposite of my original question. That is, they're about fictional items in the form of essays. That said, I agree that some standardized way of dealing with those would be good. (And I'd agitate for a shortfiction of appropriate length, with a note describing it. Not quite sure what to do about any which are too long for shortfiction, & there must be some, though.) Hmm. And I see that I left this one: "The Present State of Igneous Research" as an essay. Hmm.
I think I'm just going to quietly decide that the ones I originally asked about are shortfiction, misclassified on the basis of being listed in TOC as "Special Feature". And I suspect Campbell did that because they're not, exactly, SF; though he didn't always let that stop him from including non-SF in with the short fiction categories, mind you. But I think these are just pure & simple short stories—plot, character, dialogue as you'd expect—included in Analog because Campbell liked them & they dealt with arcana of a type he found interesting & thought his readers would find interesting. But essays of whatever type they're not.
I'll make a series to include them. I wish I had more of the issues containing these items. Dave (davecat) 11:08, 4 Feb 2008 (CST)

I ran into this problem about a year ago when I was entering some of my 1930s pulps. A couple of popular science essays were thinly fictionalized and after some head-scratching we decided to enter them as short fiction. It can be a really fine line, though, since the amount of fictionalization can vary a great deal. Ahasuerus 15:14, 4 Feb 2008 (CST)

In these cases that I asked about, the story is probably a much greater factor than in those you cite. (The words "thinly fictionalized" wouldn't occur to me.) I've met cases where I'd have left the Essay type in without even thinking to question it, myself, though. Dave (davecat) 08:32, 5 Feb 2008 (CST)

I skimmed through the issues of Astounding/Analog containing the Lockhard stories that you listed above. I believe they should be classified as fiction. The story in the Sept. '62 issue was in fact listed as a short story. I believe if we generate a "fictional essay" category we'll be headed down the slippery slope, as many well known novels would easily fall into that category. Moby Dick and the fifty or so pages on Cetology in the center of the story comes to mind. Mack Reynolds and those interminable political lectures he inserted into almost every novel that he wrote, not to mention Ayn Rand, and Heinlein. Then there are the world builder novels, written by Hal Clement, Poul Anderson and others, containing long lectures on planetary science.--Rkihara 15:29, 4 Feb 2008 (CST)

I agree 100%, as far as this kind of thing goes. (Whether we should have a type for things that are in format essays but are in fact entirely fictitious (which other people have kept bringing up here) seems to me a separate question; & I think that idea has a lot more merit.)

Rkihara, you said you skimmed through those issues. All of them without exception? Can you verify that all of them were this same series, as I understand you to say? I only have issues with (I think) three of them, so I was speculating that at least those with "Profession" or "Professional" in their titles (which I listed) were connected. Thanks (either way)! -- Dave (davecat) 08:32, 5 Feb 2008 (CST)

You have a good point, creating a separate "type" for such works would a) be a good deal of coding work when that is already a bottleneck, and b) lead to significant cans of worms that are perhaps best not opened. However, i would like soem agreement on dealing with the pure, plotless, in-universe PoV infodup, presented under a separate title. This is a work that explains in some way what is going on within a fictional setting, written as if the fictional setting were real. it is most often included in a collection or anthology,or as an afterword or appendix to a novel or long work of shortfiction. I propose as a guideline: in future, is that such works be classed as "shortfiction" (in part because that will cause them to be listed as part of any fictional series to which they belong), and that editors are encouraged to use a tag of "Fictional essay" for such works. What do you think? -DES ^Talk 15:39, 4 Feb 2008 (CST)

Davecat, I "skimmed" all of the Astounding/Analog issues you didn't have, I didn't check the one issue of Fantastic Universe, as I only have a few copies and they're buried away. I only read enough to determine that I would have classified them as fiction, but didn't pay any attention to plot, setting, or characters. They're pretty short, so I suppose I could sit down for an hour and read them all. If you have more that you would like checked, I'll read them more carefully and get back to you.

DES, I think that the dividing line between "hard" science fiction and a "fictional essay" is pretty vague. How would you differentiate between the two? Consider "Ralph 124C41+" or other utopian novels, "The Cold Equations," or other stories revolving around a scientific or mathematical solution.--Rkihara 12:29, 5 Feb 2008 (CST)

Yes, the dividing line in such matters can be tricky. But there is a specific and IMO recognizable sub-type that i think is worth noting. I am referring to works that are:

1. Shorter than book length;
2. Contain no or effectively no plot themselves;
3. part of a larger series or overall work;
4. Written from an in-universe point-of-view -- that is written as if the events in the work of fiction were true;
5. Serve primarily to explain the background of the series or larger work to which they are connected.

Had Asimov given us entire articles from the Encyclopedia Galactica instead of brief excerpts, they would fit the criteria I am listing. I mentioned above the "interludes" in Drake's Hammer's Slammers and they are precisely the kind of thing that I mean. So are the Appendices to The Lord of the Rings considered as separate works. Another possible example is the "history" that serves as an introduction to The Syndic. There is an essay on the development of the Bolos in one of Laumer's Bolo books that is an example. These are all presented as if they were essays or articles, but essays written within the fictional world of a larger fictional world or series. That is the specialized sub-type that I feel is worth noting is some way, probably by way of an agreed-on tag, rather then an entire new type such as "essay" and 'Shortfiction" now are. -DES ^Talk 13:52, 5 Feb 2008 (CST)

Now things are getting confusing. Davecat was talking about factual essays presented in the form of fiction. If I understand things correctly, you're talking about fictional essays about fictional or speculative subjects. I have to think about this. I'm inclined to think that marking such essays in a larger work is not a good idea, it would add unnecessary complexity. Would the "Prologue to The Stars My Destination" explaining the discovery of Jaunting and it's effects on society be such an essay? Maybe this discussion should be split under the two types to reduce confusion?--Rkihara 14:30, 5 Feb 2008 (CST)

Maybe it should be split, as the two types, while not unrelated are in a sense almost inverses. As to whether to record the sort of things I am speaking of, when the appear as separate entries in a ToC, particularly in a collection (as they do in the Drake work) or when they have separate titles and are long enough that they seem to merit separate entries, and particularly when they might be published separately, or when some editions of a work include them and others don't. In short, i am speaking of things we are already recording as separate works, but with no indication of their special nature. Frankly i don't consider the sort of things Davecat was discussion as anything but fiction that is rather dry, and has a didactic or educational purpose, and that could include a great deal of SF indeed. But Bill's first response in this thread seemed to point to the sort of think I am speaking of: works that are essays or articles in form, contain effectively no fictional plot, but are written as from within a fictional setting, and are separate enough from the fictional work or works to which they relate that a separate listing is justified. -DES ^Talk 15:07, 5 Feb 2008 (CST)

(unindent)For those unfamiliar with Bill Contento's classification, he has a http://contento.best.vwh.net/abbrev.htm variety of codes] to account for various shades of gray. In this case the most relevant ones appear to be:

• lk linking material
• fa facetious article
• vi vignette (under 4 pages, under 1,000 words) [sometimes used to tag introductions that assume that the described fictional universe is real]

Ahasuerus 23:27, 5 Feb 2008 (CST)

Tracking relationship of magazine cover art to interior story

The first part of this (the 2007 entries) was posted in the Community Portal and then (fairly soon) archived here:

Reposting a question from a new editor for general discussion:

Hi! I'm new. Question for you - are we tracking whether or not publications were illustrated in the original, and especially which stories inside a magazine have associated cover and interior illustrations, by what artists?

I'm particularly interested in the question of tracking printings, reprintings, and online availability of short fiction and am in conversation with some folks about automatically saving TOC - related info for collections that are being digitally scanned. Would love to coordinate efforts.

Netmouse 13:10, 10 Dec 2007 (CST)

(My answer is on Netmouse's Talk page). Ahasuerus 16:10, 10 Dec 2007 (CST)

We do not have a specific field designed to document the story illustrated by the cover of a magazine. As a result the only place to document this information is the magazine notes field which means that the inclusion of such data is not required of the ISFDb editor and not accessible for listings. I always try to include such data.--swfritter 20:20, 10 Dec 2007 (CST)

As I've been entering all these Analogs I've wished many times for this; all (or very, very nearly all) of those covers are intended to illustrate a specific story or article, & it would be good to put this in. I have not, however, been putting this in the notes. (For Analog, at least in the years I've got, in almost all cases, the first story or article after the editorial is the cover feature. And it wouldn't take very much to determine this just from the pub listings, which include the online cover images. Not much need to go back to the original magazines.)
It would really be nice to have a cover-title field in the pub record; I think that would be better than a separate content item, but I could be wrong. Dave (davecat) 11:07, 11 Dec 2007 (CST)

Well, the Coverart Title record is editable after entry, you don't have to leave it as "Cover: Pub Title". In which case making this visible on certain other screens could be the feature request, rather than adding a new field. I've noticed this isn't tightly coupled, as if you correct a title the Coverart record doesn't change. There's a 'title_ttype' of 'BACKCOVERART' allowed for too, which doesn't appear to be used. I wonder what Al intended for that? BLongley 13:21, 11 Dec 2007 (CST)

Aha. I think I'd once observed that there is a coverart record for magazines, but forgotten it. But the format does seem quite standardized. What would be a reasonable replacement? Maybe:

 Cover: Satan's World (Part 1 of 4) (Analog, May 1968)

or something like that? Dave (davecat) 05:46, 12 Dec 2007 (CST)

(unindent & end of first reposting) When this issue later came up on my talk page, Mhhutchins said this:

I honestly feel we shouldn't be changing the title record that's been assigned by the system. Updating the dates, of course, but that's not what we're talking about here. We're talking about changing the title of the title record. If you feel the system's way of assigning a title record is wrong, that's a matter for discussion. And Al von Ruff would have the final say in what changes would be made. I can't imagine the time and effort it would take to go back and change all the cover art titles to match the piece of fiction that suggested them. But that's up to the people who will be making the changes. Perhaps taking this discussion over to the Rules and Standards page would bring more points of view into the mix. Mhhutchins 11:33, 6 Feb 2008 (CST)

To which I replied:

Whereas my own feeling is that, when a magazine's cover art is titled (on the cover!) & intended to illustrate a particular story, the cover art's title should indicate that. And no automatically-generated title will do so.
I do agree that the title should also indicate which issue of which magazine it is a cover for; but we have a canonical form for doing that. Dave (davecat) 13:40, 6 Feb 2008 (CST)

I don't believe that art on magazine covers is titled. Often it illustrates a story that is highlighted on the cover. But one can't assume that the art automatically illustrates the most prominently mentioned story. I've been updating the early 70s Worlds of If. Take a look at this cover. No that doesn't illustrate Simak's story. And this cover illustrates that issue's serial by Colin Kapp, not Pohl and Williamson's "Doomship". What happens when a work of art is used multiple times? We have to give it the name of the pub on which it appears. Mhhutchins 18:48, 6 Feb 2008 (CST)

I have seen on a number of magazine issues, a TOC cover art credit that specifically attributes the art to a particular story, as well as indicating the cover artist. In other cases it is fairly clear what story the cover art illustrates. In yet other cases it probably does not illustrate any story in particular. -DES ^Talk 19:00, 6 Feb 2008 (CST)

Michael, I never said, & would never have said, that all magazine cover art is titled on the cover is titled, or that it all illustrates specific stories. There are plenty of times when it isn't & doesn't. My comment said "when a magazine's cover art is titled (on the cover!) & intended to illustrate a particular story . . . ".
As it happens, that's almost always true of Analog during the Campbell years, and mostly true of the issues of Astounding I've personally got (which is not very many of those). Sometimes the illustrated story was not the lead story, but the title on the cover went with the story illustrated. (And, yes, in these issues one story or article was listed on the cover. All this changed after Campbell died.) In at least one issue of Astounding January 1959, AFAICS the cover relates not at all to any contents of the magazine, but the title on the cover goes with the cover art not with any contents, so I'd say that in this case it's absolutely an artwork title. -- Dave (davecat) 20:00, 6 Feb 2008 (CST)

For the moment that's the state of the discussion as I know it. (Or rather, it was before more discussion got entered above. -- Dave (davecat) 20:00, 6 Feb 2008 (CST))
(Um. Swfritter & Mhhutchins did also bring in the fact that the Coverart record is invisible from the pub listing. I'm not trying to misrepresent the discussion, but I think that's a side issue except insofar as it means that editors are likely not to think to update a coverart record.) I think it would be good to get a wider range of opinions on this. -- Dave (davecat) 13:57, 6 Feb 2008 (CST)

The only changes I would make to the Coverart records would be to make the titles match the pub titles and to enter the correct date. My preference would be to remove the capacity to edit Coverart records and programmatically insure that they are updated whenever the pub record is updated.--swfritter 18:02, 6 Feb 2008 (CST)

I'm in agreement with swfritter here. If we don't have access to the coverart record in the edit mode, then the least the system can do is update the record's date when the pub's date is changed. And the same thing goes for the inaccessible editor records for mags. Mhhutchins 18:39, 6 Feb 2008 (CST)

The problem with Editor records is that most of the magazine Editor records have been merged - something that makes no sense for Coverart records.--swfritter 18:43, 6 Feb 2008 (CST)

I think it does make sense to merge Coverart records at times: e.g. here some verified covers with identical art (but different blurb) have been merged. It keeps the clutter down, which with frequently reprinted book titles becomes quite useful. Obviously not with Magazines, except maybe ones like Destinies. (Which I see has given us Award problems... but that's another topic.) However, I think either the Coverart Title has to be made visible or it should be dropped, this halfway house is helping nobody. (No point automatically updating them if they're just going to remain as "Cover: <Pub_Title>" or even "Cover: <Title_Title>", generate them on the fly.) A workaround might be to enter Coverart titles as Interiorart entries with a page of 'fc' if needed.

Editor records are another pain entirely. BLongley 13:26, 7 Feb 2008 (CST)

After having added and merged and seriesed over 800 missing Editor records, I agree about that. The fc was also one of my thoughts but another possibility is making a variant title with the story title linked to the Coverart title. Having two records for one cover might get a bit confusing but the linkage between the Coverart title and a variant title might be less so. If we are going to merge artwork we need a policy that requires either physical access to the artwork to be me merged or very reliable images.--swfritter 17:55, 7 Feb 2008 (CST)

I'll leave Editor records to the experts: merging is a clearly good way of shortening the Editor's Author page, but that strikes me as a short-sighted solution causing problems now. Trying a search for a magazine edition from the front page is useless, and I'd like that fixed. Having to know the editor before you can find the magazine is just plain WRONG. BLongley 18:14, 7 Feb 2008 (CST)

With an Editor Series it seems to me to be relatively easy.--swfritter 19:45, 7 Feb 2008 (CST)

Better, but not ideal. A new user should just be able to put the name in the basic title search and find a Magazine record straight away, same as they can enter a Novel and find the Novel straight away (well, so long as the Coverart and Reviews don't swamp the results so much it's not on the first page). The "Editor" records confuse and I only found my first magazine by tracking it down from Coverart, then couldn't do that for another because it had a photograph for a cover! We can definitely improve things further. BLongley 13:18, 8 Feb 2008 (CST)

Coverart entries are a pain we can fix more easily I think - the exact cover IMAGE is recorded at pub level, but when we have some reliable images there we can merge the coverart entries for same artwork. But we need to stop trusting/using Amazon ZZZZZZZ entries first. BLongley 18:14, 7 Feb 2008 (CST)

(unindent) Another branch of this discussion is now on the Community Portal talk page. -- Dave (davecat) 10:59, 14 Feb 2008 (CST)

Discussion moved Community Portal has been moved here.

(Thank you!) -- Dave (davecat) 19:40, 14 Feb 2008 (CST)

Linking Cover illustrations to the stories they illustrate

This sample magazine has a possible solution. I have entered an interior illustration on page 'fc' and made it a variant of the Coverart record. This is the way it looks on the artist's page. The primary downside is that the the Coverart record shows the pub twice which might not be a big deal because it is not likely to be accessed that often. Technical issues? Display issues?--swfritter 19:04, 13 Feb 2008 (CST)

This is a possible solution to what? Marc Kupper (talk) 23:55, 13 Feb 2008 (CST)

See this discussion. We have already discouraged one potential editor because we don't have this capability.--swfritter 09:35, 14 Feb 2008 (CST)

I tried to move that discussion off my talk page to Rules and standards discussions. Unfortunately activity continued, somewhat separately, both places.
swfritter, a problem I see is that it still says "interiorart". But your idea would go some distance toward meeting my own reasons for wanting the coverart record to have a real title.
(But not all. I find it, um, really strange to see dozens and dozens of records that say only that Analog had a cover each month, when those covers are in fact (almost, if not absolutely, without exception) story illustrations; though I guess I tend to overlook the fact that they also record the artist. And granting that other magazines apparently normally have random covers not related to contents.) -- Dave (davecat) 10:52, 14 Feb 2008 (CST)

I tend to agree with Dave (davecat) here. When an illustration in fact illustrates a specific work of fiction, that fact should be tracked here and it would seem msot natural to record that as the "title" of the illustration. Obviously, that it was the cover art for a given magazine is also important and should be tracked. Moreover, when such art is displayed (for example in SF convention galleries) or printed seperately (for example in "collected art of..." books) it often has a title given by the artist separate from the work it illustrates (made up example: "Moonrise -- Cover for The Moon Wizard"). When such titles are known (as often they are not, particularly for older art) shouldn't they be recorded somehow, also? -DES ^Talk 11:25, 14 Feb 2008 (CST)

I might also note that back cover and end page art are defined as interiorart. Once again, short of software modifications this is the only method I could think of that would accomplish the task. If this is not an immediately acceptable solution I would suggest we drop the issue until an acceptable solution can be implemented.--swfritter 12:09, 14 Feb 2008 (CST)

It looks OK to me - I suggested "fc Interiorart" before, the variant as well doesn't matter too much to me. The question of whether you also want to use this to record "as signed" artist versus "as credited" artist or "canonical" artist remains, but that can be dealt with separately. BLongley 13:13, 14 Feb 2008 (CST)

I picked up the first part of the idea from your suggestion. I think there might be technical issues involved as the implementation of invisible Editor and Coverart records may be modified. If this methodology fits in with those changes it might be a solution and I would suggest we don't implement it without full knowledge of any technical factors.--swfritter 14:57, 14 Feb 2008 (CST)

For future technical factors that may interfere, always ASK Al. For future features we want, TELL Al! ;-). BLongley 16:08, 14 Feb 2008 (CST)

"In Memoriam"

After entering this rather depressing set of entries I found myself wondering once again if we should be recording Obituaries in a special way. They're often quite informative, and may be as useful as "bibliography" essays, but I think I'm leaning more toward a category that puts them up there on the author page along with "interviews". Please discuss. BLongley 18:14, 6 Feb 2008 (CST)

That makes sense to me, provided that the change is not too hard to implement. Come to think of it, there are occasionally explicitly biographical essays that are not obits. If we do create a special category, should these go into it as well? -DES ^Talk 18:25, 6 Feb 2008 (CST)

There is an outstanding feature request that I filed on 2005-07-25. It asks for "secondary bibliography" support so that we could link books and articles that contain bibliographic information about Authors -- as opposed to Reviews which cover Titles -- to their respective Author records. Perhaps we could have expand the request to cover both bio- and bibliographies, obituaries being a subspecies of biographies? Ahasuerus 20:42, 6 Feb 2008 (CST)

Anything about an author that isn't by the author. Dana Carson 21:06, 6 Feb 2008 (CST)

Current Help on foreign language books

The current Help:How to enter foreign editions page is badly out of date. I suggest we add/modify a few things to summarize our current unwritten policy as follows:

• English is currently a/the privileged language in the ISFDB in that foreign language translations of English language books do not get their own title records; they are entered as Publications under the English language title instead
• For works that originally appeared in a foreign language, the master title is the title of the work as written by the author(s) (as opposed to as published to account for Bulmer, Sheffield, et al). English language translations, when available, are entered as Variant Titles. Translations to other ("third party") languages do not get their own title record and are entered as Publications under the main title record instead.
• Eliminate the section about entering non-Latin (i.e. Japanese, Cyrillic, etc) characters in titles since at this time our software's support for them is very poor and searching is not supported. (This means that we will need to hunt down and transliterate all victims of our unsuccessful attempts to use Cyrillic in early 2007, not that there are that many of them).
• Eliminate the comment about creating pseudonyms for Anglophone authors whose work has appeared in other languages. We don't want to have Author records for "Исаак Азимов", "אסימוב אייזק", etc just because Asimov's books have been translated worldwide.

Is this a fair summary of what we are currently doing with foreign language titles? Ahasuerus 23:47, 12 Feb 2008 (CST)

Well, what I'm currently doing is avoiding the problem as much as possible... :-)

Still, when I'm just whacking about an unloved but highly-represented-here author's page, I often don't put foreign titles under the English title as I have no idea which came first and I'm never 100% sure that my translation skills have got the right titles matched anyway. In those cases I'll make my best guess a variant title so it still visibly needs work and I can see where to put other editions of it in the meantime. The suggestion for where it ends up is fine by me for the moment but I'm going to want the foreign titles hideable at some point (same as for multiple near-identical English printings - the foreign variants can just be another 'edition' when we sort that concept out).

I don't understand title of the work as written by the author(s) - can you expand on what you mean there?

Yes, please let's get rid of as many non-working characters as we can! I can just about cope with Philip José Farmer and Élisabeth Vonarburg but not much more... BLongley 13:12, 13 Feb 2008 (CST)

The "written by the author" caveat is included to cover situations when a foreign language translation of an English language book appeared first. For example, Charles Sheffield's Convergence, volume 4 in his Heritage Universe, was delayed while he was transitioning from Del Rey to Baen in the mid-1990s. In the meantime, a publisher in Belarus (?) translated the book and published it in Russian in 1995, two years before it appeared in English. This was a recurring question on rec.arts.sf.written in the 1990s and even made the FAQ. If we were to use the language of the first publication to determine the canonical title, then we would have to enter the Russian language title of Convergence and make "Convergence" a variant title, which seems counter-productive.

In addition, there is a number of books and series by English language writers which were dropped by their US/UK publishers in mid-stream but were continued/completed abroad. The two best known examples are Ken Bulmer's Dray Prescot series, dropped by DAW (when Wollheim got sick and his heirs took over) and continued in Germany, and "Ansen Dibell"'s The King of Katmorie books, with 2 of the 5 volumes only available in French and Dutch. We currently list the English language titles of Dibell's books even though they never appeared in English. We don't list any Prescot books post-Warlord of Antares at all, but that's just due to an (unfortunate) gap in our coverage.

The only current exception that I know of is Robert F. Young, whose novella "The Quest of the Holy Grille" was later expanded into a novel, La Quete de la Sainte Grille, which has appeared in French but not in English. We currently list the French title, in part to avoid confusion with the eponymous novella, but we may want to use the English language version instead to be consistent. Ahasuerus 14:29, 13 Feb 2008 (CST)

OK, I'm still confused as to the "English language translations, when available, are entered as Variant Titles" part: is this only for when a title was written in English, published under a foreign title, then printed in English under a different name? It seems a lot to ask people to know what the original unpublished English title was, and certainly no help in my efforts to just plain ORGANISE some neglected authors' titles. BLongley 15:55, 13 Feb 2008 (CST)

I took that to mean that English language titles of translations of works originally published in non-English languages (such as Lem's work) should be entered as variant titles of the original work. Is that correct? -DES ^Talk 16:06, 13 Feb 2008 (CST)

Looking at the examples mentioned e.g. this one, I guess that the intention is that a work written in English should get an English title even if it first appeared under a foreign name, but I don't see an example of it later being republished under an English title that needs to be a variant of the original one? BLongley 16:30, 13 Feb 2008 (CST)

Perhaps an explanation of Deathworld 4 would help? BLongley 16:32, 13 Feb 2008 (CST)

Look at Podróż dwudziesta druga of which The Twenty-Second Voyage is a varient. -DES ^Talk 17:15, 13 Feb 2008 (CST)

I rather suspect that Deathworld 4 is an example of how we don't want things done in future, but I may be mistaken. -DES ^Talk 17:16, 13 Feb 2008 (CST)

(unindent)Bill is quite right that "the intention is that a work written in English should get an English title even if it [is] first appeared under a foreign name". This rule would cover all cases of delayed publication from Sheffield to Bulmer to Dibell. I suspect that we may be running into this issue more often as we go forward: novelizations and tie-ins are usually tied to the release of the underlying intellectual property (usually a movie) in the target market, so even though a novelization may be written in English, it may appear in a foreign language earlier if the movie is released in that market first.

To address the other part of Bill's question, I haven't thought of the class of cases where a text would be "later ... republished under an English title that needs to be a variant of the original [English language] one". And it's a good question too since there is nothing preventing US/UK/Oz editors from changing the title to something completely different if and when they decide to publish the books in English. Besides, the English language titles that we would use under the proposed rules may be literal translations of the known foreign language titles and not what the author originally had in mind. I can't think of any SF examples off the top of my head although some Harry Stephen Keeler mysteries that appeared only in Spanish and Portuguese in the 1950s and were finally published in English in the 2000s may qualify. I guess in a case like that we would have to create VTs for the newly minted English titles.

Another complication to consider is that it's not always immediately clear what the original language of a story/novel/essay may have been. For example, Stanislaw Lem's German was very good, so we enter his German language essays as they appeared since we assume that he wrote them in German. However, it's conceivable that he first wrote them in Polish and then translated them into German, in which case, according to the proposed rules, the first Polish language publication should be used to create the primary Title record. I can't think of any way to safeguard against this eventuality aside from documenting our assumptions in the Notes field.

To address the final point, Deathworld 4 is probably one of the most complicated cases that we have. As we discussed early last year, there were a lot of "sequels-by-other-hands" done in the ex-USSR in the early-to-mid 1990s when many ex-Soviet countries provided no IP protection for foreign works originally published prior to 1973 and copyright violations usually went unpunished anyway. Any remotely popular universe -- from Conan to Edmond Hamilton's Starwolves to Sterling Lanier Hiero (!) -- could (and often did) get a hastily written sequel. We even have some of them in the database.

The Deathworld case, however, was apparently different. As far as I could determine by running Internet searches, the publisher (EKSMO-Press) was one of the bigger players in the Russian market and things were getting more stable in Russia in the late 1990s, so they decided to play by the rules. They hired a reasonably well known writer (who writes SF as "Ant Skalandis") and secured Harrison's permission to use his name and the Deathworld universe to write sequels as by "Harry Harrison and Ant Skalandis". The devil, unfortunately, is in the details and the details are murky. Some sources claim that Harrison approved an English language outline of (at least) the first volume while Skalandis did all of the writing. Other sources suggest that Harrison wasn't involved at all and the only condition that he had was that the sequels should not appear in English (although some have been translated elsewhere, e.g. in Poland). BTW, Skalandis wrote the first three sequels and then another writer (who writes SF as "Mikhail Akhmanov") wrote two more volumes. I suspect that Harrison's input went from "little" to "none" as the series progressed, but that's just a feeling that I get.

Given these uncertainties, should we enter Harrison as a co-author or should we change the attribution to "by Ant Skalandis [as by Ant Skalandis and Harry Harrison]"? Should we make the Russian language title the primary one or, assuming that the English language outline did exist and was reasonably extensive, should we use the English title (which appears on the copyright page of the Russian language edition) instead? I guess it's a corner case and you can't design a bibliographic application to account for all corner cases, so it's Notes time again.

BTW, it looks like Latin-2 and Cyrillic now work with Title searches, although "42" still returns all of our Cyrillic titles in addition to any legitimate matches. Non-Latin-1 Author searches still seem to be broken.

OK, enough masochism for now, I think I need some aspirin... Ahasuerus 23:22, 13 Feb 2008 (CST)

Given that the nature and level of Harrison's input is unknown I'd go by what's stated on the publications. If someone in the know (Harrison, Skalandis, or the publisher) gives an interview that provides hard data then we can document it in the notes. That said:

• Translate the titles to English but also append something like "(translated title)" to the title and add a title-note explaining the real title and that the English version is a literal translation for ISFDB.
• Credit the authors as stated if they have Latin alphabet names. If it's a Русский Муравей then you may want to translate that too. Save some of that aspirin for everyone though. Part of my thinking here is I do want the title to show up on Harry_Harrison even if it can be shown that he had zero input other than his name. Marc Kupper (talk) 00:16, 14 Feb 2008 (CST)

OK, here's a test title Doctor Zamenhof. POSSIBLY originally published in Esperanto, Esperanto title used for the English publication, Italian translation reprint. Marc, I suspect you want "(translated title)" added rather than just a Title note. Ahasuerus: have I guessed your intents correctly? Everybody else - HELP! I don't speak Italian so this pub needs treating with a LOT of caution. Blame me, or Blame Google translations - I daren't create variants for the remainder. BLongley 15:05, 14 Feb 2008 (CST)

To address Marc's point, I agree that ghostwriters are a separate can of worms. I suspect that for canonical title purposes it's generally safer to enter ghostwritten stories/books as collaborations between the ghostwriter and the Big Name Writer (Shatner/Goulart, del Rey/Fairman, etc) whether they are published as explicit collaborations or not, in part so that they would appear on both Authors' Summary pages. Except when the Big Name Author is deader than V. C. Andrews, that is. However, that's fodder for another discussion, one that I may start either later today or on Tuesday, after I am done with verifications and return to wandering. Unless someone else beats me to it, of course :)

Re: adding "(translated title)", one downside that immediately comes to mind is that it would break our review links. Perhaps the best bet would be to use the transliterated form of the Russian language title as our canonical title and explain the gory details in Notes the way we do it with Lukyanenko. After all, the only major difference between Skalandis and Lukyanenko is that the latter has been published in English and the former hasn't. And using Cyrillic, Hebrew, etc to enter Author names is out because it would kill our searches. Some non-English characters are OK, e.g. "é" in "Philip José Farmer", but most can't be found using our Search logic, which is why we use an Anglicized version of Stanisław Lem's name.

Re: La Lingua Fantastica, wouldn't most of it fall under Rule of Acquisition 6:

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?)

? Up until now we haven't done any serious work on foreign language writers unless their work has appeared in English or they have other ties to English language SF (e.g. see Skalandis above). Do we want to start now?

Finally, it sounds like we are in agreement re: bullet points 1, 2 (except the "as written" part, which we can safely omit for now without affecting 99% of foreign language Titles) and 4 of my original post. Is that right? If so, we may want to give it another day or two and then update the Help page with the bits that we agree on. Then we can continue arguing about our beloved corner cases :) Ahasuerus 23:14, 14 Feb 2008 (CST)

Re: adding "(translated title)" - Yes, it breaks review links. But we don't have a LOT of reviews of translated titles. And when we do have them, it's a toss-up as to whether the review is of the original or the English title... or both, separated with a "^". The review links are very fragile, but after working on them a lot recently to clean up some Authors that only exist due to reviews, I hope people are noticing when there IS a Title match with a Review before updating a Title record? (I hope so, especially now we CAN edit the Review more easily.) In the Long term I'd prefer Review Links to match more solidly and be updated automatically - but then again, I'd also like the option to link reviews to a Publication (or two), or the wonderfully-elusive concept of "Edition". BLongley 14:53, 15 Feb 2008 (CST)

Re: La Lingua Fantastica and Rule 6: Yes, many of the contents probably do qualify for exclusion. Only the Harry Harrison introduction and story seem to count as IN - and the Interiorart by Karel Thole might be of interest, but I don't know where in the pub it is. Would it be better to delete the entries we don't understand, even if they're SF? (We have several Playboys with just English SF entered for "first publication" reasons, so incomplete titles aren't totally unwelcome it seems - but I can't be sure an Italian editor won't turn up tomorrow and tell us we've missed a Foreign pseudonym of an author that IS here - e.g. "William Auld" doesn't sound Italian to me.) BLongley 14:53, 15 Feb 2008 (CST)

And yes, I'm fine with 1, 3 and 4, and 2 probably just needs a little clarification. BLongley 14:53, 15 Feb 2008 (CST)

(unindent)Help:How to enter foreign editions as per the discussion above. Corrections and additions are more than welcome. Help:How to parse data in library catalogs created, although Bill's comments about 17.5 cm UK books made me wonder if the UK part needs to be beefed up. Ahasuerus 00:05, 22 Feb 2008 (CST)

Probably, but I'm not sure how. I checked a few authors where I've a mixture of US and UK paperbacks and there is some truth in the suggestion that UK paperbacks are slightly larger than US ones. E.g. "Retief to the Rescue" (US Pocket edition) is only 17cm but "A Plague of Pythons" (Penguin UK) is 18cm. But "Retief of the CDT" (US Pocket edition) is 17.9cm and bigger than most other US editions. "Shield" (Magnum UK) is 18.1cm, "The Man Who Counts" (Ace, US) 17.4cm. "Robot Adept" (UK) 17.5cm, "Phaze Doubt" (US) 17cm. I don't really want to reorder all my books by year, publisher and height and do conclusive research, but my spot check suggests 18.5cm would be a nice suggested cut-off between pb and tp for British paperbacks. The variations are too wide to separate US and UK pbs though. (Well, the 16.2cm "Bow Down to Nul" (Ace, US) is small enough that I can't imagine it being confused with a British pub.) I think what I'm saying is that the 19cm guideline is likely to be wrong based on my collection - but I wouldn't change it based on my experiences alone, ask some other British pub owners with a wider range that DO happily mix tp and pb more than I do. I also can't offer much help on hardcover size - I follow very few authors at "must have first edition" status and although I can tell you that the Terry Pratchett Discworld books noticeably jumped from 22.2cm to 23.5cm, the juveniles stayed at 22.5cm, like all my Artemis Fowl hardcovers. But that's such a small sample I couldn't say that's a standard for Juvenile versus Adult hardcovers. BLongley 13:58, 22 Feb 2008 (CST)

Canonical Publisher names

The new search on publishers feature has revealed just how non-standard our entries for publisher are. For example, the db contains no less than 40 different strings containing "Ballantine", while this page lists four imprints, and ISFDB:Ballantine has not yet been created.

(There was some discussion of this previously in this thread, and some at ISFDB:Community Portal#Print series?, but it was a little mixed in with other matters, adn it seems to me this part belongs here.-DES ^Talk 16:33, 18 Feb 2008 (CST))

I suggest that we develop some general standards on how publisher and imprint names are to be represented, and then some specific lists of canonical names for each publisher and imprint.

1. I suggest that each publisher have a single canonical name, or if a publisher has change the form of it's name significantly, a single canonical form for each era. We shouldn't have "TOR", "Tor", "Tor Books" and the like, one of these should be standard, nd the others perhaps recorded as variants, just as we do for different forms of an author's name.
2. I suggest that when a publisher and imprint are listed together, the publisher should normally come first, followed by a slash, followed by the imprint name.
3. In some cases an imprint is well enough known that it may be listed separately. (Del Rey Books comes to mind) We should identify which imprints this seems reasonable for. Pending the creation of publisher biographies, this info should probably be kept in the wiki.
4. I suggest that, once we agree on standard forms, non-standard publication records ought to be converted. Some of this can probably be done by script. Some will probably have to be done manually, because info that was put in the publisher name may need to be copied to the notes field, or a judgment may need to be made about which canonical publisher or publisher/imprint string should be used.
5. A separate imprint field may well be a good idea eventually, but if we attempt to follow standards conversion should be automatable when and if such a field is available.

What do others feel about these suggestions? -DES ^Talk 16:27, 18 Feb 2008 (CST)

I'm just about finished changing all "Del Rey" and its variants to "Ballantine Del Rey", so I have mixed feelings about changing it back to "Del Rey" or "Ballantine/Del Rey". I do agree that we need some standardized rules to follow.Kraang 16:43, 18 Feb 2008 (CST)

Well, we can always debate the proper form, but changing from one specific form to another ought to be something that a script could handle i would think. My reasons for suggesting the slash is that many publisher and imprint names include multiple words, and thus spaces, and a slash will make it very clear where the publisher name ends and the imprint starts, this will make it easier for a script to move the imprint name to a new imprint field, when and if we add such a field. It will also make it easier for automated searches to use our data, i think. -DES ^Talk 16:55, 18 Feb 2008 (CST)

Sounds like at a minimum we need publisher editing tools that allow the publisher name to be changed without visiting all of the publications (analogous to changing the author's name without visiting all of the titles) which would allow easy switches between "Del Rey" and "Ballantine Del Rey", and tools for merging publishers together. That should lower the amount of manual labor involved. Alvonruff 17:16, 18 Feb 2008 (CST)

That would certianly help a lot. How hard are those going to be to implement? -DES ^Talk 17:28, 18 Feb 2008 (CST)

Probably too easy. :-( Al, give us a tool that helps us zap Manga or RPG pubs and Titles in one go before you give anybody something as damaging as this could be! BLongley 18:31, 18 Feb 2008 (CST)

I agree that a tool that allows editors to merge publishers could be very dangerous. We have caught enough well-meaning Author merges that would have combined a canonical name with its pseudonyms ("How much difference can there be between Iain Banks and Iain M. Banks??") to make me leery even of Author merges and Publisher merges could be literally orders of magnitude worse. Nothing like destroying hundreds of hours of careful work with one click of a mouse to make us go to back to the backups or see people quit in frustration. Ahasuerus 22:55, 18 Feb 2008 (CST)

Having said that, I can also see value in organizing publisher information better so that we could use the database to answer questions about publisher history. The way things are set up now, a search on "Ballantine" returns so many hits that a casual user will be disoriented, although some subsets, e.g. "Ballantine Adult Fantasy", are obviously useful. I am tempted to propose that we change the name of the current "Publisher" field to "Imprint" and then create a new and separate table that would link these imprints to actual publishers. However, how would we handle "floating imprints" that move from one publisher to another? Would we have to revisit all of their publications and add the publisher's name to the imprint? Ahasuerus 22:55, 18 Feb 2008 (CST)

Anyway, all of this may be premature. First, we have a great deal of simple cleanup work to do using the new tool, e.g. merging Pan / Ballantine, Pan/ Ballantine and Pan/Ballantine. Hopefully, the cleanup process will help us come up with better ideas. Ahasuerus 22:55, 18 Feb 2008 (CST)

I think we can and MUST leave this for a few weeks at least, probably months. I see people are using the new tools and thinking "Doh! I really SHOULD have used THAT name instead!" and they're busy "correcting" their past "mistakes". That's fine by me, it reduces the number of imprints and publishers we'll have to merge/make variants for in the long term. I think we need the "Imprint" and "Publisher" fields established BEFORE we allow ANY automated merges though - that can be done now, just copy all "publisher" field entries to the new "imprint" field and let people fix imprints for a bit, then gather them up and we can see how the "publishers" work. BLongley 18:23, 18 Feb 2008 (CST)

I'm really AGAINST "Publisher" "Slash" "Imprint" standardisation as I feel we should be taking care of the imprints, which change hands more often than people may think. And "Publisher" is arguable - do we want the company quoted on the book, or the parent company that owns them (which might also be quoted on the book), or the Group or Division that owns them (again, this might also be quoted on the book), or the holding company that owns a publishing company or ten... what I want are details of when a Publisher got taken over, and became an imprint of another company that owned multiple publishers, and was then taken over or sold on to another company - if I came back here and discovered all of my Mayflower, Souvenir and NEL books were standardised under one current publisher I'd quit. That's the sort of vandalism of data I expect from Amazon. BLongley 18:23, 18 Feb 2008 (CST)

Yes, we probably want variant publisher/imprint names so we can record what's actually specified on the pub. Yes, we want to reduce the number of names used and have a "canonical" name. Yes, it's going to be difficult. But "NO!" - I don't want any of the changes suggested yet. 1) Define "Publisher". 2) No, create separate fields for separate data. 3). No, start from the lower levels of recorded data and work UP. That's what we have to do with printings. 4) Eventually we can clean up remaining stuff with scripts, but not because of the reasons you suggest, which IMO would have destroyed a lot of data in the meantime. 5) A separate imprint field is a MUST, not a "nice to have" eventually. BLongley 18:23, 18 Feb 2008 (CST)

I see your points, and they have significant merit.

1. If we were never going to get an imprint field, I think "Publisher / Imprint" would be the best way to handle things. However, I trust that sooner or later we will ahve such a field.

   Preferably sooner, IMO. It'll point editors in the direction we want. (If we do all want both.) BLongley 13:43, 19 Feb 2008 (CST)

2. I am not in favor of scripted changes which are at all likely to lose significant data. I see no reason why a script couldn't convert "Ballantine Del Rey" to "Ballantine / Del Rey" or "Ballantine Publishing Group" to "Ballantine Books" or "Beagle Books (Ballantine)" to "Ballantine / Beagle" with no loss of data (provided that we agreed on the desired names, these are just examples). What data loss would you see in such cases?

   It rather depends what people have been using the Slash or Brackets for so far. What if somebody has been using the slash to separate Publisher and Imprint already, and your merging a "Ballantine Del Rey" imprint with plain "Del Rey", part of the Ballantine Publishing Group? BLongley 13:43, 19 Feb 2008 (CST)

3. I think entering new publications with Publisher / Imprint, (and with any available further detail in the notes) would ease conversion when the new field is available.

   A lot of people have and probably will continue to use something like this, but as we've already got such in probably every possible permutation, I don't think it'll make things easier. The script will never know if it was entered according to the new standard or some arbitrary older one. However, introducing the new field now WITH explanations of its intended use will help - the new field can't have been affected by older conventions. Well, maybe if we pre-populate it, but then the fact that we have extra separators in both fields should be a good warning. BLongley 13:43, 19 Feb 2008 (CST)

I am not proposeing wholesale changes be made before they they are throughly discussed and agreed to, and if a new field is likely to be available soon, perhaps any scripted changes to existing data should wait on that field. -DES ^Talk 18:44, 18 Feb 2008 (CST)

Regarding slash - while it used to bother me these days I'm used to "Publisher name / SFBC" though I'm still bothered that the publications do not state SFBC other than some Doubledays that say "Doubleday Science Fiction" on the jacket though the title page states "Doubleday."

I'm personally in favor of "imprint / publisher" as the imprint name is often more visible in the publications.

Overall, the publisher names are an area where I don't have strong feelings as there are many publications where the name is quite difficult to determine. The Timescape books immediately come to mind but Fawcett gets combined with Crest and later with someone else.

I tend to enter things as stated in the publications meaning I've used "Del Rey / Ballantine" as that's the order the names appear in on the title page. I usually shorten up a name. "DAW Books, Inc." is "DAW Books"

This is a time when I wished ISFDB had automatic notification to verifiers of changes to publications as I can see people going in and making mass changes. I had thought there were some "Del Rey" books without "Ballantine" and that's one of the things I'mn re-checking as I go through my books. Marc Kupper (talk) 22:27, 18 Feb 2008 (CST)

It depends on what you mean by "without". I'm pretty sure there are "Del Rey" imprint books that aren't "Ballantine Del Rey" imprints, but the publisher is still probably listed as Ballantine Books or Publishing Group.

It rather depends on where you're taking the data from - logo, spine, front cover, copyright page, title page - and what you mean by "imprint" and "publisher" anyway. E.g. One of my acquisitions from Sunday says "Panther Granada" on the spine, but with the two words in different fonts, "Panther" before the ISBN on the cover, "Panther Science Fiction" above the title on the cover, "Panther: Granada Publishing" on title page, "Published by Granada in Panther Books" at the top of the copyright page, "A Panther UK Original" two lines later, gives "Granada Publishing Limited"s address, and at the bottom of the Copyright page lists "Granada (R)" and "Granada Publishing (R)" as registered marks. I'm definitely leaning towards "Panther" as the canonical imprint as it's about the only word guaranteed to be in it! And that can move from Publisher "Panther Books Limited" to Publisher "Granada Publishing Limited" quite easily if we separate the fields. But we might want to regularise "Limited" and "Ltd." and "Ltd" in publishers. And where do we stop with the publishers? A publishing company can still be publishing several imprints while being owned by another publishing company, or might be within a Division of a Publishing Group, or even within the publishing division... to keep things simple, I'd record the lowest level of Publisher above the imprint, and try and group ownership of publishers in the Wiki. We haven't got indefinite hierarchies sorted yet (see Series support) so I wouldn't even attempt to put those in the database itself yet. BLongley 13:43, 19 Feb 2008 (CST)

Here's an old discussion about "Del Rey"[1]. As for imprints and publishers two separate fields would be ideal, short of that I would vote for imprint/publisher as the preferred order.Kraang 19:14, 19 Feb 2008 (CST)

"del Rey" is a simple imprint case. Probably a fairly simple Publisher case too - Ballantine Books, Ballantine Publishing Group, maybe they're just "Random house" now? BLongley 15:51, 20 Feb 2008 (CST)

Del Rey Books, an imprint of The Random House Publishing Group, a division of Random House, Inc., New York. (from 2007) --Roglo 16:00, 21 Feb 2008 (CST)

Thanks! Does that mean "Ballantine" is omitted now? if so, only WE can preserve that history! (Ok, dozens of other sites could too, but WE are special, aren't we?) BLongley 17:10, 21 Feb 2008 (CST)

Not quite omitted; there is Ballantine Books on the spine above the DEL REY logo, Ballantine Books, New York on the title page, but it is not mentioned on the verso of the title page where they state 'published by' with description as quoted previously. --Roglo 17:28, 21 Feb 2008 (CST)

Here's how Locus lists publishers and imprints & publishers[2].Kraang 22:21, 19 Feb 2008 (CST)

Good for that year (2004), I see. If there's similar for other years this might be a good source of data for imprint/publisher ownership changes. BLongley 15:51, 20 Feb 2008 (CST)

(unindent) I don't really see that the higher levels of the corporate hierarchy are of much interest, at least not to most of our users. I think if we document the imprint name (where there is one) and the publisher in the DB records, and leave the history of which imprints belonged with which publishers, and which publishers belonged to which parts of which corporate entities in the wiki, we will be providing for most needs. (Del Rey is an easy case, it has never been a separate firm, and it has never belonged to any publisher but Ballantine, and most of the minor variations in how the name is displayed on book covers are not significant of anything -- "Del Rey Fantasy" and "Del Rey Science Fiction" were separate lines with different editors (Lester and Judy-Lynn del Rey) for a time, as I understand things. Other cases are more complex.) For a given publisher or imprint, we should probably regularize, at least where differences are of no significance: for example "TOR", "A TOR Book", "Tor", and "Tor Books" do not indicate any meaningful distinction. Likewise "Putnam", "G.P.Putnam", "G. P. Putnam", and "Putnam Books" probably are not significantly different, but in this case there is enough history that it is possible one of these forms might be a clue to dating. -DES ^Talk 18:05, 20 Feb 2008 (CST)

Well, with several family businesses like Putnam and Collins there's definitely a chance to separate some generations of the publisher and clarify dates, but the periods are too wide to put an exact year on them (so we need to improve on '0000-00-00') and most of the books are so old and from secondary sources that I'm very suspicious of the Publisher name being entered correctly at the moment. We could definitely regularise "G.P.Putnam" and "G. P. Putnam" though, as we do with authors. BLongley 14:15, 21 Feb 2008 (CST)

One good reason to have SOME link to the higher levels of the corporate hierarchy is that that is where we can find some of the history that the publishers still admit to, even after an imprint has died: and they're actually quite good at mentioning which of their many imprints and subsidiaries and divisions are SF. But I agree, it's not going to be of common interest and I think we can leave this to Publisher's pages (if enough notes and multiple web-links are possible there) or the Wiki. BLongley 14:15, 21 Feb 2008 (CST)

One other data element to consider, strange as it may sound at first, is the book's distributor. There are two scenarios when it's useful to know the distributor. The first one is when a much larger company serves as the distributor for a smaller SF publisher the way Simon & Schuster does it for Baen. This information can help explain overlapping ISBN ranges and also clear up confusion when we import data from places like Amazon.com which sometimes confuse the distributor and the publisher. After all, they got that box in the corner from Simon & Schuster, so S&S must be the publisher, right? :)

This could be handy with US books. Not so useful with British books, which usually don't have the distributor printed on them - and ADDING a known distributor could add to confusion, e.g. most of my newly-acquired US Books from the time I lived in London could ALL be credited to Titan Distribution. Even though they may have only distributed to one store (Forbidden Planet) and may not have had exclusive distribution rights. BLongley 14:15, 21 Feb 2008 (CST)

The second reason why it may be useful to know the distributor is to help our users to find obscure SF books. Some UK based small presses have an official US distributor and providing that distributor's contact data (often a PO box in the middle of nowhere) can be very useful. I would hesitate to put this information in the Wiki because so few passive users ever find the Wiki, but perhaps we could make it a free text field in the Publisher record? Ahasuerus 23:47, 20 Feb 2008 (CST)

That aside, it sounds like there are three data elements that we are interested in: "place of publication", "publisher" and "imprint". One obvious issue with the "place of publication", which we mentioned briefly above, is whether we want to capture it in the Publication record, in the Publisher record or in the Wiki. I can see how capturing the data at the Publisher level would help reduce clutter in Publication records, but on the other hand, if a publisher moved some time during 1986 and published 12 books that year, how can we tell which books were published where? More importantly, if we keep this information at the Publisher level, how do we distinguish between "New York: Random House", "London: Random House, "Toronto: Random House" and "New York-Toronto-Berlin-London-Tokyo: Random House"? Or do we create separate Publisher records for each geographic location? Ahasuerus 23:47, 20 Feb 2008 (CST)

I'd argue over "place of publication" if you mean which city the publisher is based in. The only times it's been of use to me is to distinguish a country - and a currency symbol in the price field is shorter and quicker. Or you need to go to a lower level of detail, e.g. when researching Orbit at the address level I could spot when an imprint/publisher moved in with another publisher, and go look for indications of a buyout or a merger - but despite 8 to 10 "different" addresses (I think they renamed the buildings occasionally, without moving) they would all be "London". At times looking at where a book was PRINTED looked useful, but the same US book reprinted for the English Market by Scottish and German printers didn't make things any clearer. BLongley 14:15, 21 Feb 2008 (CST)

I tend to suspect that "London" or "New York" are probably not of much use, and bill says. Howver "Basking Ridge, NJ" may be of use in tracking a small publisher. i cna see how the street address (when listed, which it often is not) might be of use in tracking merges and other changes of large publishers, which might sometiems be of use. But most peopel won't want to see this level of detail IMO, so if we have a publication-level field for publisher address, we need to have a user option to display or hide such info, I think. -DES ^Talk 14:37, 21 Feb 2008 (CST)

I don't think we want a Publication-level Publisher-Address. Even when I was feeling masochistic enough to go check, I started from ONE sample from each year, and only checked others when there was an obvious change from one year to the next. If there's some doubt about whether one publisher/imprint is the same as another, then spot-checks like this can help - but this is a case where the pains of recording lower-level data in the short-term don't really give us equal benefits in the long-term. I don't want to destroy overly-detailed entries here in the interest of simplicity IF there's a use for them, but I also don't want to make everyone go back and check EVERYTHING they've ever entered to populate a new field of low value. Still, Publishers like this (same as this one?) might need the detail in the short term - at the moment a clearer address should probably go in the Wiki, but are we beginning to have some ideas about what fields we want on the database Publisher records? "Notes" seems a good start! BLongley 15:09, 21 Feb 2008 (CST)

As far as "publisher" and "imprint" go, how does a naive editor (i.e. one that is not familiar with the history of various SF publishers) know which is which? To pick a book at random, "The Einstein Intersection" (Ace #19681) says "Ace Books, a Division of Charter Communications, Inc." Should our naive editor assume that "Ace Books" was the imprint and Charter Communications the publisher in this case? And if only one data element is present, e.g. if it says "FPCI" on the title page, how is our naive editor to know whether FPCI was an imprint or a publisher? Come to think of it, even I am not sure what its legal relationship with Crawford Publications was and I probably have more books/magazines published by the Crawfords than most editors. Was it a successor? An imprint? A separate entity set up by the Crawfords for legal/tax reasons? Since we currently do not distinguish between imprints and publishers and simply enter as much or as little information as the publication provided, it's not a problem at the moment. However, if and when we create two separate fields for this data, it may become a major headache, so we may want to think about it ahead of time. Ahasuerus 23:47, 20 Feb 2008 (CST)

I think this is the least of our worries. The naive editors are also the moderated editors, and WHEN we can agree on what we're recording, most moderators will either know their imprints from their publishers (and of course some names will be both), OR know where to go look for it. BLongley 14:15, 21 Feb 2008 (CST)

And once we have this sorted out, we will develop a help page or series of help pages, at which editors wanting to become less naive can be pointed. But there is no point tryuing to develop a help page until we decide exactly what info we want to collect and how it should ideally be entered. -DES ^Talk 14:37, 21 Feb 2008 (CST)

Well, we're not good at updating help even when we DO agree on something. :-/ At the moment I'm happy if we start clearing up the obvious problems (when we understand which ARE obvious, e.g. regularising some names is good IF we start simple with spacing problems, but start checking before we merge all slashed entries with unslashed similar names). We could probably even cope with some ("suggested" for now) "Canonical" Publisher or Imprint names to argue over, just so that we can indicate, for instance, that somebody HAS recognised that "Tor" and "Tor Books" are linked somehow - nothing major, just showing the possible variants for now and if it turns out that Tor was used for one period and "TOR" for another and "Tor Books" and "TOR Books" covered other years then they'll be fairly easy to switch later. Canonical Author Names are a pain to switch, and I already see Canonical Artist names being assigned too early. Indicating the links is fine, and adding data to each is fine, it'll help in the long run. I think I'm beginning to want "Notes" at Publisher level ASAP for this, as we're two clicks away from the Wiki at the moment and some people ARE lazy enough to ignore that as "too difficult". :-/ BLongley 15:26, 21 Feb 2008 (CST)

Going back to the proposed split of the "Publisher" fields in two, one for "imprint" and one for "publisher", I am sure we could easily come up with ways to handle Tor, Corgi, Putnam etc., but there are literally thousands of obscure imprints/publishers that can't be easily pigeonholed as one or the other. Sometimes it may require a fair amount of research to clear the issue up, at which point we would have to go back and repopulate the right fields in previously entered Publication records. Although probably workable in 90%+ of all cases, this would be a departure from our general WYSIWYG approach to publications. Since we create all other unstated-in-the-Publication-record data associations at the Title level, be it variant titles or pseudonym attributions, this would a significant architectural change.

Having said that, I recognize the value of collecting imprint- and publisher-level information, but I am hesitant to record "unstated" data at the publication level. I wonder if we would be better off if we continued collecting stated publisher information in a single field and then linked it to the Publisher/Imprint table in some fashion? Perhaps the filing logic could automatically create a "publication-publisher" link that we could then clean up whenever warranted? Ahasuerus 20:43, 22 Feb 2008 (CST)

OK, I've spent most of today whacking some British Imprints and Publishers around, and STILL can't see any problem with splitting the publications we have into Imprint and Publisher. Not sure if it's an imprint or a publisher? Put the same data in BOTH fields. Small publishers often ARE their own imprint. BLongley 19:02, 23 Feb 2008 (CST)

Sticking with a single field is NOT going to add any value even if we could agree if it's "Publisher / Imprint" or "Imprint / Publisher" or any other convention. We need to slim down the number of imprints and publishers and link them better - or just continue to let "Publisher" be a meaningless free-text field. I've already encountered imprints owned by multiple publishers SIMULTANEOUSLY - and I'm not giving up yet. BLongley 19:02, 23 Feb 2008 (CST)

Clearly, slimming down the number of "publishers" is a laudable goal. If we can come up with reasonably clear guidelines (well, as clear as most things get around here :-) for populating the two proposed fields, I am sure we will go to Al straight away and recommend that he create a new field. I am just not sure how easy it will be to write something simple yet comprehensive.

Let's take a simple example, Kim Harrison's The Outlaw Demon Wails, which is hitting the bookshelves as we speak. It's a hardcover book, volume 6 in a very well known series from a major publisher, so it should be as unambiguous as it is going to get. And yet when I look it up in OCLC, I see that 22 libraries have it cataloged as by "Eos" and 148 as by "Eos/HarperCollins" even though all of them are working off the same OCLC-provided guidelines. If professional bibliographers, who do this stuff for a living, can't get it right more often than 85% of the time, what are the chances that our motley crew of volunteers will do better than 70/30?

No "HarperCollins/Eos" variations though? Great! "Imprint/Publisher" seems to be the way to go if we want to keep stuff in one field. Or we now know how to separate them if we want two, if we're checking with OCLC. But we should have enough data here soon to record Eos as the imprint and HarperCollins the publisher. In fact, I see you mentioned Eos was the imprint when you added it to the Sources of Bibliographic Information. I assume you'll update the Publisher page for Eos shortly then? BLongley 14:22, 25 Feb 2008 (CST)

Yes, I need to create a "Publisher" page for Eos. I have been looking for an overview of all genre publishers/imprints and I think I have finally found something useful. I have created a Publisher:History of SF Imprints page, but it needs better formatting and then we may want to move it some place else, e.g. Help. Ahasuerus 18:37, 25 Feb 2008 (CST)

I suppose even if we decide that we can't expect a high rate of accuracy during the data entry phase, we could still create an "imprint" field, tell our editors to do their best to populate it correctly and then go back and adjust the data based on our knowledge of various imprints/publishers' history. This should improve our accuracy rate to 95%+, but we would be in danger of incorrectly assigning imprints to publishers in borderline cases.

I am not crazy about these scenarios, but on the gripping hand, one could argue that we are not doing a great job of capturing imprint details anyway, so would we really be in worse shape if we went that route? <insert a picture of Hamlet or Buridan's ass here> To quote an old Analog story, there is got to be a better way (hopefully)... Ahasuerus 23:41, 24 Feb 2008 (CST)

I personally use the "if in doubt, do nothing" on individual cases, but add data where I can, it might help later. And we're a specialist case - if we CAN'T improve on what general libraries are doing, why does the ISFDB exist? BLongley 14:54, 25 Feb 2008 (CST)

I could have a little grumble about what some editors are doing - e.g. I see a lot of regularisation of Publishers going on without anyone updating the Wiki page to say what the eventual aim IS. E.g. "Pinnacle/Zebra" or "Zebra/Pinnacle"? I saw a lot of "Pinnacle" becoming "Pinnacle Books" yesterday, for instance, is that going to be reflected in the final pages? Are we going to leave those problems on the Wiki or resolve them in the database structure? I've already seen some of my Wiki pages disappear as the page I was working on for an Imprint got redirected to the one I wanted as A Publisher of the imprint (Yes, there are more I want to link the Imprint to later), and I hope that was just an accident during the namespace move. BLongley 14:54, 25 Feb 2008 (CST)

(unindented)That was me sorting out the names and identifying the Pinnacle/Tor(1981-late'83) titles before Tor became a separate Publisher/Imprint (in late 1983). As fore the Pinnacle name I went with "Pinnacle Books" since this appears to be its most widely used form. Regarding notes in the wiki I felt it best to add them in when I was finished because as you point out if the publishers name changes the notes don't move with it. You have to go and find the old wiki title and move the notes to the new name manually. Most of the publishers I've worked on was only to bring the variant titles (mostly minor order or spelling variants) under one name. Just so everyone knows my preferred order is Publisher/Imprint where it applies.Kraang 19:20, 25 Feb 2008 (CST)

Still, great activity, and I think it's NEEDED, but I'm not sure we're all pulling the same way. A new Imprint field on the publication is what I want, but if I/we don't get that then maybe an Imprint namespace on the Wiki is the way to go. But we really could do with a better aim as to where we WANT to go. BLongley 14:54, 25 Feb 2008 (CST)

Entering Reviews of Ace Doubles

The question has come up as to whether magazine reviews of stories in Ace Doubles are entered individually, or as a whole. It was my understanding that reviews went in individually, but I can find nothing on the help pages on this. I've moved the Discussion here for more input.--Rkihara 10:24, 23 Feb 2008 (CST)

It seems to me that reviews are normally reviews of titles, not of publications, and that ACE doubles are merely publications of titles which have been, or might be, published in other forms as well. If I have understood correctly, there is no reason for our record of a review of a title to be forced to link to the ace double pub, even if that was the pub the reviewer mentions. Rather it should link to the title record, where the ace double pub, and any other pubs, will be listed. Or have I misunderstood the issues here? -DES ^Talk 15:29, 23 Feb 2008 (CST)

In ISFDB terms, Reviews are of Titles. Ace Doubles are Omnibuses with one ISFDB Title consisting of two other titles separated by " / " (which may or may not appear here separately). So if you review an Ace Double, you have to make sure the review is of the joint title if you want the link. If you enter both reviews separately, you might link to reviews of separate Titles which may not be (and actually, probably aren't) the same versions as the ones in the Ace double. BLongley 16:16, 23 Feb 2008 (CST)

In the REAL world, reviews are often of Publications that we do have here - but it's not uncommon to review a Hardcover and (Trade?) Paperback edition simultaneously. They'd be reviews of an "Edition" I'd say - but we don't have the concept of "Edition" embodied here. It's possibly a bit late to point this out, if people HAVE been entering the separate halves of Ace Doubles, but if you enter a review of the Double's title you're getting closer to matching the actual "Edition". If the review says "both novels have been severely truncated to fit into this slim paperback" it's a bit misleading to point to the original full-length titles. Or in reverse, "this Ace Double edition restores all the text that the original paperback novel versions of the magazine series lost". Even if a review is only of ONE publication, the Title linkage means you don't know which publication or edition was reviewed. Date of Review and Date of Publication usually mean it's fairly clear which edition or publication were reviewed though. BLongley 16:16, 23 Feb 2008 (CST)

Personally, I like the fact that reviews are recorded, as if they review something we haven't got here we can go FIND it. But if you review something separately that only ever existed as half of an Ace Double, you're sending me on a wild goose chase. BLongley 16:16, 23 Feb 2008 (CST)

Basically, reviews as entered HERE are either fillers in other works or useful pointers to titles we do want. Making them point to something that never existed is a bad thing, IMO. I'd quite like REVIEW records to point to the Publication(s) they review, but we've already got enough suggestions that will make everyone go back to square one on their own collection and redo everything! BLongley 16:16, 23 Feb 2008 (CST)

Ace doubles are, or if they aren't, IMO, clearly ought to be, recorded as omnibuses of two titles with two distinct title records, just as any other omnibus is. DES ^Talk 17:49, 23 Feb 2008 (CST)

Yes, we spent a lot of time making sure of that last year. "Two Novels" seems to work well, there are problems when one half is a collection or an anthology. As usual, discussions petered out before we could get a consensus on how to deal with OTHER doubles. BLongley 18:38, 23 Feb 2008 (CST)

Each of the titles might be, and in many cases has been, published separately. Therefore, IMO, any review of one of the works in an Ace Double should link to the separate title record for that work. DES ^Talk 17:49, 23 Feb 2008 (CST)

I can only read that as a request to link to the WRONG edition. Please explain! BLongley 18:38, 23 Feb 2008 (CST)

Reviews do not currently link to any edition, they link to a title, and IMO that is what they ought to do, unless we actually come up with an edition-level record. If we do that, much will need to be rethought. -DES ^Talk 19:49, 23 Feb 2008 (CST)

If both included works are reveiewed, there should be two review records. DES ^Talk 17:49, 23 Feb 2008 (CST)

Why, if it's one edition? Do you want us to create separate titles for halves of something we are sure about? We have a lot of phantom entries just to deal with pseudonyms already, how does it help to add more? BLongley 18:38, 23 Feb 2008 (CST)

Yes, I do. More accurately, often a single review column will discuss half a dozen or more works, some of them related in one way or another, others not. IMO every single title discussed at any length in such a column merits a separate review record. The titles already merit separate records, and if i under stand your comments above, are already getting them. as to why reviews ought IMO to link to title records, not publication records, it is simply that often, even with the review in front of you, you can't tell accurately which specific pub is being reviewed -- indeed it may well be an ARC being reviewed, so it won't precisely match any possible pub in our DB. And a review of one publication of a title is usually also a review of other publications unless very significant revision has occurred. -DES ^Talk 19:41, 23 Feb 2008 (CST)

Yes, a review may refer to a particular version of a work, which may match some publications and not others. However, until we have a workable concept of "edition" here, there is no practical way for a review to link to all and only the publications that have the same version of a work as the reviewer reviewed. Furthermore, often a review of one version is none-the-less helpful in considering other versions, and some reviews mention changes between versions. In short, for the present, i think that reviews should link to title records, not to publication records, and that when versions are different enough that the user would want to know this, a mention in a note on the review record (or perhaps the publication that contains the reviews, if need be) is the way to go. -DES ^Talk 17:49, 23 Feb 2008 (CST)

If people added notes about "what was being reviewed" to reviews it would help a bit. They don't though, so I'm happy to stick with my view that Reviews as we currently have are just guidelines to Publications. It's not as if the reviews we record actually say anything ABOUT the book reviewed - we just say something WAS reviewed. BLongley 18:38, 23 Feb 2008 (CST)

Please note that currently reviews are in fact listed on title display pages, and not on publication display pages. -DES ^Talk 19:46, 23 Feb 2008 (CST)

(unindent)First, let's make sure that we are all on the same page, folks ;) In ISFDB terms, an Ace Double, just like any other omnibus, consists of a single Publication record which has at least three Title records associated with it. One Title record is of the "Omnibus" type and the other two Title records match the respective type of each half of the Double. The Omnibus Title is entered as a collaboration between the authors of the two halves, so an Ace Double that includes a Bulmer novel and a Hamilton novel will be displayed as "Fugitive of the Stars / Land Beyond the Map (1965) by Edmond Hamilton and Kenneth Bulmer". As Bill pointed out, this works reasonably well when the two halves are both novels, but there are some complications when one or both of them are collections or anthologies. That's fodder for another discussion, though.

The way Author Summary pages currently work, they display both the individual title representing one half of the Double (usually a novel) as well as the Omnibus title since the latter is recorded as a collaborative work. For example, the Bulmer Summary page will display the Omnibus that I cited above as "Fugitive of the Stars / Land Beyond the Map (1965) [O/2N] with Edmond Hamilton".

The way Review records currently work (although we hope that it will be changed in the foreseeable future), they are linked to individual Titles whenever the Title record and the Author record match lexically. This design seems to leave us with three choices:

1. We can decide to enter Ace Double reviews as reviews of collaborative omnibuses, in which case they will be displayed for that Omnibus Title, but not for the two Titles that comprise the Omnibus. One could argue that this is a problem since our hypothetical "naive user" may well look up "Fugitive of the Stars" and "Land Beyond the Map" under their respective Novel Titles and not under the collaborative Omnibus Title and never find the review.
2. We can decide to create two Review records for each Ace Double (one for each half), but then the review will not be linked to the Omnibus Title and if a user looks it up there instead of under the individual Titles, he will once again fail to find the review.
3. Finally, we can decide to create three Review records, one for the Omnibus and then two for the two halves of the Omnibus. This will cover all bases, but will involve additional data entry.

I suspect that the third approach would give us the most bang for the buck, but I am hesitant to propose a complete overhaul of Omnibus reviews since we expect a change in the way reviews are handled Any Time Now (tm).

As an aside, Ace Doubles tend to be more complicated than regular omnibuses since in many cases Wollheim had to cut the original text, sometimes substantially, to make two novels fit into one book, yet he still advertised the cut versions as "complete and unabridged". That's a whole different can of worms, though, and we probably don't want to go there at this point. As long as we stay away from these complications, we can discuss the way we record all Omnibus reviews; Ace Doubles will be just a subclass. Does this make sense? Ahasuerus 20:53, 23 Feb 2008 (CST)

Maybe the novels could be entered separately, but identified as part of an Ace double by the entry of the series number (Ace M-111 for the example above) into the "series" field. The two novels are now linked together, but still listed separately, something we can do now without any programming changes.--Rkihara 22:32, 23 Feb 2008 (CST)

The problem with that is that many works published in Ace Double format are in fact part of real series, and we don't currently support a work being part of more than one series. Also, series data is recorded at the title level, not the publication level, and this method does not take into account that many works initially printed as ace doubles were later reprinted in other forms.

My feeling is that we should take the second approach, or possibly the third. i don't like the 3rd title record, and I would hope that a later revision of how omnibuses are handled would render it unneeded. But for the time being it is here.

The obvious way to handle the length changes is with variant titles -- "(abridged)" or the like -- but determining which ones were significantly abridged and creating all the vt records would be non-trivial. -DES ^Talk 00:25, 24 Feb 2008 (CST)

You can nest a series, e.g., Ace Doubles/M-series/M-111, and publications can be put in series, consider the anthology series, such as Damon Knight's "Orbit" series. The publisher data will differentiate between different printings of the publications.

Alternately, the Ace doubles could be entered by series number onto a dedicated page, much like the ones we now use for magazines. I believe the doubles disappeared in the late seventies, so the page would probably be smaller than the one we use for F&SF.--Rkihara 00:57, 24 Feb 2008 (CST)

I dislike 2 as you're misdirecting people going from review to publication. I'm not keen on 3 due to the amount of rework involved. I'm not sure how far away we are from 1 at the moment. How many people go from publication or title to review anyway? Yes, it would be nice if everything linked both ways (e.g. every story in a collection could link to a review of the collection, as well as every publication of that collection, and the collection itself). But I for one only get concerned if a review points at something we're MISSING, never the other way round. Sure, it's nice to see how many times something WAS reviewed, it's a good indicator of the expectations people had of a title. But that's a "nice-to-have" presentation thing, reviews of something that doesn't exist is a real concern to me as that's what I use reviews for. So if 1 isn't acceptable, then 3 please. How many Doubles get reviewed anyway? BLongley 06:48, 24 Feb 2008 (CST)

Whereas I, at least sometimes, use the review links displayed with a title to find the actual review, and read it, either to decide about the title, or to compare my view of a work with a reviewer's, particularly if it is a reviewer I know something about. 16:38, 24 Feb 2008 (CST) —The preceding unsigned comment was added by DESiegel60 (talk • contribs) .

(unindent) here is an example: Solar Lottery. It was first published as part of Ace double #D-103, it has been reprinted many times since singly. It has multiple reviews, at least two of which, from the dates, almost surely reviewed the ace double edition. All the reviews link to the individual title record for Solar Lottery; none link to the publication record, and none link to the joint title record. IMO this is exactly the way this should be handled in most if not all cases.

Well, obviously none link to the publication record, the database doesn't work that way yet (and may never do). I'm a bit concerned that none link to the Omnibus - when we are working on MISSING titles based on reviews, there's nothing in the separate review records that could point us at a missing Omnibus, just missing Novels. Still, your example is of one with plenty of data, in which case I can see that I probably want reviews sorted by date (I know, another new Feature Request that Al won't have time for) as it was a mild pain to pick out the relevant reviews that can only have been of the Double (if we are complete in that area and there was no 1955 separate publication). These are the relevant ones, I think?

1. Infinity Science Fiction, November 1955, (1955), reviewed by Damon Knight
2. Galaxy Science Fiction, November 1955, (1955), reviewed by Floyd C. Gale
3. The Magazine of Fantasy and Science Fiction, August 1955, (1955), reviewed by Anthony Boucher

If so, I find myself wondering whether the first Magazine is correct as the other two review the other title as well. Unfortunately that verifier is inactive. BLongley 17:08, 24 Feb 2008 (CST)

It is perfectly possible that it is correct. Knight might very well have reviewed the Dick title and not the title it was bound with, if the second title did not fit into his column or was less interesting in his view. Which also reminds us that a reviewer might not choose to review both halves of a double. But it is surely possible that our listing is incomplete. -DES ^Talk 17:18, 24 Feb 2008 (CST)

Solar Lottery (D-103, May 1955) was an Ace original while the Brackett half was a reprint of the 1953 serialization, so it's entirely possible that some reviewers concentrated on the new novel and skipped the one that their reader were already familiar with. By the way, this is where Double Your Pleasure: The Ace SF Double by James A. Corrick comes in handy. He has a lot of useful data about uncredited cover art, month of publication, subsequent reprints, etc, etc. It's one of the very few books that I have with me on the road and one of these days I will make sure that his information is reconciled with ours. One of these days... Ahasuerus 18:12, 24 Feb 2008 (CST)

Here's an example of where I'm coming from. This review appears to be of a Double. If it's not - separate review entries please. If it is - well, we've got some more hunting to do. If people entered all reviews as for the individual titles, we'd not even be inspired to go look. BLongley 15:21, 25 Feb 2008 (CST)

Hmm. After Rkihara's comment to me, I did a quick scan of reviews with slash in the title, & concluded that the only ones likely to be Ace doubles were ones I'd entered or "fixed". Obviously my eye skipped right past that one, so possibly some others.
I think I'm pretty much in agreement with Bill on this, but maybe not, so let me stick in my $0.02, especially as I guess I'm the original source of all this debate. DES said, way back, "It seems to me that reviews are normally reviews of titles, not of publications, and that ACE doubles are merely publications of titles which have been, or might be, published in other forms as well.", & followed this up with some similar comments when Bill kept talking about editions. But there's no question but that we're talking about titles; the double is a pub, as is any other edition of the same work, & that pub has a title. The reviewer is working from a specific publication, whose (individual) contents may in fact differ markedly from other editions of the same works.
Moreover, these are recorded as omnibuses; but is there really any fundamental difference in this between an omnibus & a collection? When a reviewer reviews a collection of short stories, and discusses some or all of them individually by name, does this mean that we should be entering a separate review for each story? Even though those stories have their own title records and exist in other pubs as well, conceivably even as chapterbooks on their own?
I haven't myself run into any cases where a reviewer reviews only one half of an omnibus, but others have cited them. In such a case, whether it's entered as a review of the one part or as a review of the omnibus, I'd want a note. But where all the contents are discussed under one common heading for the omnibus, it seems to me really misleading to enter it as two separate reviews. (I have yet to see a case where there was a pub comment saying these reviews were together the reviews of an omnibus; had that been done, I'd probably have followed suit. But as it was, it seemed merely to be a clear case of making the entry reflect the actual contents of the pub.) -- Dave (davecat) 10:27, 26 Feb 2008 (CST)

As to series issues, consider The Beast Master published as part of ace double #D-509, but also part of the Hosteem Storm series. -DES ^Talk 16:35, 24 Feb 2008 (CST)

Oh, sure, there are many novels that appeared as Ace Doubles and were (or later became) part of a series. However, the proposal, assuming I understood it correctly, was to set up a new Series for Omnibus title records, not for the individual Novel titles. The series would be similar to the way we set up Anthology series like Orbit even though some of the stories published in Orbit may have been part of another series.

Having said that, I am not sure if setting up an "Ace Double Omnibus Series" would help us specifically with reviews. Of course, it would make it easier to track Ace Doubles and if there is enough interest in the subject, some of our users may find it useful. Ahasuerus 18:12, 24 Feb 2008 (CST)

I thought, but perhaps was mistaken, that it was suggested above that the two titles in a double should be put into a series named for the the catalog number of the double involved, to indicated their link. -DES ^Talk 19:02, 24 Feb 2008 (CST)

Now that I have re-read Ron's proposal, I believe that your interpretation is likely correct. Unfortunately, as you pointed out, the software doesn't let us associate the same Title with more than one Series, which has been known to cause problems, e.g. R. A. Salvatore's Servant of the Shard is volume 3 in his Paths of Darkness series, but also volume 1 in The Sellswords series. Since we don't have software support for this scenario, The Sellswords series currently starts with volume 2 :-(

Also, it's worth pointing out that we have occasionally asked Al to add support for "publication series", which would let us organize Publication records in a series of sorts; it would be used to link publications like the Del Rey "Discovery" series, Ace SF Specials, etc. If and when this level of support is added, it will be easy to create a "publication series" for Ace Doubles. Until then, we use this Wiki to document things like Series:Macmillan's Best of Soviet Science Fiction Series. Note that at one point an editor tried to create a regular Title series for it, but stopped after the first book, presumably because he realized that some of the books (e.g. some of the Strugatsky books) already belonged to a Title series. Ahasuerus 19:36, 24 Feb 2008 (CST)

Speaking of Ace SF Specials, it seems someone has dealt with these by different Publishers: "Ace SF Special [1]", "Ace SF Special [2]", "New Ace SF Special" (and a stray single title under "Ace SF Special"). Does anyone know who did this, and if so can they be informed that the square brackets appear to cause problems with the Publisher Page Wiki Link? BLongley 12:25, 25 Feb 2008 (CST)

I think it was Scott Latham's idea, but I am not 100% sure. Too bad Scott is currently inactive, so we can't ask him. Ahasuerus 13:01, 25 Feb 2008 (CST)

Linking to Author images

Reposting from User talk:DESiegel60:

I have approved the addition of Tom Shippey's picture since it is hosted by his academic site and I doubt that it will cause Saint Louis University bandwidth problems :) Still, if we are going to link to some types of sites without first securing their explicit permission (which I assume we don't have in this case), we probably want to update our policy. I wonder if it is safe to say that all .edu and .gov sites are OK? Ahasuerus 14:32, 24 Feb 2008 (CST)

I assumed that an author's picture, hosted by that author's site, would be normally permissible, as it would be unlikely to cause bandwith problems, and that explicit permission was more important when a site would be the source for multiple images. But if you think better, I won't insert such a url again without seeking explicit permission from the site. -DES ^Talk 16:01, 24 Feb 2008 (CST)

I think we got burned with Wikipedia images a while back -- we had no idea what the WP policy was and were quite surprised when the issue came up -- so we may have overreacted a bit by requiring that we get explicit permission from all sites that we link to. Let me post this on the Standard board and see where the discussion may lead us. Ahasuerus 20:00, 24 Feb 2008 (CST)

Audio Bindings

In this discussion on the community portal, BLongley pointed out that listing audio books simpl;y as "audio" in the binding field is to lose possibly significant format information. I therefore propose that our standared ought to be "audio (format)" in the binding field for auudio books, with available choices being "Audio (cas)", "Audio (CD)", "Audio (DVD)", and "Audio (ebook)". Other choices can be added as they arise. "Audio" without a format listed should be coirrected whenever possible, and should not be entered for new records unless there is a note explaining that the format is not known. What do you think? -DES ^Talk 06:56, 1 Mar 2008 (CST)

Well, let's have a look at what we currently have before we go defining exact choices: here's the counts of '%audio%' and '%tape%' and 'CD' bindings from the last backup I loaded: my opinions in third column.

audio			182	Too imprecise
audio CD		143	OK
audio cassette		88	OK
audio MP3 CD		8	OK
audio mp3		7	Not OK - is this a physical CD/DVD or a download?
audioboo		4	Too imprecise
audiotape		1	Change to 'audio cassette'?
1 livre + 1 CD audio	1	Not sure what to do about Book + CD
OverDrive Audio Book	1	Not OK
Digital Audio Cassette	1	Should be 'audio digital cassette' maybe?
audio bo		1	Too imprecise
audio ca		1	Change to 'audio cassette'?
Audio MP3-CD		1	Change to 'audio MP3 CD'?
Audiobook		1	Too imprecise
tape			5	Change to "audio cassette?
CD			19	Too imprecise
CD-ROM			3	Need to check - could be games
MP3 CD			9	Change to 'audio MP3 CD'?
mp3cd			2	Change to 'audio MP3 CD'?

I can't see an example of 'audio MP3 download' or any other non-physical formats. BLongley 09:11, 1 Mar 2008 (CST)

Non-physical formats may be outside our scope - I know I've only spent a few hours on it, but I've not yet found one with a stable record number of any kind. The audio Project Gutenberg items may deserve it when they arrive, but I've not seen examples yet. BLongley 19:12, 1 Mar 2008 (CST)

I also suspect a lot of our more recent 'unk' formats are audio books, and there's a couple of thousand of those to look at. One other thing I'd like people to consider is recording Abridged versus Unabridged versions of Novels, or if they're recordings of Plays rather than the original novel. We should already be recording Narrator/Reader/Performer(s) in notes - and moving them out of the Co-Author status I often see them in. BLongley 09:11, 1 Mar 2008 (CST)

There are a number of audio ebook versions at Project Gutenberg. These are not yet entered, but will be in time. Note that "Digital Audio Tape" in that order, is the proper name of a specific format (often known as "DAT"). But I would in any case use the general format of "Audio (format string)" whatever the "format" string turns out to be -- it might be "DAT" or "MP3-CD" or "MP3-ebook" in proper cases. -DES ^Talk 12:42, 1 Mar 2008 (CST)

Also see the complete list of binding types as of October 2007 posted above. Ahasuerus 13:18, 1 Mar 2008 (CST)

In there interest of keeping the binding field relativly short and cannonical, of the forms you list above, I would change:

• audio CD -> audio (CD)
• audio cassette -> audio (cas)
• audio MP3 CD -> audio (CD-MP3)
• audiotape -> audio (cas)
• Digital Audio Cassette -> audio (DAT)
• audio ca -> audio (cas)
• Audio MP3-CD -> audio (CD-MP3)
• tape -> audio (CD-MP3) (subject to verification that this is accurate)
• CD -> audio (CD) (subject to verification that this is accurate)
• MP3 CD -> audio (CD-MP3)
• mp3cd -> audio (CD-MP3)

The others you list are too imprecise to convert without more data, except that "1 livre + 1 CD audio" I would probably handle like two separate publications forming a "boxed set". -DES ^Talk 13:26, 1 Mar 2008 (CST)

Well, unless you're volunteering to add brackets to the majority of our current standardized books I'd stick with the top ones for now, and hope Al can convert via a script to the FINAL preferred choice if we eventually reduce Binding down to something we can offer as a standardized list rather than free-form text. It'd be easy to list the irregular ones as a separate project page if people want to work on them, but as I've knocked a few dozen around a bit today that should probably wait till after the next backup. For now, it's easy to combine this with the Publisher standardization - e.g. I've just canonicised "Books on Tape" and there's plenty of publishers/imprints with "Audio" in the name that are a bit of a giveaway as to what they usually publish. ;-) BLongley 15:51, 1 Mar 2008 (CST)

Looks like about 250 edits can handle the ones where the info is there, and the transformation would be straightforward. That shouldn't be overly onerous. But what I most want is standards for going forward. Once those are agreed, I'll be happy to join in the manual conversionb work -- it this point it seems as if waitign for Al on anythign that others can do is a poor idea. -DES ^Talk 17:13, 1 Mar 2008 (CST)

OK, I'm lazy - I'd rather convert "audio" or "unk" to "audio CD" or "audio cassette" as appropriate (which I think is useful), rather than "audio CD" to "audio (CD)" and "audio cassette" to "audio (cas)" which adds nothing. I believe we can regularize usefully. Use of brackets can wait, and if we regularize then it's easier to convert eventually. The case doesn't matter much, as even if we get searches by binding, by default it will be case-insensitive search. BLongley 19:12, 1 Mar 2008 (CST)

I whacked a few more audio publications around a bit today, and discovered one new oddity. Bookcassette®. I'm open to suggestions on how to record these - see here for an example - I suspect we've classified some as plain cassette versions though, and they obviously aren't if you need specially equipped cassette players. Not a huge problem (as I don't think we actually have any Editors or Mods looking at these) but something to bear in mind. BLongley 15:34, 2 Mar 2008 (CST)

Audible.Com? They have been around a long time; just purchased by Amazon.Com -DRM though. Emusic? Just recently in non-DRM mp3 format. Most downloadable books are derived from a physical counterpart.--swfritter 18:43, 29 Mar 2008 (CDT)

Amazon Blogs

I keep coming across such for active but little-known authors, e.g. Brandon Massey. The URLs don't look especially intuitive so I wonder if they're stable enough for use here? (We can have LOTS of websites for an Author, that's not the problem - Garth Nix has SIX already for instance - but stability is an issue for me.) BLongley 16:48, 29 Mar 2008 (CDT)

Amazon has been attracting more and more writers and editors lately; the attraction is quite understandable since that's where their potential customers are. This can be quite useful, e.g. Lou Anders' Amazon "listmania" page has various lists of books that Pyr has published (or will be publishing). I added a pointer to his page to our Wiki's Pyr page just a few hours ago.

Stability is certainly a concern, especially given Amazon's image handling history, but I would guess that they would be reluctant to upset their "resident" writers by butchering their blog URLs. I could be wrong, of course, but it seems to be worth a try.

Another concern that we had in the past was linking to fan sites -- as opposed to author-authorized sites of living writers -- but it doesn't seem to be a problem with Amazon's blogs. Ahasuerus 05:05, 20 April 2008 (UTC)

Two Variants of something we don't record

It seems we have two novelizations of the same film (Capricorn One) here by Ken Follett/Bernard L. Ross and here by Ron Goulart. Clute and Nicholls imply they're different versions, although I'm not so sure as the UK one refers to the US Fawcett edition: would they do that if it was significantly different? Either way, it's a bit of a mess: if they're the same, then Ron Goulart is a pseudonym of Ken Follett as well as being a real person himself (allegedly): if they're different, then they're both variations of the Screen-Play (which we don't have).

1. Does anyone have a copy of the US version so we can compare?
2. If they ARE different, how do we record the relationship?

For once, I'm tempted to create variations BOTH ways round and watch the ISFDB go into a recursive loop. Suggestions welcome! BLongley 14:27, 31 Mar 2008 (CDT)

I have Goulart's Capricorn One in my collection and should be able to check on Saturday. Do you want to post the first sentence of three random chapters (assuming the novel has chapters) here so that I could compare? Ahasuerus 00:57, 17 April 2008 (UTC)

Sure. Chapter 1: If a city is a lady, Houston is a whore. Chapter 6: Charlie Brubaker had a red-white-and-blue trundle bed and red-white-and-blue wallpaper, done in an airplane motif, in his bedroom. Chapter 9: The long black limousine pulled in outside the air terminal. (I don't think those three lines will spoil anything for anyone. Unless they're fond of Houston. ) BLongley 17:59, 17 April 2008 (UTC)

Nope, nothing like what Goulart wrote. My 189pp mass market paperback consists of 35 short chapters with a lot of blank space on half the pages. Goulart was a very busy man in the 1970s, trying to juggle as many projects as he could -- with mostly predictable results. Ahasuerus 05:12, 19 April 2008 (UTC)

OK, clearly different. Quality issues aside - do we want to record a relationship of some kind? Or let people figure it out for themselves? BLongley 23:56, 19 April 2008 (UTC)

I think the safe thing to do would be to add a note to both Title records explaining that there are two different novelizations of this film by different writers and that they have been confirmed to have different texts. Safety first :) Ahasuerus 04:56, 20 April 2008 (UTC)

Published Bibliographies

The Book of Andre Norton contains a bibliography from 1975 of her works, compiled by Helen-Jo Jakusz Hewitt. It references a number of publications not included in the ISFDB, particularly German titles and publishers. The only information beyond title, publisher, year and language is pages for the first publication. Is this worth entering? How to credit? Any volunteers - 11 pages. -- Holmesd 22:17, 3 Apr 2008 (CDT)

I'd certainly enter missing English titles, I personally wouldn't bother with the German ones but then I'm biased against languages I can't understand. ;-) Does it do better than this site though, which quotes that article and might have used it as one of the sources anyway? BLongley 13:42, 4 Apr 2008 (CDT)

The referenced site has the information from the bibliography, but isfdb does not. Sounds like your recommendation is check the publications here and create new ones as appropriate (leaving blank the pages, title, etc.) but add a note for the source. -- Holmesd 12:05, 5 Apr 2008 (CDT)

Yes, that worked for me. Was it OK by you? BLongley

Further question - There is a list of shorter fiction at the end. One sample entry states: "The Boy and the Ogre" Golden Magazine Sept., 1966. This title is not listed currently under Andre Norton. I'd rather avoid doing periodicals - would someone else take these? 21 titles max - the first two titles had all publications already. Holmesd 04:07, 17 April 2008 (UTC)

I'm not particularly interested (so many other things to do!) : but I'd dump them on the Author Page instead of your own talk page if you want them picked up and worked on. From dealing with other titles from the same Bibliography, they look useful. BLongley 21:19, 19 April 2008 (UTC)

Good idea. Done. -- Holmesd 03:35, 20 April 2008 (UTC)

Publishers

I think we've done a lot with the new tools, but we lack direction. I don't actually WANT to be directed, but when someone is actively promoting one variant over another it would help if we knew when we were working against someone else, at least. Mods don't have a view of the most commonly used variations (OK, they may have seen the file I provided of all current variations, with number of pubs attached, but that gets out of date VERY rapidly with the smaller publishers) and there's no way yet to guide or suggest things to Editors apart from direct messaging. BLongley 00:26, 14 April 2008 (UTC)

If you feel we lack direction and yet don't want to be directed yourself then it seems up to you to provide the direction. Marc Kupper (talk) 07:41, 19 April 2008 (UTC)

OK, first directive to everybody: "Go sort out a chosen publisher/imprint you feel fairly expert with, and then post the link here as an example of what YOU want". Once we have some examples we can establish some guidelines as to what will NOT annoy others, which is a start. We may even find some things we all agree on, or at least not disagree with - e.g. I can't see anyone complaining about a Start Date for a publisher or imprint that we can use to spot earlier misclassifications. (WOTC/TSR for example.) Yes, there'll be arguments about what the start date SHOULD be, but that can come next on a case-by-case basis. BLongley 22:32, 19 April 2008 (UTC)

• I'll take DAW... Marc Kupper (talk) 07:04, 21 April 2008 (UTC)

I thought you might. ;-) BLongley 19:42, 22 April 2008 (UTC)

I think what I'd like is that new submissions get checked against the current "master list" of publishers and new publishers have to be approved specially: maybe by making sure the Wiki page exists for new publishers? That should make people stop and think a bit. BLongley 00:26, 14 April 2008 (UTC)

I don't want to do this manually. If you want to add a tag to the publisher record saying "canonical" (publisher or imprint) then the software can take care of warning users if they are entering a non-canonical name and perhaps to suggest the canonical name. Marc Kupper (talk) 07:41, 19 April 2008 (UTC)

I don't want to do it manually either. I think we need some software changes though (as mentioned below) to take this forward - even if we tagged every current publisher for Canonicity or not, with redirects in the Wiki (I've used this a bit for some Imprints) that's far too much work to set up and still requires too much work to use. BLongley 22:32, 19 April 2008 (UTC)

Secondly, I'd like some way of people taking a bit of responsibility for current publishers and suggesting which ones SHOULD be canonicised, even if temporarily. That would need software changes to allow people to mark a "preferred" publisher and also suggest such from the "unpreferred" ones. And vice versa: suggest imprints when somebody wants to use a publisher like "Random House" or "Hachette Livre". We've thousands of "Scholastic" books here that could be better categorized. (Point Horror/SF/Fantasy come to mind.) I'm happy to take a bit of responsibility for a couple of dozen imprints I'm pretty sure I can deal with, and advise on a few score more (yes, they're mostly British ones!) but so far I can't think of a workaround unless every Mod agrees to check publishers against Wiki pages in the meantime? BLongley 00:26, 14 April 2008 (UTC)

I'm not sure what you mean by "a bit of responsibility for current publishers." Can you provide an example of what a responsible person would do? Marc Kupper (talk) 07:41, 19 April 2008 (UTC)

As it stands: Claim a publisher as something YOU'RE working on, on the Wiki page. Work on that publisher - fill in missing ISBNs, missing SF titles, research ownership of the publisher or imprint, establish earliest and last dates the imprint was used, or for the publisher existing: go ask the verifiers of pubs affected as to whether they can update to match the proposed standards or have other suggestions: link to other related imprint's Wiki pages, other publisher's Wiki pages - or clear those if it's a dead end you don't want. Bug Al for software changes to match what you want (eventually - he deserves a rest) but I still want Imprint and Publisher on each and every publication, and Imprint/Publisher hierarchies by date in the long run. In the short term - maybe we have to do that in the Wiki. BLongley 22:32, 19 April 2008 (UTC)

My priority is to move to development. Al posted the current python code and so I'll be updating my server. Marc Kupper (talk) 07:04, 21 April 2008 (UTC)

I also don't know what you mean by "We've thousands of "Scholastic" books here that could be better categorized." Scholastic is a publisher and I'd assume there are many publication records much like there's many for Ace, Tor, Ballantine, DAW, etc. Marc Kupper (talk) 07:41, 19 April 2008 (UTC)

And it seems many "Scholastic" books here are just classified that way rather than recording any useful imprint that would say it's SF or not. "Point Horror" and "Point SF" and "Point Fantasy" are imprints I can support separating out. Ballantine seems to have many that can be merged usefully, but not all. "Ace Books" and "Ace" I'm not sure of the value of separating, but "Ace (5th printing)" and such-like can be merged I'm sure. I think that is now considered a failed experiment? BLongley 22:32, 19 April 2008 (UTC)

I've tended to enter publications as stated and know that publishers are not consistent. I've personally found the "Ace (5th printing)" to be quite useful. Just yesterday I wondered why there were what looked like two identical pubs (same ISBN/price) and clicking revealed one was an 11th printing and the other 12th. Adding the printing # to the db is on the queue and once that's in then yes, the publisher name "Ace (5th printing)" will go away. Marc Kupper (talk) 07:04, 21 April 2008 (UTC)

Are the plans to just add the field or use it for sorting/grouping? I can provide examples of the headaches that the British "Numbering continues across Imprints/Publishers" causes if you want to dive in the deep end, but just recording and displaying it is a good start. BLongley 19:42, 22 April 2008 (UTC)

As it is, the single largest group is the 4930 publisher names with only one publication each. Some like "Macmillan, an imprint of Macmillan Publishers Ltd" and "Tor Books, an imprint of Tom Doherty Associates" seem easy to clean up but apparently whoever entered those thought it was important to note exactly what a publication says or perhaps that's from Amazon. Do you overwrite their work? It's a lot of grunt work to inspect each one to see if it's just a typo that could be corrected or was an attempt to accurately reflect what's stated in a publication record and if this information should be preserved in the notes or some other method. Marc Kupper (talk) 07:41, 19 April 2008 (UTC)

At the moment, my Publisher edits are mostly reshuffling information from Publisher to notes or vice versa. Or just rearranging the order of imprints or Publisher/imprint in the Publisher field. I've questioned some editors that add the "imprint of..." bits but no response so far. I'm definitely conscious of destroying data, and am avoiding that as best as possible (I hate having to record stuff in notes that I think SHOULD be in the database though). BLongley 22:32, 19 April 2008 (UTC)

What order are you using? imprint/publisher or publisher/imprint? The "imprint of..." thing certainly is clear but I agree it looks messy and can result in the publisher name getting truncated. Marc Kupper (talk) 07:04, 21 April 2008 (UTC)

I generally use whichever is most common for now if I'm not working on primary copies. The spacing around the slash can be regularized via scripts if Al is willing to run such, as can swapping publisher/imprint around: it's only on the smaller numbers of variations that I'm willing to regularise manually. However, for the British paperbacks I'm currently looking at the value of having publisher on each pub is often questionable - the imprint alone is valuable, the publisher can be recorded broadly over all pubs for that imprint for several years, and the level of publisher to record is debatable - e.g. Triad books were published by a consortium of three other publishers and the useful data seems to be which TWO imprints are on the pub. BLongley 19:42, 22 April 2008 (UTC)

Oh, and I DO override Amazon Data when I can see it's from there and no human has apparently added anything. If anyone here edits stuff and leaves it looking as if they were Amazon or Dissembler, please speak up. BLongley 22:32, 19 April 2008 (UTC)

When I add a publication record where the source is not the publication itself I always add notes explaining the source(s) for the various fields and also date it with the month-year. If I override something I often leave a note saying the record used to have "x" but the source is unknown (or seems to be Y). For something like the page count or a binding of tp from Amazon I won't bother with noting the override. My thinking that much of the data, even on Amazon, has a valid basis. For example, I held and researched a submission yesterday where someone wanted to change the artist and it turned out the Canadian and USA editions had different covers. Marc Kupper (talk) 07:04, 21 April 2008 (UTC)

The amount I'm willing to do to a publication without adding notes varies according to the reason I'm adding/modifying it, my perceived value of existing data, and my perceived reliability of the source for the changes: e.g. I've added a lot of OCLC data to the Andre Norton submissions made from the 1975 bibliography, and it's no trouble to paste the OCLC record number into those edits which could be useful if people want to go back and add other editions. The dozens of British Horror anthologies I added over the weekend (and am regretting at times due to the sheer boredom of the hundreds of merges they cause) came from specialist Fan sites, cross-checked with Amazon and one or two bookseller sites and sometimes OCLC, with me doing the regularization and over-riding obvious typos - listing all the sources used for each pub would be more work than I'd be willing to take on, and I wouldn't want to influence a verifier by making it look authoritative anyway. The sites proven useful are being recorded at Author, Publisher and Series level as appropriate though, and I'm going back and fixing the debatable parts as I do the merges - e.g. trying to avoid 27 more J. S. Le Fanu variants by sticking with proven ones here already rather than letting every single variation from a fan get through. I'll have to accept that some things are beyond non-primary entry though, series books in particular are notoriously badly classified at Amazon and absent from OCLC. But at least we HAVE some data to work with now - seeing that a series was missing volumes 6-16, easily found elsewhere, started it. Then discovering that a series LOOKED complete but actually didn't have more than a title prompted a bit more. Then spotting where the series was taken over by another editor prompted a bit more. And discovering that some of the authors in the anthologies have been under-represented here (if they even existed at all before) led me to more sites... I'll have to knock that on the head for a bit though, I'm getting very bored with merges, having the submission queue get that long seems to slow the site down (hopefully due to them throttling MY use of the site alone though - or was it slow for everyone when I had 30-40 edits waiting?) and work-pressures mean I might actually stick to JUST "books received" for a bit - some publisher work has led to me purchasing items specifically for research work on publishers. I'm glad I don't have a wife to justify my spending habits to... BLongley 19:42, 22 April 2008 (UTC)

OK, a bit of a waffle there again - sorry! I don't actually LIKE British Horror Fiction reprints much, and am frankly sick of the sight of "Green Tea" yet again. I can't promise that I won't fill in obvious gaps in our coverage when I find a good source, but I really want to concentrate not only on British publishers for a bit, but on Science Fiction in particular. Not Fantasy - not Ghost stories - not Horror stories. The downside of course is that when researching those I BUY books, whereas dealing with Horror let me leave my credit-card unused for a while. BLongley 21:46, 22 April 2008 (UTC)

Doc Savage and house name and numbering

I'm working on the Doc Savage series. Just entered all of mine so I need to start with secondary sources. There are already some of the blackmask.com reprints listed. They have some entered as by Robeson and Dent. Dent was the main writer that used the Robeson house name so having him as a co-author seems wrong. One of the ones is 18 Murder Mirage And Mystery Under The Sea. According to The Doc Savage Fact Page pulp list Murder Mirage is by Laurence Donovan. I think I should change the omnibus to be just Robeson, and then make pseudos for the novels for Donovan and Dent.

Sounds a reasonable plan! Ahasuerus 00:01, 24 April 2008 (UTC)

With 190 books in the series it seems like a totally unreasonable plan. But I've never claimed that my bookaholism is reasonable. At least I'm resisting the urge to fill in my collection with the 140 I'm missing. Dana Carson 07:28, 24 April 2008 (UTC)

Will ask on the verification page if anyone has copies of the blackmask reprints to see what the copyright page actually has.

The number of the series is Bantam numbering. The Doc Savage Magazine number is different. Should I switch to that or leave it? The Bantam series did add new books at the end so it has more than the pulp series did. Dana Carson 09:31, 23 April 2008 (UTC)

Unfortunately, series numbering is not a precise science. Going as far back as the Lensman saga, there has always been tension between the "original publication order", the "revised publication order", the internal chronological order, etc etc. And then there is Moorcock :( We generally try to use publication order, but we will adjust things if there is a good reason to. In this case I would just go with what seems to make the most sense and add a note to the Wiki page associated with the series. Ahasuerus 00:01, 24 April 2008 (UTC)

A simple example would be Narnia - same books, different order. A more complex (as in "Why, Oh Why did I start doing this, Oh my God, it'll never end...") might be the Star Trek Pocket Books. There, it seemed wisest to use the longest series as the canonical numbering to avoid unnumbered extras as best as possible. And there's a few we've left as overlapping such as The Invaders and The Girl from U.N.C.L.E. as the variants are complex enough already or there's little overlap or there are so few it's not that difficult to list both orders. But for BIG series, I would almost always go for the categorization with the longest run as the main one, and reorder the shorter series in the Wiki page. BLongley 00:36, 24 April 2008 (UTC)

Fantasticfiction has the Bantam covers and the pulp numbers. It was a little confusing when the number on the cover image didn't match the number in the caption. Dana Carson 07:28, 24 April 2008 (UTC)

One other thought is that as one version is the Magazine series and so presumably one publication per title, a manual Wiki Publication Series like this might work for that side and use a 'proper' ISFDB series for the other? BLongley 17:53, 24 April 2008 (UTC)

European graphic novels

Are European SciFi and Fantasy graphic novels (e.g. Jean Leloup's Yoko Tsuno or Didier Crisse's Kookaburra) considered to be covered by the database?

According to Definitions in the Rules of Acquisition, comics and graphic novels are excluded. In practice it's a bit different: graphic novel versions of text-only books seem to be mostly left alone as other editions, and certain Big-Name authors like Neil_Gaiman and Alan_Moore seem to have some of their Graphic-only pubs included for completeness. I for one rarely go to the trouble of deleting SF Graphic Novels and Manga when I find they've sneaked in, but I do occasionally have a clean-up of Non-SF ones. I certainly wouldn't recommend adding new ones at the moment though, other editors are a bit more zealous. BLongley 12:43, 3 May 2008 (UTC)

Alright. I'll probably just clean-up the Bernard_Werber stuff (on Graphic Novel version of Empire of the Ants and Exit, a three-volume series he wrote) then. Too bad though. There's a lot of great stuff out there. Circeus 17:05, 3 May 2008 (UTC)

I'm also not sure exactly where the line is drawn on "graphic novels", though. it seems to me that soemthign like the original version of Niven's Patchwork Girl, with one full-page illo to every 5 or 6 pages of text, should be in, but that was marketed as a "graphic novel", IIRC. -DES ^Talk 00:18, 4 May 2008 (UTC)

Eric and The Last Hero have art on every page and I'd count those in. I guess I'd draw the line at the point where there was no separate text, or where a lot of the sense was in "speech bubbles" or something. BLongley 09:48, 4 May 2008 (UTC)

How about the 2000 graphic novel version of Guards! Guards!? The German translation of it is (oddly enough) there ("Ein Scheibenwelt-Comic" = "A Discworld comic"). Circeus 01:57, 6 May 2008 (UTC)

"Graphic novels" are a rapidly evolving area, so I suspect we'll keep bumping into examples that we never anticipated. The original intent was to exclude comic books since they are really a different breed and not what fiction readers expect when they are looking for "fiction". Besides, we would need to add a number of new fields to support "pencillers", "letterers", "inkers", "colorists", etc. Manga seemed to be pretty close to comics, so it got excluded as well, but now there are illustrated "light novels" and other borderline phenomena that will need to be addressed. Who said the life of a genre bibliographer would be easy?! Ahasuerus 04:38, 6 May 2008 (UTC)

Graphic novel-oriented databases (such as Bédéthèque) and European reference books only use 2 or three of those as relevant (writer, artist, and sometime colorist), so it's not that far from the current cover artist/writer situation. This is much the same that goes in Japanese manga studios: there are often several secondary artists doing minor inking stuff (sometimes even the editor has to do error correction on the art in Japan), but only the primary artist is typically credited. Circeus 01:35, 8 May 2008 (UTC)

subsequent multi-volume "serialisation"

How does one deal with the case where a book gets a subsequent multi-volume publication? Stephen King's It was republished by J'ai Lu in a three-volumes paperback set (coll#s 2892-4, boxed set #6904), as was Asimov's The Robots of Dawn published as two volumes (#1602-3). Denoël republished Asimov's Mysteries (or at least disordered parts of it) in two volumes (#113-4) etc. This is fairly common for these publishers because they stick to a fairly rigid bracket of page numbers. Circeus 22:36, 3 May 2008 (UTC)

I would be inclined to enter those as "It (vol 1 of 3)", "It (vol 2 of 3)", etc. But perhaps there is a better phrasing. In general abridged, revised, expanded, etc works are handled by adding a phrase in parentheses, so this seems like it might follow the same pattern. Other views? -DES ^Talk 00:15, 4 May 2008 (UTC)

I just found out the situation is even more messed up: the original Albin Michel French publications of It and The Stand were in two volumes, as were the republications by LGF. Re-publications by J'ai Lu, however, were in three volumes, eventually collected in boxed sets (I own the It set, a friend owns the Stand one). How the *** am I supposed to deal with that? Circeus 03:35, 6 May 2008 (UTC)

It might be best to wait a few days before entering any of the titles. I'll give it some thought and if no one else has a solution than we can enter one of the two volume sets and the reprinted three volume set and I'll play around with them.Kraang 03:54, 6 May 2008 (UTC)

Unfortunately, our support for Big Fat Books (BFBs), which later get split into 2+ volumes, has always been weak regardless of whether there is more than one language involved :-( The only discussed approach that could conceivably work within the current framework would be to retroactively convert the original BFBs into Omnibuses once they appear as 2+ volumes. We could then create a Series, which would include the original BFB and its offspring. Unfortunately, the cure is probably worse than the disease in that it changes legitimate novels into omnibuses, does it retroactively and unexpectedly, and, besides, doesn't help when the same book has been broken down in more than one way -- see Circeus' example above.

For now, we have been mostly stuffing these multi-volume reprints as Publications under the parent Title -- see Brian Jacques' Martin the Warrior, which includes 2 volumes in German and 2 volumes (out of 3) in French. If you can think of a better system -- short of rewriting the application -- please speak up, because the current one is admittedly suboptimal! Ahasuerus 04:54, 6 May 2008 (UTC)

Or there is the non-solution used for Cyteen where the parts are simply recorded as separae novels, with no clear indication of the relationship. i don't advocate this. -DES ^Talk 05:01, 6 May 2008 (UTC)

Cyteen looks particularly horrible, with FOUR Number 1s in the Series, not even showing in the right order. Even the notes are misleading it seems: "split into three sections for its inital [sic] publication" doesn't match our data. BLongley 19:34, 6 May 2008 (UTC)

The problem is that there are really two novels in that series, it just happens that one of them exists in three sets of covers for soem cases (was it for the inital MMPB pub?) I know that CJC was reported to have sworn that it would never be reprinted except between a single set of covers. I own copies of the split version, BTW. -DES ^Talk 20:50, 6 May 2008 (UTC)

On the other hand, it is pretty well arguable that thsoe two novels dont really form a series, anyway. They are in the same settign, occur at about the same time in internal chronology, and there are about 3 paragraphs in Cyteen mentioning the events of 40,00 in Gehanna. There are no common characters, and the events and thems also incluence other works set in that universe. -DES ^Talk 20:50, 6 May 2008 (UTC)

Indeed I wouldn't know how else to proceed. *shakes head*. And I'm not intending to add these particular publications for the time being. I intend to first get through Werber's stuff, than my French Asimov, Pratchett, Sheckley and Silverberg stuff. Circeus 05:05, 6 May 2008 (UTC)

just fictionally speculating . . .

Just to satisfy my overgrown curiosity, does Reading Oprah: How Oprah's Book Club Changed the Way America Reads really have a place in ISFDB?

I don't think so, and i have submitted a delete publication edit. I reveiwd the description provided by amazon, and this is a non-fiction book that is not SF, not about SF, not by an author noted for SF, nor about such an author. It is relevant only in the sense that SF books are also books, and any book about how readers read is in some slight degree relevant. -DES ^Talk 21:35, 5 May 2008 (UTC)

Although the Rules of Acquisition are still questionable, I have to say that DES has got it right. The only way I'd include it is if we had a REVIEW of it that the person providing the review was unwilling to change to "essay". And only then if we made it clear that it's only here because of an SF pub review, and we should not encourage further entries by the same author, same publisher, etc. We can accommodate SOME non-fiction - for instance, bibliographies and SF encyclopedias are handy to know about - and some science works are here as sources sited for SF works. But we do need to draw the line somewhere. BLongley 23:31, 5 May 2008 (UTC)

I am willing to include critical studies of SF. For example, IMO Heinlein in Dimension belongs in, and so do at least the more well-known book-lenght studies of Tolkien. But this isn't such a study, either. -DES ^Talk 23:44, 5 May 2008 (UTC)

I approved the deletion, maybe some other data base(recycle bin) will find it a home. I wonder how may forests were chopped down to produce it?Kraang 00:45, 6 May 2008 (UTC)

I thought it was a subtle dystopian alternate history. Alvonruff 17:29, 6 May 2008 (UTC)

I think it is just straight-up fantasy. Fantasy becomes reality when enough people believe in it.--swfritter 18:41, 6 May 2008 (UTC)

Magazines reprinted as Anthologies

Take a look at this reprint of Captain Future, Summer 1942, which is currently listed as an anthology edited by Edmond Hamilton. It will be easy to change the editorship to Oscar J. Friend, but the underlying problem is that this is really a reprint magazine in disguise. And, as we know from our past struggles with trans-Atlantic reprints of various magazines, we don't support multiple versions of the same magazine issue particularly well. Any ideas? Leave it as an anthology just like we did with this 1981 reprint of Astounding Science Fiction, July 1939 ? Ahasuerus 05:19, 7 May 2008 (UTC)

I believe there is also a PDF version Amazing Stories v1n1 and physical reprints of others. Anthology makes most sense to me for right now although it would be nice if we could somehow document the existence of the reprint pub on the mag wiki - perhaps a special section?--swfritter 20:09, 7 May 2008 (UTC)

Sure, we could add notes to each affected magazine's Wiki page. We could also update the Note fields in the Magazine pub and the Anthology pub to point to each other. Ahasuerus 23:16, 7 May 2008 (UTC)

Publisher names again

With the advent of publisher editing, and the plans for publisher merging in future, i think we need to resume our discussion of how to handle publisher names, and this time bring it to at least a partial conclusion. I propose the following steps:

1. We agree on standard forms for which I might call appendages to publisher names: for example should we use "& Co", "and Co." "and Company" or just leave the company off altogether. Should it be "X Books" or just "X". And so on. For these we ought to be able to come up with standard answers that can be applied to most if not all publishers.
2. Work through publisher by publisher.
   1. For each, those who know most about the publisher will help determine what info is significant. For example does the imprint matter? Does it matter when the publisher became a division of MegaCorp, so that we record "Jones Books" and "Jones Books, div of MegaCorp"? Does it matter when "Smith & Son" became "Smith Books, inc"? etc.
   2. Based on this, we devise a set of canonical names for the publisher and whatever lines or imprints or other variations seem worth recording. (the long list of publishers now in the DB will help with this)
3. Then we consider which existing forms can be safely converted en masse to one of the agreed canonical forms (using the publisher edit tool) and which will require an editor to do individual research before making changes.
4. Change en masse the forms that can be safely changed (and merge them once this is available)
5. Start working through the list of remaining forms doing required research on individual publications.
6. Meanwhile, once a canonical name or set of names is agreed to for a given publisher, new data should be expected to use only one of the canonical names.

Does this strike people as a worthwhile program? -DES ^Talk 20:52, 7 May 2008 (UTC)

I like the way Locus lists each publisher's books every year -- see, e.g., their list of all books published by Gryphon Books in 1999 -- and I believe that implementing a similar approach within the ISFDB framework would be useful as long as we can implement it well.

A number of our editors have already done a lot of work in this area and discussed it extensively, but there are two areas that I haven't seen addressed (perhaps I just missed the discussions) and I am not sure how we can handle them.

The first area has to do with ensuring, as David wrote above, that "new data should be expected to use only one of the canonical names". How would the average editor know the canonical name(s) that he should enter for each edition? In most cases we follow a simple "what you see is what you enter" philosophy when dealing with Publications and using canonical publisher names instead seems to be a significant departure.

The second area has to do with merging publisher records. I am sure that 90%+ of publisher merges will deal with things like "Dell" vs. "Dell Books" and other reasonably simple issues, but there will also be more complicated and/or obscure areas where a significant amount of preliminary research will be required, e.g. Australian imprints of similarly sounding UK/US companies and such. How will we ensure that we always contact our expert(s) in these fields before we accidentally mess up months of their careful work? For example, I happen to know that User:Clarkmci has massaged hundreds of Australian publications to get the publisher just right, but he rarely participates in Wiki-based policy discussions and I am not sure whether he knows about these developments. We may be able to drag him here, but what about other specialist editors who have been quietly working in their chosen niches? Ahasuerus 00:35, 8 May 2008 (UTC)

As to your first concern, it is going to have to be a matter of informign individual editors. There will ned to be a page or pages (probably in the wiki) with lists of cannonical publisher names, and thsoe lists will need to be linked to from the various editing help pages. Then the moderators will need to check and inform people when non-cannonical names are used, just as when prices are given without currency symbols, or dates not in yy--mm-dd form. Ideally, there would eventually be a warning when submit is clciked "Publisher name is non-cannonical, do you wish to submit anyway?" But that will have to wait for both programmer time and a moderatly stable set of cannonical names.

(Actually i would love a confirm display after submit, like the "preview" button on wiki edits, anyway, at elast as an option.)

As to merging of publishers, in any case where thinks aren't obviously harmless (merging "Baen Books" with "Baen" or "TOR" with "Tor", say) I would expect the merge to be proposed on the wiki and wait some days for comment. Then if anyoen knows about the xpecialist editor and his work, the comment 'Contact Joe, he knows this area well" can and should be made. Not perfect, but I see no better way.

And remember this is a fairly small community. Currently there are only 50 editors who have more than 50 edits. we can find ways to inform people if we must. And not all will get the word, but many/most will. -DES ^Talk 00:54, 8 May 2008 (UTC)

At a minimum merging should be a two step process, change name and then merge, this way if a name gets accidentally change to an existing one their not automatically merged. Changing of names at the publication level can be left as it is to general editors, but changing of names to all publications and merging should be moderators only, with no access to this function for non moderators.Kraang 01:00, 8 May 2008 (UTC)

Uncredited vs. Anonymous vs. N/A plus Reviews

As per Help:Screen:NewPub:

• Anonymous or uncredited works. If a work is credited to "Anonymous", then put "Anonymous" in the author field. The same applies for any obviously similar pseudonym, such as "Noname". If the work is not credited at all, use "uncredited", with a lower case "u". This applies to editorship of anthologies that are not credited. Omnibuses, including dos-a-dos publications such as Ace Doubles, should be given an author of "N/A".

At this point we have a grand total of 2 Essays attributed to "N/A". Is it fair to say that the last sentence quoted above is out of synch with our prevailing practices and we may want to change it?

Also, although we generally use "uncredited" when the anthology editor is not known -- as per the Help recommendation above -- a number of magazine records use "Anonymous" when reviewing anonymously edited anthologies, e.g. see the review section of Science-Fiction Plus, June 1953. Is that by design or do we wan to change all reviews to use "uncredited"? Ahasuerus 00:51, 8 May 2008 (UTC)

Collections in Omnibuses

This section has been copied (with deletions) from the thread at ISFDB:Community Portal#Collections in Omnibuses, because this is really the relevant location, IMO -DES ^Talk 21:17, 12 May 2008 (UTC).

I was under the impression that for an omnibus that included both one or more novels and a collection we typically did not list the individual shortfiction works in the collection, particularly not if the collection had previously been published separately and has that publication listed with contents. But I am not sure if we have ever really codified this policy one way or the other. (if an omnibus is a collection of collections I suspect we would just treat it as a different, larger, collection). -DES ^Talk 22:42, 10 May 2008 (UTC)

I was thinking about this the other day. I see two possible problems with using Collection Titles -- as opposed to using their constituent stories -- in Omnibuses. First, if you are looking at an Omnibus page and it contains a Collection, you need to click 2 times to get to the Collection's contents, once on the Collection's Title link and once on a Collection Publication link. Worse, once you click on the Collection Title link, you are then faced with anywhere from 0 to dozens of Publication records. And, of course, each Collection Publication may have somewhat different contents -- essays, forewords, afterwords, etc -- so how do you know what exactly the Omnibus contains? Second, if you are looking at the Title page of a story published in a Collection that later appeared in an Omnibus, you will not see that Omnibus publication, which defeats the purpose of listing all Publications for that story. It seems safer by far to list the constituent stories and avoid these problems. Ahasuerus 23:12, 10 May 2008 (UTC)

I see. A problen is that there is no concept of hierarchy to the contents of a publication, including an omnibus. If one could list both the collection title and its contents in such a way as to make clear what belongs to the collection, it would be much better. There are other cases where this would be helepful, too. For example it is not uncommon for a collection to contain groups of stories with a common group title, as well as individual titles. "X: Three tales" consistign of "A", "B", and "C". without a "group title" element, one is tempted to reder these as "X:A", "X:B", and "X:C". Sometimes this can be done with a series, as in Distant Friends, but this doesnt always work well. In any case, we have often entered omnibus pubs that contain collections without the short fiction -- Indeed I rather thought it a rule that no title in an omnibus could be of type short fiction. For example, see Mountain Magic which contains the collection Old Nathan or 3xT which contains two collections and a novel, but lists no shortfiction (and if it did, how would the user know which short works were part of which collection?) or Annals of the Time Patrol which contians two collections, and many in simialr state. If we decide that the individual works of short fiction should be enterd, i will do that. Help:Screen:NewPub is not very helpful on this point at present. -DES ^Talk 00:05, 11 May 2008 (UTC)

I think we have to enter the short-fiction. :-( More work, I know. Reasons: 1) We'll never be sure WHICH version of the collection is included, so which variant titles etc are in it. Or even contents: some collections are different in different countries due to having to split them up into separate volumes, or because one or two stories aren't free to be published in that country. (I have MANY books that say "For copyright reasons, this book is not available in the USA" - although Canada normally gets permitted.) 2) People use the pagination for Length-determination at times - personally, I'm not that bothered, but you can't get it from Magazine versions. E.g Story X may be entered as starting on page 22, another story starts on page 25, so it's 3 pages long and therefore Short Story? Not if it's "Continued on pages 125-148" - which those lazy Magazine editors never mention. ;-) It's rare (but not unknown) that a Book publication splits contents that way. ("Destinies" comes to mind though, as a Magazine in paperback format that we've had to deal with as "Anthology" - not sure if it ever did such magazine content-splitting though.) I think Ace Doubles of a pair of Collections have been the worst to deal with so far - workarounds such as Roman Numerals for one side and Arabic for the other have been suggested, but as usual never made official. BLongley 22:58, 11 May 2008 (UTC)

I see your point, and perhaps it is the best way, but it seems to me that at the least the collection titles ought to be entered in addition to the shortfiction titles. If full page numbers are entered, this will make it at least fairly clear what stories belong to what collection title. But note, lots of omnibuses are not entered in this way, even ones that are verified, and if this is to be the standard, the help should be changed. I am going to raise this on the rules and standards page, I think. -DES ^Talk 15:33, 12 May 2008 (UTC)

I'd favor entering both the collection & its contents. A problem with doing this, which sometimes shows up in other places (book review columns, say) is that we have no way of forcing two titles with the same page number to display in a particular order. But that's relatively minor.
The ideal solution would be a whole slew of programming changes, with some database changes involved, which would always expand collections in searches & displays. I kind of shudder at the changes needed, though. -- Dave (davecat) 16:47, 12 May 2008 (UTC)

I agree about the ideal, but we need a rule for the present.

1. One way is to go forward as at least some editors have been working, entering the collection title but not its contents in any omnibus publications. This makes the pub displays shorter and simpler, but has all the problems Ahasuerus and BLongley warned of above.
2. The second is to clearly adopt the standard that when an element of an omnibus is a collection (or, I suppose, an anthology) both the overall collection title and the individual shortfiction (or essay or whatever) contents should be entered in the contents list of the omnibus. If we agree on this, the help should be changed to document this as the standard, and we should be aware that many existing pubs, including verified ones, will not meet the standard.
3. The other option, I suppose, would be to enter the individual contents of the collection without its overall title. IMO this is the least acceptable option, but maybe other editors will favor it.

Well, what do others thing on this issue? -DES ^Talk 21:17, 12 May 2008 (UTC)

I've been practicing (2) on my own publications for some time. Yes, we do get caught out when a Novel becomes a Collection, and much rework ensues and more data (pagination and exact titles) is required: it's still quite a bit of work to go the other way but at least you can do that without help (and not completing the task doesn't seem to cause problems anyway, leaving the constituent short-fiction in a Novel doesn't throw up any significant warnings in the database, although some clean-up projects would flag it.) This is not a common problem though, it's less than once a month in my experience. BLongley 22:17, 12 May 2008 (UTC)

There's probably a few pubs still where someone has verified "yes, this is an omnibus containing that collection", and that was possibly OK at the time or we didn't make the current standards clear - many of us new editors verified happily, it was easy when we started, and the questions about what verification actually MEANT came later. (I'm still not entirely sure what it means, if it doesn't "fix" the publication in that state then any edit or outside change like Amazon changing the default coverart-image should probably unverify it "a bit". But that's another discussion, that we've had many times already, and probably needs new blood before it can be resolved.) BLongley 22:17, 12 May 2008 (UTC)

Anyway, my vote is for (2). If someone can find an Anthology example it might help people decide, but I can't think of one. I know I've fixed at least one (1) example recently but can't recall that either. (3) looks really undesirable. BLongley 22:17, 12 May 2008 (UTC)

I verified Mountain Magic quite recently, following the rules I thought applied. if we go with (2), I can easily add the additional contents. I will ned to think of an example of an omnibus containing an anthology, if I can. (2) would be very acceptable to me, does everyone else here prefer it? -DES ^Talk 23:00, 12 May 2008 (UTC)

(2) seems to be the intuitive choice, so I am inclined to go with it as well. If we find problems with this approach, we can always revisit the issue. Ahasuerus 01:11, 13 May 2008 (UTC)

I vote for (2). (But still the problem what is an omnibus remains, e.g.: Isaac Asimov's Science Fiction Treasury.)--Roglo 08:27, 13 May 2008 (UTC)

I would call that an omnibus, and if we are going to follow (2) converting it to one would be trivial. -DES ^Talk 14:11, 13 May 2008 (UTC)

My vote would be for (2), it's what I've been doing when I catalog my books that fit this case on Librarything.CoachPaul 01:20, 14 May 2008 (UTC)

Given that in a week no one has supported any other alternative, and several experienced editors have supported (2) above, I will taken this as consensus, and edit the help page appropriately. -DES ^Talk 17:01, 19 May 2008 (UTC)

I have made a change to Template:TitleFields:Title to record this decision. Please see if this is acceptable to all. -DES ^Talk 18:24, 19 May 2008 (UTC)

Looks fine to me. Of course, the problems caused by page-numbers restarting for each component will still be an issue, but that can be left for another discussion. BLongley 18:59, 19 May 2008 (UTC)

Merging Audio Dramatizations with Stories?

Someone else may have a better idea but this, I expect, might be an audio collection of dramatizations of stories rather than actual readings of the stories. Matheson's "Death Ship", for one, has been merged with print versions.--swfritter 16:39, 19 May 2008 (UTC)

The alibris description says "Adapted for audio from original scripts written for the television show created by Rod Serling that aired in the 1960s, these radio dramas feature music and sound effects and a distinguished lineup of guest stars". Based on that, i don't think this belongs here at all, any more than DVDs of TZ episodes based on written SF stories. -DES ^Talk 16:59, 19 May 2008 (UTC)

To add to what DES says if we allow these, then we'll need to add old X-1 and Dimension X episodes which used stories from the 40's and 50's to the db too. I do think that we can show some of this information though, by adding a Tag to these stories showing that they were TZRD episodes as I have done with Dimension X stories here.

I disagree that we'd NEED to add such. At the moment it's a case of "these sneaked in, they've been organized a bit as deleting is a multi-step pain". There's no obligation to ADD anything - e.g. I cleaned up Daisy Meadows a bit over the weekend, but felt no need to go add the information that "Sunny The Yellow Fairy" is actually a variant of "Saffron The Yellow Fairy". This is the slippery slope started when audiobooks were permitted - e.g. we have Doctor Who Audiobooks with no paper editions known - yet. At the moment Video seems right out still, but I know I've listened to "Radio Dramatizations" of short stories here that were single-reader performances, and others that were multi-reader performances, both close to the original text that I recalled. (BBC 7, for those that can receive it - some are rather good.) But demanding a dead-tree version isn't going to help here if there WAS one initially: can we agree a "two-steps removed is too much" rule? Or that Music and Sound Effects rule some audio out? Or do we just deal with them if they arrive and don't encourage more? BLongley 18:55, 19 May 2008 (UTC)

I'm not so sure that we can make generalizations here, and probably for now should go with a case-by-case basis. For instance, these TZRD cd's are filled with radio plays formatted to fit into a "radio hour", in much the same way as old time radio shows were. "The Castle in the Attic" has an audio version, that is complete with sound effects, and each part is voiced by a different voice actor, but the reading is exactly that, a reading of the book, and as such is another edition of the novel and should be here. (I believe the Redwall audiobooks are the same.) Therefore if we say that it shouldn't be here because it has multiple readers, and sound effects it shouldn't be here, we are missing out on editions that IMHO should be here. My thoughts are that if dialog from the story is changed and story elements are either left out or paraphrased, then it shouldn't be here. If it a reading of the story, no matter how many voices and special effects are used, then it should. Partially, if you hear someone utter the words "He said", (feel free to change "he" with any pronoun or proper noun that may fit,) then it is most likely a reading and not an audio play and should be included.CoachPaul 22:08, 19 May 2008 (UTC)

Maybe I should organize my thoughts better, I started the above by saying we shouldn't make generalizations...and ended with one.CoachPaul 22:10, 19 May 2008 (UTC)

I lean towards general acceptance of these (but not encouragement to enter more) when they are purported to be the same story. Otherwise we might have to delete all the "Abridged" audiobook CDs, cassettes, etc. If it's retitled ONLY for an audio publication then I think we can leave it out, they're not important enough to us to create variant titles. I'd discourage inclusion of non-SF material in the same work though - I'm happy with partial-contents entries for fringe publications: e.g. many SF works were published in non-genre magazines, and it's nice to have a reasoned entry for the first publication: we can cope with a "Saturday Evening Post" or "Collier's Magazine" or "Strand Magazine" with SF content only listed. We have "Playboy" first publications for some stories too, but I've never seen one verified yet... perhaps editors get too distracted? ;-) BLongley 22:33, 19 May 2008 (UTC)

Treating them as variant titles and adding "(Audio Dramatization)" to the variant title might be an idea. Abridged stories are sometimes treated the same way. These are actually entries that I find valuable. Movies and TV shows are well documented in The IMDb but this is data that is not as easy to find. I guess we would have to categorize them as shortfiction. If there is no acceptable option for including these then we should remove them. Keeping them is likely to result in setting a standard of inclusion.--swfritter 00:10, 20 May 2008 (UTC)

I would be inclined to include them as well for all the reasons listed above. As far as verifying old Playboys goes, some of them are collectible items, but not necessarily the kinds of collectibles that SF collectors would be interested in, which is somewhat unfortunate since Playboy has published a number of stories by big name SF writers over the years. After all, they had more money to throw around... Ahasuerus 02:25, 20 May 2008 (UTC)