Rules and standards discussions/Archive/Archive10

This is an archive page for the Rules and standards discussions page. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from April - September 2012‎.

Archive Quick Links
Archives of old Rules and standards discussions.

1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21

Expanded archive listing

Graphic Novels in 2012

Should "GRAPHIC NOVEL" be added to Pub Type? It's not always clear a publication is a graphic novel when only NOVEL is shown, i.e. "V for Vendetta" or "Watchmen".--Teddybear 13:36, 12 April 2012 (UTC)

I'd say no, it would encourage people to enter them. BLongley 14:57, 12 April 2012 (UTC)

With rare exceptions (most notably, the two you mention), graphic novels are not accepted into the database. See the definitions and inclusion policy. Mhhutchins 17:08, 12 April 2012 (UTC)

It would appear that there are two separate issues here. The first one is whether graphic novels are cataloged by ISFDB. As Michael said, with rare exceptions, they are not.

The second question is how to display them in those rare cases when they are included. Currently they are listed as NOVELs and you can't tell that they are really graphic novels until you examine each Title record. I think it would be beneficial to have a separate title type so that users could easily tell graphic novels from regular ones, especially when a bona fide novel is later turned into a graphic novel. Alternatively, we could add a "yes/no" field that could then be used by the Summary Bibliography page to display these titles appropriately. The latter approach would also allow us to handle "graphic collections" and "graphic anthologies" should the need arise.

That said, it's not a high priority at the moment. However, graphic novels are becoming more popular and, if the trend continues -- many recent bestsellers got a "graphic novel" treatment shortly after it became clear that the book is popular -- we may want to consider adding a new data element. It could be as simple as an additional checkbox on the New Pub and Edit Title pages. Ahasuerus 01:51, 13 April 2012 (UTC)

Wouldn't that require changing the current policy first? Mhhutchins 04:12, 13 April 2012 (UTC)

Well, if we are to follow the stated policy verbatim, then we should simply delete all graphic novels and the problem will go away. That was my very first project in May 2006 when I went on a "search and destroy" mission to delete all comics, RPGs and graphic novels. However, when we last discussed this issue a few years ago, some editors (including, I believe, Swfritter) felt rather strongly that certain graphic novels, e.g. Hugo winners, should be "in". No decision was made, but the existing "rare cases" were left alone until we could decide what to do with them.

Over the last 6 years my opinion of the merits of including graphic novels in the database has shifted somewhat. There are quite a few graphic novels related to bestselling SF books -- "Erin Hunter", George R. R. Martin, most popular urban fantasy series, etc -- and they can be hard to distinguish from "real" novels, so listing them may help clarify things for our users. The primary use case that I envision is "How come I have never heard about this Dresden Files book? Let's see what ISFDB has to say about it. Oh, it's a graphic novel!" It would be similar to a user wondering why he has never heard of this book by John Barnes and discovering that it is a non-genre novel. In my mind it's a disambiguation thing more than anything else.

For this reason I would be willing to support a Policy change to include graphic novels related to speculative fiction (and a software change to enable it), although I don't feel strongly about it. What I do find unfortunate is the current inconsistent situation which results in listing graphic novels as "novels", e.g. see the Neil Gaiman page. I believe that the important thing is to be consistent, which in this case means either evicting the riffraff or changing the policy and the software to let a well-defined subset in and display it unambiguously. Ahasuerus 02:13, 14 April 2012 (UTC)

I've always felt that the ISFDB policy regarding graphic novels was narrow-minded, even provincial. But like most editors I feared the slippery slope that would allow comic books into the db. The line between them is too blurry for my satisfaction. If the form could be clearly and unequivocally defined, I could support a change in policy to include graphic novels. Perhaps this should be moved to the Rules and Standards Discussion page. Mhhutchins 03:32, 14 April 2012 (UTC)

(unindent) Good idea, moving to Rules and Standards. Ahasuerus 06:40, 14 April 2012 (UTC)

I too could support a change in policy. I'm not sure whether "above a certain threshold" or "award winner" is clear enough. I would definitely not want to see every single issue of a mini-series entered, as the above examples could have, but I can support notable graphic collections and definitely graphic novel versions of 'proper' novels. However, I think we should leave them as one title record rather than have every printing, and rather than record Writer, Penciller, Inker, Colorist, Letterer, Editor and Cover Artist, just link to a better-suited resource like ComicBookDB.com. BLongley 11:40, 14 April 2012 (UTC)

I'm afraid it will mean the opening of the Pandora box. The word "notable" is too much tricky for me to favor such a change of policy even if (or because) I'm also a "bandes dessinées" (= graphic novel) collector. Perhaps can we limit inclusion to works where one of the authors (or artist) is already present in the ISFDB, allowing books like this one. Hauck 14:25, 14 April 2012 (UTC)

I favor the inclusion of somewhat more graphic novels than we currently do, but would prefer to try to define some limits. For example, I would really rather not include all of the X-Men graphic novels -- even if they were written by established ISFDB authors. With respect to, say, superhero graphic novels, I think filtering by who has also published spec fic would be somewhat artificial. I'd rather filter by content type (e.g., exclude graphic novels whose characters come primarily from the comic book world). I recognize, though, that such a filter could be substantially more difficult to define. Chavey 14:34, 14 April 2012 (UTC)

Proposal for adding other titles:

Graphic novels that are written by persons well-known within the sf field (basically our current allowance for inclusion of Neil Gaiman's work.)
- Or works that were adapted from a text version, for which we already have a record in the database. Whether the author of the new work is the author of the older work is irrelevant. If author "Joe Smith" adapts Asimov's "Nightfall", it would be included in the db, but under "Joe Smith."
- Or works that have achieved a "certain threshold" within the sf field without regard to its notoriety in the comic book/graphic novel field (such as our current inclusion of Watchmen.) This decision, as nebulous as a "certain threshold" may be, is open to opinion, but should weed out graphic novels by newer authors who have no published text work.
Only the complete "collected" version of a novel, and not the individual comic book issues in which they first appeared, such as Watchmen, would have records (Bill's suggestion).
Only the first edition of a novel would be included in the db (Bill's suggestion). (An aside: I've always felt this should have been our policy for including nongenre fiction and associational nonfiction works such as science, astronomy, space exploration, films, etc., but that's for another discussion.)
The only role credited would be the author. (This is debatable. What if the artist is already in the db for his cover and interior art work?)
A separate type for graphic novels would be created (Ahasuerus's suggestion).
A field for linking the record to the Comic Book database would be created, such as we now have for Wikipedia and IMDB. (Bill's suggestion.)

That should be a good start for discussion. Mhhutchins 14:48, 14 April 2012 (UTC)

I quite like this, although it might still be a bit too inclusive for my tastes: e.g. I think the graphic novel versions of Coraline and Stardust are OK, but all the various collected issues of Sandman are not - they're just subsets of an ongoing comic series and may or may not be a complete work in themselves. I'd quite like to wait until Award editing is fixed/improved if we're going to allow all award-winners in, or they'll just remain unlinked or unlinkable. I'd credit the artists but I think inclusion should be based on the writers notability alone - we are a fiction database first and foremost and although we are gathering more SFnal Artists we could overwork ourselves if we start allowing everything in by artists over the "threshold". BLongley 15:15, 14 April 2012 (UTC)

Other comments: it doesn't have to be the Comic Book database, that was just my first thought, and maybe it would be nice to discuss mutual cross-linking with them. It may be that if we agree our borders then we don't need to enter much (or anything apart from the CBDB link) in ISFDB and they don't need to worry about, for instance, Chris Claremont's books. BLongley 15:15, 14 April 2012 (UTC)

It might also be time to declare that I used to be an avid Comic Collector and could provide 1st comic printing details for Watchmen and V for Vendetta and others, but I don't want to have that level of detail here. BLongley 15:15, 14 April 2012 (UTC)

I think Mike's set of rules would be a great definition. I'd have to think about Bill's concern about Sandman. By the way Bill, I also am a former Comic Collector. Until I sold my complete set of the 1940's "Captain America" to put my kids through college. But I still have about 15,000 issues. Which, of course, do not belong here. But some of my graphic novels probably do. Chavey 16:04, 14 April 2012 (UTC)

I'm still thinking that the best thing for now is "NO graphic novels" (or "NO comics") due to the extreme vagueness (is it an english word ?) of Michael's third criteria even if his others criteria (like the fifth one) are perfectly sensible in limiting the mass of data. But I'm not feeling very sanguine about the whole business so my opinion can easily be ignored. Hauck 16:29, 14 April 2012 (UTC)

(unindent) Good to see that there is interest in this topic. I think that one of the basic questions that we have to address first is what qualifies as a "graphic novel". It will be important to define the term unambiguously if we are to include (some) graphic novels, but exclude comic books. I am not a comic book/graphic novel collector (I think I own two graphic novels by Goulart and that's about it), so take it with a grain of salt, but here is my current understanding:

An "illustrated novel" is a novel whose text can stand on its own even if you remove all illustrations, e.g. in a later edition.
A "graphic novel" is a novel (or, I suppose, a collection/anthology/etc) whose text can't stand on its own.
A "comic book" is different from a graphic novel in part because it's not necessarily complete and because the binding is different -- this is the part that I am having the most trouble with since I don't know much about the subject.

Corrections welcome! Ahasuerus 00:13, 15 April 2012 (UTC)

So would you say that the summary immediately above is an accurate description of the differences between "illustrated novels", "graphic novels" and "comic books"? Is the line between the last two sufficiently unambiguous for us to distinguish between them reliably? Ahasuerus 02:31, 5 May 2012 (UTC)

Between "comic book" and "graphic novel" is still unclear. The individual issues of a comic book are almost always differently bound (stapled magazines), have issue numbers, and are significantly cheaper. However, the collected issues of such are usually in trade paperback format, and it's those that I generally want to keep out. The only exceptions I want in are where the collected issues define a complete work, i.e. the individual comics could be considered parts of a completed SERIAL. BLongley 13:50, 5 May 2012 (UTC)

I would like to see an "In" for graphic novels that are part of an existing print series -- even if they are just included as a link. Two examples: The Exile: An Outlander Graphic Novel is a part of the Outlander series and is written by the same author. OK, it's a retelling of part of one of the novels (Outlander), but it's the sort of thing visitors to the ISFDB are likely to be interested in. Second, Blood Work and Blood Crimes are part of the Rachel Morgan/Hollows series. Again, written by the same author -- and in fact, they are brand new stories. For the sake of completeness of series, I think it would be good to include at least some sort of mention of these. BungalowBarbara 23:11, 6 November 2012 (UTC)

You can create publication records for graphic novels that are actually written by an author who is above the threshold (using the CHAPTERBOOK type with a SHORTFICTION content record for the work). You should not create records for graphic novels based on characters, novels, settings by an author above the threshold who is not explicitly involved in its creation (credited as "creator" and not the writer). Diana Gabaldon's The Exile is eligible for inclusion. There is already a record for Blood Crime by Kim Harrison. These exceptions should not be used as a foot in the door to enter other graphic novels written by authors below the threshold or based on work by authors who are above the threshold. We can't be everything to everybody, and there is already a couple of databases for comic books and graphic novels. If you have trouble in determining an author's status (the threshold test), just ask on the Moderator page. When entering records for eligible titles, make sure to add a "Note to Moderator" why you feel such work is eligible for inclusion. Mhhutchins 15:52, 7 November 2012 (UTC)

We really need to have a separate title/pub type if we are going to allow even a limited subset of graphic novels in. FR 3586620 has been created. Ahasuerus 07:30, 13 November 2012 (UTC)

I have a new Graphic Novel adaptation of A Wrinkle in Time, by Madeleine L'Engle, who I believe is above the threshold. The pictures are new; the text is from the original novel. L'Engle was not "directly" involved (she's dead), but the text throughout is by her. Does this qualify for inclusion? Chavey 08:09, 13 November 2012 (UTC)

How is it credited on the title page? Is it credited as "written by Madeleine L'Engle"? Or is it "based on the novel by..." or "adapted by ... from the novel"? If the former it would qualify. If the latter it would not, because Hope Larson is not "above the threshold". Basically, a work would have to pass the same rules to qualify as any other non-genre work. The rules that allow a non-genre work in the db would apply to allowing graphic novels. Breaking the rule for this work would allow any speculative fiction graphic novel into the database. Mhhutchins 08:29, 13 November 2012 (UTC)

Title page: "Madeleine L'Engel's; A Wrinkle in Time; The Graphic Novel; Adapted and Illustrated by Hope Larson". Copyright page: "Text copyright © 1962 by Crosswicks, Ltd.; Pictures copyright © 2012 by Hope Larson; ... Designed by Hope Larson and Andrew Arnold; ... LCC Publication Data; Larson, Hope; A wrinkle in time / Madeleine L'Engle, adapted and illustrated by Hope Larson". Both Amazon and WorldCat list it as by both L'Engle and Larson. Thoughts? Chavey 06:29, 14 November 2012 (UTC)

As much as I adored the L'Engle book as a child, I don't think this adaptation would be eligible for the database. Basically for one reason: according to credits you've given here, L'Engle's name would not be entered into the author field of the publication record. BTW, look again at the OCLC record and you'll see she's not credited as the responsible author. You're not supposed to use the "Author" field of an OCLC record to credit a work on the ISFDB. You should always use the "Responsibility" field. The names of the OCLC fields differ from those of the ISFDB. This help page explains that ISFDB editors should use the "Responsibility" field. Mhhutchins 07:06, 14 November 2012 (UTC)

Point taken. I'm not sure I really understand their use of [ ] around L'Engle's name in that Responsibility field, but that clearly "demotes" it from a true authorship claim. I'll leave the book out. (Although I'm going to read it Real Soon Now™.) Chavey 15:34, 14 November 2012 (UTC)

OCLC uses brackets to indicate an unstated assumption. Brackets round a date, as [1994], means the book is undated, but the person who created the record has strong enough evidence to give the year of publication. Brackets round a page count, as in [32], means that the book's pages are not numbered, but that this is a manual count of pages. Brackets around a name in the responsibility field, as in [Madeleine L'Engle], means that she is not explicitly credited as author in the book, but that there's an assumption that she was responsible for at least some of the work. In the last case, ISFDB rules don't allow for assumed roles in authorship. An assumption of authorship (there is no stated credit) must be backed up from a reliable secondary source and then a variant title record can be created (as in this case). For years I've been wanting a field to enter other roles of responsibility for a work and maybe one day we'll see that. That includes translators, editors of non-anthology types (NOVEL and COLLECTION), and adapters. Mhhutchins 16:06, 14 November 2012 (UTC)

It's very much on the list of things to implement, but it will be a HUUUUGE change to the way the software works, so it won't happen overnight. One improvement at a time... Ahasuerus 18:03, 14 November 2012 (UTC)

Publishers and quotes

Copied from Zoltar's Talk page:

Just checking to see if Czytelnik and "Czytelnik" are the same publisher. Ahasuerus 06:49, 14 April 2012 (UTC)

Yes, they are :) I used "Czytelnik" where the publisher himself wrote it this way. This word means simply "reader". The full publisher's name is "Spółdzielnia Wydawnicza 'Czytelnik'". Zoltar 06:54, 14 April 2012 (UTC)

Hm, I think it would be better to use the same name for both "Czytelnik" and Czytelnik (without quotes) so that users could see all related publications on the same page. Let me copy this discussion to the Rules and Standards page to see what others think... Ahasuerus 01:01, 15 April 2012 (UTC)

(unindent) In addition, we have S. W. "Czytelnik" and Spółdzielnia Wydawnicza "Czytelnik". "Spółdzielnia Wydawnicza" means "publishing cooperative" in Polish and the years of publication overlap, so all of them are presumably the same publisher. The differences are cosmetic, so I wonder if we may want to merge the four publisher records under Czytelnik. Ahasuerus 07:27, 15 April 2012 (UTC)

I'd personally like to see some effort towards regularizing publisher names but also know it's a real pain. I started the Verified Publishing Names project with this in mind and realized it's a level of effort that puts a burden on many people as we would need to both enter and verify data with a lot more attention to detail than we have needed for story titles and author names. "Czytelnik" and Czytelnik (without quotes) are examples of the detail we'd need to pay attention to. Another example is the publisher "Ace." It used to be a standalone company, was later a division of Charter Communications, and today is an imprint that all use variations on the "Ace" name.

These days I'm shifting towards less detail simply as it results in less hassle for everyone.

With publishers I don't believe we have agreement on if the name on the title page, copyright page, back cover, etc, has precedence. We have agreed to not include the city name. For example, "Publisher, London" usually gets recorded as "Publisher."

I've thought about a feature request for variant publisher names but have held off as I suspect the level of work for editors goes up more than the long term value. I believe variant publisher names would make sense if we got into the habit of scanning and uploading the front cover, title page, copyright page, back cover, and perhaps the spine of all publications. That would cut down on the number of verification requests to see if a publication is a "Czytelnik" or a Czytelnik. --Marc Kupper|talk 20:13, 23 April 2012 (UTC)

Typoes and pseudonyms (and when not to)

I spotted a publication, and reprint, name that credit the cover art as Tim O'Brian. I e-mailed Tim O'Brien and he confirmed that it is his work. The following are silent on how to handle typos and spelling errors in publications.

My personal inclination is to credit Tim O'Brien and to add a publication note:

The copyright page states "Cover illustration by Tim O'Brian." In April 2012 the artist Tim O'Brien confirmed via e-mail that this painting is his work. As the "Tim O'Brian" credit appears to be a typo, rather than an effort to create a new pseudonym, this ISFDB publication record credits Tim O'Brien as the cover artist.

The other alternative is to use the "exactly as stated" rule meaning we'd credit Tim O'Brian, add notes explaining it's been confirmed that the "Tim O'Brian" cover art credit is a mistake, and to add a pseudonym linking Tim O'Brian to Tim O'Brien.

There's a similar issue with publications that credit the wrong artist. Usually it happens when they switch to a new cover. From what I recall, in those cases we have credited the correct artist and added a note rather than attempting to make artist "A" a pseudonym for "B."

Occasionally we'll see a typo in a title or author credit in one part of an anthology or magazine and it's correct elsewhere. I usually add a note explaining the discrepancy and enter the correct title/name in the database.

Though it has not shown up on ISFDB yet apparently there's a publication that has the wrong spelling for Asimov's name on the cover though I don't know about the title page. Asimov mentions this in Opus 100.

Either way, I suspect the help pages should be updated to explain what to do when faced with what appears to be an error on the editor or publisher's part rather than an explicit effort to create a new variant title or a pseudonym. --Marc Kupper|talk 19:37, 23 April 2012 (UTC)

I think we've traveled down this road a bit too far to change the policy now. It's been in place for at least the five years I've been here, and I'm unable (and unwilling) to go back over 5,000 verified records to determine if a credited variant should be reversed because it's a typo. Five years ago, I would have agreed with you, but this rule was drilled into me from day one. (And when did we ever create a variant based on a typographical error on a book's cover?) Mhhutchins 19:46, 23 April 2012 (UTC)

I did not think I was asking that policy be "changed." Right now it's 100% silent on how to handle errors. --Marc Kupper|talk 20:29, 23 April 2012 (UTC)

Sometimes it's a fine line between what's an error and what's an intentional credit, whether it's editorial or not. If a story is credited to "B. W. Aldiss" who decides if this is an error or an editorial choice? What happens when a user who has that copy of the story, looks up "B. W. Aldiss" in the database and can't find it because the name was "corrected" by an ISFDB editor? Mhhutchins 00:15, 24 April 2012 (UTC)

I don't know of any instances of creating a variant based on a typographical error on a book's cover. I'm not sure why you brought that up as we already spell out that the title page has precedence. --Marc Kupper|talk 20:29, 23 April 2012 (UTC)

You brought it up. I quote from your message: "there's a publication that has the wrong spelling for Asimov's name on the cover". Mhhutchins 00:10, 24 April 2012 (UTC)

Yep, and I would not intend at all that a pseudonym be created in that instance though in reading what I wrote a day later that's not apparent. :-) I have many books where the cover does not match the title page and for those I add a note to the publication record explaining the discrepancy. --Marc Kupper|talk 17:27, 24 April 2012 (UTC)

Let me ask you this. What would you do if you found a book that said "Isaac Asimov" on the cover, the copyright page, the top of every page in the story, and the author blurb. However, the title page has "Isac Asimov." Do you create a pseudonym? --Marc Kupper|talk 20:29, 23 April 2012 (UTC)

Yes. Any other entry would be against ISFDB policy. Mhhutchins 00:10, 24 April 2012 (UTC)

Let's use another example, you find a book that credits Jack Gaughan for the the cover but you discover that it's a Michael Whelan cover and that later printings credit Michael Whelan. Is Jack Gaughan a pseudonym for Michael Whelan? --Marc Kupper|talk 20:29, 23 April 2012 (UTC)

That's been brought up before. When the publisher incorrectly credits the cover art (which happens more often than you think it should), the defacto standard, whether it's stated or not, is to correct the record and then add an explanation to the note field. That's not the same thing as crediting a book's author based on the title page. If the Asimov book was credited to Poul Anderson on the title page, you wouldn't create a pseudonym, but you would credit the record to Anderson and then create a variant record which credits it to Asimov. I don't think this would actually happen, but if it did, the policy in place would cover it. But occasionally something like this happens. Mhhutchins 00:10, 24 April 2012 (UTC)

Marc said that the current policy is "100% silent on how to handle (typographical) errors.". That's not quite right. An argument can be made that the policy could be adding more explicitly in a few additional places, but it occurs now in two places: Verification Rules and How to Record a Variant Title. On the former page, the policy says: "A true typographical error in the publication should indeed be listed on the publication record, but should always be accompanied by a note explaining the situation." On the latter page, the policy is discussing what happens when there are two entries for a single work, and you're trying to decide whether an existing variant should be left as is or corrected. List item #2 says that if it appears that the person entering the data made a typo, you should correct it, but if this was "really how the book was published" then the variant should be left as is and "the publication should have a note against it saying why the spelling is different". (Well, re-reading that section, it's not as clear as it should be, but this is what's implied.) Chavey 09:15, 24 April 2012 (UTC)

As Michael notes, there's a defacto standard to correct errors in the artist credit. Chavey, thank you for finding the verification pages. I had not thought of looking there. I suspect all of us are on the same page but have run out of time and so will take another look this evening. --Marc Kupper|talk 17:54, 24 April 2012 (UTC)

Unicode titles

Help:Screen:NewPub makes no mention of non-Latin/Unicode (i.e. Japanese, Cyrillic, etc) titles. Help:How to enter foreign language editions says "When entering the name of an Author or a title that uses non-Latin characters, use the English (i.e. transliterated) form of the name/title when known."

We have been experimenting with Unicode titles for the last few months and, as far as I can tell, it's best to enter them using the original alphabet and then add a Latin transliteration in parentheses. The main advantage of this approach is that it allows users to search for titles using either the original alphabet or the transliterated form, so I propose that we adopt it as the official standard. One possible alternative would be to add a new field for "transliterated titles" to the Title record, but it would be time consuming to implement.

Here are a few examples of what the proposed approach will look like:

Ivan Yefremov - Russian
Nagaru Tanigawa - Japanese
Ljuben Diłow - Bulgarian

One gray area is mixed-alphabet titles, e.g. Cor Serpentis (Сердце змеи). The first part of this title is in Latin ("Cor Serpentis" means "The Heart of the Serpent" in Latin) while the second part is a Russian translation of "Cor Serpentis". We currently list this title as "Cor Serpentis (Сердце змеи) (Cor Serpentis (Serdtse Zmei))", which looks busy, but I can't think of a better way to represent it.

P.S. Author names are a separate can of worms to be tackled later. Ahasuerus 04:37, 3 May 2012 (UTC)

Why not enter the title in the original alphabet and make the transliteration a variant? Wouldn't that also solve the search problem, avoid clutter, and be a little more friendly in its treatment of the mixed-alphabet case? --MartyD 10:07, 3 May 2012 (UTC)

Another variant??? The title record display pages are cluttered enough as it is. I'd rather see the titles entered as is and have a search system that finds them regardless of what alphabet is used. And then let the user have the ability to choose what pub records are displayed. The push to be all-inclusive is making the db lose focus, and less user-friendly, at least to the majority of users, who, whether you want to believe it or not, only speak English and use the Latin alphabet. (I'm steeling myself for the charges of anglocentrism.) Mhhutchins 15:00, 3 May 2012 (UTC)

The problem with as is is that we could end up with even more titles, if the same title has been transliterated various ways. And I don't think we'll ever be able to improve searching enough to enter one form and have search find the others. So I'm quite happy to go with Ahasuerus' suggestion, and let the users of such thrash out which form of transliteration is preferred. I'm stretched enough with western languages with funny accents to have much advice on funny alphabets - except that KISS ("Keep It Simple, Stupid!") seems a good principle to apply here. As for "Cor Serpentis (Сердце змеи) (Cor Serpentis (Serdtse Zmei))" I don't see the need to repeat the Latin name. BLongley 15:57, 3 May 2012 (UTC)

Changing the search system to find the original ("as is") title based on its transliterated form would be very difficult to implement. For example, "Cor Serpentis (Сердце змеи)" is stored as "Cor Serpentis (&#1057 ;&#1077 ;&#1088 ;&#1076 ;&#1094 ;&#1077 ; &#1079 ;&#1084 ;&#1077 ;&#1080 ;)" in the database. The search logic would need to know that "Serd", the Latin representation of "Серд", should match 1057, 1077, 1088 and 1076, but only when the title's language uses Cyrillic -- in other alphabets the Unicode numbers are totally different. And then there are different transliteration schemes (more than a dozen for Cyrillic alone!), so it would be a nightmare to code. Ahasuerus 17:10, 3 May 2012 (UTC)

All of this parenthetical stuff drives me nuts. Record it as it is in the publication is so much simpler for everyone. What would we want to do about disambiguating a non-Latin "generic" content title within a non-Latin titled publication? "Введение (Vedyenie) (Война и мир (Voyná i mir))"? Sorry, but ick. Maybe a simpler approach is no transliterations unless published that way (and then, yes, we'd have the variant, but how often will it really happen?). It's probably easier these days for someone to look up the native form of a title and copy-and-paste it than it is for them to come up with the proper transliteration anyway. --MartyD 17:04, 3 May 2012 (UTC)

That's a valid point, but on the other hand limiting the canonical title to the original alphabet makes it unreadable unless the user happens to be familiar with the alphabet in question. Upon reflection, it occurs to me that we may have different use cases here. I'll try to think them through and respond again later. Ahasuerus 17:10, 3 May 2012 (UTC)

I also see the information in parenthesis as somewhat superfluous. For me, the best way to have a transliteration (possibly the English way) would be to give it in the title notes. Stonecreek 17:15, 3 May 2012 (UTC)

That's a better idea. And notes can be searched, at least using Latin characters. If you're unable to enter characters from a different alphabet and language then you shouldn't be creating records. We had enough trouble with robots picking through the internet looking for pubs, so we sure don't need a human automaton doing the same thing. Mhhutchins 18:30, 3 May 2012 (UTC)

I could live with that. We could also use it to cover some useful-for-the-Anglo-centric ideas, like the literal English translation of the original title. We already have a few examples, search notes for "literal". BLongley 02:15, 4 May 2012 (UTC)

(unindent) It looks like we all agree that we want to record titles using the original alphabet. The only outstanding issue is whether/how to capture their transliterated (or, technically, "romanized") versions. Let's take a step back and examine a likely scenario:

The user knows the (or rather "a") transliterated version of a certain title and is trying to find it in ISFDB. This can happen when working with secondary sources, e.g. library catalogs, which usually use romanized titles. Similarly, the SFE3 article on Ivan Yefremov mentions the following titles:

Chas Byka ["The Hour of the Bull"] (1968; exp 1970)
Lezvie Britvy ["The Razor's Edge"] (1963)
Tais Afinskaia ["Thais of Athens"] (1968)

If we don't provide transliterations, then many users will find it hard to associate "Chas Byka" with "Час быка", "Lezvie Britvy" with "Лезвие бритвы", etc. And if you can't (easily) find the canonical title, then you won't be able to find its English, French, German, etc translations.

Four ways of recording transliterations have been proposed so far:

Enter the transliterated version in parentheses
Create a transliterated variant title with no pubs
Enter the transliterated title in the Notes field, presumably using some standard notation like "Romanized title: Chas byka"
Create an optional "Romanized title" field in the Title record and possibly in the Publication record. (We could also add a User Preference which will control whether the value of this field is displayed on the Summary page or just on the Title page.)

After reviewing the discussion above, I think the last option would work best in the long run. It's not particularly hard to implement, but it will have to wait until other changes have been tested and installed. For now, we may want to adopt the third option as a palliative solution. What do you think? Ahasuerus 03:37, 4 May 2012 (UTC)

Wait, this is starting to sound like a "decision".... :-) Sounds ok to me. --MartyD 11:08, 4 May 2012 (UTC)

For me, too - a fine idea. Stonecreek 17:55, 4 May 2012 (UTC)

With Ahasuerus' list of 4 options, let's not forget a 5th, the current policy, which is:

5. Transliterated title in the Title / Pub record + original language title in the notes.

I agree that option #4 is better than this, but as a temporary fix I'm not sure there's an advantage to #3 over #5. And should I then go back to all the titles I've done using #5, change them to #3, then come back and change them to #4 when that's ready? Chavey 13:26, 5 May 2012 (UTC)

I would not recommend reworking until the software is ready for it, you'll just end up working 3 or 4 times harder for the result. So I would wait for "Unmerge Foreign Title" at least, and possibly translator support. (I'm already seeing identical Italian translations that might be mergeable if they're by the same translator, but we can't be sure yet.) Maybe Ahasuerus can give us some idea of timescales and priorities on the outstanding and desired changes? BLongley 13:39, 5 May 2012 (UTC)

I too would advise against changing transliterated titles just yet. My proposal is more modest -- change the Help pages to make #4 the new standard so that it would be used going forward.

As far as the timescales and timelines go, well... As mentioned earlier, I have been trying to retire and concentrate on ISFDB for some time now. The last attempt (about 18 months ago) didn't pan out, but I am giving it another shot now. I'll know more in a couple of weeks. Fingers crossed... Ahasuerus 03:13, 6 May 2012 (UTC)

Disambiguating authors having the same name

See this discussion for background. A situation has arisen where we have records for two authors sharing the same name. One author is more established/widely published but is dead. The other is obscure but living. According to this help article, the recommended solution in this case is to:

... append the birth and death dates to the name of each of the authors, where known.

e.g., Ima Writer (1901-1999) and Ima Writer (1951-). Some folks feel (my paraphrasing) that the help's recommendation notwithstanding, de facto ISFDB practice is to leave the popular author's name unadorned and to add the disambiguating information only to the names of the more obscure authors. For example, what would we do if a second Stephen King appeared? And the above help page's example of Robert Frazier does not even match our actual treatment of that name: Robert Frazier for the one born in 1951, which the help says should be "Robert Frazier (1951-)", and Robert Frazier (active 1954-1955) for the one whose birth date is unknown, which the help says should be "Robert Frazier (?-)".

Which is it? Adorn each, or adorn only the more obscure/less popular/less widely published?
If one may be left unadorned, how do we decide which it should be?
Do the same rules apply when disambiguating by profession?

I don't have a strong preference, except that popularity vs. obscurity is subjective and certainly can be difficult to assess by anyone not familiar with the principals or their works. I don't know that the downside of having disambiguation on a popular name is, so maybe someone could speak to that. I'm happy to fix up the help to match whatever we want to be doing. --MartyD 11:24, 8 May 2012 (UTC)

I believe in such cases, that the author with the more prominent profile (fairly well-known within the spec-fic field), regardless of the extent of his/her productivity or whether he/she is still alive, should retain the non-disambiguated name. In twenty years, if it turns out that the newer author has risen to a more popular position, it would be fairly simple to reverse the two. (In 1984, I thought it was strange that an author used the name "William Gibson", when such a well-known playwright was still actively producing work.) If the editor or moderator are unfamiliar with either author's work, a posting on one of the community pages should give enough feedback to help in the decision.

As for disambiguating by profession, it works when someone sticks to one role: artist, photographer, journalist, playwright, etc but gets sticky when they don't. I'd rather use the IMDB method of adding parenthetical roman numbers (I), (II), etc, but I can see the value in adding the profession, or birthyear if both have the same profession. Mhhutchins 13:48, 8 May 2012 (UTC)

I agree the help doesn't match practice. Recently, we fixed some "John Grant" 123992 and left the "major" author unadorned. Deciding which to keep unadorned is one of those judgement calls - being dead is no longer a barrier to publication, given the number of Wells, Baum, Doyle, Verne etc publications I see submitted. Which reminds me, I need to see if people have expressed a view on that Vector editor... BLongley 13:53, 10 May 2012 (UTC)

BTW, a Stephen Graham King has appeared. :-) BLongley 17:03, 16 May 2012 (UTC)

The middle name makes that irrelevant to the discussion. :-) Mhhutchins 02:24, 17 May 2012 (UTC)

It does indeed help when Authors disambiguate themselves. I wonder if M. K. Hutchins added the K because of you? BLongley 19:38, 17 May 2012 (UTC)

May be a lost cousin? It's a fairly uncommon name. There are more people named Hutchinson than Hutchins, but that would make sense if they are our "sons". Mhhutchins 00:30, 18 May 2012 (UTC)

Sometimes their editors get involved, e.g. Lawrence W. Evans's first appearance was as by Lawrence Watt-Evans because his editor pointed out that there was another writer whose name was "Lawrence Evans". At luck would have it, the other Lawrence Evans stopped writing shortly thereafter, but by then it was too late to change the "Watt-Evans" byline. Ahasuerus 19:47, 17 May 2012 (UTC)

I will post a proposed rewrite when I get a chance. --MartyD 01:03, 17 May 2012 (UTC)

I would propose however that if neither was/is prolific or well known... that disambiguating both could also still be allowed..... Kevin 13:39, 18 May 2012 (UTC)

(unindent) Sorry about the delay. Proposed text posted in a new section below. Once people here agree, I will update the help page and post a notice on the Community Portal for additional review. --MartyD 13:44, 20 June 2012 (UTC)

Numbering unnumbered pages

When I questioned a new editor's use of brackets to enter the page numbers of interiorart records (maps that were printed within the pages which were numbered, but on unnumbered pages), a section of this help page was brought to my attention: Any page for which a number is derived by counting, possibly from a previously or subsequently numbered page, but the page carries no actual page number, should be listed in [brackets].

This was my response:

Thanks for bringing that help section to my attention. It doesn't make clear that this method should only be used for unnumbered pages that fall before the first numbered page or after the last numbered page. If a work is printed on an unnumbered page facing, for example, page 26, it should be entered as "27". That applies to any kind of work. Quite often book designers will omit page numbers for a titled page (with story title or chapter number/name) if they've chosen to number the pages at the top. (And if the page is an illustration, most of the time the page is not numbered.) Take a look at any book in your collection that prints the page number at the top and you'll see this. Sometimes they'll move the number to the bottom, but often don't number the page at all. In these cases it wouldn't make sense to bracket the numbers. I'll bring this up on the Rules and Standards page and see about having the help page changed.

I think the help page should be clarified. Mhhutchins 15:25, 19 May 2012 (UTC)

I agree with your conclusion -- intermediate unnumbered pages, between numbered pages, should not be bracketed. While you're at it, what should be done about interior art on unnumbered pages that are inserted between page 27 and 28? This often happens with inserted photographs, but I see it on regular art occasionally as well. My inclination is "27a", but your wording above would imply that 3 successive pages might be numbered "28". Chavey 21:47, 19 May 2012 (UTC)

Those are plates, not pages, a fine distinction, I admit. So, no, you wouldn't give page numbers for them. The only way I know to handle plates is just to create unpaged content records like this pub. Inelegant in display, yes, but I wasn't able to find a better way. Mhhutchins 00:38, 20 May 2012 (UTC)

That does seem a bit clumsy. What would be wrong with using something like 27a, 27(a), 27.1, or something like that? It seems that any of those would probably put the pages in the correct ordering. Chavey 06:25, 21 May 2012 (UTC)

Because numbers with letters or other characters aren't displayed in numerical order. Mhhutchins 17:58, 21 May 2012 (UTC)

I used to insert plates as in this publication and made a comment about the pages whereafter they appear. Sorting of those bracketed page numbers could be a bit better, but otherwise I think it doesn't look too bad. In addition, it emphasizes the plates - overall ordering isn't ensured, though. Stonecreek 09:05, 21 May 2012 (UTC)

As good as a solution as I've seen. I wish an editor was able to give an order without page numbering being involved. That's the best solution. Mhhutchins 17:58, 21 May 2012 (UTC)

Displaying page numbers in the right order is already a software headache - Roman Numerals before Arabic ones, special codes like "fep" - my usual frustration about the current situation is with poems or short reviews where there are multiple titles (by our definition) on the same page. Not ideal, agreed, but probably not a priority given the backlog of more important software changes. BLongley 22:49, 21 May 2012 (UTC)

Magazine series and regular series.

There are three Doc Savage series. Doc Savage, Doc Savage Doubles, and Doc Savage Magazine. Is there any reason to not have the magazines be a child series of Doc Savage? Dana Carson 20:07, 20 May 2012 (UTC)

No reason that I would know of. This whole area is rather messy at the moment: there are dozens of Kenneth Robeson titles that need to be VT'd, omnibuses that need to be moved to the omnibus-specific sub-series, etc. Ahasuerus 21:11, 20 May 2012 (UTC)

Proposal: Bilingual books

We have no formal policy on how to handle the languages for a bilingual publication. I propose:

Books published as bilingual (or multi-lingual) texts should be treated as if they were multiple different stories/novels within the same publication. Those two stories would belong to title records with the appropriate languages, and one would be a VT of the other. If the two language versions are inter-laced in the publication, they should still be listed as separate works, but starting on the same page number. Note that this might convert a "Novel" into an "Omnibus", for longer works, or a "Chapterbook" into a "Collection", for shorter works.

This was the solution suggested before, and implemented with Sergio and Ingrid: Pionniers de Mars / Pioneers to Mars. Based on that example, I did the same with a bilingual edition of Smith of Wootton Major. However, in neither of these cases did we actually treat the bilingual book as if it had two stories in it, i.e. the first three sentences of the suggested rule above have been implemented (at least sometimes), but not the last sentence.

I also propose that we add, as a supported language, "Multilingual", and that this language be limited (in practice, but not enforced by software) to collections, anthologies, chapterbooks, and omnibuses in which major portions of the fiction content are in each of at least two different languages, or nonfiction works in which major portions of the content are in each of at least two different languages. (Thus an English collection with a prologue essay in French would still be listed as being "English".) This overcomes the language chauvinism of having the two bilingual works listed above both having title recs that claim they are in "English", even though only some of their contents are in English. Chavey 21:20, 21 May 2012 (UTC)

"Multilingual" looks like a useful addition for container titles. I think I would prefer to include Novels and not count the same work in different languages as multiple contents to convert "Novel" into an "Omnibus" or a "Chapterbook" into a "Collection". But this isn't a strong preference - the only book I have that it would apply to would need "Klingon" added to languages as well. :-) BLongley 22:43, 21 May 2012 (UTC)

Yes, "Multilingual" would be helpful in cases as the above. Should it lead to an "Omnibus" or a "Collection"? Hmmm, yes and no. You would have two novels, for example (but after all, it'd be the same one). If you take a look at them as two titles, internal logic would be more on the side of "Omnibus" , I guess. Stonecreek 14:47, 29 May 2012 (UTC)

Merging SERIALs

According to Help:Use of the SERIAL type, "If a work has been serialized more than once, those records are normally not merged either". However, it looks like at least a few "(Complete Novel)" SERIALs have been merged, e.g. Quelli di Kaluiki (Complete Novel).

If we agree that SERIALs shouldn't be merged, then I propose that we remove the word "normally" from the Help text quoted above and unmerge the affected titles. Ahasuerus 18:56, 27 May 2012 (UTC)

I understand that SERIALs are often abridged and there is no easy way of telling whether any two abridgements are identical, but in case of declared reprints I think that they should be merged. --Pips55 20:41, 27 May 2012 (UTC)

I agree with Pips55 here. If a serial is published by the same publisher, in the same number of parts, and, in the case of these Italian publications, translated by the same translator, I see no reason not to merge them. If research shows they are the same, why not merge them? If the editor hasn't taken the time to research, only then should you hold off accepting the submission to merge them. And as of this moment, we can't use the variant function to define a difference in text. In fact, the rules allow us to merge title records which have the same author and the same title without the editor having to research whether one has been revised, expanded, or abridged. Why should serial-type records be the exception, if an editor has substantial evidence that they are basically the same work? Mhhutchins 03:28, 28 May 2012 (UTC)

I also think they should be merged when we believe them to be the same. --MartyD 10:07, 28 May 2012 (UTC)

Yep. If we think they are the same, then they should be merged. If however when we have some evidence that says they are not (or are probably not) should they remain separate. Kevin 13:58, 28 May 2012 (UTC)

OK, we can do that, but do we also want to merge multi-part serializations, i.e. (Part 1 of 4), (Part 2 of 4), etc when we are sure that the parts match exactly? Ahasuerus 03:33, 30 May 2012 (UTC)

Absolutely. If the editor is certain that the parts are identical there's no reason not to merge them. It's up to moderator to question such merges to make sure that the submitter knows exactly what he's doing. Mhhutchins 04:51, 30 May 2012 (UTC)

Thanks to everyone who has responded -- I will change Help and approve the submissions. Ahasuerus 05:06, 8 June 2012 (UTC)

Counted page numbers and []s

Putting page numbers in brackets when they are arrived at by counting makes sense. However in cases where the count is trivial I think not putting the brackets makes sense. In Under the Moons of Mars: New Adventures on Barsoom they don't number the first page of each story or the page illustration for each. This means that following the rule fully means all the numbers are bracketed and the TOC looks strange. For the first printing 379712 I removed the brackets and I think it reads better.

Opinions? Dana Carson 05:00, 29 May 2012 (UTC)

I agree this should be the approach. It's only logical. This was brought up only ten days ago here. The few that participated in the discussion agreed that the help pages should be clarified to exclude unnumbered pages within the main body of the work. But, as usual, so few editors responded that the discussion just petered out. I could have unilaterally updated the help pages, but I've seen the wrath pour down on those editors who do that. Perhaps the rules should be changed that require a discussion period and a general consensus arrived at before a standard can be changed? But if I brought that up for discussion, I'd probably get the same response. I'm not coming down on those editors who don't offer their opinions or participate in discussions. They may have seen how little is accomplished despite so much smoke and fury. It does bug me that moderators sit out such discussions, but there's nothing I can do about that either. Mhhutchins 13:49, 29 May 2012 (UTC)

Concur. Trivial page counting should be normally listed. Brackets should be used only when there is an extended block of unnumbered pages (As when describing an unnumbered block preceding regular pagination, or when the whole book is not numbered). Kevin 14:18, 29 May 2012 (UTC)

We're not talking about the page counting. We're talking about assigning numbers to contents which appear on unnumbered pages within the main body of a publication. Please re-read the original post. Mhhutchins 00:21, 30 May 2012 (UTC)

I'm all for a discussion period in such matters. Those who don't respond are assumed to have no interest in the discussed matter - which may be a misleading assumption (quite often it's just a matter of time), but it really would help to get things done or into a pipeline.

And on counting page numbers: I also concur, but would like to extend the implicated page number counting to the (mostly) Roman numbered introductory pages: Quite often there is an introduction (e. g. on pages numbered v-ix), but there could be acknowledgments on the following (unnumbered) page, a map on the page after and some extended copyrights on the next page. In this example, the assumed novel would begin on p. 1. Now, it'd be logical for me to number the unnumbered pages as x-xii. Stonecreek 14:32, 29 May 2012 (UTC)

This has nothing to do with the page count field. We're talking about assigning numbers to contents which appear on unnumbered pages within the text of a publication. Please read the original post completely. Go to the links that Dana gave and look at the comparison he's trying to make between the two records. Mhhutchins 00:21, 30 May 2012 (UTC)

I always took it that roman numerals preceding, would 'own' the block of non-numbered pages immediately following. So a very complex page count could go like this [1]-[9], [i]-[ii], iii-ix, [x]-[xii], [1], 2-100, [101-120]... in other words, 11 un-numbered pages, followed by the first page with roman numeral iii. This continues to printed roman numeral ix. Four un-numbered pages then fit in between the numbered roman numeral and the first arabic printed page, but the first arabic numeral is 2. Then after a block of 99 numbered pages, there follows a last block of unnumbered pages 20 long at the end of the volume. - Yes this ends up with duplicate page [1], but otherwise seems logical. Kevin 21:14, 29 May 2012 (UTC)

But I would not want to add any roman numerals unless at least one page had a roman numeral printed. Otherwise all preceding pages would default to [arabic] counts. Kevin 21:14, 29 May 2012 (UTC)

Sorry, Kevin. I think you've misunderstood the original problem and the purpose of the discussion. There were pages that contained artwork and the first pages of stories which were not physically numbered, so the editor gave the page number of what it should have been in brackets. The point Dana was trying to make was that it's not necessary to give brackets for content items where the page number is derived by counting forward or backward from the closest numbered page. Or maybe I misunderstood the original point of the post. Mhhutchins 00:21, 30 May 2012 (UTC)

I did understand the question, but I also see where this can appear to be a tangent. My example was limited to blocks of numbers. Only brackets appear on the ends of a defined block (never in the middle of a defined block). That was the point I had in mind that it appears I never arrived at. Once you are within an identified block of numbered pages, it is trivial to count forward or backward to identify the rare unnumbered page. Only outside of those blocks should brackets be employed. Kevin 02:14, 30 May 2012 (UTC)

I'd even read that other discussion and forgot about it. Yes, Use brackets for large enough blocks that you might miscount or where there are no numbers in the first place. If the story starts on a page numbered other than 1 count back if you need numbers before that otherwise count forward with roman numerals if those are used already. Maybe we should go to a system like some of the open source projects use. If enough people agree and no one disagrees the change happens. If there is disagreement then you use a more formal system. Dana Carson 23:34, 29 May 2012 (UTC)

This discussion makes it clear why nothing ever gets accomplished here. They go off into tangents and devolve to the point where the original meaning of the post has disappeared. As for your last comments, Dana, the problem is that for any particular issue there may be only 4 or 5 editors participating in the discussion. On the average, on any given number of days, there are 15-20 editors updating the database. So what would be "enough people" to form a quorum? We tried a "formal" system awhile back, which lasted about a week. Mhhutchins 00:21, 30 May 2012 (UTC)

I don't know why Andonsage numbered the contents of this record thusly, unless he was confused by the wording of the help page. It is common practice for a book designer not to give the page number for the first page of a content, especially if he has chosen to display the page numbers at the top of the page. A title page looks strange with a running header at the top of the page, and many book designers simply drop it entirely. Just pick up the closest anthology or collection you can find and you'll see what I'm talking about. That doesn't mean that we can't interpolate what the page number is, but the rules as stated may be confusing to some editors. That's why I proposed that the rules be clarified. (Hopefully now we're back on subject. Please, no more discussion about the page count field.) Mhhutchins 00:29, 30 May 2012 (UTC)

Well, I think we accomplished something. We (the participating discussants) all agreed to Dana's proposal, I think. And I'd make another proposal: We'll wait til Sunday (3rd of June, 2012) and then close the discussion and change the help pages accordingly.

I'd think, though, there's some kind of connection between counting pages and assigning numbers to contents. It would be especially helpful for new editors when they'd like to add contents if there already was a complete page numbering done (and if there were notes of how the assigning of numbers was done, if necessary). Stonecreek 09:06, 30 May 2012 (UTC)

There are always connections. It's a database and everything is connected. My point is that we stick to one specific, well-defined topic, and work it out before moving on to other areas, not matter how much it's related. So I present the following:

The current help page for entering contents (here) states:

Page - The page on which this item of content can be found. The page number refers to the pagination provided in the publication, not the actual count of pages from the start.

I propose that it be clarified as:

Page - The page on which this item of content can be found. The page number refers to the pagination provided in the publication, not the actual count of pages from the start. If the content item (SHORTFICTION, ESSAY, INTERIORART, etc.) begins or appears on an unnumbered page, you can interpolate the page number by going to the nearest numbered page and counting either forward or backward to determine its page number. Do not bracket this number, unless the content appears in a block of pages which are completely without pagination. Blocks of unnumbered pages might appear before the first numbered page, or after the last numbered page, and any content which appears on those pages should have its page number bracketed. The page number of any content which appears within a block of numbered pages should not be bracketed.

The remaining instructions for this section can remain intact (it's about page numbers in magazines). Any suggestions for improvement? (We can open a new topic about page counting when this one's been completed.) Mhhutchins 15:36, 30 May 2012 (UTC)

I can't think of an simple way to word it, but I think we should be explicit that 1-n is not a "block" when the first numbered page is n+1. For example, if the text starts on numbered page 7, and a 2-page foreword appears on unnumbered pages before that, and a full-page drawing appears on the page before that, use "4", not "[4]" for the artwork and "5", not "[5]" for the foreword. Assuming that's what we want, of course. --MartyD 10:19, 1 June 2012 (UTC)

Concur (For unnumbered pages preceding a numbered block). How many unnumbered pages after a block should we allow before we start bracketing? Kevin 11:28, 1 June 2012 (UTC)

I also concur with this proposal. Stonecreek 05:23, 2 June 2012 (UTC)

Disagree with the last. Any separate content before the first numbered page or after the last numbered page should have any assigned page number bracketed. Nothing in between should have brackets. There are hundreds, if not thousands of records that would have to be changed if brackets were removed from pre- or post- pages. --~ Bill, Bluesman 22:26, 1 June 2012 (UTC)

Bill - What about this example...(I'm just throwing darts at the wall to aid discussion). The novel starts on numbered page 3. There is a map two pages prior to that. There is Frontispiece on the title page which is on the page before the map, which is the 3rd printed page after the end papers. So the frontispiece is on page [3] counting forwards from the end papers. The next page has a map, and is the 1st (unnumbered) page of the novel block of numbered pages... so you give it a [1], and then the novel appears on page 3. But now the contents appear and will be read as [1] Map, [3] Frontispiece, and 3 Novel. Is that the effect you want? If so, this is different from the help pages today which state (in the page count section) "At times you will need to count backwards from the first numbered page to see which is page 1 and then would count the unnumbered pages that are before this." I just think a consistent application of the same rules would apply to figuring out the pages for contents items. Kevin 01:23, 2 June 2012 (UTC)

I have to disagree with this. In the example, the frontispiece should have pagenumber "bp" (bp -- unpaginated pages that precede pagination). I do agree with Michael's suggestion for the new helptext. --Willem H. 08:29, 2 June 2012 (UTC)

Agree to the update for the help text. Dana Carson 01:13, 3 June 2012 (UTC)

Agree to any reasonable consensus. Kevin 02:04, 6 June 2012 (UTC)

I very much like the suggested help page improvements. I agree that pages intermediate between numbered pages should not be bracketed. (I have seen several cases where the novel ends on p. 300, there are 8 unnumbered pages with some miscellany, then some appendices begin with numbers that link correctly. Those 8 pages should not be bracketed.) However, I agree with Bill that pages before the first numbered page, and those after the last numbered page, should all be bracketed. (And hence I agree with Willem's conclusion about the frontispiece.) I would be willing to make an exception for an unnumbered page 1 prior to a numbered page 2, but I would prefer to have no more exception than that. Chavey 04:07, 15 June 2012 (UTC)

Proposal: Pages on contents listing

First, hello to all. Because this is a hot spot for me, I jump in and make a proposal. I want a change on how isfdb shows the contents listing regarding page number. I want something like this, which means that for every piece of pub, in the contents to show me first page and last page of that pub. Pro argument: please take a look at this and tell me if is possible that the essay "The Clubhouse (Amazing Science Fiction, October 1974)" to be 59 pages long? Thanks Calin Hila 19:07, 4 June 2012 (UTC)

Probably because some other text (the Vance ?) restart somewhere after the essay. The indication of the pages of the text is perhaps a good idea but could lead to ever more cluttered displays, particurlarly with US magazines where a text can be split in multiple fragments. Note that your french example corresponds to this iSFDB entry and more importantly to an anthology, an easier case where texts are usually in one block and that the page numbers from noosfere are not to the ISFDB standards (they're counted from the first page of text, not from the title page). Hauck 19:20, 4 June 2012 (UTC)

It's not a bad idea, but I think it's several years too late. With Anthologies the last page can be guessed to within a page or two, with magazines there are so many "continued on page X" we've totally given up on using page numbers to guess lengths. BLongley 07:12, 5 June 2012 (UTC)

Hauck, most likely is the Vance. Cluttered ? maybe if we can handle the interior art differently we don't have cluttered displays. How about this: 5-16 • Story Title • shortstory by The Author • ill. by The Big Artist on page 5, 10, 15 (idea stolen from noosfere, too). It is easy to understand that the title of the story is on a page followed by a blank page then by the text of the story from noosfere than from isfdb, no offence.

Well, if you're so keen on noosfere, let's talk about this magazine. The short story by Dominique Douay is on the following pages : 19, 20, 21, 22 & 30 (meaning that the end of the text is separate from the beginning, like in some US or UK magazines), you'll note that this little fact is just omitted from the data. Finaly, it seems that noosfere encounters exactly the same limitations. Hauck 15:46, 5 June 2012 (UTC)

A solution: 19-22 • story by Dominique Douay and 30 • story by Dominique Douay (2) like interior art is listed when an artist has two or more pieces for one story. Another solution: 19-22 • story by Dominique Douay and 30 • story by Dominique Douay (continued) Calin Hila 16:39, 6 June 2012 (UTC)

BLongley, better too late than never, i say. To me is more important to record that the story/essay is from page to page than to know on which page is every piece of interior art (i saw that someone had mentioned a scribbling on the header), map and if there is a picture of the author on the back flap or not. Calin Hila 08:14, 5 June 2012 (UTC)

Let's imagine an issue of Locus entered this way, terrifying. Hauck 15:55, 5 June 2012 (UTC)

What is terrifying (1) to record that the story/essay is from page to page (2) to know on which page is every piece of interior art (i saw that someone had mentioned a scribbling on the header), map and if there is a picture of the author on the back flap or not.Calin Hila 16:39, 6 June 2012 (UTC)

We have 22,585 magazines that may need rework. I don't want to do any of those. Some people are clarifying Interior Art - I don't want to work on that either. As I mentioned, it's not necessarily a bad idea, but after several years of working the old way there really needs to be a very good reason to change. At the moment, I think improving language support is the top priority - we have editors waiting on those features. Award support next - at least two editors want to improve our coverage. Your sugestion is understandable but I can't really foresee it becoming a high priority. BLongley 20:30, 6 June 2012 (UTC)

Same as Bill here. It's not that your proposal doesn't have any merit (more precise data is always a bonus) but IMHO the way you presented it was not optimal (by highlighting another site's way, which in the end proved as imperfect as we are or by repeating twice the same arguments). If this kind of data is important to you, feel free to add it, at first perhaps by expanding the note field. Hauck 08:04, 7 June 2012 (UTC)

Every pub that has been verified is reworked by 2nd, 3rd, 4th or 5th verifier, so you may not need to reworked them. I see that there are other priorities and i understand that. Regarding award support, on the net you can find infos about awards, almost every award has a site of his own and Locus does a pretty god job for covering the rest of them. I came here to find data unavailable elsewhere not for infos about awards. (just my opinion, i don't want to be mean) But, if it has been decided that award support improvement is a very important one i will support it and i will contribute to it if every editor has rights to do it, not only the mods.

Hauck, i give that link to noosfere to help you understand what i want, but it seems to me that it turns out ... topsy-turvy. And one question, If you have an optional field to add the last page of a story, will you added or not? If your answer is yes then i lost by 1-1.

It mostly depends on the ease of entering the data. As I said, all information is valuable but this category od data seems to me particurlaly costly to implement. Hauck 14:02, 9 June 2012 (UTC)

Conclusion, i wait for the next time...Calin Hila 12:02, 9 June 2012 (UTC)

As partial support for Calin, I think that there might be room to allow "xx-yy" as an option in certain cases. The cases where I have wanted to use that format have been when entering an anthology or collection that includes one or two spec fic stories, and where I don't want to include the non-genre stories. While magazine page numbers can't be trusted for story length, anthologies usually give you a reasonable idea. I'd like to be able to maintain that rough sense of story length in anthologies with partial contents listed. That would only be possible with the "xx-yy" option. And using it as an option means we don't have any obligation to go back and "fix up" previous entries.

As an aside, and as one of the editors who's waiting for more complete Award support, I can tell you that the kind of support we can give to awards is well beyond what individual award pages or even Locus are able to give. Not everyone is interested in awards, but many visitors are -- especially when they see an award posted with a book they're looking at. And that's where most folks will see the awards: they don't go to the award and then look for the winners, they go to books they're interested in and then notice the awards that book has won. And they see all of the awards for that book at once. Is there anywhere else in the SF award world that someone can see that? (In a few cases, that happens on Wikipedia, but that's only for some books, and they usually don't include the smaller awards.) Wouldn't it be cool if you could go to an author's page and see all the awards they've won without having to click each book? Wouldn't it be cool if you could go to a single page and see all the books that have won some award this year? We may not get there, but I'm pretty sure none of those other award sites has any better chance of getting there than we do. Chavey 04:30, 15 June 2012 (UTC)

The page number field of an ISFDB record gives the page on which a work begins, just as a table of contents does. I don't recall ever seeing a table of contents which gives the range of pages on which a work appears, although I suppose it's possible. I don't see much value in going back through thousands of ISFDB records in order to give this data. Even if this issue had been discussed back in the days of the ISFDB's creation (and it may well have been), I would have argued against it. BTW, Darrah, I like your expansion of the ISFDB's award functionality. If they were implemented, I might be interested in that side of the database. Mhhutchins 22:16, 18 June 2012 (UTC)

Re: "Wouldn't it be cool if you could go to an author's page and see all the awards they've won?" - we do have that functionality, it's just one of the lesser known areas. Few people try out "Other Bibliographies For This Author" so just use "Summary" rather than "Awards" displays. E.g. Michael Bishop's Awards. Let me know if you spot any bugs/omissions in it - some of the untitled awards may mislink. BLongley 11:31, 19 June 2012 (UTC)

Thanks, Bill! I never knew such a list existed. Mhhutchins 18:35, 19 June 2012 (UTC)

I knew I had seen that list somewhere! But I couldn't remember how to find it again. As the awards are maturing, maybe we should consider making this more prominent, e.g. having an "Awards" link at the top of the Author Summary whenever that author has at least one award. I, at least, would find that a better use of that "above the fold" real estate than the long lists of tags and interviews that we sometimes see up there. Chavey 23:32, 19 June 2012 (UTC)

We'll need a little more development for that, and "books that one SOME award in a year" too. Especially when the outstanding changes go in and we can add any new award we think merits inclusion. BLongley 12:59, 20 June 2012 (UTC)

Chavey, thanks. A little suppport is welcome and it seems to me that you have understood me very well. I am addicted to awards/lists, but I don't find isfdb as a reliable/useful source for those.

Mhhutchins, have you seen the amazing link in my first post? Do you feel like it is complete/verified/trusted? Calin Hila 16:42, 22 June 2012 (UTC)

Yes, because I trust the editor who did a primary verification of the record, so therefore it is complete and a reliable source for the starting page of every piece in the issue. I can see how a newcomer to the database could be confused and think that the Clubhouse feature could be sixty pages long, but I'm not in a position to judge the logical thinking of the average database user. And to be clear, like Chavey, I understood your point very well. Understanding has nothing to do with agreement. Mhhutchins 17:40, 22 June 2012 (UTC)

BLongley, i know that functionality, but it would be better if we have some sorting options. We want Michael Bishop's Awards or Michael Bishop's Award? The answer is we want something better than both. (Mr. Hauck, please forgive me for linking to another imperfect site)

For all, I think that the final goal of a verified pub is TO LEAVE THE USER WITHOUT QUESTIONS AND WITHOUT DOUBTS. I have seen too many verified pubs that have differences between the cover and the data on the pub and the verifier doesn't bother to mention this. As a user I see only the cover (if there it is one) and the data confirmed by the verifier and if I see this differences I have doubts. And if that pub has two or three verifiers I consider that pub as not verified. And I would like to think that ISFDB it is not a closed circle.Calin Hila 16:42, 22 June 2012 (UTC)

As do we all. But your definition of primary verification is far too encompassing. Primary verification simply means that the editor has determined that everything present in the record is also present in the book, or if not, sourced in the notes. In a perfect world, no one would have questions or doubts. There's always going to be someone who questions data regardless of the extent of the detail. We can only hope that the majority of users find the data useful, not necessarily perfect. Attempting to cover all the bases, and to answer all the questions before they are asked is an unwise, and in the long run, unproductive course of action. Mhhutchins 17:40, 22 June 2012 (UTC)

Proposed name disambiguation help text

Here is a proposed help text revision to address this discussion above. Text in italics is unchanged (see Help:How_to_separate_two_authors_with_the_same_name).

Occasionally an author's name is identical to the name of another, different author. Although this is rare, it causes problems for the ISFDB if care is not taken to distinguish the authors.

For any data entered where the "author" field exactly matches an existing author, the ISFDB assumes that the newly entered author is the same as the existing author. The data is connected to that author, behind the scenes, without any intervention on the part of the editor.

In almost every case this is what is needed. However, there are some authors, such as Robert Frazier, for whom it is not correct. There are (at least) two different Robert Fraziers who have written works that appear in the ISFDB. One is a poet, born in 1951. The other wrote book reviews for Fantastic Universe in the mid-1950s. These are clearly two different people.

To distinguish between two (or more) people with the same name, the names must be made unique. This is done by appending parenthetical "disambiguating" information to the end of some or all of the names involved:

If one person is significantly more prominent/widely known in the Spec-Fic world than the others, e.g., Stephen King, that person's name should be left as-is, with no disambiguation appended. Otherwise, the names of all people of the same profession should be disambiguated. If you are unsure of a person's prominence in the Spec-Fic field, ask on the Community Portal or Moderator noticeboard.
If the people have different professions -- for example, one is an author, the other an artist -- append the profession in parentheses. If none is particularly prominent, add the disambiguation only to the non-author(s). For example, Chris Lynch the author and Chris Lynch (artist) the artist.
If the people have the same profession but work exclusively out of different countries, append the country in parentheses. For example, Chris Lynch and Chris Lynch (Australia).
Otherwise, use a range of birth-death years: "(1910-1990)". If the person is still living, omit the second year: "(1990-)". If the birth year is unknown, use a question mark: "(?-1990)" or "(?-)". For example, Colin Harvey (1960-2011) and Colin Harvey (1971-).
If no disambiguating birth, death, or residence information is known, the decade(s) during which the person produced works may be used as a last resort: "(1990s)". For example, James Cooper and James Cooper (1950s).

To separate works by multiple people credited to a single, ambiguous name, choose one new unique name according to the guidelines above, and go to each title and publication (if applicable) that should be credited to that new name and change the author name accordingly. The ISFDB will automatically create an author bibliography using that new name. If more than two people are involved, repeat this process for any others. When the original ambiguous name remains with only one person's works credited to it, that name can be changed if it, too, should be disambiguated. You can do this in a single step by Edit Author Data and modifying the author's Canonical Name.

Comments/suggestions welcome. --MartyD 13:42, 20 June 2012 (UTC)

Looks good, except that for number 5, I'd rather that the name be disambiguated with a roman numeral, similar to the IMDB method. In fact, I'd prefer this method to any other sort of disambiguation, but I seem to be out-voted in this case. Mhhutchins 14:21, 20 June 2012 (UTC)

Couldn't have written it any better, Marty!

With roman numerals you don't have an idea which author is which - the other solutions have always an implication. But relax, there will certainly be a case where all other disambiguating parameters don't fit (other than Romans). Stonecreek 18:57, 20 June 2012 (UTC)

I was trying to go by what has been past practice, while avoiding things that looked one-time only. I completely missed Roman numerals. I could strike #5 completely or remove "last resort" and save it for a new #6 about Roman numerals. --MartyD 01:28, 21 June 2012 (UTC)

If there's an IMDB entry with roman numerals, I think it would be nice to match their usage. It would also be nice to put in warnings about the "last name" field, we often end up with a profession/date automatically placed there, which takes people out of the author directory (at least until I run the cleanup script). BLongley 12:30, 21 June 2012 (UTC)

(unindent) How about if I take away #3 (Countries) and #5, replacing them with this new text as the final bullet:

#4 If no disambiguating profession or birth date is known, or these are still ambiguous, other unique text should be used. Some widely-employed differentiators include:

Country (in English) -- If each author worked exclusively out of one country, and those respective countries are different. For example, Chris Lynch and Chris Lynch (Australia).
Decade of work ("1990s") -- The decade(s) during which the person produced works. For example, James Cooper and James Cooper (1950s).
Roman numerals (I, II, and so on) -- Consider this especially if the person has a IMDB entry using a roman numeral differentiator, and then use the same one.

In general, attempt to follow existing precedent. If in doubt, ask for suggestions.

Does that do the trick for now? --MartyD 15:05, 21 June 2012 (UTC)

I would encourage us to not use Roman Numerals. If I'm entering a record about Chris Lynch, and I see two listings for that name, I can probably figure out which one applies to the data I'm entering. But if I see "Chris Lynch (I)" and "Chris Lynch (II)", I can't tell without actually going to their summary pages. I would also encourage us to err on the side of having disambiguation markers for almost all names that conflict, and keep the "unmarked" name somewhat rare. When someone enters a new record for "Bill Evans", without noticing that it's a duplicated name, the record will automatically go to the unmarked name. In this case, that's probably the wrong person for the record. This is generally true if the unmarked name is someone no longer writing, and whose bibliography is essentially complete. As an example, the bibliography for Bill Evans currently has a novel published 26 years after he died. If every Bill Evans required a disambiguation, then such data additions would, at worst, be collected in an "unknown Bill Evans" waiting for appropriate assignment. It would be reasonably easy to occasionally run a script to find all such "ambiguously assigned" works; but that becomes increasingly difficult if the number of ambiguous names with no disambiguation increases. (But yes, "Stephen King" should still be the Stephen King.) Chavey 16:52, 21 June 2012 (UTC)

Actually, when you're entering data, you should never consider whether the credited author already has records in the database. You enter it as stated. It's only afterward that you try to disambiguate the credit. If I come across a piece credited to "Bill Evans" I'm going to enter it that way. Mhhutchins 17:51, 22 June 2012 (UTC)

Ok, so in light of this discussion above, what do you suggest for wording? I'm not trying to resolve (or even rekindle, to be honest) a Roman numerals debate, I'm trying to fix the help text to reflect ISFDB practice. Roman numerals are used for disambiguation at the moment, and more widely than country names from the look of it. --MartyD 11:24, 22 June 2012 (UTC)

I'm OK with the wording reflecting current practice, and am happy to leave Roman Numerals to one of the last resorts. BLongley 11:40, 22 June 2012 (UTC)

I agree with Mr. Longley. Chavey 15:56, 22 June 2012 (UTC)

So do I, as long as Roman numerals remain an option. Mhhutchins 17:51, 22 June 2012 (UTC)

I agree to this proposal, Romans shoulkd remain an option. Stonecreek 18:17, 22 June 2012 (UTC)

(unindent) Done. Thanks everyone for the input. See Help:How_to_separate_two_authors_with_the_same_name. Feel free to edit that, should it need tweaking. All royalties will be shared equally. --MartyD 00:09, 24 June 2012 (UTC)

Lesson Plans

This pub is currently listed as containing The Picture of Dorian Gray and Dr. Jekyll and Mr. Hyde. However, per the publisher's description, it is actually lesson plans for teaching those novels. Technically, I supposed that qualifies as "Works about speculative fiction", but I'd lean toward not including lesson plans. Anyone else have an opinion? -- JLaTondre (talk) 22:24, 29 June 2012 (UTC)

We lean toward inclusion in such borderline cases, so I'd personally keep it in the database (after removing the fiction titles and unmerging it from the fiction title record, crediting it to Jayne R. Smith, the responsible party according to the OCLC record.) Mhhutchins 23:02, 29 June 2012 (UTC)

Good enough for me. Publication fixed (was waiting on decision to keep). -- JLaTondre (talk) 00:27, 30 June 2012 (UTC)

Gollancz SF Collectors' Editions

I know moderators have already been over this chestnut last year (and elsewhere), but I think we need greater accuracy on what constitutes a "Gollancz (US dist. by Sterling)" publication, or if we should do away with that variation altogether. The info trawled from Locus1 (eg. for titles here, here, here) indicates there were several reprinted editions for the US/Canada market that differed from earlier printings by the use of cover illustrations - eg. John Sladek's Tik-Tok: UK cover, US cover; Eric Frank Russell's Wasp: UK cover, US cover. The US editions are also physically different from those that came earlier by the removal of the front/back cover dustflaps. The majority of these titles are unverified. At the moment it's unclear if all these US editions had US/Canada prices included on the back cover or if they were price-stickered by Sterling over a solitary UK price. In one verified case, The Ophiuchi Hotline, the title's price sticker was placed over the printed US/Canada price just to increase the Canadian price by $1. My copy of that particular pub states "Distributed in the United States of America by Sterling Publishing Co., Inc." on the copyright page, but I really don't see that as a suitable justification for a "dist. by Sterling" appendage to the publisher, especially as the book only ever appeared in this edition in both territories: many later titles in the series saw a dropping of the typographic covers and dustflaps in the UK and were released with the cover illustrations, UK/US/Canada prices on the back cover and a "Distributed by" statement on the copyright page (eg. The Ophiuchi Hotline, All My Sins Remembered and Damnation Alley: these are currently listed as being published by "Gollancz" only.

Therefore at the moment we have duplicate unverified records for several titles that only differ unnecessarily between how the publisher has been identified, eg. The Masks of Time (Gollancz) and The Masks of Time (Gollancz (US dist. by Sterling)). They are the same book, and never had any earlier UK-only edition. Therefore I propose that either:

(1) the "Gollancz (US dist. by Sterling)" variation is kept only for the few US titles with cover illustrations that varied from their earlier, purely 'typographic' UK covers;

(2) The "Gollancz (US dist. by Sterling)" variation is done away with altogether, and for those few titles that did appear with BOTH covers, simply have a Note under the US edition stating something like "An edition produced for the US/Canada market, distributed by Sterling Publishing Co., Inc.". A verifier then can input any varying price sticker information, and leave it at that.

My preference here is (2), but I could also understand the bibliographic reasoning behind (1), or something similar. As the series appears at the moment I think it certainly needs some tidying up, and as I have most of this series in either UK or US format I'd be willing to work on this. PeteYoung 06:09, 6 July 2012 (UTC)

If it's certain that a title was reprinted for the US/Canada market by Gollancz, with the US/Canada prices actually stated on the book, and distributed by Sterling, then there should be a record, and it should credit only Gollancz in the publisher field, and the distribution can be noted. If a UK printing was stickered with new prices for the US/Canada market, and distributed by Sterling, there should not be a new record. The record for the UK printing can note that some copies of the printing were stickered and distributed in the US and Canada. In other words, the physical copy itself should dictate how it should be handled. I think this confusion grew because of the listings in the Locus database. So I would go with the second choice. Mhhutchins 15:34, 6 July 2012 (UTC)

I have a few of these and think like Michael, no distributor should be mentioned in the Publisher field. It would be like having a DAW / New American Library of Canada for every Canadian DAW printing, or adding Paperjacks to multiple US publishers [they distributed tons of paperbacks in Canada in the 70s]. Distributors come and go and are quite irrelevant in a Bibliographic sense, as they change nothing except a price. I agree a different cover means another record, but that's about as far as it should go. My 2¢ [CDN of course!!] --~ Bill, Bluesman 00:22, 8 July 2012 (UTC)

I side with Michael and Bill: no distributor should be mentioned in the Publisher field, thus (2). I've tried to clean up as many of those as possible, but they keep appearing again! (I guess somebody adds them from Locus1). The awkward publisher "Gollancz (US dist. by Sterling)" was created only as a temporary measure (somewhere on my talk page should be a discussion b/w myself and Michael about this) for unverified titles that arrived from Locus1. I also have quite a few titles in the series both with and w/o pictorial covers so would be happy to help. Cheers, P-Brane 01:40, 12 July 2012 (UTC).

Emailed stories: eligible or not

Awhile back, some editors (among them Swfritter and Dwarzel) decided that stories delivered by email should be considered eligible for the database. I don't recall if there was ever any discussion about that. In particular, it was the stories from Daily Science Fiction that started it. If it's been determined that this is an actual publication of the stories, I would like to enter the stories written by Bruce Holland Rogers for his short shorts subscription service. I've subscribed since mid-2005, and have received over 200 stories using this "publication" method. I think this subject deserves more discussion before I dive in. BTW, unlike the Daily Science Fiction service, there is a fee to subscribe to Bruce's stories. And his name is more highly visible in the field than the authors whose stories are published by DSF. Mhhutchins 17:10, 8 July 2012 (UTC)

For me, it's no to avoid opening the floodgates to all the pieces of fiction existing only in electronic form. If you wish to use the concept of threshold (of visibility or of reputation) to limit inclusion, it will be the same problem as for the Graphic Novels, which has, AFAIK, not been resolved. Only my 2 cents (of Euro).Hauck 17:35, 8 July 2012 (UTC)

And to a certain extent, I agree. But then, aren't there thousands of titles in the database already that exist "only in electronic form", ebooks, for example? We've made exceptions for certain webzines as well, and that hasn't opened the floodgates (thank God!) The problem with graphic novels is not in correspondence with this one. Allowing some graphic novels into the db is the same as allowing some nongenre novels into the db. They both have the same standards. Mhhutchins 17:56, 8 July 2012 (UTC)

Well, as an old-fashioned guy, I think that all "only-electronically-avaliable" texts shouldn't be included. For the graphic novels, they should be automatically included as one "half" of the work, namely the narrative part (in French the "scénario") in opposition to the drawing part, is clearly "fiction" and "speculative".Hauck 18:33, 8 July 2012 (UTC)

Under your "only-electronically-available" scenario, ebooks which have never seen print would be ineligible for inclusion. And that was debated and settle many moons ago. Mhhutchins 18:59, 8 July 2012 (UTC)

I don't think i understand you. Why do you think that "only-electronically-available" texts should NOT be included? We have been including at least some of them for many years now.

As to Graphic Novels (defined as works where the text would not convey the meaning without the images) we generally exclude them, with a very few special case=by-case exceptions. Partly this is the fear that they would lead to the inclusion of "Superhero" comic books and similar forms, and partly that we are not set up to capture the essential data elements to properly record such works. (There may be other reasons as well.) -DES ^Talk 18:50, 8 July 2012 (UTC)

All the above was my opinion (if I'm entitled to have one), which doesn't interfer (too much, I hope) with my duties as moderator and my contibutions as an editor. My problem is just with the principle of the "case=by-case exceptions". I believe that the best rules are the simplest. If graphic novels are out, no problem, it's a common policy that I'll stick to (I can use this specialized site). What always puzzle me are the exceptions ("All graphic novels are out, except those scripted by Gaiman"). What was decided moons ago could (and sometimes had) change. Hauck 14:46, 9 July 2012 (UTC)

You've misunderstood if you believe this is an exception just for Gaiman's work. I'm not going to waste time trying to point out the distinction. But let's just say that if the policy actually states "All graphic novels are out, except those scripted by Gaiman", then that would be the policy. Stating an exception doesn't muddle the policy. It clarifies it. I am forever surprised, even though by now I shouldn't be, just how often a discussion about one single issue evolves into something beyond the original issue. Maybe one day, we can have a single-issue discussion, make a decision about it, and then go on to another issue. I'm not holding my breath. Mhhutchins 21:29, 9 July 2012 (UTC)

Always nice to have found a real living compass, capable of bringing us back on the true path of the discussion. Now that you found it, permit me to leave, too much rightness is not that interesting. Hauck 14:05, 10 July 2012 (UTC)

And it's reassuring to know that they'll always be those who will try to steer in a different direction just for the sake of being contrary. We need them along to stir the shit so that the rest of us can make informed decisions about where this craft is going. Mhhutchins 14:48, 10 July 2012 (UTC)

"Robust debate" is all good and well, but let's not get dragged across the line separating it from "clash of personalities". It helps to keep things in perspective: whether Nora Roberts' non-genre works and Gaiman's graphic novels are included or not will not affect 99% of the project, but personal animosities can set it back. Ahasuerus 15:13, 10 July 2012 (UTC)

It's very difficult to participate in a debate when one side is unable to keep on subject. That's all I was trying to point out. And when I'm personally attacked for this, I'm far too human not to react. This will be my last post on the topic. Maybe one day we'll be able to have a civil discourse on the subject. Mhhutchins 15:21, 10 July 2012 (UTC)

I'm sure such has happened, but agree it's rare. BTW, do you have an opinion on - oh wait, forget I mentioned it. BLongley 22:20, 9 July 2012 (UTC)

You're certainly allowed to have an opinion, and the times we try and stretch the Rules of Acquisition are good times to express such - especially as people like us will have to moderate them. I'm fairly inclusionist, but leave a lot of submissions to people with a firmer opinion. BLongley 19:56, 9 July 2012 (UTC)

As long as you take responsibility for them (as presumably only subscribers could verify) I'd be OK with them. BLongley 18:02, 8 July 2012 (UTC)

It seems to me that the issues and possible problems are much the same as with webzines, and they should be allowed only under similar conditions. -DES ^Talk 18:26, 8 July 2012 (UTC)

Infected Author pages

"www.pinknoise.net" is Leonid Korogodski's Web page, but some of the sub-pages apparently host viruses (or so Avira tells me.) Should we (a) ignore the URL, (b) add it to the author's Wiki page with a warning, or (c) enter it in the main record? I am thinking that (b) may be the best choice in case it's a false positive. Ahasuerus 03:49, 10 July 2012 (UTC)

I'd err on the side of caution and not add it as a link till it's been cleaned up. You have notified the webmaster I presume? BLongley 11:33, 10 July 2012 (UTC)

I have now :) Ahasuerus 15:20, 10 July 2012 (UTC)

Plays - title date

For plays, should we use the date that the play was originally performed or the date when it first appeared in print? For example, Jules Verne's Voyage à travers l'impossible was first performed in 1882, but the manuscript was only found in 1978 and wasn't published until 1981. We currently list 1981 as the Title date, but I wonder if we should change it to 1882. Ahasuerus 06:19, 10 July 2012 (UTC)

I think we should stick with publication date. Performance dates are possibly worth noting - e.g. if we didn't have Wikipedia links for all our Twilight Zone, Star Trek and Buffy Stories then original broadcast dates could be of interest. BLongley 11:31, 10 July 2012 (UTC)

I agree that first publication date should be used. Many of the Twilight Zone scripts weren't published until decades after they aired, and as Bill said, those original broadcast dates have been recorded in the title records' note fields. Mhhutchins 14:51, 10 July 2012 (UTC)

I agree. publication date in print is what should be used, with a possible mention of first performance date in the notes if it is known. Should there be a mention of this in the help? -DES ^Talk 15:21, 15 July 2012 (UTC)

I doubt more help can do any harm. Our big problem is that people usually only read such once.:-/ BLongley 18:45, 15 July 2012 (UTC)

The changes look good. I also went ahead and rewrote the first section of the Help page for clarity and to bring it up to date with the current standards. For example, we don't use author-specific Project pages much, so it's better to ask questions about VTs and their dates on the Help page. Please take a look-see when you have a chance. Ahasuerus 22:14, 15 July 2012 (UTC)

Looks pretty good to me. However, by "the Help page" do you mean the Help Desk? if so, i think we should be specific, and should also link there. Otherwise people may think this means to discuss on one of the Screen Help pages where the template is transcribed, which would surely be a mistake. -DES ^Talk 23:22, 15 July 2012 (UTC)

Thanks, fixed. It's been my experience that no matter how many times you proofread documentation, you always miss something obvious :) Ahasuerus 01:27, 16 July 2012 (UTC)

"/ BCE" on publisher of Book Club editions?

While reviewing an edit to a (non-SFBC) book club edition of The Last Space Ship, I noticed no trailing "/ BCE" on the publisher. I was going to make a comment to the editor and provide a help reference, only to find Help:Screen:EditPub#Publisher (et al.) only calls out this treatment for SFBC and makes no mention of non-SBFC book club editions. A quick search shows use of "/ BCE" is quite widespread. Do we want to consider "/ BCE" a standard treatment that should be codified? If so, how about "/ QPBC", and are there others? --MartyD 10:16, 16 July 2012 (UTC)

Use of /BCE or /QPBC or any similar club (if it isn't common a comment in the notes would be good) seems reasonable to me. -DES ^Talk 14:00, 16 July 2012 (UTC)

BCE started as primarily the appendix for publications that state the words "Book Club Edition" on their front flap, and the editor is somewhat certain that it was never a selection of the Science Fiction Book Club. Doubleday printed this on the flaps of all of their book club printings, regardless of the club for which it was intended. This practice ceased in 1988 when Bertelsmann bought Doubleday and its book clubs. I've noticed that " / BCE" has been used for non-Doubleday book club printings, but as long as the editor makes it clear why they believe the book is a "book club edition" I see no problem with its use. Whether this use should be documented...well, probably, but I'm not going to be a part of that end of the discussion. Mhhutchins 14:34, 16 July 2012 (UTC)

Forgot to add, it if wasn't already apparent by my comments, "BCE" is technically not a club. "SFBC" and "QPBC" are actual book clubs. Mhhutchins 14:35, 16 July 2012 (UTC)

I should have written "Use of '/ BCE' or better a specific club indication such as '/ QPBC'..." above -DES ^Talk 14:43, 16 July 2012 (UTC)

Fantastic Art Portfolios

I have a fair number of art portfolios by Frank Frazetta, Virgil Finlay, Hannes Bok, Stephen Fabian and similar artists where the artwork is primarily from books or magazines in the data base. In addition, I have dozens of art books containing the work of single artists. I would like to see a new PUB TYPE: PORTFOLIO. It would be similar to the COLLECTION form. Under PUB FORMAT I would expect to see "hc" or "pb" for the single-artist books or "plates" for true portfolios. Complications could be how to include both the artist and the editor if they are not the same person, and perhaps how to handle the cover artist. I can't think of any off hand, but there might be a case where the cover artist is not the featured artist. It is tempting to just assume the "Artist" is the featured artist and to handle a different cover artist in the notes. Another consideration for PORTFOLIOs is the size of the plates, since larger size usually means finer detail. There are some PORTFOLIOs where the plate size varies. Is it reasonable to replace the page number with plate size (where there are plates), or is the software change too complex? Comments or thoughts? Bob 17:36, 19 July 2012 (UTC)

Reviews of books otherwise OUT

copied from User talk:MLB#Talesbones #18 again -DES ^Talk 21:06, 19 July 2012 (UTC)

The quick ISFDB definition of "nongenre" is a FICTION book that is not speculative fiction. Reviews of nongenre works by nongenre or marginal authors should be entered as an ESSAY type, not as a REVIEW type, along with reviews of films, recordings, comic books and graphic novels. The recipe book and beer guide are NONFICTION, but they are by writers who may not be "above a certain threshold", an admittedly nebulous qualification. In my opinion, you should only make REVIEW type entries for nonfiction books if the book itself is eligible for inclusion in the database, but the stated rules aren't so clearly defined. There was a discussion that opened the door to allow all books (whether nongenre or nonfiction) reviewed in genre magazines to be eligible for the database. I consider this a slippery slope, so I personally have a probably one-man war to make these reviews into essays, so that publication records aren't created for them. Creating publication records for such titles was the solution posed by others so that the reviews could be linked and to avoid stray authors, i.e. authors in the database without publication records. I think my method keeps the integrity of the database intact, and I feel comfortable with this unilateral decision. None of the changes I make to the database result in lost data. Mhhutchins 19:55, 19 July 2012 (UTC)

I thought it was an accepted standard that works reviewed in major genre publications, and perhaps in ANY genre publications were IN and should be included. I have added a number of publications of such works, and intend to continue doing so when such items come to my attention -- I don't plan to form a project and seek them out -- unless there is a clear consensus against this. In fact if I encountered such an "ESSAY" review, at least in a pub I had in hand, I would be inclined to convert it back to a review and enter the publication. -DES ^Talk 20:48, 19 July 2012 (UTC)

I copied this here from User talk:MLB to get a wider consensus if possible. -DES ^Talk 21:07, 19 July 2012 (UTC)

Then you should create publication records for books about teacup fortune telling, self healing, reincarnation, and the politics of energy, all by authors that have no publication records in the database. What is the point of having such publication records in the database? Simply because a REVIEW record can link to it? Seems rather pointless to me, and compromises the integrity of the standards. You have the ability to create any publication record you want, but do me a favor though. Don't convert the ESSAY records to REVIEW records without notifying the primary verifiers. I have to point out again that my converting these reviews to ESSAY records do not result in any loss of data, and in fact clear up such anomalies as this. Mhhutchins 22:00, 19 July 2012 (UTC)

I did say that I would tend to do this on works I have in hand. Of course if such pubs are primary verified I would not make changes without consulting the PV first.

What I am really after, and why I copied this discussion here, is an agreement on whether or when such items should be entered as REVIEWs and the corresponding publications entered, and when they should be left as or converted to ESSAY type. Then we can all work under the same standards. If the consensus is that such itmes should not be linked to via REVIEWS, I will of course comply. -DES ^Talk 22:34, 19 July 2012 (UTC)

I prefer the ESSAY approach for anything that wouldn't be "in" on its own, the exception being that I think the existence of the review should tip the balance in favor of inclusion in borderline cases. I think having a clear rule along those lines would work out better than trying to decide individual cases based on the merits of the publication in which the review appears and/or the nature of the non-genre/non-fiction publication (or its author) being reviewed. Do we really want the collection of kids' letters to Barack Obama included because Stephen Theaker reviewed it in a TQF? I don't think so, regardless of whether any editor has the pub in hand. So, then, how would we exclude it? TQF not "major" enough? --MartyD 12:58, 20 July 2012 (UTC)

I don't think the publication in which it appears should have any bearing on the matter. It seems pretty clear to me: if a book would not otherwise be eligible for inclusion then the review should be made into an essay. But in borderline cases, as has become the de facto standard, I lean toward inclusion. But when you come down to it, it's always going to be taken on an individual case-by-case basis. If it were clear cut, then there would no reason for discussion. Mhhutchins 16:05, 20 July 2012 (UTC)

I don't like dead-end links, hence the cleanup script. When it comes to inclusion of the reviewed works, I tend to include science books, when I can find details. The more nebulous the science the less likely I am to include them - so fortune telling is right out in my opinion. Nongenre is in only if the author has reached the certain threshold. BLongley 18:47, 20 July 2012 (UTC)

"I don't like dead-end links, hence the cleanup script." I think you mean authors without publication records. The script doesn't look for dead-end links or un-linked reviews. It looks for authors that are in the database with reviewed titles but have no publication records. I wish there were a script for un-linked REVIEW records. But I imagine the results would be an extreme number of records. Mhhutchins 19:17, 20 July 2012 (UTC)

It would be fairly easy to add; I think the only reason why it hasn't been created yet is that Bill wanted to go for the low hanging fruit first. Ahasuerus 05:13, 21 July 2012 (UTC)

Yes, it's trivial, but you still have quite a backlog of software changes to work on. I could post it as a project page for now? There's about 3,000 of them. BLongley 11:41, 21 July 2012 (UTC)

Actually, there's less than that, I'd forgotten Variants of Reviews. Maybe 1200. BLongley 12:35, 21 July 2012 (UTC)

[unindent] I'd be interested in seeing such a list, but would be afraid that someone who take this as a call to add nongenre publications in order to link the reviews. Please create a project page when you get a chance and one of us will add a caveat about its purpose. There must be plenty of reviews that don't link simply because the author and/or book title didn't match exactly to an SF title already in the database. Thanks. Mhhutchins 13:46, 21 July 2012 (UTC)

OK, created. Feel free to break it up into sections if multiple people want to work on them. BLongley 14:54, 21 July 2012 (UTC)

I can't wait to see if someone creates pub records for the "reviews" in this publication. Somewhere Steven H Silver is laughing at us for the uproar over his middle initial. Mhhutchins 16:27, 21 July 2012 (UTC)

Nongenre short fiction

Please see User talk:Mhhutchins#Nongenre short fiction for a recent 2-person dialog on this.

Currently, ISFDB:Policy#Rules of Acquisition says (#4) "Works (both fiction and non-fiction) which are not related to speculative fiction, but were produced by authors who have otherwise published works either of or about speculative fiction over a certain threshold (see below). This includes any non-genre works published as standalone books as well as non-genre short fiction, but exclude non-fiction which was not published as a standalone book." (emphasis added).

However, Help:Entering non-genre magazines#Why one should enter non-genre magazines says: "...many well-known (and many not so well-known) [SF] stories were first published in non-genre magazines. In such cases it is desirable to record such publications in order to make the bibliography of the story complete. However, the ISFDB is not a general-fiction index and the non-speculative-fiction contents of such a magazine should not be entered." (emphasis added).

Some moderators, relying in part on the non-genre magazines help page, and in part on their sense of what belongs in the ISFDB, have, I understand, been instructing editors not to enter non-genre fiction found in non-genre magazines or anthologies even if it is by an author clearly above the "threshold". This contradicts RoA #4 above. Perhaps RoA #4 should be changed.

It seems to me that the non-genre magazine help page quote above was not intended to restrict the RoA, but merely to warn against automatically entering the entire contents of a non-genre magazine (or anthology) just because there was a single work of SF present. (As the drafter of that page, I am certain it was not intended to restrict the existing RoA.)

I suggest that either 1) the "above the threshold" exception be written in to the non-genre magazines help page, or 2) RoA #4 be amended to exclude non-genre short fiction. I tend to favor #1, but could live with either, as long as there is a general consensus that we can all follow.

We don't have a NONGENRE type for short fiction at this time, which makes #1 a bit trickier, and if we choose #1, there might be more urgency about adding one. But such a type has been in the plans for quite a while, which implies a need for it, and if we go with #2, no such type would ever be needed. This suggest to me that #2 is more of a change in our previous vision than #1 is.

In any case I think this is a clearcut yes-or-no choice, one we can and should make relatively promptly. Perhaps, especially if we go with #1, we would need to create ISFDB:Authors considered "Above the Threshold". The page ISFDB:List of Authors with Non-Genre Work in the DB‎ could be a starting point, with lots of deletions. -DES ^Talk 19:50, 21 July 2012 (UTC)

Give me a NONGENRE SHORTFICTION type and I can support #1. Until then, I suggest a third option: follow the current rules concerning NONGENRE magazines, and ignore RoA #4's allowance of non-genre short fiction for "above the threshold" authors. That's what most of us have been doing anyway. I've not seen any clamor for adding all of Fredric Brown's mystery stories, or even Edgar Pangborn's for that matter. Mhhutchins 21:11, 21 July 2012 (UTC)

Your #3 would be basically adopting #2 without explicitly changing the RoA and revisiting the matter when/if there is software support for non-genre short fiction. Have I understood you correctly? -DES ^Talk 23:02, 21 July 2012 (UTC)

Yes, that's what I suggest: leaving the status quo until the software can support more aggressive handling of non-genre work. Mhhutchins 23:43, 21 July 2012 (UTC)

I agree that including non-genre short fiction is currently counterproductive because there is no easy way of telling genre stories from non-genre stories on Summary pages. I think we should update the RoA to state that non-genre short fiction will be included at a latter point when software support becomes available. Ahasuerus 05:06, 22 July 2012 (UTC)

(unindent) what would you say to the following revised version of RoA number 4?

In - Works (both fiction and non-fiction) which are not related to speculative fiction, but were produced by authors who have otherwise published works either of or about speculative fiction over a certain threshold (see below). This includes any non-genre works published as standalone books, but excludes non-fiction which was not published as a standalone book. Thus, Poul Anderson's mysteries and his non-fiction book about thermonuclear weapons will be included, but Gregory Benford's and Robert L. Forward's professionally published scientific articles will be excluded. Non-genre short fiction is excluded until software support for such works is implemented.

Would that cover the case? -DES ^Talk 14:12, 22 July 2012 (UTC)

We may want to clarify that we shouldn't be deleting pre-existing non-genre short stories. Perhaps something along the lines of "non-genre short fiction is currently on hold until software support for such works has been implemented"? Ahasuerus 04:49, 23 July 2012 (UTC)

I agree with the suggested re-write of In, and with Ahasuerus' amendment. Chavey 04:51, 23 July 2012 (UTC)

How about

In - Works (both fiction and non-fiction) which are not related to speculative fiction, but were produced by authors who have otherwise published works either of or about speculative fiction over a certain threshold (see below). This includes any non-genre works published as standalone books, but excludes non-fiction which was not published as a standalone book. Thus, Poul Anderson's mysteries and his non-fiction book about thermonuclear weapons will be included, but Gregory Benford's and Robert L. Forward's professionally published scientific articles will be excluded. New entries of non-genre short fiction are on hold until software support for such works is implemented. Existing records of non-genre short fiction will not be deleted because of this hold, but such records should ideally carry a title note indicating their non-genre status.

Will that do? -DES ^Talk 16:25, 23 July 2012 (UTC)

I'm a little confused about several statements above stating that we don't have software support for nongenre short fiction. The help pages for NewPub and EditPub clearly indicate that we should use NONGENRE for such stories. Under Entry Type:

NONGENRE. Used for fiction which is not speculative fiction. Note that there are stories that are widely known within sf, such as "Everyone Knows Joe", by C.M. Kornbluth, which are not science fiction by any means, but which are often included in genre collections for one reason or another -- in this case, because of Kornbluth's prominent reputation in the field. These stories can be marked NONGENRE. You do not generally need to read every story to check if it is a genre story or not; stories in genre publications can be assumed to be genre unless you have some specific reason to believe otherwise.

And under Length:

Length - The length of the item in words. Note that if you are cloning a publication, this field is not editable for existing content records. The length field should be left blank except for SHORTFICTION, and in a couple of other cases where it makes sense: NONGENRE and NONFICTION, for example.

I've searched through the R&S archives and can't find anything stating that NONGENRE shouldn't be used for short fiction, but that certainly seems to be the understanding that everyone is is working from. Even the specific example in the help pages appears to have used SHORTFICTION instead of NONGENRE. Working on the same assumption that everyone else has (i.e. NONGENRE means non-genre novel?), I would question what we should do when we encounter a story by an above the threshold author and the genreness of the story is not known, especially when working from secondary sources. Should the presumption be that a story is speculative and in? --Ron ~ Rtrace^Talk 22:04, 5 August 2012 (UTC)

The NONGENRE type should not be used for SHORTFICTION because there's no way for a title record to be both. Try it. It can't be done. Same situation with a NONGENRE COLLECTION or NONGENRE ANTHOLOGY. If you edit a SHORTFICTION, COLLECTION or ANTHOLOGY type record to NONGENRE, then the piece will appear on the author's summary page as NONGENRE and there's no way for the user to know whether it's a novel, a short story, a collection or anthology. As the type was first designed, and the only way that it works without any problem, is for the user (and editor) to assume that NONGENRE means NONGENRE NOVEL. Any other use of the type results in improper sorting and display. So, in other words, despite what the Help pages say, there is currently, nor have there ever been, software support for NONGENRE SHORTFICTION. Mhhutchins 23:35, 5 August 2012 (UTC)

To answer your question: as the help section explains, you can assume that a piece of SHORTFICTION in a genre publication is genre (sf), thus "in". Inversely, you have to assume that SHORTFICTION in a nongenre publication is nongenre (not sf), thus "out". If you've read the piece or have data from a reliable secondary source that the SHORTFICTION is in fact speculative fiction, then you can create a record for the publication, but only enter the spec-fic SHORTFICTION work and nothing else. The "thresholdness" of the author should not be a factor in this decision. That's the point of this discussion. There will be nongenre SHORTFICTION by these "over the threshold" authors, but unless they're printed in a genre publication(a spec-fic magazine, collection or anthology), you should not create a record for the work until there is software support for NONGENRE SHORTFICTION. Mhhutchins 23:43, 5 August 2012 (UTC)

Will that do? - Yep. Works for me. Kevin 02:07, 6 August 2012 (UTC)

Non Genre Magazines - Cover Scans and Pub Format

The subject of linking to cover scans of non-genre magazines came up in the Community Portal. I also ran across this discussion from 2009 that doesn't appear to have ended in a resolution. I'd like to raise the issue again about including cover scans for non-genre publications and additionally question the prohibition in the help page on entering the Pub Format for these publications. I can't think of any reason for prohibiting the entry of Pub Format. There were a few reasons given in the earlier discussion for not including cover scans (unless they illustrate SF content). The concern about the storage space for such images could be addressed by either limiting to or preferring that we link to external sites that have granted us permission. The concern that we should attribute the artist if know (and not an SF artist), but we don't want to list the cover artist could be handled by attributing such artists through the publication notes. I feel that linking to the cover scan is valuable to identify the magazine. I would propose that we allow such images, and the Pub Format to be included for non-genre magazines. The reason the recent discussion came up is that I have linked to cover scans when the SF content was merely mentioned on the cover (example). I had misunderstood the requirement that the image had to illustrate the cover, as opposed to the text in the cover referring to the SF story. If we can't reach a consensus to include covers for all non-genre magazines, perhaps allowing covers that mention the SF author or story would be a good compromise. --Ron ~ Rtrace^Talk 00:03, 6 August 2012 (UTC)

I believe the original reason why the covers should not be linked to the record is still valid (I don't think the issue of storage even came up but I could be mistaken. You might ask David Siegel who wrote the first draft of the rules on nongenre magazines.) This is a speculative fiction database, and we only allow nongenre publications under specific circumstances. The record is only in the database because of the speculative fiction content. There is really no reason to have any nongenre magazine record in the database, and there are ways to enter the data in the database without a publication record, but I lost that battle long ago.

Linking the cover to a nongenre magazine gives very little value in identifying the issue for the user who comes to the database for speculative fiction data. Shouldn't the issue date be enough to identify it? More importantly, adding the cover image is just one more step to adding other elements of the publication. The next thing (as you mention) is that someone will want to record the artist of the cover. After all, they'll say, we are using the artist's cover art to identify the issue. The next step is to add nonfiction which is only tangentially related to spec-fic: an article about the movie Aliens. Why not? It's sf. How about an interview with David Cronenberg? He makes those weird movies, right? Why don't we add an article about Arnold Schwarzenegger? He acted in several sf movies. The point I'm trying to make is that adding the covers of nongenre publications is the thin edge of the wedge. I don't want to have to explain to new editors (who've come across an ISFDB record with a cowboy on the cover) why we don't allow a publication record for a movie magazine with an alien on the cover and has an article about a science fiction movie. This is not hypothetical. It happened and the discussion went back and forth for days, taking much time, effort, and goodwill between both the moderator and the editor. Those of you who don't work with on new editors' submissions don't have this problem. It's one of the reasons I'm severely curtailing my moderator duties. Mhhutchins 00:41, 6 August 2012 (UTC)

I'm in favor of having covers for non-genre magazines. Perhaps along with a simple note in the publication record "Cover Artwork shown for identification purposes only." An added line in the help stating for non-genre magazines "The artist (if known) should not be entered into the index, but may be entered into the publication notes" would also seem appropriate. Kevin 02:14, 6 August 2012 (UTC)

Rather than just being in favor of it, perhaps someone can give a reason why we should link nongenre covers to records for nongenre magazines in a speculative fiction database. Mhhutchins 03:20, 6 August 2012 (UTC)

As stated above... for identification purposes. My specific example is so that someone can recognize a magazine (or confirm a magazine quickly) during a live sale event (like at a con), or in an online listing (like on ebay), especially when multiple magazines are pictured poorly and with little detail in the listing. Kevin 03:37, 6 August 2012 (UTC)

Hmmh? So the images at Galactic Central and FictionMags aren't good enough? A listing on eBay wouldn't have an image? Someone would come to the ISFDB to identify a certain issue? We know they wouldn't come here to get the contents (unless that's the next step: adding all the contents), but they'd come here to get the image? An image that we have deeplinked from...Galactic Central? Someone at a "live sale event" would pull out their smartphone and go the ISFDB to see if the magazine being sold is the right one? Wow, someone had better get a better source to get their data for non-speculative fiction magazines! Mhhutchins 03:53, 6 August 2012 (UTC)

Based on previous conversations, I suspect bandwidth is more of an issue than storage. But that's really not what is being argued.

I think the real point is that the ISFDB can't be all things to all people. Some basic lines need to be drawn in the sand as to what's in and what's out. While is it easy to imagine a use case for a particular change in rules, that doesn't necessarily mean that the greater public is clamoring for that same feature to be rolled out. And the rules and workflows here are complicated enough without adding more exceptions to the exceptions, which means an even steeper learning curve for new editors assuming they attempt to make a good-faith effort to read the rules. (And sure, many don't even read them, or at least don't read the bits that would be most helpful in any given situation.) And the steeper the learning curve for editors, the more work that winds up hoisted on the moderators. Albinoflea 07:33, 6 August 2012 (UTC)

Thanks for stating my position with much more clarity and a lot less sarcasm. Sometimes it's hard to make my point when the logic parts of my brain have become overheated. Mhhutchins 14:49, 6 August 2012 (UTC)

Don't get me wrong; I submitted artwork for some non-genre magazines once before I realized that went against the rules, so I can see why some people might find them useful, but I'm not sure that including them should become a priority. And there are things like Uncorrected Proofs and Advance Reader Copies that I'd love to start cataloging, but they're out, so I can't. And that's OK, because in my opinion there's a lot of value in what is here, and there's a lot of work still to be done getting what is here into better shape. Albinoflea 07:33, 6 August 2012 (UTC)

I see no intrinsic harm in linking to a cover. I like to see covers for all publication entries. Aside from identification, being able to point to an actual cover gives a small amount of provenance to the information -- i.e., yes, it's a real publication. But where we (rightly, IMO) don't want the clutter of the artists or the storage of uploaded covers, an "all-non-SF-cover-info-is-out" approach seems clearest. --MartyD 10:23, 6 August 2012 (UTC)

I see no harm in linking, I agree that it adds some provenance. But for the sake of bandwidth and storage I think we can draw the line at the "must illustrate the SF content" before we allow a cover image onto the ISFDB site. BLongley 12:13, 6 August 2012 (UTC)

(shrug). I would like to know what the cover of Boy's Life August 1950 looked like without leaving the database. (I found it here. It's a Chesley Bonestell Cover. (There is interior artwork by someone else as well)) - Now that I've done this original research, It sure would be nice to catalog it, so the next time I want to know it's already here. Let's take it one step further, what about the 4th part, of Space Scout in Boy's Life November 1950. Nope, not spec fic related, just a picture of some kids hiking. but I've done the research, and the answer is 'no, the cover is not spec fic related'. If we don't then put the cover in the database... someone may research it again (and someone else yet again). By putting it in the database once, the matter is closed. "Here is the Cover. The Cover is mundane." Kevin 14:16, 6 August 2012 (UTC)

By not putting it in the database at all, the matter is also closed. If you're so convinced that's why people come to the ISFDB, just add a note that this issue does not have sf-related cover art. (I don't even see the point in that effort.) If the ISFDB user wants to know what a cover looks like, he can go to a source that does have the cover. They would have to do that for thousands of publications in the db that do not have linked cover images. I'd rather the effort go to linking sf cover art than linking images of boy scouts and pirates. And BTW, there is so little data in this record that it's practically a waste of effort for anyone to even click on a link to it at all. It only reinforces my belief that records for nongenre magazines add no intrinsic value to the database. Mhhutchins 14:49, 6 August 2012 (UTC)

[unindent] I just uploaded the cover image to Boy's Life August 1950. Nothing in the current standards prevented anyone from doing this. Mhhutchins 18:45, 6 August 2012 (UTC)

I have no problem with lines describing what is in and out of the database, but that isn't really what we're talking about here. Rtrace^Talk 23:25, 6 August 2012 (UTC)

Aren't we talking about whether to change the rule about allowing the linking of non-sf covers to non-genre magazine records? So it is a matter of what is in and out of the database. Even if it's not on our server, if it's visible to a user it's in the database. Mhhutchins 03:01, 7 August 2012 (UTC)

It has already been decided that nongenre magazines that contain sf are in. Rtrace^Talk 23:25, 6 August 2012 (UTC)

Not really. Only the speculative fiction and related art (including covers) published in a nongenre magazine are "in". Nothing else. The root of my objection to changing the rules is embodied in your statement, which on the surface to many editors is correct, when in fact it's not. Mhhutchins 03:49, 7 August 2012 (UTC)

My point above is that I'm not arguing that new content records be allowed in. I'm only arguing that we allow currently prohibited metadata for a container record (non-genre magazine) be allowed where the container is already allowed.--Ron ~ Rtrace^Talk 12:41, 7 August 2012 (UTC)

We're really discussing how fully we want to describe these publications. Nobody is advocating that any non-sf content (non-meta data) be added. (Inclusion of Micheal's examples of sf movie articles probably merits a separate discussion, but they're not really the issue here.) Rtrace^Talk 23:25, 6 August 2012 (UTC)

The point I was trying to make about "cowboy" covers was that allowing them to be linked to ISFDB records is the first step toward adding more information to nongenre magazine records than are currently allowed into the db, such as sf-related nonfiction. You're right, it's not directly related to the discussion, but as I was saying, linking non-sf covers might be just the toe in the door to allowing more data. The next discussion might be about adding artist credit to these non-sf covers. Mhhutchins 03:01, 7 August 2012 (UTC)

I suspect that I would probably also disagree with you about sf-related nonfiction, but that is a separate discussion. I'm not advocating for the inclusion of image urls and pub format as part of any larger agenda. Obviously, if we were concerned about slippery slopes, and I'm not particularly, we shouldn't have allowed these publications in in the first place.--Ron ~ Rtrace^Talk 12:41, 7 August 2012 (UTC)

There are 4 pieces of metadata that are treated differently for these magazines. I'm not advocating a change in how Cover Artist is handled, because despite its being publication metadata it creates a content record that is clearly out by the ROA (though, strangely, we are allowed to add an actual editor's name in addition to the "Editors of..."). The two remaining metadata fields, Pub Format and Cover Image URL are oddly prohibited. Prohibiting those fields is already the exception to the original exception of allowing a non-genre magazine to be listed at all. If the policy is to only allow the bare minimum necessary to identify a non-genre publication, then why do we allow price or page count? Rtrace^Talk 23:25, 6 August 2012 (UTC)

Read the discussions that took place in the writing of the rules here and ask David Siegel (DES) to join in this discussion. He spearheaded the integration of non-genre magazines into the db and wrote 90% of the help page. Mhhutchins 03:01, 7 August 2012 (UTC)

I have and I did. I can't find anything about the reasons for prohibiting the two metadata items. Merely that URLs are not allowed, and nothing about pub format. Perhaps David can recall his reasoning. --Ron ~ Rtrace^Talk 12:41, 7 August 2012 (UTC)

If a neophyte editor is confused by the picture of a cowboy so that he ignores the ROA and begins including Zane Grey novels, I think we've got larger problems with that editor and I don't really think the image is the cause of the problem. It sounds like storage and bandwidth may not be that much of a concern. If they are, that's probably something that ought to be addressed in an overall discussion of images. I do think that these nongenre magazines are interesting and without these records we would have no way of entering sf cover art that actually illustrates sf content. I'll also reiterate my possible compromise of allowing cover urls where the cover illustrates sf content, though the artwork on the cover may not. --Ron ~ Rtrace^Talk 23:25, 6 August 2012 (UTC)

I'm confused by this last statement: ...allowing cover urls where the cover illustrates sf content, though the artwork on the cover may not. How can a cover illustrate sf content when the artwork doesn't? Mhhutchins 03:01, 7 August 2012 (UTC)

I've taken it to mean situations where the cover artwork might be non-SF (e.g. a photo of a rock band) but there's some text on the cover "AMAZING NEW STORY by IMPORTANT SF AUTHOR" that indicates that there is SF content withing the cover, but I could be wrong. Albinoflea 05:42, 7 August 2012 (UTC)

That's exactly what I meant. As shown in this example from the opening paragraph.--Ron ~ Rtrace^Talk 12:41, 7 August 2012 (UTC)

We currently host a little over 4.5GB worth of images, mostly cover scans. They haven't presented bandwidth or storage issues so far and at the rate we are growing, they should continue to be manageable in the foreseeable future. I doubt that adding the occasional non-genre scan will make a difference should we decide to change the rules. Ahasuerus 00:21, 7 August 2012 (UTC)

I have to reiterate: storage was not an issue during the discussion about linking non-sf covers. I don't even think we discussed the reason why we shouldn't link non-sf covers. It was so logical that there was never a question that they should be linked. As Albinoflea put it so eloquently above "The ISFDB can't be all things to all people." This is a specialty database as the name makes clear. Mhhutchins 03:01, 7 August 2012 (UTC)

The secondary question of storage and bandwidth came up earlier in this discussion, so I addressed it above. I will have to think a bit more about the primary issue before I comment on it. Ahasuerus 05:45, 7 August 2012 (UTC)

I think the problem is that the name makes it clear to a certain degree, but still we all have our own bundles of prejudices, likes and dislikes, and we all collect different stuff that we like to highlight and share, and this site becomes the publishing platform for all of us passionate editors whose shelves and basements and storage units are overflowing with the fabulous, incredible stuff we've collected and accumulated over the years.

That being said, the ISFDB can't be a perfect mirror to the collections and interests of each of us. So if the majority of moderators (and programmers) can't justify the inclusion of something new, or the creation of the workflow behind the ability to include it, then that new thing needs to wait, and maybe wait a long time. Or maybe there's another site that caters to that new thing better than ISFDB ever could.

Finally, the silver lining about lines in the sand is that they enable quick decisions about what is in and what is not. The more porous the rules are, the more exceptions there are to the rules and standards, and the more we try to finesse things so that every conceivable border case is covered, the end result is the whole process is more subjective, and we all spend more time arguing about how to include the borderline cases that are most dear to our own sensibilities. And more time arguing means less time doing, and more editor vs. moderator and moderator vs. moderator stress, and that certainly can't be good for the community. Albinoflea 05:42, 7 August 2012 (UTC)

My problem with this line in the sand is that it is arbitrary. If I'm entering the metadata for a new nongenre magazine, I understand that I can't enter a cover artist because that creates a content record. I understand that have to enter the editor as "Editors of ..." because we want a special EDITOR record created for non-genre magazines. For the rest of the metadata fields I have to remember that two of them are arbitrarily prohibited (and one allowed under certain circumstances). Rules are easier to follow if they are rational, simple and understandable. When they are arbitrary and complicated, one has to go back and reread the rules in order to remember the exception (which metadata fields are arbitrarily prohibited) to the exception (non-genre magazine container records are allowed).--Ron ~ Rtrace^Talk 12:41, 7 August 2012 (UTC)

I have no idea why an editor can't enter anything in the pub format field, and can't imagine what was the logic behind that. I would agree with you if the consensus is to allow an entry in that field. But the logic behind the rule not to link non-sf covers is not arbitrary and indeed has a solid rationale: this is an sf database. I wish the software would actually be designed so that editors would not have to read the rules: there could be a NONGENRE MAGAZINE type, and its entry could be limited to what the current rules are. But that's not going to happen. So it's up to moderators to know the rules. That's going to be rough, admittedly, but that responsibility comes with the role. And all it takes is for an editor to enter a non-sf cover link one time. After he's told it's prohibited, I can't imagine that he would continue to do it. There are dozens of rules that I have to point out to new editors (and almost always the same rules), but once they're alerted, the editors follow the standards. Anyone who works with new editors know what I'm talking about. How many moderators do we have now that knew every standard from the beginning and didn't have to have a moderator guide them? I'd venture to say none. We shouldn't change the standards just to make it easier for new editors. So the argument about the arbitrariness of the current standard just doesn't fly. Mhhutchins 15:19, 7 August 2012 (UTC)

I just don't see the distinction between the various metadata fields that don't generate content records. If the cover URL is out because the sf database should not contain non-sf urls, then why allow the record to contain non-sf page counts, or non-sf prices? We even allow non-sf publishers that do create additional content. It is arbitrary because the rule prohibiting non-sf metadata is being applied differently for different metadata fields without a reason that I can see as to how those fields differ. --Ron ~ Rtrace^Talk 17:12, 7 August 2012 (UTC)

I think you're making this far more complicated than it really is by bringing in the metadata fields and the records which are generated by each one. That's not at issue. It's about having the covers of nongenre magazines with nongenre art visible in the database. This is not a database for nongenre magazines and that should be clear to any user. If they're disappointed because we don't link to covers of nongenre magazines, they'll get over it...eventually. Mhhutchins 19:20, 7 August 2012 (UTC)

I think the distinction is important. Some fields can create new objects in the database (Editor, Cover Artist, etc.), others do not (page count, price, etc.). The latter merely describe the object and cover url falls in this second category.--Ron ~ Rtrace^Talk 16:33, 8 August 2012 (UTC)

On the other hand, a large number of covers that are "in" are not SF related, as well known, non-SF pieces of modern and classical art are often used for cover illustrations of genre magazines and books. It seems reasonable to allow reciprocity in the rules.--Rkihara 16:44, 7 August 2012 (UTC)

It's not the art itself we're discussing. We're talking about the content which it illustrates. If an sf story has a cowboy as a character then it's OK to link to a nongenre magazine cover with a cowboy if it illustrates that story. I didn't think it was necessary to have to explain this. It doesn't have to have spaceships and aliens in order to illustrate spec-fic content. I thought we'd moved beyond such stereotypes. I've made my point clear, and will leave the discussion. I'll leave it to the group to make the final decision. I'm outtahere. Mhhutchins 19:20, 7 August 2012 (UTC)

That was my point. Asimov's for example has used paintings by Raphael and Jacob de Heusch, that are not SF related or illustrative of stories within.--Rkihara 23:24, 7 August 2012 (UTC)

[unindent] I think I may have figured out how I'm looking at this differently than Michael (and this is supposition, of course). I see the cover URL as a picture of an item that is already allowed in the database (Non-Genre Magazines that contain SF content). If Michael is considering the URL as a picture of the artwork on the cover that illustrates content that is not allowed (non-genre stories, or what have you), then I certainly understand his position. I would argue that if the URL is a picture of the artwork, it would more properly belong on the COVERART title record (the database object for the actual illustration). Since we apply URLs to the publication, I have to interpret the URL as a picture of the publication and not merely the cover art. I think by the way I'm interpreting it, my reasons for arguing that it be allowed become clear. --Ron ~ Rtrace^Talk 16:33, 8 August 2012 (UTC)

I agree with the above statement. My desire is to show a picture of the publication that is already allowed in the database, not to document the artwork as 'art'. Kevin 00:19, 9 August 2012 (UTC)

I believe Michael's position is that the only reason that one of these pubs is allowed into the DB at all in the first place is typically because of a single piece of SF content in that pub, and that's the only piece that really needs to be documented. The rest of the pub record should be minimal, just enough to identify the appropriate issue. In that sense the publication isn't really allowed into the DB in the same way a full-fledged SF pub is, and that in itself seems like a defensible position.

That being said, it seems a bit inconsistent that we do encourage or outright support the inclusion of additional information beyond that for these pubs. I haven't been around long enough to know the reasons behind this; I suspect most of it comes from re-using the same form for inputting both Genre and Non-Genre pubs. Albinoflea 02:48, 9 August 2012 (UTC)

Martian Chronicles redux

OK, so before I even start on this one, I'd like to get some more recent thoughts about what I should do before I enter the contents in for this copy of the Martian Chronicles.

This edition uses the new dates for the stories, which push everything back 30 years or so. (e.g. Ylla is now 2030 instead of 1999) From what I can tell, these new dates are not reflected in any of the title variants that currently exist.

For instance, this version uses the new dates, but this is only alluded to in the pub note field.

I managed to dig up this archived discussion about the matter, and was wondering if now people's opinions had changed in the light of practice.

Ultimately, is it worth my time to enter them and variant them (and for some moderator to approve them) to the appropriate canonical titles with the new dates? Or would it be preferred that I just import the contents from an edition that doesn't use the dates? I'm happy either way.

Thanks. Albinoflea 08:06, 6 August 2012 (UTC)

My opinion hasn't changed - I think the dates should be included in the title and variants made. I don't think it's just the title that's changed, the content is probably a bit different too - or it's a pretty slapdash effort on the author/editor's part. BLongley 12:10, 6 August 2012 (UTC)

If the dates appear on the title pages of the stories as part of the title, I agree that they should be included in the title, and made into variants. (I have an edition in which the full date is given in the story's illustrated header, and the year is given in the margins of each story. But it is not part of the title, so I did not include it.) BTW, variants should not be created based on a change of text, especially if it's just a slight revision. Mhhutchins 14:27, 6 August 2012 (UTC)

Right. These will be variants based on the change of dates in the title, not because of any change in the text, although the copyright info says, "This edition of The Martian Chronicles has been updated and revised."

As far as entering the dates for each piece of content, should that be 1997, the date the chronology for the stories was updated? Albinoflea 05:16, 8 August 2012 (UTC)

Is Interior Artwork ever excluded?

Moving two questions here as a result of this discussion which I'll post as separate topics.

This template speaks to what interior artwork should be included. Was the intent to exclude other artwork, or is it left to the editor's discretion on whether to include items? --Ron ~ Rtrace^Talk 02:22, 14 August 2012 (UTC)

I'm puzzled. What "other artwork" is being excluded? Isn't it always left to the editor's discretion about what artwork to include. Nothing in the rules require the editor to enter interioart at all. Mhhutchins 03:47, 14 August 2012 (UTC)

These books reproduce the covers of the magazines or books in which Williamson's stories originally appeared on the endpapers with several covers to each end paper. They are individually identified and credited (though frequently incorrectly). Ron K.'s objection, I believe, was that the artwork is too small and should not be entered. My understanding is the same as yours insofar as interiorart is not required. However, I've never understood it to be prohibited.--Ron ~ Rtrace^Talk 11:54, 14 August 2012 (UTC)

I think it's pretty much understood that unless something is specifically prohibited, it's the editor's decision about what to enter. The problem is that two editors with the same edition have to compromise, and the ISFDB leans toward inclusion. I would personally agree with Ron K. if the reproductions are especially small, or if they're only there for reference and not to illustrate. But that's me. Should we be more specific about what should be excluded? Probably. Mhhutchins 15:46, 14 August 2012 (UTC)

I think a more important question concerning the discussion you link to is whether we should create separate records for PC copies. Unless these are included in a limitation statement, I don't feel there should be separate publication records for them. It's the same as creating records for review copies and those sent to friends of the author (PC=presentation copies), or misprints/printing errors (copies missing a limitation page, or mistakenly unsigned). Mhhutchins 03:47, 14 August 2012 (UTC)

I wasn't sure what to do with my copy. It's got the limitation page, is signed but is not numbered. The copyright page has both ISBNs (as I would expect). The dust jacket has only the trade ISBN, but has the prices for both the trade and limited editions. I don't have the slipcase (nor is it offered with the P/Cs that Haffner is currently selling). Additionally, "P/C" is penciled in on the half-title page, likely by a reseller (I got it on eBay and it was only advertised as signed). Since the publisher is currently selling these "P/C" copies on their site with a description that matched mine, I figured that is what I had. I'll happily delete my pub in favor of either the trade or the limited. However, should I consider my copy a souped up trade, or a limited with parts missing? --Ron ~ Rtrace^Talk 11:54, 14 August 2012 (UTC)

That's the problem: it's neither edition. But the copies look like intentional "errors", and it's rather unethical, in my opinion, for a publisher to do this and then sell them. But since it exists as a third state of the edition, there's no problem with there being a ISFDB record for it. Mhhutchins 15:46, 14 August 2012 (UTC)

FYI, I received a response from Haffner, re. "P/C Copy." Here's what he had to say:

"P/C" stands for either "Publisher's Copy" or "Presentation Copy." These are usually remaining autographed copies outside of the stated limitation (1/100, A-Z, etc.) that are either sold or given as gifts to friends of the Press. "P/C" is marked in the limitation space so that a blank edition doesn't enter the market and some enterprising fool doesn't write in #1 (or #101 or "A" or whatever)."--Rkihara 23:08, 16 August 2012 (UTC)

Additional comments from Haffner:

". . . Don Grant did this as far back as RETURN OF THARN from 1957 and there are a few of King's books in THE DARK TOWER series (at least the early ones) that have this as well."--Rkihara 00:00, 17 August 2012 (UTC)

I know what PC means, as one who has been on the receiving end of more than a few. And it's not just Haffner and Grant that print extra copies for reviewers and friends. Almost every small press publisher does it. But I still think it's unethical to sell them. The most grievous reason is that the author doesn't get royalties on them. A publisher could intentionally print more than the stated number, sell them and receive all of the profits without having to report the sales on a royalty statement, because supposedly PCs are given away, not sold. How can any publisher conscientiously do this? Haffner's response has actually made me change my mind about whether these kinds of publications should be allowed in the database. If they'd announced an unlimited trade edition in which a few had been mistakenly bound with a signed limitation sheet, thus not being part of the announced number of signed copies, then I'd understand their practice of selling them for more than the trade but much less than the limited. But selling unslipcased copies for $500? Un-fucking-believable! The ISFDB listing looks outrageous at that price. If there is no price printed on the book, it should not be given in the price field, as this appears to be the "after market" asking price, not a retail price. And after all, PCs aren't sold, so they really don't have a list price. Mhhutchins 00:16, 17 August 2012 (UTC)

Maybe it wasn't obvious but neither RTrace or I knew what a P/C copy is. Now that I know, some of your comments are a little less puzzling.--Rkihara 01:17, 17 August 2012 (UTC)

I didn't know the meaning of P/C (though I guessed Publisher's Copy), and I appreciate the comments and the research. Regarding the price, I made it clear in the notes that it is the current price it is offered at and may not have been the original offering price. The jacket has both the trade and limited prices. Nothing is marked in the numbering spot and whether or not Haffner is unethical, I'm not going to write anything in that spot, even if I wanted to resell it, which I don't. In any case, if we're uncomfortable with the price, I'll happily take it off and just leave it in the notes.

Perhaps we should start a glossary page? P/C, Headbands, Perfect Bound and Colophon. --Ron ~ Rtrace^Talk 01:31, 17 August 2012 (UTC)

Why not? We already have a page explaining the number line and ISFDB:Foreign Language Abbreviations :) Ahasuerus 03:35, 17 August 2012 (UTC)

Appendices and Magazine Departments

Second question as a result of this discussion.

How should Appendices and Magazine Departments be handled? Should they be handled in the same manner? There are examples where a separate item is entered for the Appendix or Department listing all the authors of the contained titles, each of which has its own record (e.g. Probability Zero! here or the appendix here). Other examples prepend "Appendix: " to each title, whether it appears that way on the title page or not. I'll also propose adding a series of the name "Appendix (<title of book>)" to the title of each item contained in the appendix with no additional title for the appendix as a whole. This is how I've handled magazine departments. --Ron ~ Rtrace^Talk 02:32, 14 August 2012 (UTC)

I oppose the creation of an arbitrary content record, just for the sake of creating a group title, especially in the case of the Probablity Zero "essay" record which is "credited" to four authors. First, it's not an essay because it represents four works of fiction, and second, the four authors did not collaborate on it. The "Appendix" I'd be more willing to live with, because at least it is an essay record. The problem with most appendices is that they're either not credited at all, or each section of the appendix could be credited to separate authors. In that case, why create a record for the entire appendix? Mhhutchins 03:35, 14 August 2012 (UTC)

Probability Zero was a semi-regular department of Astounding. If we were to eliminate that, then within the magazines, all departments should be stricken, since they serve only demarcate regular features. So, the Letters department should probably go, not to mention the Editorial, and book reviews, and the F&SF contest department, which often contains poems, one paragraph stories, and the like. The actual essays will lose their context. I suppose we could keep the departments and subordinate the contents as we do for the book reviews.--Rkihara 07:26, 14 August 2012 (UTC)

Why not just create a group title record? That's the problem with the current software. We're always squeezing in extra purposes for a function to the point where it bursts with meaninglessness. In the case of the Probability Zero pieces, they're handled as a title series, when in reality, they're a publication series. But the software can't put title records into a publication series. In fact, all of the regular magazine columns that are now in title series are actually publication series. Update the software so that we can place title records into publication series/group titles, and that would resolve the issue. Mhhutchins 15:57, 14 August 2012 (UTC)

A couple of points (although perhaps I should finish recovering from what appears to be a mild heat stroke first -- apparently bibliographers shouldn't venture outside in August.)

Anyway, re: "We're always squeezing in extra purposes for a function to the point where it bursts with meaninglessness". Indeed, it is a very real and serious problem. However, we are not the only ones thus affected; librarians have been struggling with this issue for decades. No matter how many fields and drop-down lists and relationships your software supports, the real world is always more complex than your representation thereof. Hence the need to stretch existing functionality to accommodate unforeseen cases, sometimes "to the point where it bursts with meaninglessness". The only answer is to increase the complexity of the software, which, in turn, makes it harder to maintain -- quite a conundrum, really.

Re: "all of the regular magazine columns that are now in title series are actually publication series", I don't think it's quite that simple. The key thing about departments like "Probability Zero" is that they apply to Title records only within a given publication. If you put the affected titles in a regular series, then that series will appear next to them even when they are reprinted in a different publication and wreak havoc. For example, Forrest J. Ackerman's and Robert A. W. Lowndes's "Dhactwhu! - Remember?" is currently listed as part of the Shock Short series because it was reprinted in the "Shock Short" department in Perry Rhodan #44: The Pseudo One. Since it's a part of a Title series, it also appears that way in Super Science Stories, April 1949, which makes no sense whatsoever.

I think what we need is a new field for "departments". It will be a part of the publication record and be associated with individual Titles in the pub just like pages are currently associated with individual Titles. We could then do various fancy things with them, e.g. show a list of departments for a magazine series similar to the way we show a list of publication series for a publisher. Ahasuerus 02:28, 17 August 2012 (UTC)

I still insist that "all of the regular magazine columns that are now in title series are actually publication series". It's a series that only applies to the publication within which it appears. Would you agree that Shock Short and Probability Zero are not title series? Whatever we call it, "departments", "groupings", etc. I think we're talking about the same thing. That was the point I was trying to make: create a new field at the publication level and the problem is resolved. How that's done is up to the programmers. And I'm not diminishing the effort it's going to take to bring this about. I'm only saying that is the solution of the original poster's dilemma. Mhhutchins 02:53, 17 August 2012 (UTC)

For the most part I agree, except that I don't think it's quite fair to say that "It's a series that only applies to the publication within which it appears". Like most other magazine departments, Shock Short and Probability Zero spanned many issues and there is value to being able to easily find all stories that appeared in them. Presumably the main reason why Title series were used in the first place was to capture this connection. For example, take a look at this Wiki page for Analog, which lovingly lists all "Essay Series", including "Probability Zero". If we implement "Departments" (or whatever we call them) the way I described them above, we should be able to preserve this cross-issue information while eliminating the undesirable side effects of using Title series. It will be a little tricky to implement, but not insurmountable. Ahasuerus 03:29, 17 August 2012 (UTC)

You've misunderstood what I said, or I failed to make myself clear. By "publication" I meant the title, not an individual issue of that title. Can we agree that Analog is a publication? And that the feature "Probability Zero" appeared in several issues of the publication known as Analog? Mhhutchins 04:27, 17 August 2012 (UTC)

Ah, I see! Sorry, I misunderstood what you meant by "publication". I have been dealing with "publication" records and "title" records in the ISFDB sense of these words for so long that I have a hard time recognizing other usages of these words :-) So yes, we were basically saying the same thing. Ahasuerus 16:05, 17 August 2012 (UTC)

To implement a "Department" field in the way you describe it, Ahasuerus, seems to me great and much better than the status quo. Stonecreek 05:15, 17 August 2012 (UTC)

OK, FR 3561728 has been created. Ahasuerus 18:09, 25 August 2012 (UTC)

Nongenre magazine with spec-fic related nonfiction

I'm holding a submission to add a nongenre magazine (The Paperback Fanatic) that contains some spec-fic related nonfiction. (My discussion with the submitter is here.) As I interpret the current rules, only the fiction of nongenre magazines should be allowed into the database. I don't think we should have records for nongenre magazines just to add nonfiction, albeit spec-fic related. Would we allow a record for an issue of TV Guide with an article about Battlestar Galactica? I see the current submission as being the same as that. Mhhutchins 05:44, 16 August 2012 (UTC)

I agree. Spec-fic related nonfiction is everywhere. I have an issue of "Oprah" with a review of the James Tiptree, Jr. biography. We have 10 reviews of that book in genre magazines, but I've always assumed that including such non-genre reviews of spec-fic related materials would be outside our scope. Since I think it would be impossible for us to be comprehensive about such material, I think we should avoid "spotty coverage". Chavey 12:20, 16 August 2012 (UTC)

I'm not against it, if the nonfiction relates to written SF. So TV Guide reviews of TV shows out, but useful bibliographies and interesting interviews or reviews should be in. Not that I intend to submit any, I've got quite enough to read. BLongley 18:20, 16 August 2012 (UTC)

Yes, but who's going to parse the articles to determine what percentage of it deals with "written" sf? I'm not, and I don't think most editors who make such submissions are going to go to the effort either. For example, a Neil Gaiman interview in Entertainment Weekly which deals mainly with the movie version of Coraline, but includes other discussion about his graphic novel work, and slightly less about his fiction? There was a recent discussion about nongenre magazines and nonfiction database "creep". This submission is a perfect example of what I was warning. Does anyone else not see the slippery slope? We allow this issue in the database, and then another issue which has slightly less related nonfiction, and then another editor wants to add all other issues for completeness. Should we just throw our hands up and say "I give up", without making any judgement about what's entered into the database and gradually getting to the point where the Rules of Acquisition have no meaning? It's really becoming a bore arguing such matters. I'm releasing the hold on the submission. Mhhutchins 18:47, 16 August 2012 (UTC)

I'm quite happy with an "if in doubt, leave it out" policy on borderline cases. The Paperback Fanatic sounds like a magazine of interest to bibliographers though, so slightly less nongenre than others. Eventually I might add the Book Collector magazine that dealt with Terry Pratchett's collectable books. But only that part of it (it's the only reason I bought it). BLongley 21:42, 16 August 2012 (UTC)

My preference is for more inclusion, even to the point of nonfiction articles about sf media occurring in non-genre magazines or books. However, I don't think the policy currently supports that view. I understand the slippery slope argument, but it doesn't worry me. I don't think that by allowing in these more borderline cases will ever lead to requests for inclusion of clearly non-sf fiction or non-fiction (excepting ROA 4 and 12). Even if it were to lead to the request, I wouldn't think it would ever be agreed to. I'm also not worried if we aren't comprehensive in our coverage of any area of the database and I think there are many areas where we aren't. If we were, there would be nothing left to edit except new publications. This seems to me, not so much a question of what content is allowed in the database, but rather allowing content based solely on where it is published. Nobody would question the inclusion of these items, even the TV Guide article about Battlestar Galactica, if it were collected in The Big Book of Battlestar Gallactica. To give a real example (that I haven't entered in detail), Harlan Ellison's review of 2001: A Space Odyssey is eligible for inclusion because it is collected here, but we wouldn't allow the same essay in it's original publication in Trumpet #9. We could note it, sure. But if it were not collected, it would be lost to us entirely. If a work (essay, story[if the were a type for it], novel[nongenre]) is considered in, I'd advocate that we include everywhere it is published (and I'm not talking about web publishing, that's a whole other can of worms). --Ron ~ Rtrace^Talk 01:04, 17 August 2012 (UTC)

Back when we created the Rules of Acquisition which state that "[w]orks about speculative fiction" were in, I assumed that this rule included all reviews of (and articles about) speculative fiction whether they appeared in genre or non-genre publications. It has been my understanding that the only reason that we list so few articles appearing in non-genre pubs has been insufficient manpower and I have been looking forward to a time when we could start adding them on a more consistent basis. Perhaps that was just my interpretation, though -- I have been surprised before, e.g. when it turned out that "[d]issertations" were not seen as belonging in the database.

In any event, I would support the inclusion of reviews of/articles about speculative fiction in non-genre pubs. Ahasuerus 01:11, 17 August 2012 (UTC)

And I do support it, too. Sadly, the day when we would consider ISFDB as complete is an utopia (or pure speculative fiction). We may achieve that day for genre magazines sometime, but even that is hard to achieve as we are still striving. Even the fiction outside of genre publications is only partially included - and this will forever be the state of things, I fear. But if there are users who specialize in certain authors or even one special work I imagine they'd be thankful for any information on non-fiction published anywhere (at least that's the case for me). Stonecreek 04:31, 17 August 2012 (UTC)

This discussion reminds me of the issue of Rolling Stone with the Paul Williams interview/article with/about Philip K. Dick. I have the magazine somewhere, and will probably someday want to enter it in the database, because it's an important article (it can be read/downloaded here. I agree with Michael, but could we adapt the rules for things above "a certain threshold"? --Willem H. 20:04, 17 August 2012 (UTC)

Advance Review/Readers Copies

I've always assumed that ARCs are not allowed into the database, and even recall a previous discussion here which came to the same conclusion. (Although I'll never be able to find it.) But there is nothing in the policy explicitly forbidding them. Some one accepted this record into the database without question. So what's the policy? Mhhutchins 18:31, 1 September 2012 (UTC)

I found earlier discussions from 2007 and 2008. I think ARC's and uncorrected proofs should not be allowed in the database, and always thought that was the rule. --Willem H. 19:44, 1 September 2012 (UTC)

A more recent policy discussion on this issue, from 2011, is here. It includes the first 2008 link that Willem listed, and later links in 2008 and 2009. Chavey 20:59, 1 September 2012 (UTC)

(Arguing for) I know of one book where the ARC run was ~2500 copies. This is a respectable edition size in Trade Paper size. - Kevin 16:19, 2 September 2012 (UTC)

(Arguing for) I know of one book where there are both an ARC (Trade paper format with a glossy cover with final artwork) and Galley Proof (trade paper size, construction paper plain printed text cover) editions floating about the collectible market. The book was later released in HC. - Kevin 16:19, 2 September 2012 (UTC)

(Arguing against) This is just a big pile of ambiguously dated records waiting to happen. - Kevin 16:19, 2 September 2012 (UTC)

(Arguing for) If the Arc represents an 'earlier' form of the book, with either: recognizable different contents/title, obvious changes in coverart, or if the publisher expended an ISBN on it, it should probably be included. If the Arc represent the 'only' form of a book (ARC released, no commercial edition ever sold), then it should also be included. - Kevin 16:19, 2 September 2012 (UTC)

(Arguing against) If the Arc brings nothing new to the table, other than to document an edition with a 'plain' cover, and that edition is properly labeled itself as an ARC or galley proof (In other words, if you have it in your hand, it will tell you it's different / special), then there is little benefit to inclusion for anyone other than the compleatist/archiver. (And for some authors I do count myself in that category). - Kevin 16:19, 2 September 2012 (UTC)

(Arguing for) We do a lot of things here that are of little benefit to anyone other than an highly focused collector. - Kevin 16:19, 2 September 2012 (UTC)

There are going to be exceptions, and we can understand the benefits of a few exceptions. But they have to be extremely exceptional. We just can't allow an unexceptional one in the db so that it can be used as a precedent for an onslaught of more submissions. I'll leave a note on the verifier's page and ask him what he thinks is exceptional about this one. Mhhutchins 17:24, 2 September 2012 (UTC)

I recently entered and verified a number of ARC's, unaware that they were controversial. Besides the one cited above, they include Sword Woman, Horror Stories, El Borak and Bran Mak Morn. I can't see that any of them are "extremely exceptional", or even mildly so. On the other hand, I don't see any reason not to include them in the data base. How are they different from multiple printings by the same publisher? Or different dust jackets or boards? Or trade and book club editions from the same publisher. They fit with my personal use of the data base, but then I tend to be one of Kevin's "highly focused" collectors for some authors. Bob 18:19, 2 September 2012 (UTC)

Why shouldn't we include them? Because ARCs are not published (they're printed) and Rules of Acquisition #1 requires that the work be published. Mhhutchins 18:42, 2 September 2012 (UTC)

I am not in favor of including ARCs either. A book may go through many incarnations (including ARCs) before it's officially published. If we were to list all of them, it would make the database almost unmanageable. If there is something special about a particular set of ARCs, e.g. they were sent out two years before the book was published, we can always explain what happened in Notes. Ahasuerus 01:03, 3 September 2012 (UTC)

(Otherwise we'd allow unsold manuscripts into the database.) Mhhutchins 18:42, 2 September 2012 (UTC)

It may be worth pointing out that we create 8888-00-00 titles/pubs when a book/ISBN was announced but never appeared, but that's a different story. Ahasuerus 01:03, 3 September 2012 (UTC)

[unindent] A search for the word "advance" and another for "uncorrected" in the note field returned the following records, some of which are primary verified (PVed):

A Fisherman of the Inland Sea, an advance hardcover edition of the uncorrected proofs "to be distributed as gifts of the author and publisher"
A Woman of Destiny, Part 1, an excerpted ARC of a nongenre novel by O. S. Card
Amber and Ashes, a possible ARC with a different ISBN from the trade edition
Bad Brains, an ARC that gives a full number line, the first trade printing dropped the "1" [PVed]
Brak Mak Morn: The Last King, advance uncorrected proof, with the same ISBN as the trade release [PVed]
Daughter of Elysium, a possible ARC with a different ISBN from the trade edition
Deepsix, possible ARC ("Special Edition") trade pb with a different ISBN than the trade hardcover edition
Dracula Unbound, ARC trade pb with the same ISBN as the trade hardcover edition [PVed]
Eternal Light, possible ARC trade pb with a different ISBN from the trade hardcover edition
Master of Earth and Water, possible ARC trade pb with a different ISBN from the trade hardcover edition
A Hitchhiker's Guide to The Hitchhiker's Guide to the Galaxy, promotional book with no corresponding trade edition [PVed]
The Horror Stories of Robert E. Howard, advance uncorrected proof, with the same ISBN as the trade edition [PVed]
Mine, possible ARC trade pb with a different ISBN from the trade hardcover edition
Neverwhere, ARC which is recorded in the publishing history of all later printings as "Special Printing" [PVed]
Pendragon, possible ARC trade pb with a different ISBN from the trade hardcover edition
Replica, ARC with the same ISBN as the regular release, which was published 7 months later.
Seeing Redd, an ARC trade pb with a different ISBN from the trade hardcover edition
Storm from the Shadows, an ARC ebook which was sold to the public by the publisher
Sword Woman and Other Historical Adventures, advance uncorrected proof with the same ISBN as the trade edition [PVed]
The Bloody Crown of Conan, advance uncorrected proof with the same ISBN as the trade edition [PVed]
The Iron Dragon's Daughter, possible ARC trade pb with a different ISBN from the trade hardcover edition
The Last Centurion, an ARC trade pb with a different ISBN from the trade hardcover edition [PVed]
The Wolf's Hour, an ARC mmpb with a different ISBN than the first mmpb edition
Weaveworld, an ARC mmpb with a different ISBN than the first mmpb edition
Frederik Pohl's Favorite Stories, cancelled title that exists only in an ARC edition
Sea of Glass, cancelled title that exists as one publisher's ARC, but later released by another publisher with a different ISBN
New Dimensions 13, cancelled title which had advance copies mailed to reviewers before the remainder of the print run was pulped
Talking It Over, an ARC with a different ISBN from the published edition, published 7 months later
Tarzan Forever, a advance promotional excerpt, with different cover work than the published edition [PVed]

I would accept all into the database except for 5, 8, 12, 16, 19, and 20, all of which are ARCs with the same ISBN as the released edition. The others, in my opinion, are exceptional for different reasons. (Full disclosure: I created the records for 4 and 14.) Mhhutchins 17:18, 3 September 2012 (UTC)

I would have to agree that entering all of the ARCs would be an intractable job and a colossal waste of time. Most of the ARC's Michael cites that are "exceptional" are not verified, most of the ones that are "unexceptional" are verified. Would it be unreasonable to include verified ARCs, exceptional or not, and otherwise limit ARC stubs to the truly exceptional? Bob 16:49, 4 September 2012 (UTC)

I don't believe the fact that they're primary verified makes them exceptional. Just because an editor has a copy and can confirm its existence doesn't make them eligible for the database. The ISFDB should not be used as a personal catalog of an editor's collection.

And I disagree that only the "unexceptional" ones are verified, just the ones you verified (5, 12, 19 & 20). The ones that haven't been primary verified might be due to their rarity. I've love to have a copy of New Dimensions 13!

I would continue to support the inclusion of "exceptional" ARCs, including those that have different ISBNs, or were cancelled before publication. And the exclusion of "unexceptional" ARCs: those that have the same ISBN, and are otherwise identical to the trade issue except for their binding. Mhhutchins 17:20, 4 September 2012 (UTC)

Michael, in the interest of accuracy, somehow you missed the fifth ARC I entered. It's shown in an earlier response I made above. I never claimed (in fact, I explicitly denied) that any of my entries were exceptional (again see above). But having ARCs in the database fits with my personal use of the database: I use it to identify pubs that exist but that I don't own. I don't need the database to be a personal catalog, I have my own catalogs. I just want to help others like me. Bob 23:15, 6 September 2012 (UTC)

Poetry Collections

I would like to request a new feature, Poetry Collections. I've established such a title for Robert E. Howard, and there are a dozen pubs there so far, with more to come. But that title falls among other titles for series of related publications, which the poetry collections clearly are not. The database needs something like "Magazine Editor Series", "Nonfiction Series", "Chapterbook Series", "Short Fiction Series" and "Essay Series". I suggest we need a similar "Collection Series". I was surprised when I entered the series title and it didn't appear above "Collections" on the author's page.

So far the only place there is a "Poetry Series" is for Howard, but other authors where such a series would be useful include HPL, CAS, Frank Belknap Long, Poul Anderson, Sprague deCamp, Frank Brunner and Roger Zelazny among others. Bob 16:43, 4 September 2012 (UTC)

Perhaps you meant to say "the database already has [the other types of series]." Collections that are part of a title series are already displayed under those series (such as Howard's Conan collections).

I think you're wanting to create a new publication type instead of a new kind of series. That seems to be the best way to handle poetry collections in order that they be displayed separately from story collections. Placing titles into series doesn't affect publication records, which is probably where the distinction should be made. Creating this new category would then make it simpler to display poetry collections on an author's page. The current series into which you've placed the Howard collections is neither a title series nor a publication series according to the current ISFDB definition. If it were a true series (which I don't believe it is) then all other poetry collections by all other authors would have to be placed into it. Giving them their own publication type, and then converting all of the hundreds of other poetry collections in the db, makes more sense to me. Mhhutchins 17:09, 4 September 2012 (UTC)

Yes, the only meaningful way to handle collections of poetry differently than 'normal' collections - if that's what we want to do - would be to create a new pub type. Title series are defined primarily by their content (for example a common protagonist or a future history). But poems are as different in content as are short stories or novels.

So, I'd favour such an implementation but oppose the use of a title series for collections of any kind.Stonecreek 18:46, 5 September 2012 (UTC)

To propose such an alternative publication type, decide how we would handle a collection which has lots of poems and some stories, e.g. A Guide to Folktales in Fragile Dialects, with 28 poems and 8 short stories. Is that a regular collection or a poetry collection? Where's the line? Chavey 00:07, 6 September 2012 (UTC)

I guess we'd have to use the same criteria when a work has both fiction and nonfiction: majority rules. In any case, we're jumping ahead. Neither of the developers have joined in the discussion. They may not be able to implement such a request. Mhhutchins 04:05, 6 September 2012 (UTC)

It took me a while to respond because I am not sure I fully understand the proposal, so please bear with me, folks. If we want to make poetry collections appear in a separate section on the Summary Bibliography page, then what we need is a new title type rather than a new publication type. Granted, we will presumably also want a new publication type to go along with the new title type, but that's just to keep things consistent and avoid mismatches like the unfortunate NONGENRE-NOVEL situation.

If my understanding of the desired functionality is correct, then yes, it's certainly possible to implement. The only obstacle is the amount of work that it will take: a quick search finds 59 unique places in the ISFDB software that reference COLLECTIONs in some way. Changing the code won't be that hard, but re-testing everything that may be conceivably affected to confirm that we didn't break anything would be rather time consuming. Ahasuerus 06:00, 6 September 2012 (UTC)

We would probably want to sort out NONGENRE SHORTFICTION and other desired types at the same time. It's certainly possible to do but it's best to do it only once. BLongley 10:55, 6 September 2012 (UTC)

And should probably determine if there are other title types we should add. For example, it seems to me that we should have NONGENRE NONFICTION. We have lots of non-fiction books in the system about "Science", but which have nothing to do with "Science Fiction". Chavey 15:39, 6 September 2012 (UTC)

It was briefly discussed before that we should add LETTER as a type of entry; there are entire pubs devoted to letters, so a title type of LETTERS might also be of benefit. Right now such pubs are NONFICTION, but the letters can contains a lot of poetry and fiction. Bob 22:57, 6 September 2012 (UTC)

The originally proposed approach to resolving the NONGENRE-NOVEL mismatch was to add a new field, "Nongenre", to each title record. That way we will be able to enter Nongenre novels as well as Nongenre collections, anthologies, non-fiction, short fiction, essays, etc.

As far as letters go, there are two ways to handle them. We could create a new title type or we could create a new subtype under Essays the way "short story", "novella" and "novelette" function as sub-types under SHORTFICTION. Personally, I am in favor of the first approach since sub-types are something of a pain to implement and maintain. Ahasuerus 04:02, 7 September 2012 (UTC)

Well, I think letters are well-placed as nonfiction, there's always the possibility to make up a nonfiction title series for an individual author, for example 'Letters by Robert E. Howard'. Stonecreek 14:13, 7 September 2012 (UTC)

Letters only work as ESSAY at the moment, not NONFICTION. If we can thrash out exactly what we want - and that's not always easy when you have non-programmers confusing the terminology - then we can have a big stab at fixing it for good. (OK, for another few years at least - nothing is ever totally settled here.) But I'm waiting for Ahasuerus to catch up on last year's fixes before I do any more development. It appears that I will be separated from organised bookshelves for some time so development might be a good waste of time when you can all cope with this untrained loose cannon again.... BLongley 19:26, 7 September 2012 (UTC)

Linking online reviews to title records

I accepted the submission adding a webpage to this title record without checking the website. I see now that it's an online review of the novel. I'm not sure if this was the original intention of adding this field to title records, but I can find no policy statement about the use of the field. I'm concerned that a title record could be inundated with links to various online reviews, professional or otherwise. Should there be restrictions on these links? Mhhutchins 22:04, 11 September 2012 (UTC)

It was me that placed the link, and I think that it falls within the rules that have been suggested for this field:

* Web Page - A field for the URL of a Web page related to this title. Examples of related Web pages include legally posted versions of the title's text and Web pages that discuss this title. This is a multiply occurring field, so if one Web page has already already been entered, you will be able to add another page.

It is not just a page by some nobody that just have happened to read the title, it is a professional online review. If the link I gave -www.sfsite.com- is not permitted for that field, than what kind of webpage would be?.--Dirk P Broer 00:16, 12 September 2012 (UTC)

As usual, I think the stability of the site is one of the most important criteria. There's no point creating hundreds of links to a resource that goes bust a few months later. SFSITE seems fairly stable though. BLongley 10:26, 12 September 2012 (UTC)

I also approved a webpage link submitted by Dirk a few days ago to another review at SF Site, and I don't believe it goes against the letter of the rule as it is currently stated however that rule is perhaps not explicit enough: there's a perceivable difference here between the words "discuss" (Wikipedia-style, etc.) and "review". Ideally, online reviews should appear down there under "Reviews", as they already do for reviews at Strange Horizons, for example. The problem I have at the moment is that its webpage listing gives no indication that it's actually a review being linked to. PeteYoung 07:05, 13 September 2012 (UTC)

In general, I don't have a problem with using the "Web Page" links to record pointers to places with more information. I think it's useful. On the other hand, IMO we certainly wouldn't want links to blog or forum postings, and we also wouldn't want links to online versions of print-published reviews, which should be captured in a different way. I suppose the other issue is that while they may contain facts, reviews are opinion, and any attempt to treat one set of opinions (e.g., reviews at SFSite) differently from another set of opinions (e.g., reviews at flaming*hole.com) is problematic: is the treatment truly based on mechanics, or is it incorporating a value judgment on the viewpoints? I don't have any bright ideas about how to reconcile all of those thoughts. --MartyD 10:38, 13 September 2012 (UTC)

There seems to be a Feature Request in here somewhere. Add "Type of Site" to web-page entries maybe? If we had that, then we needn't restrict ourselves to one Wikipedia entry when there are versions in several languages, for instance. I'm not sure a ban on blog postings is desirable, author blogs often have one or more articles relevant to release dates or table of contents. BLongley 16:10, 13 September 2012 (UTC)

I also hesitate to accept any sort of restriction against blog postings (Though comments to blog postings may fall below the line). Here is an example of a blog post Vessel - Sarah Beth Durst on Scalzi's Whatever that is reasonably stable, and has info specific to the title and the author. Kevin 14:41, 17 September 2012 (UTC)

Sometimes a blog is our only source of data for old/obscure titles, so I agree that a blanket ban would not be a good idea. Ahasuerus 06:08, 18 September 2012 (UTC)

There are some very good blogs dedicated to author research, e.g. Bear Alley. I see we've linked there over 50 times, sometimes for some very tricky authors. BLongley 09:46, 18 September 2012 (UTC)

Sorry to have taken this off track. I wasn't suggesting we ban blogs. --MartyD 10:25, 18 September 2012 (UTC)

Rules exception request

I've completed the indexing of the whole run of 30 issues of the Philip K. Dick Society Newsletter, that Paul Williams sent to me a few years ago. There is one issue I have left out, however: issue "#9/10", which was issued only as an audio cassette and never as a paper-based issue, so while it's in "audio (CAS)" format it is not however an audiobook, and therefore not eligible under current rules. Its contents are: Side 1 – a conversation with Paul Williams, part of which formed the basis for Williams's title Only Apparently Real, and Side 2 – a rambling, mostly untranscribable (and dare I say unintelligible) monologue by Dick for a work-in-progress, which Dick taped instead of typed because his arm was in a splint. Therefore I'm requesting permission to add issue #9/10 to complete this series. Comments very welcome. PeteYoung 07:46, 14 September 2012 (UTC)

I have no strong objection to making this exception, and even find there's value in having a complete run of the newsletter. Mhhutchins 15:45, 14 September 2012 (UTC)

I think we can forgive you this aberration - just don't make it a habit! ;-) BLongley 17:45, 14 September 2012 (UTC)

Seems like a good idea to me, too. If you wanted to, you could do the submission and then post a note on the Community Portal telling everyone you've done it and asking for anyone with strong objections to respond. If someone did respond and did make a good case against, it could always be deleted. I doubt very much you'll run into any flak. --MartyD 11:04, 15 September 2012 (UTC)

I'll consider that for any future "aberrations"! Cheers. PeteYoung 01:17, 16 September 2012 (UTC)

Changing the help page for the "Note to Moderator" field

The current help page for the "Note to Moderator" field states:

Note to Moderator - A field where you can enter additional notes that will be viewed by the approving moderator. These notes will not become a part of the permanent database record once the submission has been approved.

I propose the following change:

This field should be used to provide additional information for the moderator to help in the decision to accept the submission. This note is seen only by the moderator handling the submission and does not become a visible part of the database record. For example, it can be used to inform the moderator that this submission is the first of several steps in updating a record. You can use this field to inform the moderator that you're working from a book-in-hand which will speed up the acceptance of the submission.

Do not enter any vital bibliographic information which you believe adds value to the record. This includes the source of the data used to update or create the record. That data should be recorded in the "Note" field.
Do not ask questions of the moderator in this field. You will get a faster response at the Help Desk, and by asking there, you may avoid needless effort and unnecessary submissions.

Any objections/suggestions/improvements? Mhhutchins 16:16, 14 September 2012 (UTC)

I'm OK with it, but I'm surprised you didn't mention "source of data" as a common use. But we do need to address the issue of bibliographical notes in the Note to Mod field being lost, or conversely submission-specific notes being added to the permanent notes. BLongley 17:51, 14 September 2012 (UTC)

Doh! Thanks for the reminder. I've made changes to the original proposal based on this suggestion. Mhhutchins 21:02, 14 September 2012 (UTC)

A bit of an aside: I've now picked up the keys to my new abode, so should be able to get back to coding in a couple of weeks after I've moved 5000 books and a score of bookcases. (If Ahasuerus can cope, of course.) I'm aware that "Sourcing made simple" and "Mod notes on remaining submission types" were considered quite desirable when I posted mock-ups. In the absence of proper prioritisation, feel free to poke me with your most desired features and I'll see if I can fit them in with Ahasuerus' plans. BLongley 17:59, 14 September 2012 (UTC)

I like it. --MartyD 10:59, 15 September 2012 (UTC)

Well written. Chavey 20:01, 15 September 2012 (UTC)

Yes, this should clear up things for new submitters. Thank you, Michael! Stonecreek 04:43, 16 September 2012 (UTC)

A conclusive and helpful changing! Rudam 06:33, 16 September 2012 (UTC)

Parenthetically expanded author names

While reviewing some variant submissions, I noticed we accepted The Prentice Hall Anthology of Science Fiction and Fantasy, with its parenthetically expanded author names. For example, expanded initials: A(lfred) E(lton) van Vogt or additional name components: Fredric (William) Brown. I think we should either augment the normalization rules or suspend the exactly-as-in-the-pub rule in this situation to allow the parenthetical expansions to be dropped or at least be relegated to the notes. I don't see any benefit to having these names as author records and having to have all of those variants because of them. Thoughts? --MartyD 10:54, 20 September 2012 (UTC)

I agree. It's jarring to the reader, and in most cases, in this anthology for instance, they just seemed arbitrary. I have been collecting anthologies for over forty years and I have never seen anything like it. It's especially bad as many of these authors have never used anything but their initials (like James P. Blaylock instead of James P(aul) Blaylock, or the A. E. van Vogt example used above) or authors like Stephen King, Jane Yolen, Damon Knight, etc suddenly having their names spelled out this way, causing a variant the authors never wanted. If these authors had wanted their whole names used, they would have used them. This leads to an extremely unnecessarily cluttered and hard to read listing. Still, I'm new to this and there may be good reasons to continue with the practice as is. MLB 11:45, 20 September 2012 (UTC)

This happens more often than you would think, especially in an academic setting. We're going to run into it more often when someone starts adding text book anthologies that have only a few spec-fic stories. But I agree, it is cumbersome and displays sloppily. A problem will arise if we don't handle such situations on a case-by-case basis. I'm afraid that if we codify it, it can be used as an "exception to the rule" that will be used for a different, but similar situation. Mhhutchins 14:09, 20 September 2012 (UTC)

I agree that we should make an exception to the "exactly-as-in-publication" rule for cases like this. Such variants should not be used. In fact, it seems to me that such a form is telling us exactly what the "real" attribution is supposed to be, and we should "honor" that request and use the implied attribution form, e.g. "A. E. van Vogt". Chavey 13:04, 20 September 2012 (UTC)

To go slightly off topic: Should the current "as-is" policy be dropped when translated publications screw up the author's canonical name? Lately, I've seen so many variants based on these errors, it's frightening. For example. Mhhutchins 14:00, 20 September 2012 (UTC)

On the original topic, in this case it would be a good thing to make an exception. I don't think there will ever be a user searching for "Fredric (William) Brown".

Off topic: We already have a rule about the Jr. suffix, All we have to do is enforce it. It is true that translations cause more variant names (there's a Dutch Tiptree collection that drops the Jr., so now James Tiptree is a pseudonym) but this example should i.m.o. be regularised. --Willem H. 15:23, 20 September 2012 (UTC)

I can get behind an exception to the 'as printed' policy. When Parenthesis appear in the credited name, Ignore the parenthesis and the contents. Kevin 21:42, 20 September 2012 (UTC)

I support the exception. BLongley 11:31, 21 September 2012 (UTC)

"Prematurely" translated titles

Here is what the Note field of "The Graveyard Night" (2011) record currently says:

According to the author's Web site at http://www.williamking.me/the-graveyard-night/ , "a slightly different version" was first published "by editor Martin Fajkus in an anthology of dark fantasy stories in the Czech Republic" in the 2000s. The unnamed anthology appears to be "Legie nesmrtelných" published in 2006 by Fantom Print, hardcover, 490 pages, ISBN 8086354741. The cover lists "W. King" as one of the contributors.

So it looks like the story was written in English ca. 2006, but only a Czech translation appeared at the time. The original was eventually made available (as an ebook) on Amazon in 2011.

Help:How to enter foreign language editions has the following to say about these cases:

If a book was written in one language, but a foreign language translation was published first, then the original language title is entered as the canonical title and the translated title is entered as a Variant Title.

Since Help doesn't say anything about dates, I wonder if the canonical title's date should be 2006 or 2011. Ahasuerus 06:02, 21 September 2012 (UTC)

It should remain dated 2011, the date of its first publication in the language it was written, which is the ISFDB standard regardless of the language, and regardless of any translation, whether before or after. If an author's work was first published as written, in 2011, then that should be the date of the ISFDB record. If we changed the date of "The Graveyard Night" to 2006, the average user may believe we're missing a publication record, if the first one we have is a 2011 publication. The title record's note field can have a note to explain the earlier translation (just as it does now.) Mhhutchins 06:20, 21 September 2012 (UTC)

It's going to look like an error if the first pub date under Canonical title is later than title date, but that's probably the best we can do. I'll leave it up to the other-language editors to see if they want to change the rule - it looks rather English-centric. BLongley 11:38, 21 September 2012 (UTC)

Untrue. The same rules would apply regardless of the language. If a French author's story is first published in English, the variant will have the date of the first English translation's publication, while the parent title will continue to have the later date if a French publication followed. So the rule would not be anglocentric. Just the opposite, in fact. Mhhutchins 15:03, 21 September 2012 (UTC)

Perhaps I should say English-centric by default, as the vast majority of our authors are here because of their English publications. Or Nationality-Centric, as there seems to be an assumption of which language a person writes in. I really can't be sure which works Harry Harrison wrote in Esperanto for instance, although I was quite interested to find 'Bill the Galactic Hero' first appeared in Italian. BLongley 10:44, 22 September 2012 (UTC)

Marion Zimmer Bradley's The Door Through Space was first published in English in 1961, and the title rec is listed that way. However, a VT of it is shown as Raubvogel der Sterne [German] (1959), indicating the earlier publication in German. That seems to me to be a fairly clear way to indicate a "prematurely translated" title. Yes, it's odd to a see a VT with an earlier date than the main title, but I suspect a reader can quickly figure out what's going on, even before looking at the notes for the VT. Chavey 12:54, 21 September 2012 (UTC)

OK, sounds reasonable. I guess we will want to update the Help page to make sure that the outcome of this discussion doesn't get lost in the mists of time. We may also want to document that the current policy can result in somewhat peculiar behavior: if a translation appears (and is entered in ISFDB) first, it will be used as the canonical title for a while, but when the original text is published, the translation should be turned into a VT. Ahasuerus 02:12, 22 September 2012 (UTC)