Rules and standards discussions

ISFDB Discussion Pages and Noticeboards
Before posting to this page, consider whether one of the other discussion pages or noticeboards might suit your needs better. If you're looking for help remembering a book title, check out the resources in our FAQ. Please also see our Help pages.
Help desk Questions about doing a specific task, or how to correct information when the solution is not immediately obvious. • New post • Archives	Research Assistance Help with bibliographic projects. • New post • Archives	Rules and standards Discussions about the rules and standards, as well as questions about interpretation and application of those rules. • New post • Rules changelog • Archives	Community Portal General discussion about anything not covered by the more specialized noticeboards to the left. • New post • Archives	Moderator noticeboard Get the attention of moderators regarding submission questions.   • New post • Archives • Cancel submission
Roadmap: For the original discussion of Roadmap 2017 see this archived section. For the current implementation status, see What's New#Roadmap 2017.

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

Rules and standards changelog

Every rule change that comes out of a discussion here should be added to the Rules and standards changelog.

Advance Reader Copy

I'd like to take Marty's direction and ask you to have a look at this thread with particular reference to guidance in the Help, acquisition policy, and consensus opinion. Any thoughts on how things stand? Thanks, Kev. --BanjoKev (talk) 09:19, 26 January 2023 (EST)

I think they should remain out, per long-standing practice. The reason for that is that they aren't considered a final published product, even if they are being sold. They are basically proof copies created to check for errors that need fixing prior to sending the final version tot he printer. ···日本穣 · 投稿 · Talk to Nihonjoe 12:32, 26 January 2023 (EST)

Orthogonal to the "in" vs. "out" question: Given the relatively high frequency of this topic's appearing, if ARCs are indeed "out", I think it would be a good idea to add some sort of obvious statement about them to the policy. We did that for audio recordings. The policy's definition of published is a set of qualifications, not a definition, so the Included section does not help. If someone considers them published, that list would make them "in". And if someone considers them unpublished, #1 in the Excluded section can be read to allow them nonetheless, if they fall into one of the Included section's points -- for example, if they are issued by a mainstream publisher. One approach would be to add something like Manuscripts, advance reader copies (ARCs), and similar compilations produced prior to official publication for purposes of proof-reading or marketing are not considered "published", even if offered for sale by the publisher. That could be a qualifier in the Included section or an early bullet in the Excluded section. One idea, anyway. --MartyD (talk) 15:31, 26 January 2023 (EST)

I like your proposed wording. It is clear and concise. We could also add something to the ISFDB:FAQ about it, too (with a link directly to the SCOPE). ···日本穣 · 投稿 · Talk to Nihonjoe 17:55, 26 January 2023 (EST)

I have a fundamental problem with ARC's being out. These are copies that exist out there, are genre, people have them, can be bought/sold, so need to be recorded (as a separate edition) imo. MagicUnk (talk) 04:02, 27 January 2023 (EST)

Printed ARCs are not supposed to be sold. They clearly state that on every one I've ever had. They also tend to have on the cover some marketing/bookseller info to help with ordering and whatnot. eARCs generally can't be sold, and the only ones I know of that can be bought are those from Baen (and I believe the purchaser receives a copy of the ebook once it's finalized, though I could be remembering incorrectly). If we do add them, we should have some standardized way of marking them so they clearly show up as an ARC so people don't get confused when viewing the editions of the books. ···日本穣 · 投稿 · Talk to Nihonjoe 13:03, 27 January 2023 (EST)

A couple of thoughts.

First, ISFDB:Policy#Included doesn't really define the term "published". For paper books, it says "paper books published by ..." and it's silent on the topic of ebooks and online publications.

The common definition of "published" is "made available", but "available" can be ambiguous. In most cases it means "available to the public at large", but it can also mean:

• available to all members of a book club
• available to those who have subscribed to a limited edition
• available to members of an APA (Amateur Press Association)

Wild Talent and The Time Masters by Wilson Tucker is an extreme example:

• "No publisher, no place, no date [about 1953], 1953. Hardcover. First edition. Typed sheets, rectos only, of the two novels, a total of 260 pages. Professionally bound with headbands and title and author printed in gilt on spine. This special book was typed from the magazines by Stuart Hoffman, science fiction fan and editor of Index to "Unknown ," probably before the books were published in 1954. " - from Robert Gavora, Fine and Rare Books, ABAA

Is this really a "publication" or just something put together by a dedicated fan for his own use and perhaps for his friends?

Second, a number of authors make new works available to their Patreon subscribers before they are "officially" published on Amazon. For example, Glynn Stewart's Discretion was officially published on 2022-07-26 and cost $5.99 on Amazon.com, but Patreon subscribers could read it 4 weeks earlier, on 2022-06-28, and it cost them only $5.00. It's not clear whether these types of "early releases" should be considered separate publications or ARCs. At one point I proposed that we call anything with a price a "publication", but the Rules and Standards consensus was against it. Ahasuerus (talk) 14:10, 27 January 2023 (EST)

A few more random notes (after resolving an edit conflict):

• Dating: we cannot date the title records based on an ARC date or any search of titles from a specific month/year becomes meaningless. Which means that we either need a new special date (7777-00-00 for example) or we need to consider allowing publications with titles in the future (which is a big no-no now).
• Multiple ARCs are produced for some books - usually some of them are for specific projects, sometimes they come at different times. As such they can have different covers and more importantly contents - images and excerpts may or may not be included; other bonus material like extra stories or essays may also appear in some and not in others. Do we record them all separately? If not, how do we decide which one takes precedence and whose contents to add? Annie (talk) 14:19, 27 January 2023 (EST)

(unindent)Two advanced searches of notes fields resale and not for sale reveal some interesting anomalies. Here are a few random-picked from a search on "advance": [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16]. Kev. --BanjoKev (talk) 15:59, 27 January 2023 (EST)

In my opinion ARCs and proofs should be excluded from the ISFDb. The general sense of the word "published" means an item that is made available for sale to the general public. That is not the case with ARCs and proofs although I am aware that there are rare exceptions (eg Baen eARCs, as Nihonjoe says). They are designed to be distributed freely to proofreaders and reviewers. Some of these recipients then choose to sell them but by then they are second hand goods. I have no problem with Ahasuerus' ambiguous examples because a member of the general public can join a book club / subscribe to a limited edition / etc. Regarding the Wilson Tucker extreme example, well... there will always be extreme examples.
Works available to Patreons may need to be treated separately. There was a discussion in Rules and Standards in 2021 with no clear resolution.
Whichever way this goes, I definitely agree with MartyD that ARCs and proofs should be explicitly mentioned in the Included or Excluded section of the Rules of Acquisition Policy. Teallach (talk) 19:02, 27 January 2023 (EST)

Re: "a member of the general public can join a book club / subscribe to a limited edition / etc.", it can get complicated. For example, I was thinking about entering a Russian language edition of George Orwell's 1984 earlier this week -- see filial.shpl.ru/wp-content/uploads/2020/10/Oruell-555x800.jpg . As far as I can tell, it came out in 1984 and was the first edition of the novel published within the Soviet Union. The catch is that even though the publisher, "Прогресс" (Progress), was a major Soviet company and published books in many languages, which were then sold all over the world, this particular edition was limited to trusted members of the Soviet government as indicated by the statement that you can see at the top of the cover. Was it really "published" as we define it? Ahasuerus (talk) 14:06, 28 January 2023 (EST)

I don't have any preference about this, but I suppose movies might provide a precedent and useful analogy. Pre-release screenings for cast/crew or critics are not considered the "release" of a movie (and, indeed, what gets released may be different). ARCs seem very much like those pre-release screenings. --MartyD (talk) 08:00, 28 January 2023 (EST)

I also have no preference but desire clarity (having entered one and run across multiple). Abebooks has a page delineating advanced copies, galleys, proofs & other pre-first edition books which our Help/Policy pages should cover in explaining what we mean by 'publish'. Using the general sense of "published" isn't good enough, we have our own definition for several terms (e.g. Chapbooks) and the interpretation of 'publish' is context sensitive (looking at the number of online definitions). A couple of (random) thoughts - given the 'publication' of material on the internet, does something have to be 'sold' for a 'price' to be considered published? We consider the change of advertisements (e.g. in Ace editions) to be additional 'printings' and worry about how many editiorial changes constitute a new 'edition' - so where are ARC's on this spectrum? Are they a valid 'printing'? ../Doug H (talk) 11:23, 28 January 2023 (EST)

Re: "does something have to be 'sold' for a 'price' to be considered published?", I don't think so. We have numerous pubs which have no price associated with them: webzines, fanzines, certain promotional editions like this one. Ahasuerus (talk) 16:21, 28 January 2023 (EST)

Shouldn't we be better of discarding the 'published' notion? As discussed above, there's not a definition to be found that will not spawn exceptions. Rather, shouldn't we consider any single "version" (to not have to use the word published), and record that? MagicUnk (talk) 16:27, 29 January 2023 (EST)

Well, we have a "publication date" field, which implies that a pub has been "published".

Also, if we were to use "version" instead of "publication", how would we distinguish "versions" produced by the author or the publisher internally from versions delivered to the intended audience, whatever it may be? An author and the author's editor(s)/publisher(s) may go through multiple iterations of a manuscript before it's finalized. And sometimes it's never finalized as was the case with the famously unfinished The Last Dangerous Visions. We currently list it as "unpublished", but multiple versions of the text existed at various points in the past. Ahasuerus (talk) 15:15, 30 January 2023 (EST)

Perhaps it would help to view these as types of publication and to designate them for inclusion/exclusion on that basis. That would even allow for some variation of treatment, instead of one-size-fits-all, with a default policy that's in or out and a small set of exceptions to cover the rare cases we agree ought to go the other way. Sort of like we have done for online publications. --MartyD (talk) 15:56, 30 January 2023 (EST)

I am not sure I fully understand the proposal. When you say "type of publication", do you mean that we could add a new "pub type" or a new "format"? Ahasuerus (talk) 17:27, 4 February 2023 (EST)

Not really a proposal, just sort of thinking out loud.... Originally I had in mind "pub type", but as I write this, I'm wondering if "title type" might be more appropriate; e.g., a la SERIAL (e.g., A serialization of a title is IN, but an advance proof of a title is OUT). --MartyD (talk) 14:32, 6 February 2023 (EST)

(unindent) After reading Ahasuerus' post below, I'd like to bring it into this main thread, as I think it posits a necessary wider view that I think we need in order to make progress which will stand the test of time. Quote:

"After re-reading this section, I think we may be inching closer to a working definition of "publication". We have a few common scenarios:

• A work is made available to "the public at large" as is usually the case on the internet
• A work is made available to all residents of a certain geographical area, e.g. the UK, the US, The European Union, etc
• A work is made available to subscribers, where "subscribers" can be members of a book club, people who have pre-ordered a limited edition, Patreon/Kickstarter patrons, etc
• A work is made available to libraries only

All of these scenarios meet the following criteria:

• the work is made available to people and to organizations outside of the publisher (including self-publishers) proper
• the work is made available for reading as opposed to for reviewing or other editorial and technical tasks

Are we getting closer? Are there additional scenarios that I am not thinking of? [Ahasuerus]"

As a sidebar, I have two genre short stories by the same author; one was published by Faber and the other on the author's publisher's website, free to view complete. If Ahasuerus' two summary criteria were to be applied, I would be able to enter both these works. Kev. --BanjoKev (talk) 18:26, 4 February 2023 (EST)

I should note that I was only trying to formalize the currently used divide between ARCs and other "internal to the author/publisher" versions of books on one hand and and "published" books on the other hand. I didn't mean to suggest that we should change the rest of the "Included" section, especially as it relates to online publications. There is a lot of complexity there that we would have to consider before we make any changes. For example, Web serials can contain hundreds (in some cases thousands, e.g. Forty Millenniums [sic] of Cultivation) of chapters and can be published over the course of many years. We don't have a workable way of capturing this information short of making each chapter a SERIAL in a CHAPBOOK, which would be a massive headache for a number of reasons. Ahasuerus (talk) 19:02, 4 February 2023 (EST)

Patreon Editions

If it's OK, I'd like to break out the Patreon edition discussion into a separate topic from the Advanced Reader Copy topic so that it doesn't get lost again. I think they are significantly different enough that they need to be handled separately.

The last inconclusive discussion in 2021 can be found here. Since then I have been adding the Patreon edition info for Glynn Stewart's books in the title notes. There are about 30 to date with a new one about to be added. In Glynn's case, the covers are the same as the ebook covers but have PATREON EDITION prominently printed on them; the title pages also have PATREON EDITION printed on them. (I'm also keeping a set of cover images for each of them just in case). They are released on Tuesdays anywhere from two to four weeks prior to the public release of the book and may contain some typos that are fixed in the public release. They consistently cost $5.00 regardless of the public release price.

In my mind, they should have separate pub records. That said, I suspect that Patreon editions by other authors may be less distinct but I have no proof one way or another. In addition, I'm not sure if Kickstarter editions should be handled the same way but they would seem likely candidates as well. Phil (talk) 12:37, 28 January 2023 (EST)

I also think that "Patreon editions" are really separate pubs -- as opposed to ARCs -- with a separate publication date, a separate price and (sometimes) a slightly different cover. I think we should be able to create separate publication records for them based on the following statement in ISFDB:Policy#Included:

• [Included:] Internet-based publications which are downloadable as electronic files in any number of ebook formats (ePub, Mobi, PDF, etc).

In some ways, they are similar to limited editions, which we create separate publication records for. Ahasuerus (talk) 15:52, 30 January 2023 (EST)

I had the same thought about these being like limited editions. --MartyD (talk) 15:58, 30 January 2023 (EST)

I forgot to comment on Kickstarter editions. Kickstarter campaigns can result in regular, i.e. publicly available, editions, limited editions or a mix of the two. They even call them "limited editions" on their Web site, so I think "limited edition" would be the best way to treat exclusive editions which are made available to Kickstarter "backers". Ahasuerus (talk) 08:52, 31 January 2023 (EST)

(unindent) After re-reading this section, I think we may be inching closer to a working definition of "publication". We have a few common scenarios:

• A work is made available to "the public at large" as is usually the case on the internet
• A work is made available to all residents of a certain geographical area, e.g. the UK, the US, The European Union, etc
• A work is made available to subscribers, where "subscribers" can be members of a book club, people who have pre-ordered a limited edition, Patreon/Kickstarter patrons, etc
• A work is made available to libraries only

All of these scenarios meet the following criteria:

• the work is made available to people and to organizations outside of the publisher (including self-publishers) proper
• the work is made available for reading as opposed to for reviewing or other editorial and technical tasks

Are we getting closer? Are there additional scenarios that I am not thinking of? Ahasuerus (talk) 17:43, 4 February 2023 (EST)

This seems workable to me. Phil (talk) 15:16, 6 February 2023 (EST)

I like this definition. It's clear and reasonably concise. ···日本穣 · 投稿 · Talk to Nihonjoe 19:51, 6 February 2023 (EST)

Is this waiting for the adoption of the wording for Defining "Published" in the next section? There haven't been updates to either of these topics in a month. Phil (talk) 12:13, 6 April 2023 (EDT)

Yes, this discussion spawned a number of related discussions below. We were getting close to a new set of Policy definitions, but then the discussion stalled, possibly because I was distracted with other issues. Thanks for the reminder. Now that the majority of the other issues have been addressed, I'll see if I can get us to the finish line, although it may take a day or two. Ahasuerus (talk) 13:58, 6 April 2023 (EDT)

Sorry about the delay. I haven't forgotten, but I was busy with something else. I hope to get back to this discussion shortly. Ahasuerus (talk) 09:07, 10 April 2023 (EDT)

Just reminding you this is still hanging fire. :) Phil (talk) 14:50, 25 April 2023 (EDT)

Yup! The prioritization phase of the monthly Fixer run was finished earlier this afternoon and the latest reported bug was fixed 5 minutes ago, so Rules and Standards discussions are next on my list :-) Ahasuerus (talk) 17:54, 25 April 2023 (EDT)

Defining "Published"

Discussions of "Advance Reader Copies" and "Patreon Editions" immediately above suggest that we may be getting close to a consensus. How about we replace the following sentences at the beginning of the ISFDB:Policy#Included:

• Included
• 1. Published works of speculative fiction, regardless of whether they are published within or outside the genre. "Published" is defined as follows:

with something like:

• Note that for a work to be considered "published" in ISFDB terms it must be made available:
  • to people and/or organizations (e.g. libraries) outside of the publisher
  • for reading as opposed to for reviewing or other editorial or technical tasks
• Included
• 1. Published works of speculative fiction of the following types:
• [the rest of the section which deals with paper, electronic and audio publications will remain the same]

This change would mean that Patreon editions would be considered "published" and eligible for inclusion as separate publication records. Regular ARCs would remain "out" while "pseudo-ARCs", which are offered for sale by publishers like Baen, would be "in".

Also please note that the text above removes "regardless of whether they are published within or outside the genre" because it's obsolete, not to mention somewhat confusing. Originally it was supposed to indicate that we wanted to include books published both by "SF-only" publishers like Ace, Tor and Baen and books published by mainstream publishers like Random House. However, this distinction is pretty much moot now that we have a detailed list of included subgenres. The rise of self-publishing, which doesn't follow the "genre/non-genre" divide, is another reason not to mention it any more. Ahasuerus (talk) 18:24, 11 February 2023 (EST)

That works for me. I suggest though that the cases you have quoted ("Patreon editions", "Regular ARCs", "pseudo-ARCs") should be explicitly listed in the policy as examples. That will make it clearer and easier to interpret for someone who looks at the policy in the future but who hasn't read this discussion. Teallach (talk) 06:42, 12 February 2023 (EST)

I like the spirit of this, but I have two inter-related suggestions:

1. I'd prefer to see the wording less passive and a little more comprehensive (The ISFDB considers a "published" work to be...). In addition to purpose, there is the permanency aspect.
2. I think more prominent placement and/or labeling could be useful. What about a third section, between Definitions of Speculative Fiction and Rules of Acquisition, something along the lines of "Definition of Published" or "Published for ISFDB Purposes" or even just "Published"? Then use of "published" in the rules of acquisition could be hot-linked back to that definition.

I don't have any strong feelings about either of these suggestions but figured I might as well mention them. --MartyD (talk) 08:39, 12 February 2023 (EST)

Good points. Currently, the "Definitions" sections is organized as follows:

• 4.1 Definitions of Speculative Fiction
  • 4.1.1 Inclusions
  • 4.1.2 Exclusions

We could change it to:

• 4.1 Definitions
  • 4.1.1 Speculative Fiction
  • 4.1.1.1 Inclusions
  • 4.1.1.2 Exclusions
  • 4.1.2 "Published"

Ahasuerus (talk) 12:28, 12 February 2023 (EST)

I like this. It's logical and clear. Phil (talk) 12:00, 15 February 2023 (EST)

I like it. Annie (talk) 12:15, 15 February 2023 (EST)

It sounds like we have consensus then. I am a bit under the weather today, but I hope it's nothing serious. I'll try to post the final language below once I feel better. Ahasuerus (talk) 17:18, 18 February 2023 (EST)

Proposed Policy language (Defining "Published")

Here is the proposed Policy changes based on the discussion above:

• Add a "Definitions" section to the ISFDB:Policy page. It will be a part of the "Contents/Project Scope Policy" section and will include two sub-sections: "Speculative Fiction" and "Published"
• ISFDB:Policy#Definitions_of_Speculative_Fiction will remain the same except it will be renamed "Speculative Fiction" and shifted one level within the section hierarchy
• The "Published" sub-section will contain the following text:
  • The ISFDB considers a work "published" if it has been made available:
    • to people and/or organizations (e.g. libraries) outside of the publisher (including self-publishers), and
    • for reading as opposed to for reviewing or other editorial or technical tasks
  • This includes limited editions, book club editions, editions restricted to subscribers (including Patreon/Kickstarter editions), and editions advertised as "e-ARCs" if they are offered for sale by the publisher
• Remove:
  • regardless of whether they are published within or outside the genre. "Published" is defined as follows
• from the "Included" section of ISFDB:Policy#Rules_of_Acquisition

Ahasuerus (talk) 16:54, 20 February 2023 (EST)

Looks ok to me. One minor suggestion "e-ARCs" -> "ARCs". I know we're thinking about Baen's e-ARCs, but I don't think format is relevant to the spirit of this definition. --MartyD (talk) 11:18, 21 February 2023 (EST)

Looks ok to me as well. And I agree with Marty's note - but maybe instead of dropping the "e", change "editions advertised as "e-ARCs"" to "editions advertised as advanced Copies (under the name "e-ARCs" or any other similar name)". eARC is very Baen specific but I won't be surprised that other publishers come up with their own thing... and what we are trying to say is that an Advanced copy is an edition as long as it is for sale by the publisher.

On a procedural question and before we change the language - will the publication date of the novel/story be the eARC date in that case and how does this map with awards eligibility for example? Annie (talk) 12:48, 21 February 2023 (EST)

" ..."published" if it has been made available: "

" * to people and/or organizations (e.g. libraries) outside of the publisher... "

" * This includes limited editions, book club editions, editions restricted to subscribers (including Patreon/Kickstarter editions), and editions advertised as "e-ARCs" if they are offered for sale by the publisher "

Annie, I don't see any specific wording in the above to warrant " an Advanced copy is an edition as long as it is for sale by the publisher. ". Or am I missing something? Kev. --BanjoKev (talk) 13:04, 21 February 2023 (EST)

How do you read "editions advertised as "e-ARCs" if they are offered for sale by the publisher" in the proposed language if not that way? Annie (talk) 13:25, 21 February 2023 (EST)

My point is that the door seems to be open for free ARCs to be considered published if they are made available to people ... and for reading... as opposed to...etc. At the same time e-ARCs are included if they are offered for sale by the publisher.

We know there are editors who advocate strongly for free book ARCs to be included in the DB and authors who advocate strongly against it. How does this look from either perspective? I don't mind a resolution either way, but the above looks like "free-ARCs-in" and "paid-for-e-ARCs-in" to me. Is that really the case? Is that what we're heading towards?

Authors have said they get a lot of feedback from fans by giving out free ARCs and that that can influence the later 'official' market edition. They and the publishers are making increasing use of this technique, so we can expect more of them not fewer.

I'm just a bystander trying to understand the struggle to control the issues in these threads and I'd like to end up with an intuitive feeling for what's correct or not in any given situation. That's not happening at the moment; I'm not seeing the black and white. Kev.--BanjoKev (talk) 15:08, 21 February 2023 (EST)

Again - look at the proposed language: "editions advertised as "e-ARCs" if they are offered for sale by the publisher". I am not sure which part of the proposed language makes you see "free-ARCs-in" as an option (the Patreon/Kickstarter and so on are a different animal). If you are reading that statement in that way, we need to change it because I think we are trying to make a differentiation between the ARC being sent to people for various reasons and the ones being essentially first editions in disguise (Baen's e-ARCs for example).

If you want to propose to drop the "offered for sale by the publisher" and replace that with "offered to the public by the publisher", feel free to make the proposal but Ahasuerus's proposed language literally says "if they are offered for sale by the publisher". I won't be in favor of allowing any ARCs being sent out though... Annie (talk) 15:22, 21 February 2023 (EST)

Putting aside the Patreon/Kickstarter and Baen e-ARCs subjects for the moment, as I said before, "paid-for-e-ARCS" look in to me, and the proposed wording "offered for sale by the publisher" is succinctly and well stated (call it point 2). It's dawning that if I can read "free-ARCs-in" into points 1a and 1b when that is not what is intended then the right wording for what is intended hasn't crystalized yet. Maybe it's too passive, maybe it's using too few words to describe a lot of perameters, maybe it needs to clearly state what is excluded (or did I miss that somewhere else?). I can't put my finger on it at the moment. I understand why you are not sure, as someone with a good overview of the whole subject, how I can see "free-ARCs-in" in points 1a and 1b. It's like one of those 'what colour is this dress' things :) I suggest we can agree that the wording does need to change though. Kev. --BanjoKev (talk) 16:16, 21 February 2023 (EST)

Aha, that explains where the disconnect is - it is not in that last statement, it is higher up - you do not see "for reading as opposed to for reviewing or other editorial or technical tasks" as excluding all ARCs. In my mind, an author making an early draft/ARC/whatever available and seeking opinions and feedback (or reviews and publicity and blurbs and so on) falls under "made available for reviewing or other editorial or technical tasks" so it is explicitly being disallowed. If this already allowed all ARCs, we won't need to call out eARCs specifically at the bottom as they are indeed ARCs after all. But yes, if you do not read it that way, we need to figure out how to say it so it matches what we are trying to say. Annie (talk) 16:29, 21 February 2023 (EST)

(unindent) Riiight, now we're getting somewhere. Your last prompts me to see that all the points should be chopped up and rearranged. Let's try to say the same things in a different way -

• The ISFDB considers a work "published", and therefore included:
  • If it has been made generally available to people for reading.
  • The term 'people' includes organizations outside of the publisher, such as libraries and self-publishers.
  • If it has been made available as a limited edition, book club edition, an edition restricted to subscribers (including Patreon/Kickstarter editions), or editions advertised as "e-ARCs" if they are offered for sale by the publisher.
• The ISFDB considers a work "unpublished", and therefore excluded:
  • If it has been made available for the purposes of reviewing, or other editorial or technical tasks.

Each point is one-topic, making for easier understanding and future editing. Any good? Kev. --BanjoKev (talk) 18:14, 21 February 2023 (EST)

Keep in mind that "published-unpublished" and "included-excluded" are different axes. There are "unpublished" works that we include:

• Unpublished works of speculative fiction which have been:
  • announced as forthcoming within the next 90 days
  • announced but never published (entered as "unpublished")
  • published only in translation (the original should be entered as "unpublished")

and there are "published" works that we exclude, e.g.:

• Works published in a web-based publication and available exclusively as a Web page -- such as blogs, author-run sites, fan fiction, web serials, etc -- unless listed in the Included section

We are currently trying to fine-tune the "published-unpublished" axis without touching the "included-excluded" axis. Ahasuerus (talk) 15:45, 22 February 2023 (EST)

(unindent) It looks like we have three outstanding issues: 1. Clarify the use of the term "e-ARCs". Two clarifications have been proposed:

• MartyD: Change "e-ARCs" -> "ARCs" [in order to cover other types of ARCs offered for sale]
• Annie: Change "editions advertised as "e-ARCs"" to "editions advertised as advanced Copies (under the name "e-ARCs" or any other similar name)"

I think the second proposal is more explicit and will help our users who may not be familiar with the term "ARC".

I agree. I was only thinking about not limiting to eARCs. Annie's suggestion achieves that while also clarifying what we're talking about. I'm all for that! --MartyD (talk) 08:43, 26 February 2023 (EST)

2. Annie: "will the publication date of the novel/story be the eARC date in that case and how does this map with awards eligibility for example?" I have been thinking about the date issue for the last couple of days, but I have been unable to come up with a definitive answer, although I have some ideas.

If Baen makes an "e-ARC" version of a first edition available for sale in October and a regular (aka "trade") version available in November -- and if we consider the "e-ARC" version a separate "publication" for ISFDB purposes -- then it stands to reason that the title date of the book should be its first publication date, i.e. the e-ARC date. On the other hand, the trade version will likely say something like "First edition: November 20xx", which may be confusing. My current thinking is that it would be best to make "October 20xx" the title date and add a note to the "trade"/November publication record to clarify that even though the book itself claims that it is the first edition and that it first appeared in November 20xx, an earlier version of the book was offered for sale in October 20xx and refer the user to the e-ARC publication records for details.

3. BanjoKev: Clarifying the e-ARC definition and restructuring the proposed Policy language. I will comment on it after taking a break. Ahasuerus (talk) 12:24, 24 February 2023 (EST)

Can someone who buys the Baen eARC look into a December one (when the book is out officially in January) and see what date/year is actually printed in the eARC book? Or any book if they print a month - but with a December/January, there will also be a change in the year. I don't have any so I cannot check but looking at any other ARC I have around the house, the insides of the book carry the actual publication date and the ARC date may or may not be on the back cover somewhere. So if that is the same for eARCs, the availability date is getting overridden by the date inside of the book anyway (as it won't agree on month and/or year). The only case where the eARC date will stand as the date of the title is if there is no date inside of the eARC or if the date there is the eARC date. Thus me thinking that we should check what Baen are actually doing.

We still need to figure out what we want to do in the generic case though... My thinking is that we use whatever date we know and add notes but that will hide the book from looking at titles published in a certain month/year or general searches... which I am not sure how we can solve... Annie (talk) 12:54, 24 February 2023 (EST)

PS: Changing the header title slightly while I am here so when it shows on a Wactchlist or changes list, it is clear which language we are working on. Annie (talk) 12:56, 24 February 2023 (EST)

I may have a few old e-ARCs in one of my offline archives. I'll poke around once I finish Fixer prioritization. Ahasuerus (talk) 12:23, 26 February 2023 (EST)

The good news is that I have all of Baen's "giveaway" CDs (as ISO images.) The bad news is that I can't find any e-ARCs among the hundreds of files that the images contain, but I only sampled things.

Poking around the internet, I see a few Baen e-ARC files floating around. Comparing the e-ARC version of P. C. Hodgell's Honor's Paradox with what Amazon's Look Inside shows for the final version, I see that the e-ARC file has the same ISBN, number line and publication date ("First Baen printing, December 2011") on the copyright page. The main difference -- minor capitalization quirks aside -- is that the "Library of Congress Cataloging-in-Publication Data" section is empty in the e-ARC file. There is a "t/k" placeholder in its place, which is editor speak for "to come". There are some capitalization differences in the body of the text as well. There is no back cover and no separate "e-ARC publication date". Ahasuerus (talk) 10:30, 27 February 2023 (EST)

I am confused now. How does this constitute a separate edition for us then? It will have the same date as the other ebook (even if it is out in October, the printed date will be the governing one), it will have the same cover and all other details). We don't record minuscule changes in ebooks as separate publications outside of this scenario... And that sounds like almost every ARC I had ever received from a publisher - possibly missing illustrations, possibly missing table of contents, before the last proofreading and so on passes... The only difference is that this one is paid. If this is why we make it a separate edition, then fine but we need to be explicit about that in the language. Annie (talk) 11:45, 27 February 2023 (EST)

Well, it's possible that this particular e-ARC file is not representative of what Baen usually does. Or perhaps they changed the way they do e-ARCs at some point. It would be better if someone who has more experience with e-ARCs commented on them. Perhaps Nihonjoe has more information? Ahasuerus (talk) 12:42, 27 February 2023 (EST)

An archived Baen Web page posted in December 2013 says:

• As of December 16th, Baen ebooks are now available for sale at other vendors as well as at this site. ... Available exclusively at Baen.com are eARCs and serialized monthly bundles at our previous prices. There is a significant change from our previous practice in the availability of the bundles: once the books for a given month are printed and the books made available individually at other sellers, these bundles will not be available for orders. For example, books labeled January 2013 were available for sale only until December 2012.

This seems to suggest that Baen e-ARCs were/are deliberately "labeled" with the publication date of the final product. If so, then I believe Annie's interpretation above is correct and their records would share the "official" edition's publication date. This is different from Patreon editions, which typically have a different publication date. Ahasuerus (talk) 14:48, 6 March 2023 (EST)

I have found an ARC version of David Weber's Mission of Honor on one of the "giveaway" CDs in my library. The copyright page says "First printing, July 2010", which is what the copyright page of the trade edition says. We now have two e-ARCs which use the same publication date as the trade edition (see P. C. Hodgell's Honor's Paradox above) plus the Baen statement from December 2013. I think it's enough to establish a pattern until and unless we come across exceptions. This means that we don't have to worry about potential publication/title date discrepancies since Baen ARCs apparently use the same date as their trade editions. Ahasuerus (talk) 17:38, 26 April 2023 (EDT)

ROA changes: unpublished non-genre and non-fiction works

(Splitting out from above) While we are reorganizing the above section, can we also move #4.2.1.2 in the ROA ("Unpublished works of speculative fiction which have been") down to become #4 and to read "Unpublished works which are eligible based on the criteria in 1-3 above..." or something along these lines? The way it is ordered now makes non-genre (from above threshold authors) and non-fiction works not eligible pre-release or when cancelled and we had never followed that rule - the current policy that had been followed for years is that if a book is eligible on publication, it is also eligible 90 days pre-publication or if cancelled. Or do we want to split this into a separate discussion? Annie (talk) 12:15, 15 February 2023 (EST)

Oh, I see. I have never considered it, but you are right: we include announced-but-unpublished non-fiction/non-genre books like Orchids for Doc: The Literary Adventures and Autobiography or Robert A. W. "Doc" Lowndes, but the Policy language would disallow them. I agree that we want to change the Policy to reflect current practice, but I suggest that we create a separate discussion section for the issue. I don't expect objections, but I am worried about overloading this section. Ahasuerus (talk) 13:53, 15 February 2023 (EST)

Here we go. I am more concerned with any genre non-fiction added pre-release (the current language allows only "Published non-fiction works about speculative fiction" - that word published in there excludes something that is not out yet) than the cancelled and never published ones but it applies to both exceptions from our "published" policy anyway. The cleanest will be to first define what we allow when published and then specify when the same books are eligible without being published (yet or ever). It matches current practice as well :) Annie (talk) 14:38, 15 February 2023 (EST)

Heh. I agree with Annie's point, and I support this suggestion, but having now re-read that whole section, I find myself compelled to suggest a further reworking of it. 4.2.1 contains a mix of specification of publication form and work type, with a dash of timing. For example, 4.2.1.1 specifies various formats but restricts its work type scope to "of speculative fiction". 4.2.1.2 specifies timing and likewise restricts its scope to "of speculative fiction". 4.2.1.3 deals solely with work type ("about speculative fiction") and fails to say anything at all about publication form or timing. 4.2.1.4 also deals with work type ("not related to speculative fiction, but were produced by authors ... over a certain threshold") and says nothing about publication form or timing.

What about extracting the work types into one section, and the other conditions into a second section, under a broader inclusion eligibility umbrella, or something like that? That is, for a work to be eligible for inclusion, it must meet...

One of these work type criteria:

• It is a work of speculative fiction <reference to 4.1.1 or 4.1.1.1>
• It is a work about speculative fiction <requirements from current 4.2.1.3>
• It is a work neither of nor about speculative fiction but was produced by an author who has published a number of works of or about speculative fiction over a certain threshold.... <details from current 4.2.1.4>

Both of these publication criteria:

• Publication form: <form requirements from current 4.2.1.1>
• Publication timing: <timing requirements from current 4.2.1.2>

I'm sure that could be worded better, but I hope it gets the idea across. This does extend Annie's proposed "rule change" (hers: publication timing criteria should apply to anything eligible for inclusion) such that all of the form and timing criteria explicitly apply to everything otherwise eligible for inclusion. --MartyD (talk) 16:02, 16 February 2023 (EST)

That works for me. That is why I was trying to push the timing down the list but yes, the form also needs to go down to make it as clear cut as possible even if we all know what we mean. :) Annie (talk) 18:28, 16 February 2023 (EST)

One thing I realized upon reading the "work type criteria" list again is that we have three alternatives for the non-SF work type:

1. It is by someone above-the-threshold (*)
2. It is contained in a larger SF work (*)
3. It contains one or more eligible SF works (*)

(*) All of these with various caveats/limitations -- I am not trying to open a discussion about their nature.

The latter two are somewhat buried or implicit, with details elsewhere. What to Include covers #2, but periodicals are only mentioned obliquely: 4.2.1 says an SF work is included whether "published within or outside the genre" and lists magazines and newspapers in the bullet about periodicals, but nothing in the policy states that a non-SF magazine or newspaper is eligible for inclusion. You have to go to Entering non-genre periodicals to find the policy for magazines, and only there is it stated that non-SF anthologies are handled the same way. That page also refers one back to the RoA page for determining whether a non-genre periodical is eligible, even though the RoA page does not say anything about including them (other than the physical form and timing parts). --MartyD (talk) 08:48, 18 February 2023 (EST)

You had to go there, didn't you? :) The ROA had always had a bit of a problem with the concept of what it calls a work - publications vs. titles in our DB. In some places it is clear what we are talking about, in some as you pointed out above, it is left for the reader to draw their own conclusion of what we mean and other help pages are needed to actually make it clear on what is eligible. Let me pull all relevant docs up and see if I can mash together something to get these sorted out and have a proposed language that reshuffles things around non-genre works. Annie (talk) 13:31, 21 February 2023 (EST)

Proposed: Allow inclusion of serials in novels, anthologies and collections

Currently, serials are limited to magazines, fanzines and chapbooks. As a result we're sometimes forced to misclassify serials as short fiction. I noticed a number of examples while working on the 'Short Fiction Title Records with '(Part' in the Title field' exception report. There were a number of novels with obvious serial installments not classified as such. John Scifibones 19:40, 22 February 2023 (EST)

Do you have specific examples of these? ···日本穣 · 投稿 · Talk to Nihonjoe 20:22, 22 February 2023 (EST)

Here are a few from the aforementioned report, example 1, example 2, example 3. John Scifibones 20:28, 22 February 2023 (EST)

Here are a couple examples where they are being used contrary to present standards, but should be allowed . example 4 and example 5. John Scifibones 20:47, 22 February 2023 (EST)

I support allowing SERIALS in all of our containers (including NONFICTION in case someone decides to add a single piece of serialized fiction in otherwise non-fiction collection of essays for example). While they used to be almost exclusively used in magazines and fanzines, these days they can popup everywhere - in anthologies, in collections and as bonus to novels. We already opened up the rules a bit to allow them in chapbooks. In addition, we had been kinda bending that rule for awhile anyway and already making some of these SERIALS when it makes sense - as the report shows. Switching them to short fiction will lose their connection to their parent/full work and there is no real reason to do it - we might as well codify the current usage and actual publishing reality.

PS: We should make a distinction between serial and excerpt when we open up the rules though - one is intended to be the whole work; the other one is intended to be a single part. Annie (talk) 10:56, 23 February 2023 (EST)

I also support this proposal. The examples are drawn from a publication series that was published weekly.

Another question: would we allow serial instalments within one and the same publication like the parts of 'Jube' here or the ones of 'Corsairs of the Second Ether' here? Christian Stonecreek (talk) 12:57, 23 February 2023 (EST)

Two different usecases on that one - one clear and one a bit murky:

• Fragmented complete text (your examples are like that): if they are ever published as a single text, then I'd think yes. If they were only published in this form, I'd make the case that they might be intended as separate texts. See this one for example - Festival Moon is technically one story that weaves through the narrative. Do we want this or the ones in your examples as serials? I lean towards yes for them as well quite honestly but if someone has a compelling reason to say no, my mind can be changed...
• More than one installment in the same book but not the complete text - definitely yes.

How would you record either if they were in a magazine? If we would record them as SERIAL, we have our answer - we really should not have different rules for SERIALs based on the type of the publication IMO. :) Annie (talk) 13:13, 23 February 2023 (EST)

(unindent) I can see how allowing SERIAL titles in collections, anthologies, novels (as "bonus items") etc would be beneficial.

One thing that we may want to clarify is whether the proposed change would also allow turning "split novels" into SERIALs. Personally, I don't think it would be an improvement compared to what we currently have (a title level "split work" flag would be better), but we'll want to make it clear either way. Ahasuerus (talk) 20:19, 23 February 2023 (EST)

The change shouldn't affect split novels. The proposal is to treat SERIALS the same as SHORTFICTION or POETRY for eligibility in our container titles. NOVELS would only allow a SERIAL as bonus, not primary, content. Any current language which prevents CHAPBOOKS from being used for split novels would still apply. If we feel further language necessary, we could add a specific prohibition in the CHAPBOOK section under publication type. John Scifibones 14:35, 24 February 2023 (EST)

I agree but I had been thinking about that and there is a slight wrinkle here:

• Republished novels which are split for publication on their own or in omnibuses (usually in translation) remain novels - including cases such as having the initial 3 volumes in a series being published in 2 volumes by splitting the middle novel between the two novels.
• The wrinkle are anthologies and collections. If a publisher is publishing a collection/anthology of 10 stories and 1/3rd or 1/2 of a novel, with the rest of the novel being published in other collections in the same pub series (or regular series) for example, do we want this novel's parts as SERIALs or as Split novel entries (as will be the case under the current rule)? I can see that argument going both ways... By being inside of an anthology/collection they feel like SERIALs to me and I think we had been using the split novel rule here because we simply had no options (not that we have that many of these admittedly - but I've seen that with some publishers in my languages). But I am not opposed to leaving these as NOVELs under the split novel rule - I am mainly thinking aloud through the different permutations when we open SERIAL to be used in all containers.

Depending on how we want to record that last group, we may need to call out separate rules for Novels and Omnibuses and for the containers (Magazines, Anthologies and Collections). Annie (talk) 17:20, 24 February 2023 (EST)

Magazine reprints are another side of the same coin. An exception report 'Anthology Publications with Invalid Title Types' would alert us to just how often serials are incorrectly used now. I have also seen magazine reprints as type MAGAZINE rather than ANTHOLOGY. My purpose is not to embarrass anyone. If a magazine serializes a novel over a number of issues, while correct, it's counterintuitive to import the NOVEL title record into every reprinted issue . I'm not advocating anything, just brainstorming with you. John Scifibones 18:30, 24 February 2023 (EST)

Hold on a second. You don't import the NOVEL title in these - you add a NEW NOVEL title, usually with a slightly different name (Part 1 or 1 added for example - mirroring the SERIAL title basically) or with a note explaining what it is and then that new title gets varianted into the main novel title as a split novel. Split novels records don't end up with the record of the complete novel inside of them, they have their own variant under the main title (the second case where we variant with difference in contents - the first being SERIALs).

Back on the reprints conversation - yes. But we can easily solve that by specifically calling the case (if the anthology/collection is a reprint of a periodical in its entirety, leave the SERIALs as SERIALs (or something like that)) even if we lean the other way otherwise. Annie (talk) 19:09, 24 February 2023 (EST)

Of course you're correct. I must be getting tired, I really do know that. Thanks for reminding me. John Scifibones 19:16, 24 February 2023 (EST)

It is Friday afternoon/evening - everyone's brain is allowed to be mush. :) Annie (talk) 19:36, 24 February 2023 (EST)

Adaptations and Abridgements

Another editor was doing some work on one of my verified publications which made me aware of changes that had been made to record subsequent to my entry of it. The title page lists "The Road to Oz by L. Frank Baum" with "Adapted by Peter Archer" in a smaller font underneath. My question is whether the adapter should have an author credit. I originally entered the record with only Baum listed as an author and with the contained title as "The Road to Oz (abridged)". The adapter was (and still is) listed in the notes. I posed the same question over a decade ago, along with others questions about how adaptations should be handled. Unfortunately, it didn't really result in a consensus of how these should be handled. My current thinking on how these should be handled (or how I thought they were handled) are:

• Adapters, abridgers and retellers should not be listed in an author credit, but instead should be listed in the notes as we do with translators.
• The adapted work is not varianted to the source work. Actually, I'd prefer that it were, just like we treat translations, but this could result in the necessity to nest variant titles (adapted, then translated).
• The adaptation should have the type and length based on its adapted form (most commonly SHORTFICTION adapted from a source NOVEL). This is not how we do translations which always keep the parent title type. I would prefer that translated variants followed this rule to, but I understand the reasons and the consensus to do it the way we do.
• Adding a disambiguator (e.g. "(abridged)") is useful, to prevent finding duplicates with the source title. I had originally done this with the title in the above publication, but it was changed by another editor subsequently.

One of the editors in the 2010 discussion mentioned not wanting to create a new policy because of a forthcoming "based on" variant feature. I believe that such a feature has proven difficult to implement. Perhaps we can agree on how best to enter these records in the meantime. Thoughts? Thanks. --Ron ~ Rtrace^Talk 08:15, 4 March 2023 (EST)

Adaptations and abridgements are not the same thing and, in my opinion, should be handled differently. Abridgements usually are just reductions in the amount of text (with maybe some editing to string the bits together). Adaptations can be anything from minor changes to complete re-works that make it a totally different story. Based on that, my thoughts are:

• Publications should always be credited per the title page. If the title page only credits the original author as the author, then that is how the pub record should be entered and the adaptor/abridger should be listed in the notes. If the title page credits as original author AND adaptor, then both names should be on the pub record.
• Abridgements should be treated like excerpts: 1) they should not be varianted to the original; 2) they should be dated as per the date that specific abridgement was published; and 3) if the title matches the original work, it should have "(abridged)" added to the title record title. If the abridger was not credited on the publication title page as an author (which is typically, I'm not sure I've seen a case where they were), then they should only be listed in the title notes (not varianted to a title record that is author AND abridger).
• Adaptations, if credited to just the author, should generally be varianted to a title record that is author AND adaptor and not varianted to the original. If the changes are truly only minor, I'm fine with merging with the original work, but I would see this as the rare case. Most adaptations are different enough they should be considered new works.

This does leave some room for judgement and debate, but we already have that case when an author revises their own story (how much change is enough for a new record) so that doesn't bother me. -- JLaTondre (talk) 09:45, 4 March 2023 (EST)

I wish we had a "Like" button. That all matches my thinking, too. --MartyD (talk) 11:11, 4 March 2023 (EST)

I agree with JLaTondre. ···日本穣 · 投稿 · Talk to Nihonjoe 13:29, 6 March 2023 (EST)

I have seen quite a few abridgements of Jules Verne and most credit the abridger. Sometimes the credit exists, but is not on the title page. It sounds like this would disqualify them from being listed as an author. (No opinion, just clarifying). As an example - Mysterious Island, abridged by Ann Abridger on the title page would have a publication title of Mysterious Island(abridged) or Mysterious Island (abridged) and two authors with no varianting to the original. If her credit is not on the title page we have the same title but only the original author and no varianting. If I have two abridgements that are clearly different (page count, non-title page credits) should they have separate TITLE records? (If so, need to prevent automatic merge). If two abridgements are the same, only differing in whether the abridger is credited on the title page or not, are these linked somehow? If the abridger is known, the name is to be put in the TITLE notes (they should also be kept at the publication level as a cross-check for people putting them in willy-nilly). If they are not known, is there a bucket TITLE for abridger/adaptor unknown? Or are they kept separate? (Possibly merged by publisher). There is the {{tr|}} for translators, should there be something similar for abridgers and/or adaptors? And I wonder why abridgements are only noted as such when titles match, why not all the time? Alternate titles can be just as confusing as matching titles when looking at an author's summary. ../Doug H (talk) 23:57, 6 March 2023 (EST)

Crediting per the publication title page is already the rule. If the publication has:

• Mysterious Island by Jules Verne and Ann Abridger, then both names should be listed as authors on the publication record
• Mysterious Island by Jules Verne, abridged by Ann Abridger, then only Verne's name should be listed as author on the publication record

Having additional templates for abridgers, editors, & cover designers would be a nice software improvement. -- JLaTondre (talk) 10:19, 12 March 2023 (EDT)

I generally agree with JLaTondre but there is a bit of a wrinkle here - good old translations. Especially a lot of the older ones are closer to abridgements and adaptations than straight translations and they are not always discloses to be abridged (and even modern ones in some languages). If we treat them as we do excerpts, the lack of connection to the source material will make our DB a lot poorer. So if we are going to spell a rule about all that, we need to call out translations (and clarify the rules there)... Annie (talk) 02:32, 7 March 2023 (EST)

Dates on Dust Jackets

There have been a number of edits recently adding a more precise publication date based on what appears to be date text appearing on the dust jacket. I have tended to approve these edits. I've now been asked to update one of my verified books based on such a date which got thinking about this. This is the book in question. The back flap of the jacket has "0388" printed directly below "Printed in the U. S. A.". The book was published by Doubleday who is the main (only?) publisher that used Gutter codes to identify manufacture dates. Only the year is mentioned on the title page and the copyright page. My question is whether we should consider these dust jacket dates to be publication dates, or, as with gutter codes, manufacture dates. I did a little research and found one blog entry that refers to them as printing dates. However, I wouldn't count that site as authoritative. While I lean towards the idea that this is a manufacture date, I have one more piece of data in this specific instance. The LOC copyright office gives the publication date as 1988-03-03 which is consistent with the jacket flap date. Thoughts on how these dust jacket dates should be used? Thanks. --Ron ~ Rtrace^Talk 09:19, 18 March 2023 (EDT)

Quick look through some of my hardcovers shows that all of the dust jacket dates I have match the publication month for the first printing. And they remain the same on later printings on the few books I looked at that are later printings. I’ve always considered them publication dates for the first printings (although I had never added a date solely because of them - there had been no need to). Annie (talk) 12:19, 18 March 2023 (EDT)

Can we get an update to the Help Screen for Magazines?

The discussion of Magazines as a publication type on the help screen makes reference to the magazine and book wiki pages. I doubt these exist and if so are not monitored. ../Doug H (talk) 12:41, 15 April 2023 (EDT)

Good point. Publication-specific Wiki-based discussion pages were deprecated some years ago, although you can still create a Wiki page for a publication and link it from the "Web pages" multi-field. They can be useful if you want to upload a scan of the copyright page or add other supporting documentation. I have removed references to Wiki-based pages from Template:PublicationFields:PubType and replaced them with references to the Help Desk and the Community Portal which can be used to discussed whether a publication is a magazine or an anthology. Ahasuerus (talk) 15:46, 19 April 2023 (EDT)

It also makes no reference to newspapers and journals, which printed a number of novels and stories in the days before magazines, pulps and paperbacks. I suppose in retrospect MAGAZINES should have been SERIALS and the SERIALS should have been ISSUES, or something along those lines. ../Doug H (talk) 12:41, 15 April 2023 (EDT)

Newspapers were officially included in October 2018, at which point we replaced most references to "magazines" with references to "periodicals" on Help pages like Help:Entering non-genre periodicals. I have updated Template:PublicationFields:PubType to clarify that this publication type includes both professionally published magazines and newspapers.

While working on this issue, I noticed that we didn't have a bullet point for FANZINEs, which were included many years ago and which are supported by all data entry pages. I have added a bullet point to Template:PublicationFields:PubType. Ahasuerus (talk) 16:10, 19 April 2023 (EDT)

Thank you. ../Doug H (talk) 23:34, 19 April 2023 (EDT)

The split novels again: Magician by Raymond E. Feist

Is there a reason I am missing for Magician: Apprentice and Magician: Master to be treated as separate novels and not as split novel parts? Under the current rules for split novels, they should be varianted under Magician. Can anyone see why these should be an exception? Annie (talk) 12:49, 18 April 2023 (EDT)

Defining "Published" - Take Two

The last proposal to define the term "Published" as used by Policy and various Help templates stalled in late February in part because we were not sure how e-ARCs were dated. Now that we have a degree of confidence that their stated publication dates are the same as trade editions' -- see my latest findings at the bottom of the linked section -- we can revisit the issue.

After incorporating MartyD's and Annie's changes and tweaking the wording, I have the following:

• Add a "Definitions" section to the ISFDB:Policy page. It will be a part of the "Contents/Project Scope Policy" section and will include two sub-sections: "Speculative Fiction" and "Published"
• ISFDB:Policy#Definitions_of_Speculative_Fiction will remain the same except it will be renamed "Speculative Fiction" and shifted one level within the section hierarchy
• The "Published" sub-section will contain the following text:
  • The ISFDB considers a work "published" if both of the following two conditions have been satisfied:
    • The work was made available to people and/or organizations (e.g. libraries) outside of the publisher (including self-publishers)
    • [wording changed 2023-04-27] The work was made available for reading and not for reviewing or in-house editorial/technical tasks
  • [added 2023-04-27] This definition of "published" excludes:
    • [added 2023-04-27] Advanced Reader Copies except for rare cases where they were produced for sale by the publisher
  • This definition of "published" includes:
    • limited editions
    • book club editions
    • editions restricted to subscribers, including Patreon/Kickstarter editions
    • [wording changed 2023-04-27] Advanced Reader Copies (under the name "e-ARCs" or any other similar name), but only if they were specifically produced for sale by the publisher
• Remove:
  • regardless of whether they are published within or outside the genre. "Published" is defined as follows
• from the "Included" section of ISFDB:Policy#Rules_of_Acquisition

I tried to clarify the sub-section which starts with the words "The ISFDB considers a work "published" if both" in order to address BanjoKev's readability concerns, but I am not sure how successful I was.

Thoughts? Ahasuerus (talk) 18:10, 26 April 2023 (EDT)

First of all, thank you all for the work that's been put into this topic! My immediate thought is that, to address your last line, and to restate that I have no preference for either yes-ARC or no-ARC, I don't think we can dodge the bullet which I would characterise thus: I'm holding an ARC tp which I might have bought in a bookshop or acquired at an author signing session. To me, it satisfies both points under "The ISFDB considers a work "published" if..." and so I would submit it.

As we know, strong views were expressed both for and against including such a book at the beginning of the whole discussion. I suggest we take a time-limited poll, a simple "yes" or "no", and let the majority take the day. Whichever way such a poll might go, the inclusion/exclusion of such a book can then be boldly stated. Thanks, Kev. --BanjoKev (talk) 19:15, 26 April 2023 (EDT)

I believe that previous discussions established a consensus re: ARC inclusion. We want to exclude the vast majority of ARCs which are printed "in house" and sent out to reviewers. On the other hand, we want to include (the relatively rare) editions which are called "e-ARCs" (or similar) but are offered for sale to the general public as a variation on the "limited edition" idea.

Based on your comments, the version of the proposed Policy text that I posted yesterday didn't make this distinction clear. I have added a couple of lines and clarified a few others -- please see the marked sections above. Ahasuerus (talk) 09:00, 27 April 2023 (EDT)

2023-04-30 version

Here is the text of the proposed Policy changes after the last iteration of edits and minor cleanup to ensure consistency across bullet points:

• Add a "Definitions" section to the ISFDB:Policy page; the new section will be a part of the "Contents/Project Scope Policy" section and will include two sub-sections: "Speculative Fiction" and "Published"
• ISFDB:Policy#Definitions_of_Speculative_Fiction will remain the same except it will be renamed "Speculative Fiction" and shifted one level within the section hierarchy
• The "Published" sub-section will contain the following text:
  • The ISFDB considers a work "published" if both of the following two conditions have been satisfied:
    • The work was made available to people and/or organizations (e.g. libraries) outside of the publisher
    • The work was made available for reading and not for reviewing or in-house editorial/technical tasks
  • This definition of "published" excludes:
    • Advanced Reader Copies (ARCs) except for rare cases where they were specifically produced for sale by the publisher
  • This definition of "published" includes:
    • limited editions
    • book club editions
    • editions restricted to subscribers, including Patreon/Kickstarter editions
    • Advanced Reader Copies (under the name "e-ARCs" or any other similar name), but only if they were specifically produced for sale by the publisher
  • [Added 2023-05-01] Note that not all works considered "published" are eligible for inclusion in the database. See ISFDB:Policy#Definitions_of_Speculative_Fiction and ISFDB:Policy#Rules_of_Acquisition for details.
• Remove:
  • regardless of whether they are published within or outside the genre. "Published" is defined as follows
• from the "Included" section of ISFDB:Policy#Rules_of_Acquisition

I am going to ping the editors who participated in the last couple of rounds of this discussion to see if we have reached consensus. Ahasuerus (talk) 16:23, 30 April 2023 (EDT)

Thank you for responding to the scenario I posited, there can be no doubt or room for miss-interpreting now. This is the result I was hoping to elicit when, on Marty's suggestion, I first posted the ARC question. My only thought now is what, if anything, is to be done with the (now non-qualifying) existing db entries? Thanks, Kev. --BanjoKev (talk) 19:14, 30 April 2023 (EDT)

What would be an example of an ISFDB record which won't qualify under the proposed rules? If we can identify a pattern in the Notes field, I can whip something up to facilitate the cleanup process.

Also, if the proposed change is approved, we will need to search Title Notes for the word "Patreon" and create publication records for newly eligible editions. There are only a few dozen of them, so it shouldn't be too hard to do. Ahasuerus (talk) 20:03, 30 April 2023 (EDT)

Following is a repeat of an earlier post under this topic, bearing in mind that I picked those at random. Some are obvious non-qualifiers, others would need more finesse.

"Two advanced searches of notes fields resale and not for sale reveal some interesting anomalies. Here are a few random-picked from a search on "advance": [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32]." Kev. --BanjoKev (talk) 20:26, 30 April 2023 (EDT)

I see. It looks like some of them may need to be deleted, but I am not sure there are enough of them to warrant a separate cleanup report. Hopefully we can handle them manually. Ahasuerus (talk) 16:58, 1 May 2023 (EDT)

re Patreon publication creation: I have the covers ready to be uploaded for 33 of the Glynn Stewart Patreon publications and will PV all of those publications as well. I'm not a mod or self-approver so it will take a couple of approval passes. Of course I'd also remove the Patreon notes and change the title date on those title records as part of that effort. Let me know if that would be a help or if it would just be better to have a moderator create the publications and change the titles with me then adding the PVs and the covers. Phil (talk) 22:35, 30 April 2023 (EDT)

Not sure what to do about indentation here.... :) This looks good to me. One wording clarity question: what is the parenthetical "(including self-publishers)" meant to modify in the first condition? It can be read as referring to "people and/or organizations", even though that doesn't seem to make sense. If it's trying to say self-publishers are publishers, well, that is redundant; do we need that emphasis here? --MartyD (talk) 11:23, 1 May 2023 (EDT)

"(including self-publishers)" was supposed to modify "publisher". I suppose my instincts still tell me that self-publishers are something unusual and require additional clarification even though it's no longer the case. I have deleted the parenthetical clause. Thanks for identifying the issue! Ahasuerus (talk) 17:02, 1 May 2023 (EDT)

I like the wording but if you read it on its own, it sounds like an author/publisher posting a story on a blog somewhere is considered published - it does satisfy both of the conditions we have for being published after all. We need to tie it with ROA Included I think - maybe reference to it or just a note such as "in a format and venue per ROA" or something along these lines. Annie (talk)

We can certainly add a bullet point along the following lines:

• Note that not all works considered "published" are eligible for inclusion in the database. See ISFDB:Policy#Definitions_of_Speculative_Fiction and ISFDB:Policy#Rules_of_Acquisition for details. Ahasuerus (talk) 18:16, 1 May 2023 (EDT)

Yep, that works I think. Annie (talk) 18:20, 1 May 2023 (EDT)

(unindent) It looks like we have consensus. I will give it another 24 hours and then update the Policy/ChangeLog pages. Ahasuerus (talk) 19:14, 3 May 2023 (EDT)

All done -- see ISFDB:Policy‎ and Rules and standards changelog. Thanks to the editors who participated in this discussion! Ahasuerus (talk) 17:45, 4 May 2023 (EDT)

Changing "Blocking Policy" to "Disciplinary Policy"

ISFDB:Policy currently has a section called Blocking Policy. However, it covers more than blocking, so it's really a "Disciplinary Policy" section. Can anyone think of a better name for this section's name? Ahasuerus (talk) 08:06, 7 May 2023 (EDT)

I do think 'Disciplinary Policy' would cover the possible actions fittingly. Christian Stonecreek (talk) 08:43, 7 May 2023 (EDT)

If you wanted something a little more neutral, "Comportment". That said, it is more a list of penalties than it is a list of behavioral standards, so "Disciplinary" seems fine to me, too. --MartyD (talk) 08:53, 7 May 2023 (EDT)

When I see the word "Comportment", I think "respectable/polite/dignified behavior". Perhaps "Conduct Policy" would be a more neutral synonym. Ahasuerus (talk) 13:11, 7 May 2023 (EDT)

Science Fiction Conventions commonly use "Code of Conduct" for this sort of thing. --Ron ~ Rtrace^Talk 14:14, 7 May 2023 (EDT)

Browsing the internet, I see that Wikipedia has Category:Wikipedia conduct policies, which includes Wikipedia:Civility, Wikipedia personal attacks‎, Wikipedia:Harassment, Wikipedia:Vandalism, etc. It sounds like "Conduct" is the most general term used across the spectrum. Ahasuerus (talk) 09:09, 11 May 2023 (EDT)

Outcome: "Blocking Policy" has been changed to "Conduct Policy". I don't think it needs to be documented in Rules and standards changelog since it's not a "rules" change. Ahasuerus (talk) 09:52, 18 May 2023 (EDT)

Appeals Policy

One thing you should think about adding to that policy section is an appeals process (or maybe add a more general "Appeals Policy" section that describes how a contributor goes about appealing any moderator/administrator action). --MartyD (talk) 08:53, 7 May 2023 (EDT)

That's a good point and I am creating a separate sub-section to discuss it.

There are two different types of appeals. Editors who disagree with moderator decisions can appeal on the Moderator Noticeboard, which I think is mentioned somewhere, but I don't recall where. Similarly, editors who are given a warning can post on the Moderator Noticeboard. There is no appeal process for getting banned. Ahasuerus (talk) 13:11, 7 May 2023 (EDT)

Outcome. ISFDB:Policy#Appeals_Policy has been created. At this point it simply says "Decisions made by moderators can be appealed on ISFDB:Moderator noticeboard". Ahasuerus (talk) 10:30, 18 May 2023 (EDT)

Clarifying "Non-constructive or Disruptive Behavior"

The discussion above also raises another issue. The current Blocking Policy includes a short list of disallowed actions -- spam, vandalism, personal attacks -- and leaves defining "non-constructive or disruptive behavior" to the discretion of ISFDB administrators. I find that "non-constructive or disruptive behavior" is hard to define because it's so vague and subjective. Perhaps we should clarify it, e.g. by disallowing "incivility" on the ISFDB Wiki. It would be a stronger prohibition than the current ban on "personal attacks". The penalties would presumably be lighter than for personal attacks. Perhaps 3 warnings first? Ahasuerus (talk) 13:11, 7 May 2023 (EDT)

Transliteration link

I suggest adding the link How does the ISFDB deal with Unicode and accented characters? to the Field-by-field guide. I only found that useful information by chance - not expecting it in the FAQ section. Thanks, Kev. --BanjoKev (talk) 13:52, 12 May 2023 (EDT)

Done. Thanks for reporting the issue. Ahasuerus (talk) 14:49, 12 May 2023 (EDT)

Many thanks, Kev. --BanjoKev (talk) 14:58, 12 May 2023 (EDT)

External ID template: additional wording needed for Audible-ASIN

I'm assuming that regular editors shouldn't make changes to templates. That said, in the Template:PublicationFields:ExternalIDs here, for the Audible-ASIN entry, could something like "If the Audible ASIN is an ISBN-10, convert the ISBN-10 value to ISBN-13 and place the ISBN-13 value in the ISBN field." be added after the sentence that reads "Also note that, unlike regular Amazon ASINs, we record Audible ASINs even when they match the ISBN-10."? Thanks! Phil (talk) 07:00, 22 May 2023 (EDT)

Dates for serialized novels revisit

Under our current rules, a novel is dated based on its first appearance in book form due to how serializations work and the possible changes (mainly in the US market). I had always interpreted as first appearance in the original language - a translation does not shift an original date. However, that creates records such as this one: serialized in 1970, first Russian publication in book form in 1982 (giving the record its date), 7 translations between the serialization and the first book publication (the first of them is as early as 1972). So questions:

• Does everyone interpret the dating rule as "in the original language"? If so, should we add that to the rules?
• Do we need different dating rules when not dealing with American magazines?

Thanks for any ideas/comments. If we decide to hold up the rule as it is, I am fine with it but... this novel is credited as a 1970 one everywhere but here. So thought I should bring it up and we can figure out what we want to do. :) Annie (talk) 19:02, 24 May 2023 (EDT)

I should say so (that we use the first book publication in the original language). This I think should be so in general; with the Strugatskys it would also reflect the difficulties of publishing some of their works they encountered within the USSR.

In the general approach we also would encounter problems justifying our dating of other works (not serialized) that were published only a great tiem later in their original language, I do think. Christian Stonecreek (talk) 06:25, 25 May 2023 (EDT)

I don't think that a serialization should change the dating of the NOVEL title, and I'm a little unclear as to the established rule. If I had come upon Annie's example, without this discussion, I would have thought the date on the NOVEL record was incorrect and that it should have been 1972. I can give a different example that doesn't include a serialization Mr. Breakfast which was published in translation in 2019, but not published in English until 2023. We have that one as 2019. Ignoring the serialization, these two titles are handled differently, so we're not consistent. My feeling is that we should determine whether a title's publication date follows first book publication or first publication in the language of composition. Since we're not dating the novel record based on serialization dates (except when there has never been a book publication), I don't think the existence of a serialization should change the date of the NOVEL. --Ron ~ Rtrace^Talk 07:14, 27 May 2023 (EDT)

This situation is addressed in the third point under 'General rules and information' here. I'm ambivalent as to whether an existing serialization should alter the rule. John Scifibones 08:19, 27 May 2023 (EDT)

A few thoughts:

1. I also interpret the dating rule as "in the original language" and would be in favor of using 1982 as the date of the canonical (Russian) NOVEL title. If we can reach consensus, we'll want to update Help.

2. I find that the first reason behind the current SERIAL/NOVEL date rule, which is described in Help:Use_of_the_SERIAL_type#Date_Rule:

• serially published novels are/were often extensively rewritten prior to book publication (e.g. the Lensman saga), so it's natural to think of these two versions as separate works

often applies to other languages as well. To use the Strugatsky brothers as an example, they had to make almost 900 changes to the original magazine text before Prisoners of Power (now The Inhabited Island) could be published as a book 3 years later. Similarly, Roadside Picnic, which was first serialized in 1972, had to be rewritten multiple times before it could be published in an omnibus in 1980.

3. Some bibliographies, including the Second edition of SFE, use the following dating convention:

• Galactic Patrol (1937-1938, 1950)

where "1937-1938" is the year(s) of the first serialization and "1950" is the year of the first book publication. We could update the software that displays Author, Title and Publication pages to examine each canonical title and display something like:

• Galactic Patrol (ser. 1937, 1950)

The reason why we would have to use "1937" instead of "1937-1938" is that we can easily find the date of the first serialization whose language matches the language of the canonical title, in this case Galactic Patrol (Part 1 of 6) (1937). However, we have no way of determining that Galactic Patrol (Part 6 of 6) (1938) was the last published part of the original serialization, so we have to use the date of the first known installment. Ahasuerus (talk) 14:24, 27 May 2023 (EDT)

Sounds like everyone is in favor of the current rule so so be it. :) I still think we should not be sorting these on author pages where we do now but that’s a different conversation. As we have the rule on file (as John pointed out above) that we don’t date originals based on translations dates, I think the only thing remaining here is to update the dating page help explicitly to state that so it is clear why it is 1982 in my example and not 1972 even if they do not find the other page. This novel can even be a perfect example on how exactly to date in such cases. :).

On a software level, we may think a bit more on how we display these dates, making 1970 visible here and on an author list or making the author list sortable by it may be a good idea. But that’s a feature request and not really about the rules of entry. Thanks everyone for chiming in. I will continue dating as we had always done then. Annie (talk) 18:37, 27 May 2023 (EDT)

Portuguese titles capitalisation / rationalisation

It's clear from a look at much-translated authors' pages, such as Philip K. Dick or Isaac Asimov, that we are not treating capitalisation in these titles consistently. The Title Regularization page gives no guidance. In another discussion, ErickSoares3 posted a link to this guide by the Fundação Alexandre de Gusmão.

Can we agree on how to treat Portuguese titles and formalise it on the Regularization page? Thanks, Kev. --BanjoKev (talk) 15:59, 31 May 2023 (EDT)

I am fine with using sentence case for Portuguese (even for the ones from before the reform - they used title case before that at least in some of the areas which speak the language). It will make it easier to deal with reprints anyway. Annie (talk) 16:07, 31 May 2023 (EDT)

I agree to follow the sentence case for Portuguese, following the reform agreement - there's no need to think on a whole new system for this site... ErickSoares3 (talk) 16:28, 31 May 2023 (EDT)

I don't have a preference one way or the other. As long as we're consistent and have clear guidelines, I'm fine however it ends up. ···日本穣 · 投稿 · Talk to Nihonjoe 19:07, 31 May 2023 (EDT)

As far as I know, sentence case is perfectly all right in Portuguese. See here for instance. Linguist (talk) 04:14, 1 June 2023 (EDT).

We seem to be agreed on sentence case for Portuguese titles so I'm asking, what happens next? Kev. --BanjoKev (talk) 22:08, 5 June 2023 (EDT)

Patience. Give people some more time (probably until the end of this week) - it had been less than a week and there was a long weekend in there (US-side). Once it is cleared up, the policy is set, the help pages are updated and we can go and fix any old titles we feel like fixing (and any new that get added incorrectly). Annie (talk) 11:48, 6 June 2023 (EDT)

(unindent)I had been thinking - if we are going to be updating the help anyway, maybe we should finally add the note for languages other from English. Unless I am missing a language (I will check again), we are doing sentence case for all languages but English. German has peculiarities (capitalized nouns) but this is their sentence case so sentence case covers that as well. We can even make it more explicit - explaining what sentence case is (or link to an online explanation). Something like: "For all other languages, use sentence case". We can just add the note for Portuguese for now but adding a complete note will help new international editors. Plus it just codifies what we are doing anyway. Annie (talk) 14:28, 6 June 2023 (EDT)

(unindent) From the looks of it, we have an agreement that Portuguese is sentence case as well and not an exception. The help page needs updating (anyone feeling strongly about the format of the note?) but the policy is clear. Annie (talk) 12:50, 27 June 2023 (EDT)

We seem to have consensus; please feel free to go ahead and update the Help page. I am still sick and I am not sure I can do it justice in my current state. Ahasuerus (talk) 13:27, 27 June 2023 (EDT)

I will take a stab at it later today and will open a new discussion with proposed language for the rest of the languages so we get the whole section fixed for all languages. Annie (talk) 13:47, 27 June 2023 (EDT)

Imprint Inconsistencies

The current publisher policy leaves it to the submitter's option as to whether to record an imprint as just IMPRINT or as IMPRINT / PUBLISHER. This creates inconsistencies in how data is entered and how publisher records are marked. For example, Gollancz ceased to be a publisher in Dec 1998 and became an imprint of Orion. Our Gollancz publisher record notes are written as if it is the record for the publisher only and we have a separate Gollancz / Orion record for the imprint. However, if you look at the Gollancz record you will find many books published after 1998 listed.

• Should we consider removing the option to submit imprints either way and pick just one way?

If not, we need to clean up the publisher notes to clarify these cases as it's bound to be confusing for non-regulars. -- JLaTondre (talk) 16:46, 10 June 2023 (EDT)

I see the Gollancz - Gollancz / Orion problems stemming from the publisher record notes. The Gollancz record needs rewording to forestall entries as just Gollancz being made after they were acquired by Hachette. The Gollancz / Orion record is clear but might be made a bit more direct in instructing editors to make the distinction.

Another wrinkle is that where post-1998 (i.e. Hachette) pubs have been entered with Gollancz as the publisher, and then that pub is cloned, the deficiency is perpetuated.

FWIW, I think the current policy is alright, leaving us to reach consensus on a case-by-case basis as these companies morph through acquisition. Kev. --BanjoKev (talk) 19:29, 10 June 2023 (EDT)

I'm not sure I'm following. You say you are fine with current policy. However, you sound like you are advocating that we change all Gollancz imprints to being Gollancz / Orion. What is the justification for telling verifiers (like Mavmaramis below who wants to use Gollancz only) they have to use Gollancz / Orion when the policy says they can choose either? If we want to say either form can be used, but once the community picks one, everyone has to use it, that would still be a change to what is currently written. -- JLaTondre (talk) 07:06, 11 June 2023 (EDT)

My 2p worth is that unless it's really obvious it's an imprint 'Gollancz / Orion' (either the cover or title page has it specifically printed as opposed to it being a bit of text in the printing history on the copyright page) then I'm minded to keep it to the notes. In the case of The Iron Wars; The Second Empire and Ships from the West they are externally Gollancz - Gollancz G logo on spine, 'Victor Gollancz' on title page. The menion og Gollancz being an imprint is relagated to a bit of text in the printing history on the copyright page. In fact even back in 1995 Gollancz was an "imprint of the Cassell Group". Keep it for the notes unless blatantly obvious. --Mavmaramis (talk) 06:46, 11 June 2023 (EDT)

Anthology vs. Collection Example Disconnect

We need a different example than Late Knight Edition under the ANTHOLOGY section of the PubType policy. PubType currently states:

'For example, "Late Knight Edition" contains stories by both Damon Knight and Kate Wilhelm, individually; this is an anthology, not a collection.'

However, Late Knight Edition's stories are all by Knight with Wilhelm's sole contribution being the introduction. This book has been listed as a collection in ISFDB since 2007. -- JLaTondre (talk) 08:24, 11 June 2023 (EDT)

Dating titles that have only been published in translation, and 8888/9999 dates in general

I was looking through the cleanup reports, and noticed that of the 6 titles in the Forthcoming Titles report, half of them are the English versions of works only published in translation. Unless there is evidence that these are definitely going to be published in their original form, would it not be better to mark them as 8888-00-00/unpublished?

Digging around, I couldn't see any documention in Template:TitleFields:Date about what to use for title dates on stuff like this; there's a link to Help:How_to_enter_translations, but that doesn't seem to have anything relevant.

Template:PublicationFields:Date talks about 8888 and 9999 being used for pubs that were announced for the past or future respectively, but doesn't cover the case where there has been no announcement of a title being published at all, so isn't of much help.

Also, I ran a couple of database queries, and it seems 9999-00-00 is only used on the aforementioned titles. This makes me wonder what the use case is for it, especially now that WatchPrePub exists to indicate doubts from editors about information they've entered.

   MariaDB [isfdb]> select year(pub_year), count(1) from pubs where year(pub_year) >  2024 group by year(pub_year);
   +----------------+----------+
   | year(pub_year) | count(1) |
   +----------------+----------+
   |           8888 |      644 |
   +----------------+----------+
   1 row in set (0.086 sec)

   MariaDB [isfdb]> select year(title_copyright), count(1) from titles where year(title_copyright) >  2024 group by year(title_copyright);
   +-----------------------+----------+ 
   | year(title_copyright) | count(1) |
   +-----------------------+----------+
   |                  8888 |      899 |
   |                  9999 |        3 |
   +-----------------------+----------+
   2 rows in set (0.235 sec)

ErsatzCulture (talk) 08:28, 11 June 2023 (EDT)

As I recall, "9999-00-00" goes way, WAY back as in the 1990s. Back then we didn't have proper support for future dates, cleanup reports, etc. Once the software and the data were cleaned up, I proposed that we eliminate "9999-00-00". There wasn't much interest in the idea at the time, but perhaps we can revisit the issue now that "9999-00-00" hasn't been used in a number of years. Ahasuerus (talk) 13:38, 27 June 2023 (EDT)

I've often wondered what this site was like pre-public editing, which started in 2006, I think. How many people were entering books in 1995-2006, who's still around from those days, and is there any way to tell which books were entered in that time period and where the info came from? --Username (talk) 13:45, 27 June 2023 (EDT)

When I checked in June 2022, this Wikipedia article matched my recollection of the project's early history reasonably well.

One of the big issues that the project faced in the late 1990s was the lack of a moderation system. Once public editing was enabled, it quickly became clear that data quality was going to be a significant problem.

The next issue was the use of automated tools to import data from publicly accessible online catalogs and similar sources, mostly in the early 2000s. The software that existed at the time was entirely too enthusiastic. It imported thousands of non-SF ISBNs by major authors like Robert Louis Stevenson because it couldn't tell what was SF and what wasn't.

Once the software was rewritten by Al in the mid-2000s and public editing was re-enabled in May 2006, things began to improve. Editors and moderators were recruited, old data was cleaned up, new software features were added, etc. Ahasuerus (talk) 22:29, 27 June 2023 (EDT)

Obscene account names and the Conduct Policy

ISFDB:Policy#Conduct_Policy disallows "Blatant vandalism (blanking, obscenities, nonsense, etc)". I interpret it as disallowing obscene account names like User:Fuckyouisbd, which was blocked earlier today. Should we add an explicit statement to the Conduct Policy along the lines of:

• User/account names containing obscenities are not allowed. Account names found to be in violation of this policy will be blocked and their account owners will be allowed to create a different, non-obscene, account name.

? Ahasuerus (talk) 15:02, 15 June 2023 (EDT)

I doubt the person who created that account has any intention of adding anything useful here; they couldn't even spell the name of this site correctly. Blocking trolls only feeds them; I would suggest ignoring it and only doing something if they make any edits which would mean their obscene name showing up again. Why can't people be professional? Shame. --Username (talk) 15:14, 15 June 2023 (EDT)

I agree with the change. I think allowing trolls to have access feeds them. Blocking the accounts prevents them from doing any actual trolling beyond creating the account. Does blocking them on the wiki side prevent them from making edits on the database side? ···日本穣 · 投稿 · Talk to Nihonjoe 19:37, 15 June 2023 (EDT)

Not at this time since we have never encountered this issue. Typically, malicious users are paid spammers who try to use the ISFDB Wiki as a platform for spreading spam. They are not interested in the database side of things.

That being said, it's easy to prevent blocked users from creating submissions since it requires only one or two changes to the software. FR 1575, "Prevent blocked users from creating submissions", has been created. Ahasuerus (talk) 21:18, 15 June 2023 (EDT)

FR 1575 has been implemented. Ahasuerus (talk) 20:11, 28 June 2023 (EDT)

Template:TitleFields:Juvenile

In this page there is the declaration for works targeted at the juvenile or Young Adult market. Does that mean up to 18 years or less? --Zapp (talk) 09:53, 10 July 2023 (EDT)

A very good question indeed. To quote this white paper adopted by the American Library Association in 2008:

• When the term first found common usage in the late 1960’s, it referred to realistic fiction that was set in the real (as opposed to imagined), contemporary world and addressed problems, issues, and life circumstances of interest to young readers aged approximately 12-18. [snip] the conventional definition of “young adult” has expanded to include those as young as ten and, since the late 1990s, as old as twenty-five.

Our original intent was to use the "juvenile" flag to cover works aimed at readers who are 18 and younger. Given the increased "fuzziness" of the definition as seen above, I think it would be better to spell out the age range:

• Check this check-box if this work is targeted at readers who are 18 or younger.

Ahasuerus (talk) 10:18, 10 July 2023 (EDT)

So what do we do if the book is marketed to YA in English but marked 18+ in Russian or Czech for example? It does not matter why they are marked that way really - but when two separate markets disagrees, which ones wins. Annie (talk) 11:46, 10 July 2023 (EDT)

Different publishers may use the same text(s) to target different age groups even within the same market. To quote Lawrence Watt-Evans's Web site:

• I had once hoped to maybe relaunch the [Ethshar] series from a major publisher as YA ("young adult"), aimed at teenagers, but that idea doesn't seem to have gone anywhere.

Some of it was pure marketing, e.g. one of the issues that the "major publisher" had was:

• ...the fact that the title character wasn't the viewpoint character.
• (Yes, i offered to change the title, but that wasn't enough to overcome the other issues.)

So it's not common, but it can happen even within the same market. I suppose the most logical way to classify books that have appeared both as "adult" and "YA" would be to adopt the classification used by the first edition. Ahasuerus (talk) 12:24, 10 July 2023 (EDT)

On a very separate note, lumping YA and the kindergarten books under juvenile makes us look even more old-fashioned and less inviting to younger contributors. If we mean "under 18", change the label to say so and let's not hide under a library word that is fuzzy for a reason - libraries had stopped using juvenile for YA for at least a decade even if older definitions still use that here and there. Annie (talk) 11:46, 10 July 2023 (EDT)

I can see how changing the "Juvenile" flag to "under 18" would be an improvement. Ahasuerus (talk) 12:26, 10 July 2023 (EDT)

I help classify books for a charity book sale, and we've come up with an approach to save time for newer books. For us, the "targetting' applies if the publisher's web contact address mentions children or teens (e.g. HarperTeens). We also use imprints (e.g. Scholastic), awards (e.g Newbery, Caldecott) and who is reviewing the book (e.g. School Library Journal). This, as noted above, can leave a title open to having publications differ in classification. I would suggest that once a title has been marketed as "YA", it is forever tainted - how can something once considered YA stop being YA? There are also translations (e.g. dumbed down Jules Verne) and abridgements (e.g. Illustrated Classics) which are treated differently in terms of varianting. Assuming variants can differ (for translations), would a YA variant work for titles with both variations? ../Doug H (talk) 15:00, 10 July 2023 (EDT)

Is it possible to put the conclusions of this discussion into here? --Zapp (talk) 14:27, 23 July 2023 (EDT)

awards

I notice there is a section for awards in my author listing. Am I allowed to list nominations? As in, I was a finalist for the Baen Fantasy Adventure Award, but did not win? —The preceding unsigned comment was added by Fabius (talk • contribs) . 17:59, 24 July 2023‎ (EDT)

We don't currently have either of the Baen contest awards catalogued here on ISFDB. We don't always include contests, though I think these would be good to include. Also the Writers of the Future and Illustrators of the Future winners. ···日本穣 · 投稿 · Talk to Nihonjoe 13:24, 25 July 2023 (EDT)

Pages - help screens and templates

There are 5 screens of help and guidance for entering page values; NewPub, EditPub, PublicationFields, PubContentFields, How to determine the value for the "Pages" field in a book.

In the light of recent discussions I think it would be helpful if, at the top of each screen,, there could be four lines (one for each of the other four screens) which includes a link to same. At present, 3 of the screens have a link to the "How to..." page but it's right at the end. The "How to..." page has references and links to the PublicationFields template (twice) and the NewPub page. Admittedly 3 of the pages contain identical wording, but knowing of the existence of them all, whichever page one first lands on is what I'm addressing. Thanks, Kev. --BanjoKev (talk) 17:21, 12 August 2023 (EDT)

It might be good to combine all of the information from each of those pages and create one page that can be transcluded to all of those locations. That way, the information on all of them will be identical, and any changes to the one location for the information will be propagated to all of them. ···日本穣 · 投稿 · Talk to Nihonjoe 13:58, 27 September 2023 (EDT)

I think that's an excellent idea Joe. Kev. --BanjoKev (talk) 07:45, 5 October 2023 (EDT)

Does inclusion in the Hugo Award Voter Packet count as a publication?

Apologies if this is an old topic, although I think this particular case might be a new spin on it.

There are (at least) 2 Chinese stories in the Hugo Voter Packet that have English translations provided. They are in PDF and/or EPUB formats. The original Chinese stories and their publications were added to the database when the Hugo finalists were announced, so these translations would be alternate titles to existing records. (Exception: some of them are stories for the Astounding Award for Best New Writer finalists, which I didn't add anything for at the time, because it seemed too hard/nebulous.)

At least one of those translations is scheduled to be an anthology due out later this year, and another I'm 99% certain will appear in Galaxy's Edge magazine at some point, so it's not as if (some of) these translations will never get recorded in the database.

After reading ISFDB:Policy#Included, I'm still unsure as to their eligibility for inclusion here. Maybe they fall under "Convention programs, guides, etc. We definitely want any convention-published "real books", but probably not the ephemera.", but as that note is marked as "Debatable", it's not exactly helpful...

Thanks. ErsatzCulture (talk) 19:38, 20 August 2023 (EDT)

I had a discussion (beginning with the first response) with Annie last year about this. We were both leaning towards adding the Hugo packet as a publication. I had (and continue to have) other priorities that I'd rather work on. However, I would still support the Hugo packet as a single OMNIBUS publication published by the Worldcon for the year. --Ron ~ Rtrace^Talk 21:03, 20 August 2023 (EDT)

If my understanding is correct, "Hugo Voter Packets" are sent to all World Science Fiction Society (WSFS) members -- see https://www.thehugoawards.org/category/voter-packet/ and en.chengduworldcon.com/help/1. Anyone can become a WSFS member (and therefore a Hugo/Lodestar/Astounding voter) by paying $50 per year.

For most practical purposes this system is similar to book clubs, APAs and other organizations which limit circulation to their members. Since we include book club editions, fanzines, etc, it seems to make sense to include these "Hugo Voter Packets". Ahasuerus (talk) 09:55, 21 August 2023 (EDT)

Just for the record - I still think it should be eligible as an e-book omnibus. Annie (talk) 12:32, 21 August 2023 (EDT)

Thanks all, I'll try to make a start on this year's some time soon.

One follow up question: for stuff like custom submissions that contain multiple stories or essays, I think it's better to group those as new OMNIBUS, COLLECTION, ANTHOLOGY or NONFICTION titles, which then get pulled into the OMNIBUS, rather than just have all the individual SHORTFICTION, ESSAY, etc imported directly into the OMNIBUS.

e.g. this year's Best Editor (Short Form) for Sheree Renee Thomas comprises 14 PDFs, which are an issue of F&SF, a full anthology, and 12 individual stories and essays extracted from F&SF and a couple of anthologies. Rather than import those directly into the "Hugo Voter Packet" OMNIBUS publication, I propose to have a "Sheree Renee Thomas Hugo Award 2023 Voter Packet Submission" OMNIBUS containing those, which is then imported into the top level OMNIBUS. This (IMHO) keeps things more consistent and tidy with for example, the Neil Clarke submission, which is a single PDF anthology of 13 stories and an essay. Objections/thoughts? ErsatzCulture (talk) 17:29, 21 August 2023 (EDT)

I'll defer to the software experts, but I'm pretty sure that an OMNIBUS cannot contain another OMBNIBUS. --Ron ~ Rtrace^Talk 18:45, 21 August 2023 (EDT)

Not under the current rules, no... And I really would prefer not to change this -- we had a discussion around that when someone was adding the Baen disks - creating artificial containers that had never existed is going to look ordered on the surface but will be a pain for an end user - aka - in order to get the complete list for the packet, they will need to open multiple non-existing publications (as you will need a publication for these internal omnibuses if you want to import in them). So I'd just import all stories/articles/whatever into the single omnibus and use Notes to explain what is what (and use the numbering to keep the separate pieces next to each other). If the concern is where the award/nomination gets assigned - this is not different from when a set of books are nominated - just add it to each of the title records - for the example - she did not get nominated for an omnibus containing these works, she was nominated because of all the separate works... Although technically speaking, as it is a nomination for her and not the works, these should not get the nomination added to them anyway - but if there is something where that applies, the logic is the same. Annie (talk) 18:51, 21 August 2023 (EDT)

Ah, no problem, I'll just chuck everything in the "top-level" omnibus.

The thought of adding the award nomination to those hypothetical "fake" title records didn't actually occur to me ;-) I agree that awards to people rather than titles should be done as untitled awards. ErsatzCulture (talk) 19:01, 21 August 2023 (EDT)

If you look at the thread Ron linked above, I was wondering at the time between an overall omnibus and a series/pub series for the different pieces -- mainly due to the fact that parts of it are distributed separately. But it is a special case and a single omnibus makes more sense I think -- and makes it easier to see what is inside (plus as with all other omnibuses containing other containers (collections/anthologies), you will ultimately want to add ALL contents pieces in the top level anyway for visibility - aka for people who want to see where the story can be found - as we do not have "indirect" lists so having the fake middle ones will be mostly so you can have visual separation more than anything...). Plus if we ever change our mind, we can always create the smaller containers. Does not change the fact that we want all visible in the big omnibus anyway - which means importing all in it as well...

As for the awards note - yeah I realized it as soon as I typed it but then there may be other pieces in there for which that applies so I left it and added the last note). :) Annie (talk) 19:21, 21 August 2023 (EDT)

Interior art - do we use artwork captions in the titling?

That's one of the questions arising from this discussion about the artwork in Project Hail Mary. Clarification of the rules would be much appreciated. Thanks, Kev. --BanjoKev (talk) 20:34, 25 August 2023 (EDT)

My opinion: The spirit of artwork record titling is that, except when published as a "standalone" piece of art, artwork is subordinate to the work or publication with which it is associated. Artwork record titles generally reflect that subordination. Here is what I think is de facto practice:

• COVERART titles should always be the same as that of the publication. (In fact, I think this is the one place we do not add disambiguation for the case of two different works of art by the same artist for different publications/editions with the same title.)
• INTERIORART titles in a publication of, or about, artwork should record the "natural" labeling used in the publication. If works are identified by title or caption, that text should be used. If works are identified by use case, then either the canonical title with " (use case)" appended or a descriptive title should be used. For example, if a plate in publication XYZ is publication ABC's cover, title XYZ's INTERIORART record "ABC (cover)". If a COVERART record for ABC's cover is present, XYZ's INTERIORART record should be made a variant of that.
• All other INTERIORART titles should usually be the same as that of the illustrated work, or of the containing publication if not illustrating a specific work. However, each of a publication's INTERIORART titles should be unique within the publication's contents. Where the use-the-publication-or-work's-title scheme would result in the publication's having multiple INTERIORART content records with the same title text, the titles should be disambiguated. Different disambiguation techniques are employed, depending on use case and information available.
  • If the same artist is responsible for multiple works of art that are being recorded separately, the title text for each must be made unique.
    • If the works have titles or captions, those may be used.
    • If the works have different use cases, append " (use case)" to one or more of the otherwise ambiguous records. E.g. "ABC (map)".
    • If no better differentiator is available, append " [number]" to each of the otherwise ambiguous records. E.g., "ABC [1]", "ABC [2]",...
  • If different artists are responsible for different pieces of art, the normal titling scheme is followed, with each INTERIORART record having the same title text but different Artist credits. Note that "use case" disambiguation may also be employed in this case. E.g., "ABC (maps)" by artist 1 and "ABC (illustrations)" by artist 2. If differing artist credit alone is not sufficient to produce uniquely identifiable records, then one of the disambiguation schemes should be applied first to produce the title text, then the appropriate artist credit should be assigned. E.g., "ABC [1]" by artist 1, "ABC [2]" by artist 2, "ABC [3]" by artist 1.

As I said, that is my opinion. I would also note that ISFDB's view of artwork has changed over the years. We used to treat artwork as much more of an afterthought/second-class data citizen than we do today. So, for example, you will see disambiguated-by-number records entered long ago where today we would use some more readily identifiable form of disambiguation. Or older single publication-wide records where today we would tend to use multiple records to document each of the individual works. Some of the help text may not be fully in tune with the times. --MartyD (talk) 07:20, 26 August 2023 (EDT)

re "I think this is the one place we do not add disambiguation for the case of two different works of art by the same artist for different publications/editions with the same title": Cover art is not a special case. We only disambiguate artwork titles within the same publication, not across publications. I agree with you on the remainder. -- JLaTondre (talk) 08:20, 26 August 2023 (EDT)

If I read Marty's reply correctly, what it boils down to is that for the art's title, the illustrated work's title is used with all the disambiguation cases etc, as explained above (and except for the bullet point 'If the works have titles or captions, those may be used [to make them unique]' - which I don't read in the current rules btw).

My interpretation of the rules is exactly that, ie. the title of INTERIORART is the same as the title of the work it illustrates - even though there are several examples currently in the DB where the actual INTERIORART title or caption are used as title, instead of the title of the work the art illustrates. The issue that I'm having with the current rules is that they are not very clear in explaining what title to use, hence should be rewritten to make them unambiguous - because right now, the rules do not clarify what do to in case there's artwork that has a proper title of its own. - cfr. the discussion here. I have two proposals to make the rules clearer:

* INTERIORART always get the title of the work it's illustrating. If the work does not illustrate any particular work, use the title of the publication the art appears in, or

* If INTERIORART has its own title or caption, use that title or caption. Else, use the publication's title instead

(+ the disambiguation cases laid out by Marty above, of course). Thoughts? MagicUnk (talk) 12:03, 28 August 2023 (EDT)

As has been noted by others, if the interior art has a caption, use that for the title. Otherwise, it should be using the title of the work plus a disambiguator as noted above. ···日本穣 · 投稿 · Talk to Nihonjoe 13:15, 28 August 2023 (EDT)

Yes, but that's not what the current rules say. Do we agree to amend the rules to make it clear that the caption should be used if there is one, and the title of the work in all other cases? (we may want to refine for artwork publications). Regards, MagicUnk (talk) 11:46, 29 August 2023 (EDT)

It should be optional, not a requirement. Same as it is optional to enter individual titles or leave it as one record for the entire pub. -- JLaTondre (talk) 18:25, 29 August 2023 (EDT)

Works for me. Anyone else who'd like to chime in? I'll try to come up with an update for the rules text to clarify that INTERIORART gets the title of the work it illustrates, and if there's a caption, that caption can be used instead. MagicUnk (talk) 05:10, 30 August 2023 (EDT)

(unindent) If "... INTERIORART gets the title of the work it illustrates" means the publication title, then I object. It would make my favored approach outside standards. The title record Winds of the Forelands (maps) covers all the maps used in a series. It clearly shows how the maps are credited, where they appear and is easily edited if additional volumes are published. John Scifibones 07:44, 30 August 2023 (EDT)

I agree with John. It's important to be able to use one record for the same illustrations (maps in particular) used in a series. Sometimes the illustrations don't have a caption or there are several possible captions. A grouping title can provide a container that clarifies the use of the illustrations without unnecessarily duplicating them. The approach being discussed doesn't seem to provide for the flexibility to use a grouping title. It also feels like the proposed approach could inflate the number of works attributed to a given artist. Phil (talk) 08:20, 30 August 2023 (EDT)

The rules currently state that artwork is only supposed to be titled per the title of the work (story or publication). The above is relaxing that rule to match how things generally are done. I'm fine adding an additional relaxation for "series" artwork as I agree combining maps makes sense. But if you are both objecting to any change, then you should realize your way of handling maps is not valid per the current rules. -- JLaTondre (talk) 17:11, 30 August 2023 (EDT)

The current standard for Maps - "Maps. These are considered interior art for ISFDB purposes and are typed as INTERIORART. The format for titling maps is "Title of Work (map)", for example: Brightness Reef (map). Optionally, if a map is titled you can use the stated title of the map without appending the name of the work, for example The Land of Nehwon (map)." (emphasis added) I interpret work as inclusive (publication, series, or story).

Note the wording in MagicUnk's proposal - "INTERIORART always get the title of the work it's illustrating. If the work does not illustrate any particular work, use the title of the publication the art appears in..." (emphasis added) The change from work to publication was the source of my objection.

If the original intent was for work to be synonymous with publication and story only, then I am indeed proposing a change. John Scifibones 19:05, 30 August 2023 (EDT)

Concerning illustrations (eg maps) repeated in multiple volumes (of a series), under the current rules there is always the possibility to variant titles. That will effectively tie them together - under the current rules there's no need to 'invent' a common title for use across a series.

Mind that I'm not saying that we can't change the rules, but the change John's (and Phil's) proposing requires more discussion before (if) we can accept the change and can update the rules accordingly. What do we do with INTERIORART that has

• a caption, artwork identical, and that caption is identical across the volumes of the series --> this is an easy one; use the caption. Will need a rules change, but per the discussion above I'm fairly certain everyone's OK with adding 'if it has a caption, you have the option to use it'
• a caption, artwork identical, but captions differ between volumes? --> since we'd make the use of the caption optional, we could decide to either use the series' title instead, or go the variant route, using the different caption titles (this latter would be my preference, as that's common practice for variant work titles anyway)
• no caption, artwork identical, --> either use the title of the work it illustrates and variant per the other volumes, or, use the series title instead
• combination of the above - might not be common, but can't be excluded either imo

and then I've not even touched John's example: how to write down the conditions to cover this case where there's a grouping of different maps involved, which are not identical across volumes?

Note that using the series title has its own challenges: what with series titles that change over the years? Are we going to go back and update all INTERIORART titles that were based on the old, no longer applicable, series title? What with series titles that we've "invented"? Those that are not to be found on or in the publication? Is using these "invented" titles for INTERIORART a good idea?

Lastly, we're now having two topics to discuss: "optional usage of caption", "usage of series title". What do you say, split the discussion in two sub-discussions? (splitting would allow us to update the rules to at least allow usage of captions...) MagicUnk (talk) 05:44, 31 August 2023 (EDT)

Splitting it seems reasonable. Phil (talk) 22:06, 31 August 2023 (EDT)

Numbering of pages numbered in the ToC but not numbered themselves

Please go read this discussion for background. Please keep comments here, though, since this discussion will be referred to regarding any outcome.

Here's the summary: For pages prior to the main content, we generally use the numbering found on the pages themselves (this is the same for all other content, too). In some cases, those pages don't have any numbering on the pages themselves. For those, we generally include the number of those pages in square brackets prior to the main page count. For example: "[12]+374" for a book that has 12 unnumbered pages of recordable content (maps, introductions, etc.) prior to the main content. In the case linked above, the table of contents gives Roman numerals to that content, so I used that in the numbering ("[x]+690+[3]") and included a note to that effect in the notes for the publication: "Although no roman numerals are printed on any pages, the Contents page lists Maps beginning on page viii."

The question is whether using the Roman numerals is what should be done here (and in other such cases). On the Help:Screen:NewPub page, it states "Caution: Do not use the table of contents to determine the page numbers of a publication's contents." My understanding of this is that it's meant to prevent us from using the table of contents page numbers when they disagree with the actual page numbers (basically, when the publisher forgets to update the table of contents when a change is made that affects the page numbers).

However, I don't think it should be applied in this case since it's the reverse of what I believe the intention of that rule is. In this case, the pages themselves don't have any page numbers on them. Rather, the only place the page numbers are given is in the table of contents. Because of this, there's no disagreement between the actual page numbers (since there aren't any) and the table of contents.

So, let's sort this out. Should we completely ignore page numbers in the table of contents in all cases? Are there cases (like the one described above and at that link) where we should use the information found in the table of contents? Is there something else that should be done?

Thanks for your input on this discussion. ···日本穣 · 投稿 · Talk to Nihonjoe 19:02, 17 October 2023 (EDT)

If I am reading this correctly, you are thinking that where Help:Screen:NewPub says:

• Caution: Do not use the table of contents to determine the page numbers of a publication's contents

it was actually originally meant to say something like:

• Caution: When a page number in the table of contents contradicts the page number in the body of the publication, use the page number in the body of the publication

Or, perhaps:

• Caution: If a Contents item doesn't have a page number within the body of the publication but has a page number in the table of contents, enter the latter in the Page Number field and put square brackets around the value

? Ahasuerus (talk) 21:58, 17 October 2023 (EDT)

I think the intent of it was the first one, as that's how I've always seen it applied in the past. ···日本穣 · 投稿 · Talk to Nihonjoe 13:00, 18 October 2023 (EDT)

The thread title misstates the fundamental problem. At question is the proper handling of unnumbered pages before page 1 which contain indexable content. Proper determination of the Pages field in the publication metadata is the source of contention. I maintain that this situation is addressed in bullet point 3, under Pages, here. User:Nihonjoe argues that Arabic numerals are not required and Roman numerals may be used instead. I see nothing in the help which allows this. The help specifically calls for Arabic numerals. The proper entry for the page field of each content title flows directly from the publication Pages field.

If we decide that Roman numerals are appropriate, bullet points 2 and 3 will need to be completely rewritten. Of course I will support any consensus decision. John Scifibones 18:06, 18 October 2023 (EDT)

Sorry if it was confusing for you, but that wasn't my intent. Perhaps the title of this discussion isn't as clear as you would prefer, but the post itself is very clear. I was trying to be concise as really long section titles can be cumbersome.

Regarding the rest of your comment, it really depends on the definition of "unnumbered" since I'm arguing that the ToC does number the pages since it has page numbers and the pages themselves do not. We need to determine if the ToC can absolutely never be used for any page numbers, or if (as I'm arguing in this case) it can be used for those page numbers when the ToC has them but the pages do not have them and the page numbers cannot be derived from surrounding pages that do have page numbers. ···日本穣 · 投稿 · Talk to Nihonjoe 18:51, 18 October 2023 (EDT)

I think that using the numbers from the table of contents, with a note stating so, makes more sense in this case than inventing new numbers and discarding information printed in the book. I've always read this part of the help in the same way as you - it is there to define what to use when the actual book and the contents page disagree not to prohibit using the TOC when it is the only source.

With this being said, I can see the other side of the argument (for consistency sake if nothing else) - but my gut feeling is to go with what is printed in the book itself. Annie (talk) 20:08, 18 October 2023 (EDT)

(after edit conflict) I am in Annie's camp. I don't have strong feelings about this, other than I think from a database user's perspective, it would be somewhat strange to have content listed as on "[7]" when the TOC says it is on "v". My inclination is to adjust the "Caution" wording slightly to say that page numbers should be taken from the numbers printed on each content item's page, not from the TOC. Then in the "Pages without a printed page number" section add a bullet stating that if the page is given a number in the TOC, that number should be treated as if printed on the page, as long as not in conflict with numbering printed on other pages or with the number of physical pages in the publication. Something like that. That should be compatible with the other rules, page count determinations, etc. --MartyD (talk) 20:21, 18 October 2023 (EDT)

My only real problem with using a Roman numeral found only in the ToC is that if a reader were to pick up the book, look at the ToC, and try to go to that page, they couldn't find it using the page reference. No matter what, there definitely needs to be note describing the situation. More than anything, I would just like a well-stated, clear rule to apply. Phil (talk) 21:34, 18 October 2023 (EDT)

I definitely agree. Having a note in these cases is very important. Having a clear and concise guideline is as well. ···日本穣 · 投稿 · Talk to Nihonjoe 12:29, 19 October 2023 (EDT)

(unintend) Let me clarify a couple of things. We are currently discussing Unnumbered pages within a range of unnumbered pages scenarios. Help currently says:

• If a content starts on an unnumbered page within a range of unnumbered pages, its page number should first be derived and then entered in squared brackets. The page number can be derived by counting forward from the first page of the section of unnumbered pages. For example, if a content appears on the fifth page in a range of unnumbered pages, enter "[5]".

If I understand it correctly, the proposal under consideration would add a sub-rule after the second sentence, something like:

• If the table of contents specifies the page number where the content starts AND that page number matches the number derived by counting forward, then use the numerals (i.e. Arabic or Roman) found in the table of contents. If the page number in the table of contents doesn't match the number derived by counting forward, then use the number derived and Arabic numerals.

The caveat after the capitalized "AND" above would be presumably needed to account for situations where the page number in the table of contents doesn't match the number derived by counting forward since we all know how bad tables of contents can be (my "favorite" example is here.)

Am I reading this correctly? Also, will this affect Unnumbered pages within a range of numbered pages scenarios which are covered by a separate Help paragraph? Ahasuerus (talk) 18:12, 20 October 2023 (EDT)

Very close to an edit conflict with Ahasuerus.

Ahasuerus: Your understanding of the discussion re: Unnumbered pages within a range of unnumbered pages is correct. The situation of Unnumbered pages within a range of numbered pages has not yet been considered.

What follows below is what I had prepared to say before Ahasuerus jumped in first. :-) Teallach (talk) 18:58, 20 October 2023 (EDT)

The ISFDb rules already have a method for assigning page numbers to unnumbered pages that are not derivable by counting forwards / backwards, namely, the use of Arabic numerals in square brackets. So we don't need to resort to a secondary source for the page number. The way Pages are denoted in the ISFDb is already horrendously complicated and if we adopt the use of Roman numeral page numbers from the ToC where no number is printed on the actual page then we introduce further complications and also open other cans of worms. Examples:

1) Should the Roman numeral be enclosed in square brackets? This is currently not supported in the ISFDb rules.

2) Suppose a map is on an unnumbered page that is derivable by counting backwards (page 4, say) but the ToC lists it on page iv? What do we do? [Ahasuerus' proposed sub-rule addresses this case]

3) Suppose there is an article on an unnumbered page that is not recordable in the Contents section but the ToC lists it with a Roman numeral page number? What do we do?

If we use page numbers from the ToC then all the consequences and implications need to be considered and documented.

I am in favour of not using page numbers from the ToC where no number is printed on the actual page.

Whichever way this goes:

i) the Help notes need updating to clarify what to do

ii) a pub note definitely needs to be added to explain the discrepancy and the Help notes should state this. Teallach (talk) 18:59, 20 October 2023 (EDT)

Here are a few questions using the publication which caused me to raise this issue, The Bavarian Crisis. Pages: '[x]+690+[3]'. L-O-C '690' pages

• Is anyone else concerned that the Pages field will differ from all secondary sources? (L-O-C in the above example). When we use bracketed Arabic numerals it's an obvious ISFDB construct.
• Looking at my copy, viii is the only Roman numeral in the TOC. I assume [x], brackets addressed by Teallach, is a count of the total pages before page 1. This differs from how we presently deal with Roman numerals. Should the Pages field be 'viii+690+[3] or would that be another explanation in the help section?
• I repeat for emphasis Teallach's point 3.
• The Pages field will become impossible for a reviewer to confirm unless they own the publication or there is a scan available. John Scifibones 14:12, 21 October 2023 (EDT)

Regarding each point:

• Our page counts already often differ from those at many secondary sources. Whether the bracketed numerals are Arabic or Roman doesn't make our way of listing page numbers any less an "obvious ISFDB construct". There are a number of things we do here which can be confusing to people outside of ISFDB (the whole CHAPBOOK thing, for example). In this case, the only reason I put the Roman numerals in brackets was because the pages themselves are not numbered, and we'd do the same thing if they were completely unnumbered (meaning no mention of page numbers in the ToC or on the pages themselves).
• The [x] is the total number of unnumbered valid content pages, derived from counting forward and backward from the one page number mentioned in the ToC for the pre-story content. Since the pages themselves didn't have any actual page numbers on them, but the page number for one of the pages was listed in the ToC, I used that.
• I don't really understand what Teallach means by "Suppose there is an article on an unnumbered page that is not recordable in the Contents section but the ToC lists it with a Roman numeral page number? What do we do?" If the content is not recordable, then we don't include the content, regardless of whether it appears in the ToC or not, and regardless of whether it has page numbers or not. We do include the page numbers, however (for example, if there's an "Acknowledgements" or an "About the Author", and the pages were numbered, we'd include them in the page count but wouldn't record the content as a separate title. I would also include a note explaining the situation.
• Unless a reviewer has a copy of the publication (whether physical or a PDF or scan of the publication in question), they wouldn't be able to confirm anything anyway. Maybe I'm misunderstanding this concern, but it seems like a non-concern from how I'm reading it. ···日本穣 · 投稿 · Talk to Nihonjoe 15:24, 23 October 2023 (EDT)

This pending edit, https://www.isfdb.org/cgi-bin/view_submission.cgi?5796089, relates to this discussion. Is the way I entered numbers the way it's been decided they're supposed to be done? Because it does mention "179" on contents page. --Username (talk) 00:49, 24 October 2023 (EDT)

Nihonjoe: here is an example to clarify my point 3).

The text of a novel starts on a page with a printed number of 1 and finishes on a page with a printed number of 999. There are ten unnumbered pages in the book before the start of the novel. A one page "About the Author" article appears on the fifth of these pages. The ToC lists the "About the Author" article and assigns it a page number of v.

Now, we don't record the "About the Author" article in the Contents section but what do we put in the publication Pages field? The possibilities seem to be 999 or v+999 or [v]+999 Teallach (talk) 18:54, 24 October 2023 (EDT)

I'd do either v+999 or [v]+999 (depending on if we want to count the ToC assigning a page number as "numbered" or "unnumbered"), unless the "About the Author" is multiple pages, and then I'd extend the Roman numeral count accordingly. In your example, I'm assuming there is no other content, recordable or otherwise, outside of the "About the Author" section? ···日本穣 · 投稿 · Talk to Nihonjoe 19:39, 24 October 2023 (EDT)

I was not looking for a solution to the example. I just provided it to clarify my case 3 because you said you did not really understand it. At this stage of the proceedings I do not consider it appropriate to start working solutions to the three cases I raised, firstly because they will not be relevant if the consensus is that we do not use page numbers from the ToC where the pages are not numbered themselves and secondly because we risk losing focus on the main issue. The existing rules for Pages are already very complicated. If we do use page numbers from the ToC where the pages are not numbered themselves then all those cases I described (plus possibly others that I and other editors / moderators have not thought of or raised yet) will need to be discussed, agreed upon and have additional rules added to the Help Notes on Pages to deal with them. This will make the rules for Pages even more complicated. I am very much against doing this unless it is necessary because the more complicated the rules are, the easier it is for editors and moderators to make mistakes. In this situation, it is not necessary. In my opinion, it's not even desirable. If we decide to not use page numbers from the ToC where the pages are not numbered themselves then we just need to add one sentence to this effect to the Help Notes and we are done. Teallach (talk) 18:41, 25 October 2023 (EDT)

I don't think it's a major change either way. Printed page numbering directs how we record the page number and the count of pages in the block where the numbered page appears. For pages with no numbers, either we always count and always use Arabic numerals, or we allow pages to be considered numbered by proxy via the TOC first, before defaulting to the counting + Arabic numeral scheme. Use of the TOC, however, would need some kind of caveat to cover the case where a TOC is reprinted from a different format edition without adjustment and does not match the layout (similar to copyright page/printing statement handling). --MartyD (talk) 06:14, 26 October 2023 (EDT)

Other Missing Values on the Title Page

It occurs to me that the "no page number on the title page" is related to other "missing values on the title page" scenarios.

What do we do if a story or an essay doesn't have a title printed on the title page, but the information appears elsewhere within the publication, e.g. in the table of contents? Help:Screen:NewPub#Regular_Titles says:

• For short stories, essays and poems, when working from a primary source, always take the title from the heading on the page where the work begins. The title shown in/on the table of contents, running page headers, index, front cover of the publication, secondary bibliography, or a promotional website listing is secondary.

However, what does "secondary" mean in this case? Does it mean that we can use "secondary" titles if no title is given on the title page? If so, then we should spell it out and also explain the hierarchy of "fallback scenarios", e.g. whether the version in the "running page header" should be used before the version in the table of contents.

Similarly, what do we do if a story or an essay has no author credit? In most cases we use "uncredited", but Help:Screen:NewPub#Regular_Titles allows an exception:

• If an individual work doesn't have an author credit, which is common in single-author collections, use the form of the author's name stated on the publication's main title page.

Essays whose authors sign their names at the end -- as opposed to on the title page -- are another de facto exception since we typically enter the signed names in the "Author(s)" field.

These scenarios are similar to "missing page number" scenarios in that they provide alternative values -- sometimes documented in Help and sometimes undocumented -- that editors use to populate "Title" and "Author(s)" fields. I am thinking that we should start by clarifying the current rules and bringing then up to date before we start changing the rules for page numbers. For authors, it could be something like:

• For Content entries, the order of locations to take author names from is:
  1. The title page if author name(s) are present
  2. The last page of the content item if signed by the author(s)
  3. For single-author collections only, the publication's main title page
  4. If none of the locations listed above list author name(s), enter "uncredited"

For titles, we will also want to clarify where the pub's main title should come from if the pub has no title page, which is increasingly common with independently published books. I have been using what's printed on the cover, but we really need to spell out what the hierarchy should be.

Once we clearly document the current de facto standard for titles and authors, it should be easier to decide what to do with page numbers. Ahasuerus (talk) 15:00, 28 October 2023 (EDT)

Don't forget TOC :). Not to start down a rat hole, but I believe we also don't follow strict order once the preferred location fails to provide a value. E.g., if TOC used one name and last page used another, and one was canonical, we'd likely use that. Anyway, it also sounds like we need to distinguish the "secondary" that is from-the-pub-but-not-in-the-official-place from "secondary" that is from-somewhere-other-than-the-pub. Perhaps "fallback" for the former? --MartyD (talk) 22:12, 30 October 2023 (EDT)

I have run a few database searches and it looks like we use the following values for works without a title:

• "Untitled" -- note the capitalization -- e.g. this story or this poem
• "untitled" -- all lowercase -- e.g. this story or this poem
• "Untitled" or "untitled" followed by the first few words in the body of the work in parentheses, e.g. Untitled ("1.6: These texts are a book about the people and their Gods ...") (SHORTFICTION) or untitled ("A gate in rubble") (POEM)
• "Untitled" or "untitled" followed by a short description of the work, e.g. "(Untitled Congratulations to Isaac Asimov)"
• The same as immediately above except disambiguated, e.g. "(untitled editorial) (ERB-dom, June 1973)"
• "[Untitled]" or "[untitled]", e.g. this story or this poem
• "(Untitled)" or "(untitled"), e.g. this story or this poem
• The same as immediately above except disambiguated, e.g. (untitled) (Twisted #4, Summer 1987) and (untitled) (Twisted #4, Summer 1987) [2\]

So a lot of different scenarios, all of them revolving around the use of "untitled". I don't think we have this de facto standard documented anywhere, do we? Ahasuerus (talk) 17:34, 1 November 2023 (EDT)

I believe for poems we also sometimes use the first line, or portion thereof, in quotes (without "untitled"). I believe I have done it, and I don't recall from where I got the practice. Of course, I believe lots of things.... --MartyD (talk) 11:06, 7 November 2023 (EST)

Relating to the third item in the listing of the order of locations to take author names from: I think it would be meaningful to also add novels to this item (to use the publication's main title page), in case there are forewords, prefaces, notes worthy to add, all of which are unsigned but obviously written by the author(s) of the novel. Stonecreek (talk) 05:34, 31 October 2023 (EDT)

Kindle Vella - In or Out?

We have two previous discussions I can find (this one and this one), neither of which seemed to come to any conclusion. Do we want to include them as ebooks, or do they not count as ebooks since they can only be viewed within the Kindle app or on an actual Kindle device? Would they be considered serials? They seem to be a bit outside the norm for what we accept here. ···日本穣 · 投稿 · Talk to Nihonjoe 20:19, 18 October 2023 (EDT)

Note: I've placed this submission on hold pending the outcome of this discussion. ···日本穣 · 投稿 · Talk to Nihonjoe 20:27, 18 October 2023 (EDT)

The first linked discussion petered out when we couldn't find a way to download Vella files. As I wrote at the time:

• With regular e-books that you purchase on Amazon, you go to "Manage Your Content and Devices", then "Digital Content", then "Books". When the desired book is displayed in the list, click "More Actions" on the right. In the pop-up list select "Download & transfer via USB" and click "Download". This will download the book as an azw3 file.
• When you follow the same steps for a Vella serial, you get to the last step, but the "Download" button is grayed out. Instead you get a "You do not have any compatible devices registered for this content. Buy a Kindle or get the free Kindle reading app." I haven't been able to find a way around it. Ahasuerus 16:49, 9 March 2022 (EST)

You then responded with:

• That's probably due to Vella still being in beta. I haven't been able to figure out how to do it, either. I'll keep trying different ways. ···日本穣 · 投稿 · Talk to Nihonjoe 20:05, 9 March 2022 (EST)

Any luck since then? I haven't touched Vella, so I am out of the loop. Ahasuerus (talk) 17:28, 20 October 2023 (EDT)

I haven't really tried since then. I don't like Vella myself. It's a pain to use and there's not enough there that interests me enough to make a concerted effort to try to figure it out. ···日本穣 · 投稿 · Talk to Nihonjoe

One of the books I recently added to the DB is also published on Kindle Vella. I tried in vain to find the the release dates for each chapter but gave it up as a wasted effort. If we can't get critical data like the publishing date, I'd say Out. Phil (talk) 22:02, 20 October 2023 (EDT)

Yeah, Amazon has not made it easy to figure out anything regarding Vella works. ···日本穣 · 投稿 · Talk to Nihonjoe 15:04, 23 October 2023 (EDT)

And I'd say Out as well, until the releases are collected into something which has identifying information and a release date. ···日本穣 · 投稿 · Talk to Nihonjoe 13:12, 7 November 2023 (EST)

Linking to third party Web pages -- defining "legally posted"

This Community Portal discussion got me thinking. Template:TitleFields:WebPage starts with:

• 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 [emphasis added]

Our goal when originally crafting this Help template was to make sure that we wouldn't become a hub for links to unauthorized copies of texts still under copyright protection. The Help language seemed self-explanatory at the time, but how can our editors tell whether a "version of the title's text" has been "legally posted"? For example, the main Luminist page justifies the fact that they host copyrighted works without permission as follows:

This collection may contain copyrighted material which has not been specifically authorized for our use. The copyright law of the United States (Title 17, United States Code) provides for making “fair use” copies of copyrighted materials under certain conditions, including that that the reproduction is not to be used commercially or “for any purpose other than private study, scholarship, or research.”

As I pointed out on the Community Portal, that's an odd interpretation of the copyright law:

The part of the Copyright Law that they cite -- "for any purpose other than private study, scholarship, or research" -- doesn't come from the "fair use" clause (Section 107 of the Copyright Act.) Instead it comes from Section 108, "Reproduction by libraries and archives". Section 108 is a lengthy section with a set of provisions that are completely different from the "fair use" provisions in Section 107. It's odd that the Luminist Web site cites Section 108 ("libraries and archives") language to support what they state is a Section 107 ("fair use") exception.

I should add that both Section 107 and Section 108 lawsuits can get complex and technical as we saw during Hachette v. Internet Archive in 2020-2023.

This stuff can get confusing very quickly, so I think we need a set of unambiguous rules that editors and moderators could use when deciding whether to add/approve a link to a third party-hosted text.

In addition, the fact that we currently link both to the US-based Project Gutenberg and to Project Gutenberg Australia -- which use different copyright rules and have different sets of texts available for download -- suggests that we interpret "legally posted" to mean "legally posted in the jurisdiction where the third party Web site is hosted". We may want to make it explicit in the template. Ahasuerus (talk) 20:49, 26 January 2024 (EST)

I agree that making it more clear in our documentation will be a good thing. I think we should generally avoid linking to full scans in cases where the item in question may not be in the public domain. This might mean removing some archive.org links as their track record of making sure things are in the public domain is questionable. On the other hand, they do act more like a library in that (generally) things that are not in the public domain can either be browsed on the site in a limited fashion or checked out for a specific amount of time for more lengthy review. Luminist does not do that. ···日本穣 · 投稿 · Talk to Nihonjoe 20:44, 27 January 2024 (EST)

Or how about not taking any links down unless a specific individual asks ISFDB to do that? Archive.org links over the last 3 years that I've added, several thousand by now probably, are mostly still working when I happen across them later on to update info but occasionally I'll click a link and there will be that message about the upload being taken down; could be lots of reasons and probably they do get complaints now and then from Harlan Ellison types who think they own everything but most (living) authors don't care with many glad to see their works available to such a wide audience because in many cases publishers have no interest in reprinting their books. Many (most, probably) copies on Archive.org are ex-library and often not in the best condition with people clearly donating them instead of tossing them in the trash because they know how hard many of the books are to find these days and they want people to be able to read them. I recently did some more MZB Sword and Sorceress edits after doing a lot of them long ago and noticed that 3 links to volumes in that series I added back then had been taken down so I removed those links since all 3 had one other copy also linked; they all had the kind of URL where it's obvious that someone uploaded the books themselves, not the typical Archive URL for books they digitized, so maybe somebody asked them to take their copies down. The issues of copyright around Marion Zimmer Bradley's works are notorious and can easily be read about online; one wishes her trustees cared less about protecting/profiting off her works and more about her (and her husband's) history re: children but that's another story. So that's my suggestion - let the Internet Archive handle requests to take certain books down, which they are clearly willing to do if someone asks them, and let ISFDB stay out of it and remain solely a research site. If anyone comes across a record with a link that's no longer working, just remove it. If you allow users of this site to decide what should be taken down you're going to create a huge mess with people taking down links to authors they don't like or links added by editors they don't like and I don't think anyone wants that. I'd still like the Moondust edit to be un-rejected if that's possible but if not at least people now know where to go if they want to read it. --Username (talk) 21:25, 27 January 2024 (EST)