Rules and standards discussions

From ISFDB
Revision as of 13:05, 16 June 2024 by Ahasuerus (talk | contribs) (→‎Round Two: Re; overlap between editors)
Jump to navigation Jump to search


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.



Shortcuts
ISFDB:RS
ISFDB:R&S
RS

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.

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)
Following up on this, it looks like Template:PublicationFields:Pages is already transcluded to Help:Screen:NewNovel, Help:Screen:NewPub, and Help:Screen:EditPub, but it is not transcluded to Template:PubContentFields:Page. Should we transclude it there, too? I don't think it needs to be transcluded to Help:How to determine the value for the "Pages" field in a book, and there is already a link from Template:PublicationFields:Pages (at the bottom) to Help:How to determine the value for the "Pages" field in a book. ···日本穣 · 投稿 · Talk to Nihonjoe 14:29, 22 March 2024 (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 ~ RtraceTalk 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 ~ RtraceTalk 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)
I went ahead and took a stab at the 2024 packet which can be seen here. I don't know if any of our editors with a better knowledge of Chinese than I are Hugo voters and have access to the packet, but the excerpts of the books edited by 姚海军 (Yao Haijun) are in a locked down PDF format that wouldn't allow me to copy and paste into Google translate. These are the Chinese titles beginning with 万物之主 (Herr aller Dinge). It is possible that the cover art is credited in the excerpts, but if it is, I was unable to determine that and thus added the covers as uncredited. Also, now that we have an example of how this can look, we can discuss if there is anything folks would like to see differently. I'll wait a while before I try to tackle another one of these. They are a little labor intensive to enter. --Ron ~ RtraceTalk 21:09, 10 June 2024 (EDT)
Thanks for working on this publication record, Ron. "A little labor intensive" indeed... Ahasuerus (talk) 22:22, 10 June 2024 (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:
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)
Let's first try to determine if there are areas that we all agree on. I can think of two scenarios that unambiguously fall under the "legally posted" clause of Template:TitleFields:WebPage:
  • Links to texts that have been made available by the copyright holder. ("Copyright holder" is important because in certain cases it may not be the same as the author.)
  • Links to texts that are out of copyright in the jurisdiction where the linked site is located. (The qualifier is important since copyright laws are different in different countries.) We could also add links either to our Wiki pages or to third party Web page explaining how to determine whether a given text is out of copyright in common jurisdictions.
This leaves us with texts that are still under copyright in the jurisdiction where the linked Web site resides, but the site owner claims some kind of exemption, whether it "fair use", "libraries and archives" or something else. The problem here is that it's hard to tell if the claimed exemption is (a) really in compliance with the relevant laws and (b) whether the site owner accurately represents the site's position on copyright.
Apparently the legality of linking to illegally posted copyrighted material has been an area of active litigation both in the US, where "contributory copyright infringement" is illegal (but the details are complicated -- see this article for a high level overview), and in Europe (see this discussion).
A recent example of how these things can go is Anna's Archive, i.e. annas-archive.org. When it appeared about a year ago, I poked around, found literally millions of copyrighted books and articles and immediately wondered whether it was legal. More digging discovered that they apparently had two lines of defense. First, they stated that:
  • We do not host any copyrighted materials here. We are a search engine, and as such only index metadata that is already publicly available. When downloading from these external sources, we would suggest to check the laws in your jurisdiction with respect to what is allowed. We are not responsible for content hosted by others.
Second, they had a DMCA page which let copyright owners request that links be taken down.
I wasn't sure whether it would be enough to make the site legal in most jurisdictions, but I am not an expert.
Fast forward to January 2024 and we have this 2024-01-08 report:
  • On December 4, 2023, the Italian Publishers Association (AIE) filed a copyright complaint against Anna’s Archive. [snip] AIE’s complaint cites over 30 books, emphasizing that this is just a glimpse of the content distributed by Anna’s Archive to which its members hold rights. [snip]
  • With no counterclaims from the contacted parties and clear evidence of mass infringement, an order was issued to Italian ISPs to disable https://annas-archive.org through a DNS block within 48 hours. Visitors to the site are now met with a blocking page in Italian.
Granted, we don't position ourselves as a "search engine for shadow libraries" the way Anna's Archive does, so we are in a somewhat different position. However, if we end up with hundreds or thousands of links to Web pages whose legality we can't easily determine, we may find ourselves in a legally questionable situation. It may be safer to simply stay away from sites of that nature. Ahasuerus (talk) 18:13, 28 January 2024 (EST)
There is a very big difference between hosting content and linking to someone else's hosted content. It is unreasonable to expect our editors and moderators to be expert enough to evaluate sites' legal claims. I think our policy should be something like: "Only links to content legally posted in the host site's jurisdiction are permitted, but the ISFDB is not qualified to make legality assessments. If ISFDB becomes aware of legal action resulting in the suspension or prohibition of a site's display of certain content, links to that site's posting of the content will be removed until the matter is resolved, or permanently, according to the circumstances." And then provide a mechanism to notify the ISFDB of host site legal issues/legal challenges to a site's posting(s). --MartyD (talk) 06:51, 29 January 2024 (EST)
There are currently a large number of edits in the queue adding links. Should these be held/skipped pending the results of this discussion? --Ron ~ RtraceTalk 09:36, 29 January 2024 (EST)
It looks like the consensus is that archive.org links are OK to add. By default, archive.org only lets you access copyrighted books' metadata, cover images and the first few pages of the text, which is similar to what Amazon's Look Inside does. You have to join their "Lending Library" program in order to be able to "check out" books. The legality of the LL program is currently under review by the courts and the last brief that I know of was filed on 2023-12-15. As long as archive.org remains a legitimate organization and complies with relevant court orders, linking to its Web pages shouldn't be an issue for us. Ahasuerus (talk) 16:15, 3 February 2024 (EST)
One other thing we could do is maintain a list of sites to which ISFDB has chosen to prohibit any content links (sort of a complement to the deep-linking-permitted list) due to concerns with the site's general compliance with applicable copyright laws. That should be clear for everyone, and the software could help enforce it. ISFDB is under no obligation to permit links, so legal precision is not necessary. There could be some transparent process for managing entries on the list (e.g., an R&S discussion with a definitive conclusion required). We could have some general guidelines for what does or does not merit being on the list. For example, we might decide that sites engaged in good-faith copyright protection and infringement dispute resolution -- e.g., such as Google Books, Internet Archive, and Project Gutenberg -- are not candidates despite any specific infringement complaints, while sites subject to multiple complaints and not obviously engaged in protection management and infringement dispute resolution -- e.g., such as the Anna's Archive example above -- are candidates. --MartyD (talk) 12:09, 29 January 2024 (EST)
I agree. Anna's Archive (and the once-popular site Ocean of PDF and all the others, many probably run by the same people under different names) pretends to be aboveboard but they're really just a dumping ground for pirated e-books and their download page is a list of shady sites, users being encouraged to become members if they want faster downloads, including the infamous LibGen that encourage bulk torrent downloads that are certainly not being used just for some light reading. Any site that has individual pages for each work, Archive.org, Luminist, Galactic Journey, etc. should be acceptable. Any site which mentions bulk or torrent or anything similar is a no-no. --Username (talk) 12:27, 29 January 2024 (EST)
Speaking of which, [1], I did a search for webpages with oceanofpdf and those 2 links were added by Zapp in 2023. I think they should be removed and, if you do decide to make a blacklist, Ocean of PDF should be on it, not only because of pirating but because it's virus city and you don't want anyone clicking on a link and screwing up their computer. There's no viruses on Archive.org or any of the other legit sites mentioned above. --Username (talk) 12:35, 29 January 2024 (EST)
The topic is expressly the Web Page field, but does all of this apply to recording the site or document in a Note field? ../Doug H (talk) 15:40, 29 January 2024 (EST)
I don't think different displayed fields -- Notes, Web Pages, etc -- should be treated differently for the purposes of this discussion if they link to the same third party Web sites. Notes are somewhat harder to control in the software, but that's a technical issue as opposed to a legal/policy one. Ahasuerus (talk) 16:16, 29 January 2024 (EST)

A blacklist/whitelist-based solution

After mulling it over, I think a "blacklist"-based solution would be viable or at least a good first step. It would require three components:

  • A couple of new Bureaucrat menu options to add, edit and delete blacklisted domain names like annas-archive.org, oceanofpdf.com, etc
  • A new yellow warning to be displayed when a submission tries to link to one of the blacklisted sites
  • A new nightly cleanup report to find links to blacklisted sites, which will automatically flag records once a domain is added to the blacklist

A similar whitelist of "known legitimate sites" like Project Gutenberg, Google Books, archive.org, etc would also be useful. If we implement it, we should be able to create another yellow warnings for links to domains that are not on the whitelist and may require additional digging.

Re: viruses, you are much more likely to run into them when accessing well-known illegitimate Web domains, but, unfortunately, there are no guarantees on the internet. When authors (or other people/organizations) stop paying for domain names, they become up for grabs. At that point it's anyone's guess whether they may end up in the hands of spammers, criminals, etc. Swapping this information with SFE and deleting bad links is part of what I do in the background. Ahasuerus (talk) 16:30, 29 January 2024 (EST)

I like the idea of creating a blacklist and a whitelist. I think both should require some sort of documentation supporting the addition to either list, even if that documentation is only visible to bureaucrats or admins (so that they have some sort of reference as to why a specific domain was added to one or the other). It may be good to have a "last reviewed" field, too, so we can somehow indicate when a site's inclusion on one or the other list was last reviewed (since, as you said, domain names can be picked up by someone else if the original/most recent owner chooses to not renew the domain). ···日本穣 · 投稿 · Talk to Nihonjoe 13:49, 30 January 2024 (EST)
I can see how a "Note" field would be a useful addition to the proposed table of blacklisted sites. Its contents could be made available to moderators reviewing the proposed cleanup report. Ahasuerus (talk) 14:39, 30 January 2024 (EST)
Hearing no objection, I have created FR 1590, "Create a blacklist of disallowed third party domain names". Ahasuerus (talk) 09:38, 4 February 2024 (EST)
As per the discussion immediately below, the wording of the FR has been changed to "Create a blacklist of disallowed third party URL patterns". Ahasuerus (talk) 08:53, 4 April 2024 (EDT)

Luminist's PDF files

Reviewing the above discussion, and until the FR is implemented, I note that we agreed there was consensus for adding links to archive.org. I'm seeing new edits to add links to pdfs hosted by wasabisys.com. This seems a different kettle of fish. Do we have consensus on whether links to downloadable pdfs from this site should be allowed? --Ron ~ RtraceTalk 17:49, 28 March 2024 (EDT)

I haven't seen wasabisys.com, which redirects to wasabi.com, before. Based on this FAQ it appears to let anyone upload and store arbitrary amounts of data. Kind of like Google Drive or Amazon's S3, right? Ahasuerus (talk) 19:34, 28 March 2024 (EDT)
This is the edit that gave me pause. I believe Username refers to the as "luminist" links. The ones he has added all appear to be served from the wasabisys domain. The question would be whether wasabisys has any safeguards to prevent copyright violation, or are they a site that will host files for bad actors. I stopped approving the addition of any links to scans of books under copyright when this topic was raised. I resumed approving links to archive.org once we had consensus to include those, but am hesitant to approve others if we haven't agreed that they are acceptable. --Ron ~ RtraceTalk 20:07, 28 March 2024 (EDT)
They're from a site at Luminist.org, the URL's contain the word luminist, and the guy who runs the site calls himself Luminist, https://isfdb.org/wiki/index.php/User_talk:Luminist. Also, links from when the site still used Adobe document links are to be found in many PV Analog records on ISFDB, [2], added a few years ago by Dave888 and approved by...RTrace. I did add the Naked Storm one, though. --Username (talk) 20:24, 28 March 2024 (EDT)
Oh, right, I remember it now. Luminist has apparently moved all (?) of his PDF files to wasabisys.com -- see our Web Page Search results. I expect that it may be a more cost-effective solution for small operators since sites like wasabisys.com and backblaze.com host files relatively cheaply, in the $6-7 per month per terabyte range.
This presents a problem from our perspective since the solution proposed above was to create a "blacklist" of sites which are known to violate copyright: Anna's Archive, oceanofpfd.com, etc. With an aggregate site like Wasabisys, Backblaze or even Google Drive, there may be no easy way of telling who the owner of the linked files is. It makes the "blacklist" approach unworkable for this type of cases. Still useful in other cases, but not as comprehensive as I hoped it would be.
I note that all Wasabisys.com links start with "*wasabisys.com/luminist/", so it may be something to pursue, although it wouldn't help with files hosted by Google Drive since it doesn't have that kind of convenient URL structure.
Going back to the Luminist situation, he hosts a variety of PDF files. There appear to be three separate types of scenarios:
  • Scans of books that are no longer under copyright protection, e.g. A Trip to Venus (1897) or The Altar of the Legion (1926), which were published before 1929 and are therefore in public domain in the US.
  • Scans of books published between 1929 and 1963. Their copyright status is often unclear since they only enjoy copyright protection if copyright has been renewed, which is rare for genre books like Zip-Zip Goes to Venus (1958). Project Gutenberg and some other sites look for copyright renewal notices in The Catalog of Copyright Entries before making their files publicly available, but Luminist doesn't seem to do it.
  • Scans of books published after 1963 and therefore still under copyright protection. Luminist justifies it 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.” By accessing files linked to this site you are agreeing to abide by these restrictions. If you do not agree, do not download. If any copyright owner objects to our inclusion of their material on this web site, please do not harass our hosting providers; just contact us with the pertinent information. We will remove contested content promptly upon receipt of legitimate requests. Readers who wish to obtain a permanent copy of any item are encouraged to acquire one from a bookseller of their choice.
This is presumably based on Chapter 1, section 107 of Title 17, "Limitations on exclusive rights: Fair use". Perhaps it may be argued that some relatively obscure books like The Tsaddik of the Seven Wonders (1971), which hasn't been reprinted since 1981, are only of interest to researchers. However, Luminist also has scans of books that have been recently reprinted, e.g. The Secret of Barnabas Collins, which has had multiple editions since 2019.
I don't think we are (or should be) in a position to decide which post-1963 books fall under "fair use" and which ones do not. "Fair use" cases are settled by the courts on a case by case basis; we don't have the knowledge or the resources to deal with its complexities. Template:TitleFields:WebPage currently allows:
  • legally posted versions of the title's text
but doesn't define "legally posted". My current thinking is that we could clarify it to disallow "texts known to be under copyright protection and made available without the copyright owner's permission". The clause "known to be under copyright protection" would exclude everything from 1964 on.
If we decide to do this, then it would be easy to create a cleanup report to look for PDF files associated with post-1963 publication records.
Thoughts? Ahasuerus (talk) 13:55, 31 March 2024 (EDT)
That sounds workable, at least for moderation and as a guide to editors for what is allowed. How hard would it be to add a yellow warning (for both editors and moderators) for this? Not a big hurry for that, but it would make things easier, assuming that others, if any, agree with handling Luminist and Wasabisys in this manner. --Ron ~ RtraceTalk 16:46, 31 March 2024 (EDT)
It would be a simple task. The process of adding new warnings has been much more straightforward since the "yellow warning" system was revamped in 2023. Ahasuerus (talk) 17:09, 31 March 2024 (EDT)
Date-based warning sounds like a handy reminder.... @Ahasuerus, you could think about a pattern-based approach to blacklist/whitelist, instead of relying strictly on domains. E.g., for the Luminist example on Wasabi, a pattern might be: *.wasabisys.com/luminist/* (or whatever pattern-specification syntax appeals to you -- regex, SQL, ...). Since the pattern itself would not be created by ISFDB end-users, but rather "internally", it doesn't really matter what the pattern syntax would be, as long as we can explain it in plain English. --MartyD (talk) 17:34, 31 March 2024 (EDT)
A good point. We'll just have to change the name of the menu option and the text of the associated yellow warning from "Blacklisted domains" to something like "Blacklisted Web page URL segments". Ahasuerus (talk) 18:06, 31 March 2024 (EDT)

Clarifying editor data entry rules in Help

Earlier today an ISFDB editor pointed out that Help:Screen:NewPub does not explicitly tell you what to put in the "Author" field for MAGAZINE publications. Template:PublicationFields:Author, which is transcluded in Help:Screen:NewPub, says:

  • If it is an ANTHOLOGY, multi-author OMNIBUS, or multi-author work of NONFICTION, credit the editor as the "author" of the publication.

but doesn't mention MAGAZINEs or FANZINEs. I am thinking that we should add something like:

  • For MAGAZINEs and FANZINEs, credit the issue editor as the "author" of the publication. Note that for non-genre MAGAZINEs/FANZINEs, "Editors of PERIODICAL NAME" may be used instead of some or all editor names if they are unknown or unclear or not of genre interest -- see Help:Entering non-genre periodicals for details.

How does it sound? Ahasuerus (talk) 20:10, 2 February 2024 (EST)

Sounds right to me. --MartyD (talk) 15:39, 4 February 2024 (EST)
Sounds good. ···日本穣 · 投稿 · Talk to Nihonjoe 19:39, 5 February 2024 (EST)
Yes, it does fill out a very minor hole in the rules, but it will actually be helpful in some cases. Christian Stonecreek (talk) 05:59, 6 February 2024 (EST)

Outcome -- Clarifying editor data entry rules in Help

Template:PublicationFields:Author has been updated with the proposed language. Thanks, folks. Ahasuerus (talk) 10:49, 9 February 2024 (EST)

Currency codes

Recently I uploaded records for an Estonian book from the Soviet Union, which cost SUR 1.40. Soviet rubles SUR were in use from 1961–1991, Russian rubles RUR were in use from 1992-1997, and now the new Russian Ruble RUB is in use since RUR was devalued to RUB at a rate of 1000 to 1.

Similarly I uploaded a Bulgarian book whose cover price said "2 лв" meaning 2 levs. But there is no single Bulgarian currency. BGJ was used 1881-1952, BGK from 1952–1962, BGL from 1962-1999, and BGN is used now since 1991.

The thing is, a currency is NOT a currency just because it has the same name. The US, Canadian, and Australian dollars are not all just dollars just because they use the word "dollar" or the dollar sign "$". Estonia na SUR, then EEK, now EUR. In that case, the names changed too (ruble > kroon > euro). In Bulgaria the word "lev" applies to BGJ and BGK and BGL and BGN, but despite the name they aren't the same currency and if our database doesn't have the correct currency for a publication then the currency field is essentially worthless apart from USD and CAD and so on.

ISFDB isn't a pricing database, but its information really must be accurate. See ISO 4217 for currency codes.

I cannot find an actual link to an actual list of Rules and Standard, but I entered BGL when I uploaded the book and one of the admins changed it to BGN, which is simply not correct. If Bulgaria were to give up the lev and take up the euro, would we change all the BGNs to EUR? No; so we should not change BGL to BGN. Evertype (talk) 11:19, 9 February 2024 (EST)

Do you have a link to the publication that was changed? Also, you can find a list of currently-supported currencies at Help:List of currency symbols. Yopu're welcome to propose additions to the list, too, if there are some we should have but which aren't on that list. ···日本穣 · 投稿 · Talk to Nihonjoe 11:39, 9 February 2024 (EST)
Here is it. And the change and explanation about why was shared on the Editor's page together with the links to the help page. Annie (talk) 11:45, 9 February 2024 (EST)
This topic has come up a number of times. The longest Rules and Standards discussions were in July 2013 and June-July 2017. Here is what I wrote about the challenges associated with using ISO codes instead of currency symbols in 2017:
  • ... the ISO standard assigns a new code when a currency is revalued, so the code for the Mexican peso changed from "MXP" to "MXN" when the peso was replaced with the "new peso" ("nuevo peso") in 1993. In 1997 the word "nuevo" was dropped, so it's now back to just "peso". However, the ISO code has remained "MXN". If we were to use ISO codes, what should an editor do when entering an undated Mexican books whose price is listed as "100 peso"? Depending on whether it was published prior to 1993 or after 1996, the correct ISO code should be either MXP or MXN, something that most of us couldn't determine without a fair amount of digging.
  • To go back to the Russian example, the ISO code for the Soviet ruble was "SUR". When the USSR was dissolved at the end of 1991, the code was retired. It was replaced with "RUR" (later "RUB" as per the discussion above) for the Russian ruble and "BYB" for the Belarusian ruble. The latter was replaced with "BYR" in 2000 and then with "BYN" in 2016.
For a bibliographic database like ISFDB to keep track of these changes over many decades and even centuries would be very time-consuming and not the best way to spend editor time.
One possible "low-hanging fruit" enhancement would be to update the mouse-over bubbles that we display for prices. They currently say things like "Lev: Bulgarian lev". We could update them to say things like "Lev: Bulgarian lev. ISO codes: BGJ in 1881-1952, BGK in 1952–1962, BGL in 1962-1999, BGN since 1991". Ahasuerus (talk) 12:40, 9 February 2024 (EST)
Is there interest in updating the mouse-over bubbles with information like "Lev: Bulgarian lev. ISO codes: BGJ in 1881-1952, BGK in 1952–1962, BGL in 1962-1999, BGN since 1991"? It would be a very simple textual change in the software. Ahasuerus (talk) 09:01, 4 April 2024 (EDT)

Appendices

The other day User:Elysdir added the following paragraph to Template:TitleFields:Title:

  • Appendices. If the page where the work begins includes a phrase like "Appendix A", then include that phrase in the work's title. For example: "Appendix B: Ashima Slade and the Harbin-Y Lectures: Some Informal Remarks Toward the Modular Calculus, Part Two".

I am moving the proposed language to the Rules and Standards page to see what other editors think of it. Ahasuerus (talk) 08:02, 10 February 2024 (EST)

Thanks for moving this here! I should note that before I made that change, I did a title search on “Appendix” and found that a large majority of the appendix titles in ISFDB (in cases where there’s more than one appendix) use the format that I mentioned. (The advanced-search version of that search shows all 900+ titles.) There are only three titles in those search results that use the format “(Appendix A) Title”, and hundreds that use the format “Appendix A: Title” (or “Appendix 1: Title” or “Appendix I: Title” or “Appendix One: Title” or etc). So my writeup was an attempt to document what I was (incorrectly) assuming was an existing policy, rather than an attempt to make new policy. —Elysdir (talk) 14:46, 10 February 2024 (EST)
I prefer the use of a colon as it is better at indicating the wording is part of the title. When I see parentheses, my brain interprets it as something not part of the title but used to clarify or disambiguate. So, I support this proposed wording. ···日本穣 · 投稿 · Talk to Nihonjoe 12:06, 12 February 2024 (EST)
Yeah, I prefer using : as well - and we do use that for subtitles elsewhere so it also makes sense. And Appendices with no other title should be followed by the title of the work in brackets (we may as well throw that to complete the rule although it derives from the standard naming of essays). So "Appendix B: The making of a world" if the title is there and "Appendix B (Book title) if it just say "Appendix B". That will also make it easier to determine when there was a printed title. The corner case is when the title is printed in brackets on the page itself (which the Appendix B part is not... not sure if we want a : there or to ignore the brackets or what we want to do. Annie (talk) 13:50, 12 February 2024 (EST)
Just to make sure we are on the same page: when you wrote "brackets", did you mean "[]" (aka "square brackets") or "()" (aka "parentheses")? Ahasuerus (talk) 12:52, 13 February 2024 (EST)
Sigh. Parentheses - I meant parentheses :) I usually use square brackets for [] to make sure it is clear which ones I mean and I do not always remember that () have their own word. Annie (talk) 12:55, 16 February 2024 (EST)

(unindent) A couple of questions/clarifications.

  • The proposed language is a phrase like "Appendix A". Would this be limited to the word "Appendix" or would it also cover alternative terms like "Addendum or Appendices? Some currently use parentheses, some use colons and some say things like "Addendum to Whirligig World". Ahasuerus (talk) 14:15, 16 February 2024 (EST)
Others may well have better answers, but I thought I might as well comment here: although I didn’t say this in my proposed language, I was focused specifically on the case where the appendices have individual subtitles as well as the general title of “Appendix A”. (So my language should be updated to say that; see below.) I was assuming that when a title consists entirely of a standard book-section name, it should follow the disambiguation rule at the end of that page, in the “"Standard" titles” paragraph: “you should parenthetically append the container title (title of the novel, collection, anthology, etc) to the title of the essay, i.e. "<generic essay title> (<container title>)" in order to create a unique title”. So for cases where there’s an addendum that’s just titled “Addendum”, I would use the format “Addendum (<container title>)”, which is also what the majority of those existing cases that you linked to already use. In the rare case where addenda also have their own individual subtitles, I would use the colon format, as demonstrated by the existing item “Addendum 1: Description of Maps”. And I would expect that the title “Appendices” by itself would also be covered by the “"Standard" titles” rule: “Appendices (A Magic of Twilight)”.
 
So maybe another way to approach this appendix-title guidance would be to reframe it as a sub-guideline of the “"Standard" titles” guidance. At the end of the page, after the “"Standard" titles” paragraph, we could say something like this (phrasing could use some further polishing):
 
Standard titles with specific subtitles. If the title consists entirely of a standard title, then use the standard titles guidance above. (Examples: “Appendices (A Magic of Twilight)”; “Appendix B (A Galaxy Unknown)”; “Introduction (50 in 50)”.) But if the title starts with a label for a standard section of a book (such as “Appendix” or “Addendum”) and then is followed by an individual subtitle for that specific section, then put a colon between the book-section name and the individual title. (Examples: “Appendix: Chronology of Technic Civilization”; “Appendix B: Closures and Openings”; “Introduction: 37 Divided by 3”.)
 
…Note that that framing does introduce a difference from how some existing ISFDB titles currently do things: it removes the quotation marks around the individual subtitle. —Elysdir (talk) 20:39, 16 February 2024 (EST)
Perhaps I am not grasping some subtleties, but wouldn't the proposed approach be the same as what the Subtitles section of Template:TitleFields:Title currently says:
  • If the title has a subtitle, enter it, with a colon and a space used to separate the title from the subtitle. For example, the 1986 edition of George MacDonald's "Lilith" has "Lilith" on the title page, and below that, in a smaller font, "A Romance". This should be entered as "Lilith: A Romance".
? Ahasuerus (talk) 10:47, 19 February 2024 (EST)
I think it’s essentially the same guidance, yes, but applied in a different case. The subtitles guidance reads to me as being about the title and subtitle of a book, as opposed to a section. If instead of adding a new section, you would prefer to clarify the Subtitles section to say that it also applies to things like an appendix or an addendum, that would be fine with me. …My goal in all of this is to clarify to editors how they should format the titles and subtitles of appendices; I’m fine with any approach y’all want to take. (…And I apologize if I’m overstepping by participating in this discussion at all—if I should step back and just leave it to you folks to decide, let me know.) —Elysdir (talk) 15:47, 19 February 2024 (EST)
Oh, no, everyone is welcome to contribute to Rules and Standards discussions! Sometimes an outside perspective reveals that Help is unclear or that it doesn't account for a certain category of cases.
The current discussion is a good example. The first three paragraphs of Template:TitleFields:Title, as currently written, are in the following order:
  • Novels
  • Subtitles
  • Short fiction, essays and poems
The way they are ordered, it's possible to assume that the "Subtitles" paragraph only applies to novels, especially since the next 2 paragraphs (SERIALs and excerpts) have special rules for subtitles and disambiguators. However, I believe the intent was to apply the "Subtitles" rule to all other title types (that do not have explicitly stated exceptions) as well.
If my understanding is correct, then we may be able to eliminate this ambiguity by moving the "Subtitles" paragraph below the "Short fiction, essays and poems" paragraph. We should probably also move "Omnibuses, nonfiction, anthologies and collections", which is currently the 6th paragraph in this template, right below the "Novels" paragraph. That way the order would be:
  • Novels
  • Omnibuses, nonfiction, anthologies and collections
  • Short fiction, essays and poems
  • Subtitles
  • SERIALs
  • Excerpts
  • Artwork
  • Etc
The 4 paragraphs preceding the "Subtitles" paragraph would all use the same subtitle rule while the paragraphs following the "Subtitles" paragraphs would have special rules. We could also make it explicit in the language of the "Subtitles" paragraph. Would this work from your perspective? Ahasuerus (talk) 16:07, 19 February 2024 (EST)
I like this idea. ···日本穣 · 投稿 · Talk to Nihonjoe 17:35, 19 February 2024 (EST)
  • Some languages -- notably French -- use a space between a word and a trailing colon. We generally follow language-specific rules for non-English titles, so is it safe to assume that we would be using " :" as opposed to ":" for French titles? Ahasuerus (talk) 14:15, 16 February 2024 (EST)
Huh, interesting, I didn’t know that about French. Given that difference, I would expect that yes, we would use " : " instead of ": " in French titles. —Elysdir (talk) 20:39, 16 February 2024 (EST)

Appendices - Outcome

Hearing no objection, I have re-ordered the first 4 paragraphs in Template:TitleFields:Title based on the order proposed above. One sentence was split into two for readability. Ahasuerus (talk) 18:47, 28 February 2024 (EST)

Secondary source artist credit in face of credit change over time

For full background, see User_talk:MartyD#Dixie_Ray. Different editions of a Ballantine Fahrenheit 451 use the same cover art but credit the artist differently. This is what we know:

ISBN Printing Date Artist credit
0-345-25027-3[-150] 40th 1975-12-00 no credit at all
0-345-25027-3[-150] 43rd 1976-08-00 Whistlin' Dixie
0-345-27431-8 44th 1977-08-00 Whistlin' Dixie
0-345-27431-8 45th 1977-11-00 Whistl'n Dixie
0-345-27431-8 46th 1978-08-00 Whistl'n Dixie

The later "Whistl'n" is canonical. The rules do not permit us to assign "uncredited" to the 40th's cover, but they do permit us to assign an identity using the later editions' credits as a secondary source. Which later edition's credit should we use here, the non-canonical "Whistlin'" or the canonical "Whistl'n"? Likely the 41st and 42nd printings will have either no credit or "Whistlin'". I'd really like to use "uncredited" to give a complete picture of the credit's evolution. It would be a little odd to have one or more earlier editions have the canonical credit, then have some later ones with a non-canonical credit, then even later ones "revert" to canonical (when in fact they progressed to canonical). For now I have gone with canonical, but I thought I'd raise the question to see if we should standardize on something else for this scenario.

Two further hypotheticals to consider: Suppose we only had the 40th (uncredited) and then entered the 43rd ("Whistlin'") and so went back and adjusted the 40th to use that. Now the 45th ("Whistl'n") gets entered, and its credit is determined to be the canonical form. What would we want done with the 40th's (now) non-canonical credit at that point? Likewise, suppose we had the 40th, then entered the 45th ("Whistl'n") and went back and adjusted the 40th to use that. Now the 43rd gets entered. What would we want done with the 40th's credit at that point? --MartyD (talk) 07:12, 17 February 2024 (EST)

Re: "I'd really like to use "uncredited" to give a complete picture of the credit's evolution."
As per Template:PublicationFields:CoverArt here is how I think we currently credit cover artists depending on what is in the publication:
What is stated in the pub What we enter in the "Artist" field
Canonical name Canonical name
Alternate name Alternate name (VT created)
Initials Canonical name if known
Artist-specific symbol (sometimes a stylized version of the artist's initials) Canonical name if known
Signature, often illegible Canonical name if known
No explicit credit, but the artist's style is recognizable The "Artist" field is left blank; Notes updated with the name of the artist and reason for attribution
No explicit credit, but a secondary source credits the artist Canonical name; Notes updated with the source
No explicit credit, but the credit is implied, e.g. a small illustration may be reproduced as a credited INTERIORART work Can be arguably considered a "secondary bibliographic source" for our purposes and treated as such, i.e. enter the canonical name in the "Artist" field and update Notes with the source
This is a tricky decision tree diagram, which, admittedly, makes it hard to "give a complete picture of the credit's evolution". I think the underlying issue here is that it would be difficult to enter artist credits the way we enter author credits, i.e. "as stated in the pub". The main reason is that signatures, symbols and barely legible stylized initials are not something that can be easily captured as text.
That being said, I think it would be beneficial to restructure Template:PublicationFields:CoverArt as a series of bullets to make it easier for new editors to parse. Ahasuerus (talk) 11:33, 19 February 2024 (EST)
I understand the current rules call for canonical when the credit is taken from a secondary source, and that is what I did. It seems wrong to me in this case, however. For argument's sake, let's assume there is also no credit in the 41st printing and the credit in the 42nd is the alternate "Whistlin'". If all printings were recorded at the same time, we would have none/canonical -> none/canonical -> alternate -> alternate -> canonical -> canonical. If instead we entered them over time we would start with none -> none, then on discovering the "Whistlin'" we might change those to: non/canonical(1) -> none/canonical(1) -> canonical(1), with the first two citing the third as secondary source. Two printings later, we would discover "Whistl'n" and realize it should be canonical, so we'd VT the existing TITLE records and end up with: none/alternate -> none/alternate -> alternate -> alternate -> canonical(2). Someone would have to know to review all previous credits to see if they came from the publication or used the source of the now-alternate credit and in the latter case change them to the (new) canonical to match what would happen if we entered them all at the same time. --MartyD (talk) 07:05, 20 February 2024 (EST)
I think we have somewhat of a grandfathered problem here. Our rules had been pretty straightforward for a long time - secondary credits of art use the canonical name. In our digital era, I'd argue that a scan that is proved to be of a certain printing should be considered primary source for this determination (and I think we had been applying it that way). However, as a practice we had often made an exception for this rule for later (and earlier) printings and even different formats altogether (audio/ebooks/paper had gotten credits based on the other formats) - we had often imported straight from the one we do know the credit for even if it is not using the canonical name because it is (usually) a good guess that most of these will match. Thus the conundrum now for the few credits where they do not match.
We have two paths: enforce the rule as written OR come up with a language that allows us a bit of creativity: "You can use the credit as found in a later or earlier printing if data for the current printing is not available, with a mandatory note on the exact source of the name used. That includes the usage of uncredited. The same applies for other formats sharing a cover (i.e. audiobooks which have only a cover and the artist may or may not be credited on it). Using the canonical name is always allowed in the cases of unknown credit (due to lack of source information or only secondary sources information) - with an appropriate note.". Feel free to rewrite/change/argue. And if we are changing the rule, can we please make it more forceably requiring a note on the decision if you are not grabbing the name straight from the book - otherwise it is a nightmare to change a canonical name for example - I am sure we had created a lot of mistakes in the DB in the process of changing canonical names of artists simply by not knowing when a credit is a direct one and when a canonical is being used.
I am leaning towards the second option - mainly because it is somewhat of a practice anyway (in the multi-formats) and it kinda covers this case here. Annie (talk) 12:10, 20 February 2024 (EST)

Clarifying Template:PublicationFields:CoverArt

Going back to the issue of clarifying (as opposed to changing) what's currently stated in Template:PublicationFields:CoverArt, earlier this week an editor asked me if we could update the template language with what I wrote above to make the instructions more clear. Here is the proposed new language to be used when deciding what to enter in the "Artist" field:

  • If the artist's canonical name is stated in the publication, enter it
  • If the artist's alternate name is states in the publication, enter it and make sure to create a Variant Title later
  • If the cover has the artist's initials, enter the artist's canonical name if known and update the Notes field with the source of the attribution
  • If the cover has an artist-specific symbol, e.g. a stylized version of the artist's initials, enter the artist's canonical name if known and update the Notes field with the source of the attribution
  • If the cover has a recognizable signature, enter the canonical name if known and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit, but the artist's style is recognizable, leave the "Artist" field blank and update the Notes with the name of the artist and reason for attribution
  • If the publication has no explicit artist credit, but a secondary source credits the artist, enter the canonical name and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit, but the credit is implied, e.g. a small section is reproduced as a credited INTERIORART work, treat it as a "secondary bibliographic source" scenario described above: enter the canonical name and update the Notes field with the source of the attribution

Does this look right? Ahasuerus (talk) 19:01, 28 February 2024 (EST)

Looks good to me, though I'd put the two "e.g." parts in parentheses. ···日本穣 · 投稿 · Talk to Nihonjoe 12:56, 29 February 2024 (EST)
I agree on the parentheses. Annie (talk) 13:19, 29 February 2024 (EST)
Spot-checking Help:Screen:NewNovel, I see that we use "e.g." inconsistently. In roughly one third of all cases we use parentheses while in the other two thirds we do not. Different grammar guides give contradictory advice. AP Style requires the use of parentheses and a trailing comma, but Fowler's Modern English Usage does not. Ahasuerus (talk) 13:45, 29 February 2024 (EST)
I find it a lot more readable when the parentheses are there. It also simplifies the reading of the sentence for non-native speakers and we have quite a lot of them - the clearer we state things and the easier we make it for someone whose English may be shaky, the better IMO. Annie (talk) 14:18, 29 February 2024 (EST)
I would also move "If the publication has no explicit artist credit, but the artist's style is recognizable, leave the "Artist" field blank and update the Notes with the name of the artist and reason for attribution" to the bottom of the list and change it to " If the publication has no explicit artist credit and no secondary or implied credit, but the artist's style is recognizable, leave the "Artist" field blank and update the Notes with the name of the artist and reason for attribution.". Otherwise it contradicts the next 2 rules in case of a recognizable artist and secondary credit for example. Annie (talk) 13:19, 29 February 2024 (EST)
That's a very good point. Here is the updated proposed order:
  • If the artist's canonical name is stated in the publication, enter it
  • If the artist's alternate name is stated in the publication, enter it and make sure to create a Variant Title later
  • If the cover has the artist's initials, enter the artist's canonical name if known and update the Notes field with the source of the attribution
  • If the cover has an artist-specific symbol, e.g. a stylized version of the artist's initials, enter the artist's canonical name if known and update the Notes field with the source of the attribution
  • If the cover has a recognizable signature, enter the canonical name if known and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit, but a secondary source credits the artist, enter the canonical name and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit, but the credit is implied, e.g. a small section is reproduced as a credited INTERIORART work, treat it as a "secondary bibliographic source" scenario described above: enter the canonical name and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit and no secondary or implied credit, but the artist's style is recognizable, leave the "Artist" field blank and update the Notes with the name of the artist and reason for attribution
Ahasuerus (talk) 13:51, 29 February 2024 (EST)
The "secondary sources" bullet does not quite match current practice. If an earlier printing has no credit and a later printing with identical artwork has a credit, we use the later credit's form as the implied/secondary credit on the earlier printings (and, in fact, we merge the records). It would not surprise me if in other secondary-source scenarios our de facto practice is close to what we do for reviews and interviews: If the name provided is something for which we already have a record, that is used, otherwise the canonical is used. --MartyD (talk) 09:47, 1 March 2024 (EST)
I think that you are right and the current de-facto practice is just to import the cover/art as is from the later/earlier printing/edition, even if a pseudonym is used - despite the clear rule saying to use the canonical in such cases... Annie (talk) 12:13, 1 March 2024 (EST)
I see. In that case, how about we insert a new bullet after the "alternate name" bullet and before the "artist's initials" bullet? Something like:
  • If the publication has no explicit artist credit, but another printing of the same book credits the artist, import the COVERART title from the printing with the artist credit, adjust the COVERART title's date to reflect its earliest known appearance and update the Notes field with the source of the attribution
We may need to further clarify this rule to account for the following class of scenarios:
  • some printings do not to credit the artist
  • (optionally) some printings credit the canonical name
  • some printings credit one or more alternate names
I assume it's uncommon, but better safe than sorry. Ahasuerus (talk) 14:59, 3 March 2024 (EST)
That seems to cover it. As a practical matter, I did run across the case of earlier editions with no credit, intermediate editions with one-form name credit, later editions with other-form name credit (all same artwork and publisher). Our current practice is "choose canonical", so I think if any of the conflicting possibilities includes canonical, we would import that one. I don't know what we do for multiple alternates only; I suppose current-practice recommendation would be to find and import the canonical, rather than any of the alternates. --MartyD (talk) 14:50, 5 March 2024 (EST)
I don't know if this is theoretical or practical, but what is current practice if artwork is only ever published under a pseudonym but that artist has a canonical identity under which other works are published? E.g., imagine "Ima Writer" who is a prolific SF novelist but dabbles in SF artwork as "Ima Painter", and we have Ima Painter as an alternate name for Ima Writer. If we came across uncredited artwork identified via secondary source as by Ima Painter, would we record it that way (and make a variant), rather than recording it as by Ima Writer? I DO NOT MEAN TO DISCUSS HOW THIS SHOULD BE TREATED. :) I am only asking what is current practice for purposes of the wording of the proposed bullet.. --MartyD (talk) 09:57, 1 March 2024 (EST)
I'd go with the canonical in this case usually - but I can remember probably one or two cases I had seen like that (all of them while I was untangling the languages when we added the field all these years ago). Annie (talk) 12:13, 1 March 2024 (EST)

Template:PublicationFields:CoverArt - Adjusted Language

Here is the adjusted language based on the discussion above:

  • If the artist's canonical name is stated in the publication, enter it
  • If the artist's alternate name is stated in the publication, enter it and make sure to create a Variant Title later
  • If the publication has no explicit artist credit, but another printing of the same book credits the artist:
    • Import the COVERART title from the printing with the artist credit
    • Adjust the COVERART title's date to reflect its earliest known appearance
    • Update the Notes field with the source of the attribution
  • If the cover has the artist's initials, enter the artist's canonical name if known and update the Notes field with the source of the attribution
  • If the cover has an artist-specific symbol (e.g. a stylized version of the artist's initials), enter the artist's canonical name if known and update the Notes field with the source of the attribution
  • If the cover has a recognizable signature, enter the canonical name if known and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit, but a secondary source credits the artist, enter the canonical name and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit, but the credit is implied (e.g. a small section is reproduced as a credited INTERIORART work), treat it as a "secondary bibliographic source" scenario described above: enter the canonical name and update the Notes field with the source of the attribution
  • If the publication has no explicit artist credit and no secondary or implied credit, but the artist's style is recognizable, leave the "Artist" field blank and update the Notes with the name of the artist and reason for attribution

Ahasuerus (talk) 16:44, 5 March 2024 (EST)

Looks good to me. ···日本穣 · 投稿 · Talk to Nihonjoe 20:02, 5 March 2024 (EST)
It sounds like we have consensus. If I don't hear any objections, I will update Template:PublicationFields:CoverArt tomorrow. Ahasuerus (talk) 15:36, 12 March 2024 (EDT)
Template:PublicationFields:CoverArt and Rules and standards changelog have been updated. Thanks to everyone who contributed to the discussion and helped clarify this thorny area! Ahasuerus (talk) 14:17, 14 March 2024 (EDT)

Subtitle help needs some minor tweaking

The help text for the the Subtitle section of the title help is inconsistent between: Template:PublicationFields:Title, Help:Screen:NewPub#Title, and Template:TitleFields:Title (there may be other places I didn't find). Template:TitleFields:Title is missing the wording "Note that the title page may show the series name, and sometimes the publication's position in the series. The present (2018) usage is to enter only the "simplified" title, for example, you could enter the title for a publication as "Song of the Dragon" and the note would have "The title page states 'Song of the' over 'Dragon' over 'The annals of Drakis: Book One'."

In addition, all three need to be updated to contain wording for the decade-old practice of not including the phrase "A Novel" in the subtitle. With the addition of wording for this practice, the Help text would likely be clearer if each of these points were listed as separate bullet points so that they easily catch an editor's attention. Thanks. Phil (talk) 18:20, 15 March 2024 (EDT)

Help:Screen:NewPub#Title transcludes Template:PublicationFields:Title, so there are only two templates that need to be reconciled. Here is what they currently say about subtitles:
Template:PublicationFields:Title:
  • Subtitles. If the title has a subtitle, enter it, with a colon and a space used to separate the title from the subtitle. For example, the 1986 edition of George MacDonald's "Lilith" has "Lilith" on the title page, and below that, in a smaller font, "A Romance". This should be entered as "Lilith: A Romance". It is sometimes a judgement call as to whether a change of font or a colon indicates a subtitle or just some creative license on the part of the typesetter. If in doubt, take your best guess and document the guess in the publication's notes.
    Note that the title page may show the series name, and sometimes the publication's position in the series. The present (2018) usage is to enter only the "simplified" title, for example, you could enter the title for a publication as "Song of the Dragon" and the note would have "The title page states 'Song of the' over 'Dragon' over 'The annals of Drakis: Book One'."
Template:TitleFields:Title:
  • Subtitles. If the title of a novel, omnibus, nonfiction, anthology, collection, short fiction, essay or poem has a subtitle, enter it. Use a colon and a space to separate the title from the subtitle. For example, the 1986 edition of George MacDonald's "Lilith" has "Lilith" on the title page, and below that, in a smaller font, "A Romance". This should be entered as "Lilith: A Romance". It is sometimes a judgment call as to whether a change of font or a colon indicates a subtitle or just some creative license on the part of the typesetter. If in doubt, take your best guess and document the guess in the publication's notes.
The reason that the first sentence of Template:PublicationFields:Title differs from the first two sentences of Template:TitleFields:Title is that we recently updated the latter and didn't touch the former.
Here is what I think we may want to do:
  • Create a new Help template for "subtitles" and transclude it in the two templates quoted above.
  • Use "If the title of a novel, omnibus, nonfiction, anthology, collection, short fiction, essay or poem has a subtitle, enter it. Use a colon and a space to separate the title from the subtitle." as the first two sentences of the new template.
  • Change:
    • Note that the title page may show the series name, and sometimes the publication's position in the series. The present (2018) usage is to enter only the "simplified" title
  • to:
    • If the title page includes the series name and/or the title's number within the series, do not enter the series name or the series number in the Title field
Ahasuerus (talk) 12:41, 16 March 2024 (EDT)
Looks good but don't forget we also need a bullet for not including the term "A Novel" as a subtitle. :) Phil (talk) 16:51, 16 March 2024 (EDT)
Thanks for the reminder! I will try to consolidate everything and post the new template language below tomorrow morning. Ahasuerus (talk) 19:15, 16 March 2024 (EDT)

(unindent) Here is the proposed language of the new, consolidated, template:

  • Subtitles. If the title of a novel, omnibus, nonfiction, anthology, collection, short fiction, essay or poem has a subtitle, enter it in the Title field using a colon and a space to separate the title from the subtitle. For example, the 1986 edition of George MacDonald's "Lilith" has "Lilith" on the title page, and below that, in a smaller font, "A Romance". This should be entered as "Lilith: A Romance". It is sometimes a judgment call as to whether a change of font or a colon indicates a subtitle or just some creative license on the part of the typesetter. If in doubt, take your best guess and document the guess in the publication's notes. Note that there are two exceptions where subtitles should not be entered in the Title field:
    • The subtitle is "A Novel". This subtitle is generic and should be ignored.
    • The title page displays the series name (and sometimes the title's position in the series) where the subtitle would normally be. The series information should not be recorded in the Title field, but you may record it in the Notes field. For example, if the title page says "Song of the Dragon" and then "The Annals of Drakis: Book One" below it, you would enter "Song of the Dragon" in the Title field and then optionally update the Notes field with detailed information like "The title page states 'Song of the' over 'Dragon' over 'The Annals of Drakis: Book One'."

How does it look? Ahasuerus (talk) 16:04, 17 March 2024 (EDT)

Looks good. Phil (talk) 16:10, 17 March 2024 (EDT)
Look at the last sentence of bullet point 2, too many overs. John Scifibones 16:18, 17 March 2024 (EDT)
Well, that's how it is phrased in the current Help template. Checking Amazon's Look Inside for the first (2010-07-00) hardcover edition, I see that the title page has three lines:
  • Song of the
  • Dragon
  • The Annals of Drakis: Book One
That said, it may be too involved for the proposed Help template. If we simplify it to read "'Song of the Dragon' over 'The Annals of Drakis: Book One'", it will make more sense to our editors. Ahasuerus (talk) 16:46, 17 March 2024 (EDT)
Agree. John Scifibones 16:53, 17 March 2024 (EDT)
You may want to tweak the phrasing such as to explicitly include the fact that there may be more than one subtitle. You could add '... has one or more subtitles...', 'separate each with a colon'. Or similar wording. I remember coming across these occasionally, but can't find an example atm. MagicUnk (talk) 10:39, 18 March 2024 (EDT)
Something like A Son of the Ages: The Reincarnations and Adventures of Scar, the Link: A Story of Man from the Beginning? Ahasuerus (talk) 11:47, 18 March 2024 (EDT)
Indeed. MagicUnk (talk) 17:05, 18 March 2024 (EDT)
Regarding "The Annals of Drakis: Book One", I'd consider that a mention of the series the book is in rather than a subtitle. "The Annals of Drakis" would be the series name and it would have a series number of "1". ···日本穣 · 投稿 · Talk to Nihonjoe 11:53, 18 March 2024 (EDT)

Subtitles: Proposed Help template language

Here is the latest version of the new template. I believe it incorporates all of the comments made above.

  • Subtitles. If the title of a novel, omnibus, nonfiction, anthology, collection, short fiction, essay or poem has a subtitle, enter it in the Title field using a colon and a space to separate the title from the subtitle. For example, the 1986 edition of George MacDonald's "Lilith" has "Lilith" on the title page, and below that, in a smaller font, "A Romance". This should be entered as "Lilith: A Romance". If multiple subtitles exist, they should all be entered and separated with colons and spaces, e.g. A Son of the Ages: The Reincarnations and Adventures of Scar, the Link: A Story of Man from the Beginning. Note that it is sometimes a judgment call as to whether a change of font or a colon indicates a subtitle or just some creative license on the part of the typesetter. If in doubt, take your best guess and document the guess in the publication's Notes field.
  • Exceptions to the Subtitles rule. There are two scenarios where subtitles should not be entered in the Title field:
    • The subtitle is "A Novel". This subtitle is generic and should not be entered in the Title field.
    • The title page displays the series name (and sometimes the title's position within the series) where the subtitle would normally be. The series information should not be treated as a subtitle or recorded in the Title field. Instead it should be recorded in the "Series" and "Series Number" fields of the Title record. You may still record it in the Notes field for the sake of completeness. For example, if the title page says "Song of the Dragon" and then "The Annals of Drakis: Book One" below it, you would enter "Song of the Dragon" in the Title field, "The Annals of Drakis" in the Series field, and "1" in the Series Number field. You could then optionally update the Notes field of the publication record with detailed information like "The title page states 'Song of the Dragon' over 'The Annals of Drakis: Book One'."

Ahasuerus (talk) 18:00, 18 March 2024 (EDT)

I think that's very clear. I like it. ···日本穣 · 投稿 · Talk to Nihonjoe 19:51, 18 March 2024 (EDT)
I like it too but it is kinda incorrect for French (for example) where they use a space before and after the colon. So maybe after "using a colon and a space to separate the title from the subtitle." insert "For languages with different punctuation rules, i.e. French where colon is preceded and followed by a space, use the appropriate punctuation for that language". Or something to that effect. Annie (talk) 21:47, 18 March 2024 (EDT)
Good point. How about the following version of the first section:
  • If the title of a novel, omnibus, nonfiction, anthology, collection, short fiction, essay or poem has a subtitle, enter it in the Title field using a colon to separate the title from the subtitle. For English language titles, the colon should be followed by a space. For example, the 1986 edition of George MacDonald's "Lilith" has "Lilith" on the title page, and below that, in a smaller font, "A Romance". This should be entered as "Lilith: A Romance". For titles written in other languages, use language-specific rules for the use of colons. For example, in French colons are both preceded and followed by a space, e.g. "Défricheurs d'imaginaire : une anthologie historique de science-fiction suisse romande".
? Ahasuerus (talk) 14:19, 19 March 2024 (EDT)
Sounds good to me. Annie (talk) 14:39, 19 March 2024 (EDT)
The only suggestions I have is a comma after "essay" in the first sentence, and some additional punctuation for the last sentence: "For example, in French, colons are both preceded and followed by a space (e.g., "Défricheurs d'imaginaire : une anthologie historique de science-fiction suisse romande"). Other than that, I think it's great! ···日本穣 · 投稿 · Talk to Nihonjoe 18:01, 19 March 2024 (EDT)

2024-03-19 version of the proposed template

OK, folks, I think we are getting close. Here is the latest version incorporating everything that has been suggested:

  • Subtitles. If the title of a novel, omnibus, nonfiction, anthology, collection, short fiction, essay, or poem has a subtitle, enter it in the Title field using a colon to separate the title from the subtitle. For English language titles, the colon should be followed by a space. For example, the 1986 edition of George MacDonald's "Lilith" has "Lilith" on the title page, and below that, in a smaller font, "A Romance". This should be entered as "Lilith: A Romance". For titles written in other languages, use language-specific rules for the use of colons. For example, in French, colons are both preceded and followed by a space (e.g., "Défricheurs d'imaginaire : une anthologie historique de science-fiction suisse romande"). If multiple subtitles exist, they should all be entered and separated with colons and spaces, e.g. A Son of the Ages: The Reincarnations and Adventures of Scar, the Link: A Story of Man from the Beginning. Note that it is sometimes a judgment call as to whether a change of font or a colon indicates a subtitle or just some creative license on the part of the typesetter. If in doubt, take your best guess and document the guess in the publication's Notes field.
  • Exceptions to the Subtitles rule. There are two scenarios where subtitles should not be entered in the Title field:
    • The subtitle is "A Novel". This subtitle is generic and should not be entered in the Title field.
    • The title page displays the series name (and sometimes the title's position within the series) where the subtitle would normally be. The series information should not be treated as a subtitle or recorded in the Title field. Instead it should be recorded in the "Series" and "Series Number" fields of the Title record. You may still record it in the Notes field for the sake of completeness. For example, if the title page says "Song of the Dragon" and then "The Annals of Drakis: Book One" below it, you would enter "Song of the Dragon" in the Title field, "The Annals of Drakis" in the Series field, and "1" in the Series Number field. You could then optionally update the Notes field of the publication record with detailed information like "The title page states 'Song of the Dragon' over 'The Annals of Drakis: Book One'."

Ahasuerus (talk) 21:39, 19 March 2024 (EDT)

If there are no objections, I plan to add this template on Saturday night. Ahasuerus (talk) 13:15, 22 March 2024 (EDT)
Just one note: Maybe we should say: "The subtitle is "A Novel" or its equivalent in the language of text." instead of just "A Novel". Annie (talk) 13:20, 22 March 2024 (EDT)
Makes sense. Also, at some point we may want to revisit the issue of "generic subtitles", but I would prefer to finalize and post what we currently have before opening another can of worms. Ahasuerus (talk) 14:10, 22 March 2024 (EDT)

Subtitle changes -- Outcome

A consolidated Help template, Template:TitleFields:Subtitles, has been created. Template:TitleFields:Title and Template:PublicationFields:Title have been updated to transclude the new template. Rules and standards changelog has been updated. This was a good and comprehensive discussion -- thanks! Ahasuerus (talk) 14:49, 24 March 2024 (EDT)

Subtitles and Variants

While everyone has got subtitles on their mind, I recently added the audio book of Le Guin's No Time to Spare which does not include the subtitle that appears on the cover, "Thinking About What Matters". At the same time I updated the title record to remove the subtitle and updated all of the disabiguations similarly. This was based on my understanding that when a container title is published both with and without a subtitle, or with differing subtitles, we omit the subtitle from the title record but include it on those publications where it occurs. I have been handling this situation since at least 2009 after having this discussion with Mhhutchins where he stated "Pulling a random book off the shelf: In the Ice King's Palace: The World in Amber, Book 2. I consider everything after the colon to be a subtitle and shouldn't be part of the title record, but have no problem with it being in the publication record." n.b. I believe this was before we prohibited series names in title fields. There was a small kerfuffle about the Le Guin book which was cheerfully resolved where an editor had added the subtitle to my publication. After I backed his edits out, he went further and made the title without a subtitle into a variant of the title with one. This caused be to realize that my understanding may not be universal. It certainly isn't documented anywhere aside from that conversation. However, there are many examples of records being handled this way. How do other editors handle this situation? I'll also note that this only works for container titles. Short fiction that appears both with and without a subtitle must be varianted to reflect how it appears. Thoughts? --Ron ~ RtraceTalk 19:14, 19 March 2024 (EDT)

My understanding of the current practice is that, with the exception of Magazine/Fanzine publications when they are combined into yearly records, we match the two titles (of the publication and title records) and then use the same for any COVERART records - mimicking what "NewPub" will create. So in a case where there are different subtitles on two publications or one has one and another does not, I'll make variants. We do have quite a few of older records where the "naked, non-subtitled" title is inside of publications that have the pub title with a subtitle though but I had not seen a lot of these being added that way in the last years. Annie (talk) 19:58, 19 March 2024 (EDT)
I make variants. Often the original publication has a subtitle and some but not all later publications omit it. Occasionally a subtitle is only included in a later publication. Audiobooks and CDs mostly seem to omit the subtitles. Phil (talk) 21:51, 19 March 2024 (EDT)
If you consider "title:subtitle" being the title proper of a title record, then any variations should be varianted. Which is consistent with current rules, and that is how I treat the case. When I come across an example Annie mentions, I will correct it and create proper title:subtitle entries that match the publication records, and do the necessary varianting. Regards, MagicUnk (talk) 12:20, 20 March 2024 (EDT)
Yes, that's why I did the varianting. The help pages for entering a publication state "The title should appear exactly as published, even though this may be different from the canonical title" which leads one (in my opinion) toward that the titles in the publication and its title record have to match (and thus an added or missing subtitle causes a new title record). (Note aside: I just ran over a title which has different subtitles in diverse publications since 1923 - of which I'll add some in the next weeks) Christian Stonecreek (talk) 12:46, 20 March 2024 (EDT)
It's my understanding that publication titles should match their "reference" titles, i.e. contained titles whose title type matches the containing publication's type. Magazines/fanzines, which use consolidated titles, are the only exception that I can think of.
I should add that we have a cleanup report, Publication Title-Reference Title Mismatches, which looks for these types of discrepancies. It's currently configured to ignore title-publication pairs where the publication's title is fully contained within the reference title's title OR the reference title's title is fully contained within the publication title. It also ignores differences in punctuation. However, that was a temporary measure. Back when the report was implemented, we had so many other mismatches in the database that we decided to concentrate on the most important discrepancies first. Now that the current report is down to 20 discrepancies, we could change the report logic to look for all discrepancies. Checking the data on the development server, I see 10,363 mismatches. Ahasuerus (talk) 13:00, 20 March 2024 (EDT)

(unindent) Are there any objections to expanding the cleanup report "Publication Title-Reference Title Mismatches" to cover all mismatches between publication title and their reference title records' titles? Ahasuerus (talk) 14:38, 3 April 2024 (EDT)

None, I think it is time. Annie (talk) 14:41, 3 April 2024 (EDT)
FR 1599 "Make 'Publication Title-Reference Title Mismatches' more comprehensive" has been created. Ahasuerus (talk) 09:24, 4 April 2024 (EDT)

Subtitles and Variants - Outcome

FR 1599 has been implemented -- please see the Community Portal announcement for details. Ahasuerus (talk) 17:09, 4 April 2024 (EDT)

Clarifying the Audiobooks entry

Moved from the Community Portal.

I've placed this submission on hold because I'm unsure if YouTube audiobooks should be included here since they are generally not downloadable (instead being streamed). The rules include audiobooks, but also exclude "[w]orks published in a web-based publication and available exclusively as a Web page" (which is pretty much what a YouTube video is), and they say nothing about podcasts.

Annie suggested that we need to expand this section to better match the electronic publications section. This is what those two section parts currently state:

    • electronic publications of the following types:
      • e-books with a unique identifier such as an ISBN, ASIN, EAN, or catalog number
      • downloadable e-zines
      • Internet-based publications which are downloadable as electronic files in any number of ebook formats (ePub, Mobi, PDF, etc).
      • Webzines, which are defined as online periodicals with distinct issues (note that online periodicals without distinct issues are not considered webzines).
      • One time speculative fiction anthologies published on the Web
      • Online publications available exclusively as a Web page, but only if:
        • published by a market which makes the author eligible for SFWA membership (listed here), OR
        • shortlisted for a major award
    • audio books, i.e. readings, but not dramatizations

In the exclusion section, the applicable point currently states:

  1. 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

Do we want to change it to explicitly include downloadable genre podcasts (which, to me, are basically audio essays or interviews, stories (some of them)), and explicitly exclude YouTube/streaming-only podcasts and audiobooks? How else should we change it?

Here are my suggested changes to the Included section (bolded):

    • electronic publications of the following types:
      • e-books with a unique identifier such as an ISBN, ASIN, EAN, or catalog number
      • downloadable e-zines
      • Internet-based publications which are downloadable as electronic files in any number of ebook formats (ePub, Mobi, PDF, etc).
      • Internet-based audio publications (such as audiobooks, podcasts, etc.) which are downloadable as electronic files in any number of formats (MP3, MP4, etc).
      • Webzines, which are defined as online periodicals with distinct issues (note that online periodicals without distinct issues are not considered webzines).
      • One time speculative fiction anthologies published on the Web
      • Online publications available exclusively as a Web page, but only if:
        • published by a market which makes the author eligible for SFWA membership (listed here), OR
        • shortlisted for a major award

Here are my suggested changes to the Excluded section (bolded):

  1. Works published in a web-based publication and available exclusively as a Web page -- such as blogs, author-run sites, fan fiction, web serials, non-downloadable or streaming audio content, etc. -- unless listed in the Included section

Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 15:33, 20 March 2024 (EDT)

I think that we should keep our ebooks and audio-books separately so instead of including the line you added up in the electronic section, maybe we should rename that section to mark it as text only (non-audio, non-video non-whatever they come up with next) and then work a specific set of rules for the audio formats, starting with the line you had but also adding a line about all physical formats. So something like this:
    • audio books, i.e. readings, but not dramatizations
      • All physical audio formats - Audio disks, MP3 Disks, Audio Players, Casettes and so on.
      • Digital audiobooks which are downloadable in any format (Audible, MP3, MP4 and so on)
      • Internet-based audio publications (such as podcasts, etc.) which are downloadable as electronic files in any number of formats (MP3, MP4, etc).
That also ensures that the "not dramatizations" applies to the podcasts and all downloadable things. I also pulled the audiobooks into their own line but I am not sure we need that - it is a matter of naming things to some extent but I do not want to call Audible.com or the audio-section of Kobo "internet-nased audio publications". If everyone disagrees, I won't insist on that though. Annie (talk) 16:03, 20 March 2024 (EDT)
Annie (talk) 16:03, 20 March 2024 (EDT)
I like that. Keeping them separate is a good thing. ···日本穣 · 投稿 · Talk to Nihonjoe 17:04, 20 March 2024 (EDT)
I agree that the proposed wording would be an improvement. That said, do we currently explicitly define "downloadable"? There are many ways to download a Web page or an audio file using Web browsers or various browser extensions, but it doesn't make Web pages "downloadable" for our purposes, right? If so, should we make it explicit? Ahasuerus (talk) 20:01, 25 March 2024 (EDT)
The proposal in the below section does that a little. ···日本穣 · 投稿 · Talk to Nihonjoe 20:06, 25 March 2024 (EDT)

(unindent) Minor editorial tweaks to bring the capitalization and wording in line with other RoA sections:

  • audio books, which are defined to include readings and to exclude dramatizations, of the following types:
    • all physical audio formats such as audio disks, MP3 disks, audio players, cassettes and so on
    • digital audio books which are downloadable in any file format (Audible, MP3, MP4 and so on)
    • internet-based audio publications (such as podcasts, etc.) which are downloadable as electronic files in any number of formats (MP3, MP4, etc).

This leaves the issue of clarifying what "downloadable" is, which is currently being discussed below, open, but I think this is a clear improvement and could be added to RoA without waiting for the other discussion to be wrapped up. Are there any issues with the wording above? Ahasuerus (talk) 17:44, 27 March 2024 (EDT)

It looks good to me. ···日本穣 · 投稿 · Talk to Nihonjoe 13:01, 29 March 2024 (EDT)
I am ok with that wording. Annie (talk) 19:19, 29 March 2024 (EDT)
If there are no objections, I will updated the Policy page tomorrow. Ahasuerus (talk) 12:53, 31 March 2024 (EDT)
ISFDB:Policy#Included and Rules and standards changelog have been updated. Ahasuerus (talk) 10:32, 2 April 2024 (EDT)

Formats help pages

When we expanded ROA to include a lot more webzines awhile back, we never cleaned up the Formats help pages here and here (and possibly a few more places) - MagicUnk noticed. As it is, the text is not wrong but it is not really useful either. :) The text now reads:

  • webzine - Used for Internet-based periodical publications which are otherwise not downloadable as an "ebook". Not all webzines are eligible for inclusion in the ISFDB. Initiate discussions about inclusion/eligibility on the Community Portal.

I propose to change that to:

  • webzine - Used for Internet-based periodical publications which are otherwise not downloadable. Not all webzines are eligible for inclusion in the ISFDB - only webzines, which are defined as online periodicals with distinct issues are always eligible. Please consult the Rules of Acquisition for some extended eligibility criteria and initiate discussions about inclusion/eligibility on the Community Portal if needed.

Thoughts? Better proposed language? Annie (talk) 11:25, 22 March 2024 (EDT)

I would suggest a slight change in the proposed wording:
  • webzine - Used for Internet-based periodical publications which are otherwise not downloadable. ISFDB defines webzines as "online periodicals with distinct issues". Only those that meet this definition are eligible for inclusion. Please consult the Rules of Acquisition for some extended eligibility criteria and initiate discussions about inclusion/eligibility on the Community Portal, if needed.
Not a huge change, but I think it's more clear. ···日本穣 · 投稿 · Talk to Nihonjoe 12:44, 22 March 2024 (EDT)
I like it :) Annie (talk) 12:48, 22 March 2024 (EDT)
I agree with the changes. Small problem I have always had with this point. For periodicals available both online and downloadable, does our wording imply webzine is not appropriate? John Scifibones 13:00, 22 March 2024 (EDT)
RoA has two bullet points that cover the distinction:
  • downloadable e-zines
  • Webzines, which are defined as online periodicals with distinct issues (note that online periodicals without distinct issues are not considered webzines)
Perhaps we could expand the first bullet point to explain how we use the term "e-zine", which would be similar to the way we explain how we use the term "webzine" in the second bullet point. Something like:
  • Ezines, which are defined as electronic periodicals with distinct downloadable issues.
  • Webzines, which are defined as online periodicals with distinct issues. Note that this includes online periodicals without downloadable issues, but excludes online periodicals without distinct issues.
Also, I am thinking that "some extended eligibility criteria" may be better as "detailed eligibility criteria". Ahasuerus (talk) 13:13, 22 March 2024 (EDT)
If they are both available inline and as downloadables, we add them as two separate publications: once as an ebook and once as a webzine (that also allows us to have slightly different contents sometimes - like the extra materials in the Lightspeed ebook compared to the webzine). Annie (talk) 13:23, 22 March 2024 (EDT)
I'm thinking about publications like AntipodeanSF. I don't think of them as two distinct formats. John Scifibones 13:34, 22 March 2024 (EDT)
But under our format definitions, they had always been considered as two separate ones (the same way how a print on demand availability of an ebook is considered a printed book so requires its own publication). Until we opened the doors for webzines, only the ebook version of AntipodeanSF was eligible to be added. Now both versions are. Annie (talk) 13:41, 22 March 2024 (EDT)
Thanks for clarifying, John Scifibones 13:48, 22 March 2024 (EDT)
So, how about this:
  • webzine - Used for Internet-based periodical publications which are otherwise not downloadable. ISFDB defines webzines as "online periodicals with non-downloadable distinct issues". Only those that meet this definition are eligible for inclusion. Please consult the Rules of Acquisition for detailed eligibility criteria. If needed, initiate discussions about inclusion/eligibility on the Community Portal.
Then the RoA parts could be changed to this:
  • Ezines, which are defined as electronic periodicals with downloadable distinct issues (i.e., PDF, epub, and so on).
  • Webzines, which are defined as online periodicals with non-downloadable distinct issues. Online periodicals without distinct issues are specifically excluded.
Just to make the wording more uniform and succinct. Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 14:11, 22 March 2024 (EDT)
Looks good to me. We may want to massage the RoA text further to clarify that e-Zines are to be recorded as ebook? The RoA has no clear mapping between what's included and what format(s) to select. It -is- listed in the format section though. Or perhaps just insert a reference to the Format template for ease-of-use? MagicUnk (talk) 10:43, 25 March 2024 (EDT)
Like this?
  • webzine - Used for Internet-based periodical publications which are otherwise not downloadable. ISFDB defines webzines as "online periodicals with non-downloadable distinct issues". Only those that meet this definition are eligible for inclusion. Please consult the Rules of Acquisition for detailed eligibility criteria. If needed, initiate discussions about inclusion/eligibility on the Community Portal.
Then the RoA parts could be changed to this:
  • Ezines, which are defined as electronic periodicals with downloadable distinct issues (i.e., PDF, epub, and so on), and should have the Format of "ebook".
  • Webzines, which are defined as online periodicals with non-downloadable distinct issues. Online periodicals without distinct issues are specifically excluded. These should have the Format of "webzine".
I included a link to the Format help page. ···日本穣 · 投稿 · Talk to Nihonjoe 14:29, 25 March 2024 (EDT)
Yup, I like it. MagicUnk (talk) 15:08, 25 March 2024 (EDT)
Reading ISFDB:Policy#Rules_of_Acquisition, I note that it doesn't specify what formats we use for any other types of publications. Adding this information for ezines and webzines only would create an exception and I am not sure it would be useful. I would only explain which formats we use for which types of pubs on the relevant Help pages. Ahasuerus (talk) 17:30, 25 March 2024 (EDT)
As you mentioned above, it might be good to have somewhere that defines "downloadable" and "non-downloadable". Maybe something like this?
  • Downloadable - Formats such as PDF, epub, mobi, MP3, MP4, and similar. These are for ebooks, audiobooks, and similar content.
  • Non-downloadable - These will generally be websites, generally only for webzines.
That should be good for a start on the discussion. ···日本穣 · 投稿 · Talk to Nihonjoe 20:12, 25 March 2024 (EDT)
Make sure to not get rid of the 'periodicals' notion as in your earlier proposal. But I do see value in clarifying (non) downloadable. And on the exception Ahasuerus mentions - my proposal to guide the user to the formats to use (either directly, or indirectly via pointer to the Format template), is because editors may get confused since webzine is defined here, and is also a format, while ezines is defined, but is not a format... At least adding clarification in RoA for those two should clarify. Also, I don't mind the exception. Don't see any harm in it. MagicUnk (talk) 12:09, 26 March 2024 (EDT)
How about something like this?
  • Downloadable - Formats such as PDF, epub, mobi, MP3, MP4, and similar. These are for ebooks, audiobooks, and similar content.
  • Non-downloadable - Generally only for periodical webzines that do not have a downloadable version of each issue.
Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 12:29, 26 March 2024 (EDT)
A couple of thoughts. First, I am not entirely sure that we are talking about the same thing. The issue that I had with the use of "downloadable" is that there are many browser extensions and other software tools that let you turn Web pages, including embedded audio files, into epub, mobi, PDF, MP3/MP4, etc files. WebToEpub is one of the better known browser extensions of this type and YouTube downloaders are also common. If you look at it from the perspective of a YouTube user who always sees a "Download as ..." button on YouTube pages -- because of some YouTube downloader that he installed years ago -- all YouTube videos may appear to be downloadable.
I am intimately familiar with this phenomenon because I read a significant amount of Web-published fiction, but it's always transformed into Kindle-compatible files first. The result is that I rarely make a conscious distinction between ebooks and Web-published fiction -- they all look the same on my Kindle.
Based on the above, my thinking was that it would be beneficial to clarify that we only allow works that are natively available as downloadable files.
Second, I think this discussion has effectively split into at least 2 separate sub-discussions and I am having trouble determining which argument applies to which sub-discussion. It may be best to have a separate section for the "downloadable vs. non-downloadable" topic. Ahasuerus (talk) 17:29, 27 March 2024 (EDT)

Defining "Downloadable"

Based on the discussion above, I would like to suggest adding a "Downloadable" section to the Definitions part of ISFDB:Policy. The proposed text is an amalgamation of what Nihonjoe and I wrote above:

  • Downloadable
    • Electronic content -- ebooks, audio books and so on -- is considered downloadable if the content provider made it publicly available as a file such as PDF, epub, mobi, MP3, MP4, and similar. It is not considered downloadable if the content needs to be converted to a file using tools such as browsers, browser extensions, or third party programs.

This would be displayed below the "Published" section. Ahasuerus (talk) 14:36, 3 April 2024 (EDT)

Sounds good to me. Very clear and concise. ···日本穣 · 投稿 · Talk to Nihonjoe 14:48, 3 April 2024 (EDT)
If there are no objections, I will add the proposed language to ISFDB:Policy tomorrow. Ahasuerus (talk) 16:48, 5 April 2024 (EDT)
ISFDB:Policy#Contents.2FProject_Scope_Policy and Rules and standards changelog have been updated. Ahasuerus (talk) 13:43, 6 April 2024 (EDT)

Baen vs Baen Books publishers redux

(Moved from Help, per Nihonjoe)

I'm reopening the Baen vs Baen Books publishers discussion as I currently have some edits on pause [3] that attempted to add the "Baen Science Fantasy Books".

The last time this was discussed, it kind of fizzled out, and so when PVing my Baen books I ended up just following the existing patterns for which Baen publisher variant to use. (I've currently have PVed 1077 Baen publications). The addition of the little used variant would follow existing patterns, but at a much lower usage.

Is this the time to get this straightened out? --Glenn (talk) 16:13, 27 March 2024 (EDT)

At one point we had an editor (User:Bluesman) who strongly believed that these were two separate publishers. He was the one who added the following comment to the two publisher records:
  • Do NOT merge this with Baen Books, there are two completely different timeframes and three different logos
He hasn't been active since December 2018, so we can't ask him why he thought that these were two separate publishers. The linked post includes the following comment by Nihonjoe:
  • I could ask Toni Weisskopf about it. She's the publisher at Baen, and has been with them since the beginning (or very close to it). ···日本穣 · 投稿 · Talk to Nihonjoe 19:21, 23 February 2021 (EST)
Let me ping him to see if he has had a chance to ask Toni Weisskopf. Ahasuerus (talk) 17:10, 27 March 2024 (EDT)
I haven't yet. Let me do so. Give me a few days. ···日本穣 · 投稿 · Talk to Nihonjoe 17:18, 27 March 2024 (EDT)
Thanks! Ahasuerus (talk) 17:30, 27 March 2024 (EDT)
Toni wrote: "It should be "Baen Books." (There was, briefly, in the '80s an attempt to separate out a Baen Fantasy line, but since it never went beyond a slight change of logo on the spine, and was only for a few months, I don't think that needs to be taken into account.)" ···日本穣 · 投稿 · Talk to Nihonjoe 21:03, 28 March 2024 (EDT)
Thanks for checking! Based on that response it sounds like we should:
We may also need to look into Baen Computer Books, which has two publications, and Baen Science Fiction Books, which has 108 publications. Both look like they could be turned into publication series under "Baen Books". The publication series Pournelle Users Guide, which contains 2 publications, is currently split between "Baen Computer Books" and "Baen". Ahasuerus (talk) 21:55, 28 March 2024 (EDT)
Maybe make "Pournelle Users Guide" into a regular series, and put both into a publication series called "Baen Computer Books"? ···日本穣 · 投稿 · Talk to Nihonjoe 23:13, 28 March 2024 (EDT)
That should work. Ahasuerus (talk) 00:23, 29 March 2024 (EDT)

(unindent) This seems somewhat parallel to how 'Ace Science Fiction Books' and 'Ace Fantasy Books' publishers got used in the mid-eighties. --Glenn (talk) 16:20, 29 March 2024 (EDT)

Back when fantasy took off in the late 1970s and early 1980s, some bookstores tried to create separate sections for fantasy books. I am guessing that Ace and Baen tried to make life easier for them by explicitly labeling SF/F books. Ahasuerus (talk) 15:34, 30 March 2024 (EDT)

(unindent) If there are no objections, I plan to implement the proposals listed above tomorrow. Ahasuerus (talk) 14:21, 3 April 2024 (EDT)

How would those books marked "Science Fiction", "Science Fantasy", and "Horror" within the "Baen XXX Books" on the title page? --Glenn (talk) 15:25, 3 April 2024 (EDT)
They would become publication series under "Baen Books". Here is how a 1990 "Baen Fantasy" publication is currently entered -- Warriorwards. Note the following lines:
  • Publisher: Baen Books
  • Pub. Series: Baen Fantasy
Ahasuerus (talk) 15:29, 3 April 2024 (EDT)

Outcome - Baen publisher and publication series records merged/reorganized

The following changes have been made:

  • "Baen" and "Baen Books" have been merged. The new publisher name is Baen Books
  • Fixer's submission mechanism has been updated to use "Baen Books" in the future
  • "Baen / SFBC" and "Baen Books / SFBC" have been merged. The new publisher name is Baen Books / SFBC.
  • "Baen Computer Books" is now a publication series under "Baen Books". All pubs have been migrated.
  • Pournelle Users Guide is now a regular series. Two non-fiction books by Jerry Pournelle have been added to it.
  • "Baen Fantasy" is now a publication series under "Baen Books". All pubs have been migrated; their primary verifiers have been notified about the migration project.
  • "Baen Science Fiction Books" is currently in the process of being migrated to a publication series. There are still 100 pubs that need to be migrated. I have run out of energy for the day; if anyone wants to take it over, please feel free. I plan to get back to the project tomorrow morning.

Ahasuerus (talk) 19:51, 4 April 2024 (EDT)

I think they've all been moved now. ···日本穣 · 投稿 · Talk to Nihonjoe 22:13, 4 April 2024 (EDT)
They have. Thanks! Ahasuerus (talk) 22:54, 4 April 2024 (EDT)
I still have 3 edits pending for the "Science Fantasy" variant. I can cancel, and convert to publication series edits. Should I proceed? --Glenn (talk) 15:34, 5 April 2024 (EDT)
Yes, please! Ahasuerus (talk) 15:42, 5 April 2024 (EDT)
Should it be "Baen Science Fantasy Books" or "Baen Science Fantasy" for the publication series? It seems inconsistent that "Books" is included in the "Science Fiction" publication series, but not in the "Fantasy", even though the title pages included "Books" in both cases. --Glenn (talk) 17:37, 5 April 2024 (EDT)
If the title page includes "Books" for both, then both should likely include it here. ···日本穣 · 投稿 · Talk to Nihonjoe 18:00, 5 April 2024 (EDT)
Luckily, since they are both set up as publication series now, we can change their names with a single edit. Ahasuerus (talk) 18:03, 5 April 2024 (EDT)
The title pages contain a hexagon, wider than tall, that contain the words "Baen" in the top half, and "Books", in the lower half. When present, the phrases Fantasy, Science Fantasy, or Science Fiction occur as a separate line between "Baen", and "Books", and is in a smaller font, sometimes in reverse video (foreground and background colors exchanged). --Glenn (talk) 18:46, 5 April 2024 (EDT)
Dug further. The series starts with the hexagon logo with "Baen|Fantasy|Books", but switches to Baen Fantasy with the dragon logo in about 1987. --Glenn (talk) 21:47, 5 April 2024 (EDT)
Thanks for looking into this. I guess there are two ways we could handle the change from "Baen Fantasy Books" to "Baen Fantasy". The first way would be to treat these books as a single publication series, "Baen Fantasy", which happened to have two different logos at different points in time. We would then document the logo changes in the Notes field of the Publication Series record.
The second way would be to split this Publication Series into two, one for "Baen Fantasy Books" and another one for "Baen Fantasy". Personally, I don't think it would be worth it, but I haven't looked deeply into it. Thoughts? Ahasuerus (talk) 10:53, 6 April 2024 (EDT)
Of those publications currently entered, 5 have "Baen|Fantasy|Books", and 20 have Baen Fantasy with dragon logo, either on title page, or spine, or both. --Glenn (talk) 18:20, 6 April 2024 (EDT)
OK, I have added a consolidated version of the descriptions above to the Note field. Hopefully it makes sense.
I have also searched Baen pubs for the word "Fantasy" in Notes and added "Baen Fantasy Books" as a publication series where appropriate. Ahasuerus (talk) 11:10, 7 April 2024 (EDT)

Clarification on Conduct Policy wording

A sentence near the bottom of the Conduct Policy states, "Note that these are general guidelines and ISFDB Administrators are not bound by them." This can be misread into admins not having to follow the rules. I think a clearer way to state this would be something like "Note that these are general guidelines, and ISFDB Administrators are not restricted to taking actions only against behavior explicitly mentioned here."

Thoughts? Better wording? ···日本穣 · 投稿 · Talk to Nihonjoe 13:31, 11 April 2024 (EDT)

It's been almost 18 years since I wrote that sentence, so I am not 100% sure, but I think that it was supposed to be read as an introduction to the next sentence:
  • Particularly egregious cases may be dealt with more promptly while repentant sinners may be given another chance.
So the idea was that administrators would apply the ISFDB:Policy#Conduct_Policy guidelines, but the exact punishment would be determined by specific circumstances. It's similar to how the law works in the larger world.
We could certainly try to clarify the intent and make the language ("repentant sinners") less playful. Ahasuerus (talk) 15:01, 11 April 2024 (EDT)
Perhaps rewording to something like this:
"Note that these are general guidelines. ISFDB Administrators may—in particularly egregious cases—deal with issues by skipping warnings, while repentant sinners may be given another chance. Administrators are trusted to use their discretion appropriately."
Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 16:35, 31 May 2024 (EDT)
I like the proposed change, but it occurs to me that it may also be useful to replace "Administrators" with "Moderators and Bureaucrats" to make it clearer who can impose these penalties. Ahasuerus (talk) 21:28, 3 June 2024 (EDT)
So something like this?
"Note that these are general guidelines. ISFDB Moderators and Bureaucrats may—in particularly egregious cases—deal with issues by skipping warnings, while repentant sinners may be given another chance. Moderators and Bureaucrats are trusted to use their discretion appropriately."
···日本穣 · 投稿 · Talk to Nihonjoe 15:09, 4 June 2024 (EDT)
Sounds reasonable. Ahasuerus (talk) 11:46, 6 June 2024 (EDT)
I think "sinners" should be replaced with "offenders" since it's less of a loaded term. Great otherwise. Phil (talk) 16:58, 6 June 2024 (EDT)
Works for me. Ahasuerus (talk) 17:39, 6 June 2024 (EDT)
Hearing no objection, I have updated ISFDB:Policy#Conduct_Policy and Rules and standards changelog. Ahasuerus (talk) 21:30, 10 June 2024 (EDT)

Definition of "advertising", and so on

Moving this here from the discussion on the moderator noticeboard, and putting it as a subheading under the conduct policy discussion above since it's pretty closely related.

Help:Contents/Purpose#What the ISFDB Wiki is not states that "The Wiki is a support tool for the ISFDB, and should not be used for anything that is not appropriate for that purpose." The Conduct Policy is also connected to this discussion. I think it boils down to the first paragraph of the policy: "Policy intent: Anything that helps make the ISFDB a more useful and more reliable bibliographic tool is encouraged. Anything that hinders this process is discouraged." (emphasis added)

In my opinion, political or religious content (regardless of which set of politics the content espouses or supports) on user pages and elsewhere on ISFDB that is not directly related to the stated purpose of "[making] the ISFDB a more useful and more reliable bibliographic tool" should be not only discouraged, but removed. ISFDB is not the place for making political or religious statements. There are plenty of other places online to do that.

Perhaps adding something like the following to the Conduct Policy would be good.

"The Wiki is a support tool for the ISFDB, and should not be used for anything that is not appropriate for that purpose. This includes posting—anywhere on the wiki—political, religious, or other content that is not directly related to making ISFDB a more useful and more reliable bibliographic tool. Such content is not allowed and may be removed by an Administrator on sight. Appeals regarding removed material can be made on the Moderator Noticeboard."

Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 16:35, 31 May 2024 (EDT)

I do support this intended move including the addition to the Conduct Policy. Christian Stonecreek (talk) 02:08, 3 June 2024 (EDT)
I like the idea, but we may want to clarify what we mean by "other content that is not directly related to making ISFDB a more useful and more reliable bibliographic tool". User pages in particular often contain information that is not "directly related to making ISFDB a more useful and more reliable bibliographic tool" yet is perfectly innocuous. For example, User:Alvonruff/ReadingLists lists all SF books that he has read since 2006.
In general, I suspect that it would be fine to have non-SF content on User pages as long as it is about the user and as long as it is informational as opposed to promotional. For example, if L. Neil Smith -- who was an SF author as well as a musician and a political activist -- were still alive and wanted to contribute to the ISFDB project, I don't think it would be objectionable if he added a list of links to his other Web sites/pages to his User page, but I don't think we would want his User page to become an advertising platform for his numerous social and political ideas or a place to sell his books. Defining this distinction in a way that could be fairly applied across the board could be admittedly tricky. Ahasuerus (talk) 21:54, 3 June 2024 (EDT)
I can see that. The main thing I don't want to see is divisive content such as political stuff (of any stripe), pushing of religious beliefs or anti-beliefs, and stuff like that. It will definitely be tricky to be explicit enough without painting ourselves into a corner. ···日本穣 · 投稿 · Talk to Nihonjoe 15:09, 4 June 2024 (EDT)

Crediting magazine editors

(copied from a part of a Moderator Noticeboard discussion)

... this discussion ... highlights a problem with our Help: we don't have an explicit definition or even guidance re: what types of magazine editors should be entered in the Author/Editor field and what types should be entered in Notes (e.g., assistant editors, associate editors, department editors, etc). Template:PublicationFields:Author currently says:

  • Editors, authors, translators, etc. ... For MAGAZINEs and FANZINEs, credit the issue editor [bolding added] as the "author" of the publication. (Note that for non-genre MAGAZINEs/FANZINEs, "Editors of PERIODICAL NAME" may be used instead of some or all editor names if they are unknown or unclear or not of genre interest -- see Help:Entering non-genre periodicals for details.)

Note the bolded part of the text, i.e. "issue editor", which is somewhat helpful, but is not very specific. Template:TitleFields:Author doesn't seem to say anything relevant either. I am thinking that we should start a Rules and Standards discussion and make our current de-facto rules explicit in the affected Help templates. Ahasuerus (talk) 15:58, 11 April 2024 (EDT)

Just let me note one other problem with crediting only the editor-in-chief that came to my mind during the last night: from time to time several magazines have allowed guest editors to edit one single issue: their respective stamp on the issue would be lost if we go strict by 'the only credit the editor-in-chief' policy. Christian Stonecreek (talk) 02:15, 12 April 2024 (EDT)
Also, for non-genre magazines it says ..."Editors...". Just sayin' ;). There exist co-editors (that are not department editors), so I would allow for them. But then the question is, can we come up with an unambiguous (set of) rules... MagicUnk (talk) 09:54, 12 April 2024 (EDT)
As a researcher I would want to easily identify who shaped / influenced the content of a publication. Christian (Stonecreek) mentioned some German magazines in his posting where only the primary editor (credited 'Redakteur') who did all the editing is currently stated - but not the editor in chief. It sounded right when I captured the resp. items - but I do appreciate that the credited editor in chief influences a magazine a consequently should have been added as well and needs to be amended. To me that's a nice and simple rule - take who is credited, i.e. editor in chief plus however else is properly stated. Everything else (text editors, writers of plot outlines and whatever else may come to mind) should be in the notes and nowhere else. Best, John - JLochhas (talk) 02:43, 20 April 2024 (EDT)
It does seem that the 'credit-only-the-editor-in-chief' rule does work for most of the big genre-defining magazines like 'Astounding/Analog', 'Amazing' 'The Twilight Zone Magazine' or 'MFSF': at least there seems nothing to be known of the other credited editors in choosing the contents.
On the other hand it does seem to me that many other magazines that are nearer to being published non-professionally are often to be found as edited by joined forces (like the newly issued "Worlds of If" or the German Exodus).
The other cited German magazines are found to be edited by one 'deputy' editor, and the editor-in-chief mostly supervising (and occasionally adding ideas or vetoing) the first one's efforts: take a look for example at the author page of Ute Müller which shows only the genre magazines, but she also is the editor-in-chief for even more nongenre magazines / chapbooks and other genre chapbook series like "Die UFO-Akten" and "Gespenster-Krimi". Christian Stonecreek (talk) 10:14, 20 April 2024 (EDT)
Dear ISFDB peeps, thank you for inviting me to this discussion as it is most interesting to read the behind the scenes conversations. I notice that Stonecreek is basing his push to include the deputy editor on a supposition of what the duties of a chief and deputy editor may or may not be in a given publication. One cannot draw such a conclusion with any degree of confidence as the dynamics of each publication are unique. I think that co-billing editors is a slippery and problematic slope, assigning responsibilities and weight of credit to job titles where a knowledge of such is not actually known or qualified in writing anywhere. The point of my edit submission and notes was why begin experimenting with conjecture and interpretation now? I also see that he says it's "joined forces." I have read the magazine and I think he is mistaken on that point as well. So I don't agree with the liberties that Stonecreek has taken with this one. Also, and I don't recall if I put this in my notes or just thought about it, but why is the new magazine listed as a relaunch when Clifford Hong's 1986 issue was not? Just seems like lots of irregularities and confusion over something that should be straightforward. Thanks, Jan ExplorerOne (talk) 01:16, 22 April 2024 (EDT)
Hi, and welcome!
To answer your statements I do think it is best to ensure that we are on the same level of knowledge regarding the editorials (it may be some time ago that you actually read them), so I'd like to ask you a few questions about them first:
1) What do the two editors (Sloane and Garnier) write in their respective editorials?
2) What do you think is the reason there are two editorials anyway?
3) Who (if any) of the editors is actually credited for choosing contents?
4) What is said about the respective preferences, and do you think they are in any way reflected in the magazine issue?
5) How is the relation regarding the editing between Sloane and Garnier described?
6) Who (if any) of the editors is writing about curating contents, and which are those?
Waiting for your answers, Stonecreek (talk) 05:31, 23 April 2024 (EDT)

(unindent) I'd like to comment on the following part of Stonecreek's response above:

  • It does seem that the 'credit-only-the-editor-in-chief' rule does work for most of the big genre-defining magazines like 'Astounding/Analog', 'Amazing' 'The Twilight Zone Magazine' or 'MFSF': at least there seems nothing to be known of the other credited editors in choosing the contents.

SF historians typically know quite a bit about the people who did the actual editorial work on the early SF magazines. For example, take Amazing Stories. We credit the following editors in the 1920s-1940s:

However, much of the actual work was done by other people. For example, here is what SFE says about T. O'Conor Sloane:

  • He was associate editor (designated managing editor for #1) of Amazing Stories and of Amazing Stories Quarterly from the beginning, and carried much responsibility for the actual running of the magazines, which involved compiling and editing the issues, subject to Gernsback's final approval, and after Gernsback's fall from power that of Arthur Lynch. He succeeded to the editorship of both journals in 1929.

Re: the 1940s, SFE says that:

  • [Howard Browne] worked 1942-1947 for Ziff-Davis where, among other responsibilities, he was managing editor of Amazing Stories and Fantastic Adventures, then under Raymond A Palmer's editorship.

Or take Gernsback's Wonder Stories family of magazines (1929-1936). The editorial work was mostly handled by David Lasser in 1930-1933 and Charles D. Hornig in 1933-1936.

This information has been known for many decades and yet we only credit the main editor. Ahasuerus (talk) 14:16, 22 April 2024 (EDT)

I do think that a person should be credited for the work he or she does or has done, and if the person did add knowledge, connections and expertise to a given magazine (for example in actually choosing contents) he or she deserves the credit for this work.
I view ISFDB as a site that also gives a concise perspective of the covered genres histories in supplying the various pages from title pages to issue grids. And since history is made by people, I think we should also cover their respective endeavors. Christian Stonecreek (talk) 06:00, 23 April 2024 (EDT)
There are different types of contributions and we credit them differently in the database. For example, cover designers and translators are only credited in Notes. Editors other than the stated editor-in-chief can also be credited in Notes or on linked Wiki pages, e.g. Series:Air Wonder Stories currently says:
  • Hugo Gernsback, president.
  • I. S. Mannheimer, secretary.
  • Sidney Gernsback, treasurer.
  • Staff:
  • Hugo Gernsback, editor-in-chief.
  • David Lasser, Literary editor, July 1929-February 1930, Managing editor, March-Mary 1930. Despite his title Lasser was in effect managing editor for all issues.
  • M. E. Dame, Associate editor, April-May 1930.
  • A. L. Fierst, Associate editor, February-May 1930.
  • C. P. Mason, Associate editor, February-May 1930.
  • Frank R. Paul, art director all issues.
Ahasuerus (talk) 08:58, 23 April 2024 (EDT)
I would go with this, except in the cases we know better: In my first post for the thread over at the Moderator noticeboard I wrote "Usually, I also would have credited only the editor-in-chief", but I also am definitely committed to the statement John posted here above: "As a researcher I would want to easily identify who shaped / influenced the content of a publication", which leads me to wish for the ones who have actually chosen the raison-d'être-contents of a given magazine (fiction pieces and poems for a magazine like "Weird Tales" or "Asimov's", essays & reviews for a magazine like Science-Fiction Studies) to be credited. Christian Stonecreek (talk) 13:00, 23 April 2024 (EDT)

Round Two

I am resurrecting this discussion since the issue is still outstanding. It's my understanding that there are two separate issues that need to be clarified in Help re: what is entered in the "Author/Editor" field(s):

  • Do we enter the name(s) of the editor(s)-in-chief only or do we credit other editors, e.g. associate editors, managing editors, literary editors, etc? The established practice has been to credit the editor(s)-in-chief only, but it wasn't explicitly documented in Template:PublicationFields:Author.
  • Do we credit issue-specific guest editors as co-editors along with the editor(s)-in-chief?

The current de facto standard that has been used by most editors is to enter the editor-in-chief only, but it is not explicitly documented in Help and some editors have been using different standards.

Here are the opinions of ISFDB editors that I have been able to find:

I will be asking other active editors to review this discussion and share their opinions. Ahasuerus (talk) 22:13, 10 June 2024 (EDT)

My opinion is still the same: enter only the editor-in-chief (except in cases of guest editors, in which case enter both the editor-in-chief and the guest editor(s)), at least until we add the ability to credit multiple types of contributors. Until then, the other types should be entered into the notes. ···日本穣 · 投稿 · Talk to Nihonjoe 22:42, 10 June 2024 (EDT)
I agree that ordinarily we should use only the editor-in-chief. For guest edited issues, we should certainly enter the guest editors. Where I differ is that I believe that in these cases, only the guests should be entered in place of the editor-in-chief. It is my recollection that the chief is not usually credited in the magazine for these issues. If both the chief and guests are present in the magazine, then they both should be entered. --Ron ~ RtraceTalk 06:26, 11 June 2024 (EDT)
While I only credit editor(s)-in-chief, I also make an exception for guest editors. In those cases, I only credit the guest unless the editor-in-chief is explicitly credited in the issue. John Scifibones 07:49, 11 June 2024 (EDT)
I can see that. I'm fine with that, too. ···日本穣 · 投稿 · Talk to Nihonjoe 12:34, 11 June 2024 (EDT)
I agree. -- JLaTondre (talk) 19:09, 11 June 2024 (EDT)
I agree as well. Annie (talk) 20:02, 11 June 2024 (EDT)
Like John, I'm still in favor of crediting the editors that we know of being responsible for selecting the contents relevant to us. In the case that brought this up (the new incarnation of Worlds of If) the editorials only give hint that Garnier is responsible for those: he is the only one who is actually credited for choosing contents and writes of himself as curating stories and poems. It seems that Sloane only had a major hand in choosing the artwork, which is also prominently featured in the magazine, & the essays and I guess he also made the connections to involve Robert Silverberg and Daniel Pomarède, but he ain't credited at all for choosing anything within the editorials. Note also that Garnier is credited as 'Deputy Editor-in-Chief', not just as a 'Deputy Editor' or 'Contributing Editor'. Christian Stonecreek (talk) 02:17, 12 June 2024 (EDT)
So, while we do have two editors-in-chief for the 2024 "Worlds of IF", the questions remain how to handle analogous cases (where there are distinct statements of other editors being responsible for choosing fiction contents) or cases like "Amazing Stories" for which Ahasuerus remarked regarding T. O'Conor Sloane: "He was associate editor (designated managing editor for #1) of Amazing Stories and of Amazing Stories Quarterly from the beginning, and carried much responsibility for the actual running of the magazines, which involved compiling and editing the issues, subject to Gernsback's final approval, and after Gernsback's fall from power that of Arthur Lynch. He succeeded to the editorship of both journals in 1929."
I do think it is our policy to eventually credit the actual editor(s), like we do with various magazines, possibly in varianting to the actual editor(s). (And analogous to the way we do for example credit the collaborative works of Henry Kuttner and C. L. Moore, like here). Now, if an actual fiction editor like T. O'Conor Sloane is credited in the copyright section of an issue of "Amazing Stories" (but not as editor-in-chief), we would variant the, say, 1928 run of the magazine to Gernsback and Sloane (relying on the dependable source of SFE), which I personally would regard as somewhat as absurd, because Sloane is credited for example as managing editor within the relevant issues. Christian Stonecreek (talk) 05:23, 12 June 2024 (EDT)
One thing to consider is that credited editors are not always the editors who put the issue together. It can take up to 6 months, sometimes more, for the new editor to exhaust the stories and the artwork originally purchased by the old editor. Credits can be handled differently depending on how the publisher and the new editor decide to approach it.
In the case of Astounding/Analog, John W. Campbell, Jr. took over as the editor-in-chief in September 1937. He is credited in the October 1937 issue, even though Astounding continued publishing stories purchased by F. Orlin Tremaine, the previous editor-in-chief, until the spring of 1938.
On the other hand, when Campbell died on 1971-07-11, the new editor, Ben Bova, wasn't credited until January 1972, when the bulk of Campbell-purchased stories was exhausted.
I believe it's an argument in favor of continuing our current de facto policy of crediting the stated editor-in-chief. Ahasuerus (talk) 13:05, 16 June 2024 (EDT)
I, too, am in the credit-only-the-"main"-editor(s) camp: editor(s)-in-chief or issue's guest editor(s). Whoever has top editorial billing, with guest editor instead of EIC unless explicitly co-credited. Those filling other credited editorial roles should be (optionally) documented in the notes until the data model is expanded to accommodate their capture in a more structured form. --MartyD (talk) 14:30, 12 June 2024 (EDT)

Cover artist credit small puzzle

I know we just finalized the Cover Art rules clarifications but I need a little more clarification for this case. In May 2022, I added a pub record for Fair Trade. When I added the Cover Artist, I used the name from the copyright page "David Mattingly" which is a variant of the canonical "David B. Mattingly". However, the canonical name "David B. Mattingly" was found on the dust jacket which I stated in one of the notes. At the time I believe this was the correct way to create this record. The latest rule says "If the artist's canonical name is stated in the publication, enter it". Since both are there, it looks like the correct entry now should be the canonical name. Question: Should any priority be given to what's on the copyright page over what's on a dust jacket when differing data is present? Should I go back and change this publication to use the canonical artist name instead of the variant artist name? Phil (talk) 15:21, 22 April 2024 (EDT)

For cover artist, I would consider both the copyright page and dust jacket / cover to be equal in preference. If they disagree, and one of them is the canonical, I would use the canonical name and put a note on the publication explaining what you described. ···日本穣 · 投稿 · Talk to Nihonjoe 16:09, 22 April 2024 (EDT)
The latest iteration of Help:Screen:NewPub#Cover_Art says:
  • For other sources of artist attribution use the following rules in the order they are listed below:
    • If the artist's canonical name is stated in the publication, enter it
    • If the artist's alternate name is stated in the publication, enter it and make sure to create a Variant Title later
which in this case means that we should credit David B. Mattingly.
In cases like this one it's especially important to document where the artist is credited and any different forms of the artist's name used in the publication. Dust jackets are easily lost, which was one of the reasons why we originally decided to use title pages as the main source of author credits. Ahasuerus (talk) 18:12, 22 April 2024 (EDT)
Since the canonical has been changed to David Mattingly, that's who you should credit. Still need to document the alternate name on the jacket. John Scifibones 18:31, 22 April 2024 (EDT)
Fixed. The note in the hardcover publication still correctly reads: "Cover art by David Mattingly" on the copyright page. "Illustration by David B. Mattingly" on the dust jacket rear flap. Phil (talk) 22:12, 22 April 2024 (EDT)

Clarifying "legally posted"

As we discussed in January-March 2024, Template:TitleFields:WebPage currently allows links to:

  • legally posted versions of the title's text

but it doesn't define "legally posted".

My current thinking is that we could clarify the template to disallow:

  • texts known to be under copyright protection and made available without the copyright owner's permission

This wording would disallow links to copyrighted texts in cases where the hosting site doesn't have the copyright owner's permission and claims "fair use" or "libraries and archives" exception under Title 17 of the United States Code. As mentioned previously, I don't think we are (or should be) in a position to decide which copyrighted texts fall under "fair use" and which ones do not. "Fair use" cases are settled by the courts on a case by case basis; we don't have the knowledge or the resources to deal with its complexities. Moreover, it's possible -- although in our case, perhaps, unlikely -- for links to copyright-infringing sites to be considered "contributory infringement". To quote this Cornell Law School Web page:

  • One who knowingly induces, causes or materially contributes to copyright infringement, by another but who has not committed or participated in the infringing acts themselves, may be held liable as a contributory infringer if they had knowledge, or reason to know, of the infringement. See, e.g., Metro-Goldwyn-Mayer Studios Inc. v. Grokster, Ltd., 545 U.S. 913 (2005); Sony Corp. v. Universal City Studios, Inc., 464 U.S. 417 (1984).

Whether knowingly linking to an infringing Web page constitutes "materially contributing to copyright infringement" is something that may be debated in court, but I don't think it's an area that we want to touch.

For US cases, the clause "known to be under copyright protection" would apply differently to 3 different periods:

  • Texts of books that are no longer under copyright protection, e.g. A Trip to Venus (1897) or The Altar of the Legion (1926), which were published before 1929 and are therefore in public domain in the US, would be "linkable".
  • Texts of books published between 1929 and 1963. Their copyright status is often unclear since they only enjoy copyright protection if copyright has been renewed. Since copyright renewal was rare for 1930s-early 1960s genre books like Zip-Zip Goes to Venus (1958), we can probably allow links as long as there is no evidence that copyright has been renewed. If such evidence is found, as was the case with H. Beam Piper's Space Viking the other day, we can document it and remove offending links.
  • Texts of books published after 1963 and therefore still under copyright protection would not be "linkable".

If we decide to do this, then it would be easy to create a cleanup report to look for PDF files associated with post-1963 publication records.

Thoughts? Ahasuerus (talk) 22:10, 22 April 2024 (EDT)

I think this sounds reasonable. This table from Wikimedia Commons may be useful, too, as it walks through how to determine if something is in the public domain in the States, or still under copyright protection. ···日本穣 · 投稿 · Talk to Nihonjoe 13:33, 23 April 2024 (EDT)
Looks useful, thanks. Including the body of the linked table (with WikiMedia-spedific templates removed/expanded) for reference purposes:
US copyrights for works first published in US, excluding audio works
Year of first publication
Note: publication is not creation
Copyright duration
  • Before [current year-95]
  • During [current year-95]–1963: without notice, or with notice but not renewed within 28 years of first publication
  • During 1964–77: without notice
  • From 1978 to March 1, 1989: without notice and without registration within 5 years of first publication
Work has entered US public domain
  • During [current year-95]–1963: with notice and renewed
  • During 1964–77: with notice</translate>
Copyrighted for 95 years after first publication
  • From 1978 to March 1, 1989: pre-1978 creation with notice, or without notice but registered within 5 years of first publication
  • From March 2, 1989 to 2002: pre-1978 creation
  • If author is known, copyrighted until the later of either 70 years pma (post mortem auctoris or "after the author's death") or Dec 31, 2047
  • If author is unknown or corporate authorship, the earlier of 95 years after first publication or 120 years after creation, but not earlier than Dec 31, 2047
  • From 1978 to March 1, 1989: post-1977 creation with notice, or without notice but registered within 5 years of first publication
  • From March 2, 1989 to 2002: post-1977 creation
  • Unpublished before 2003 (i.e. first published after 2002)
  • If author is known, copyrighted for 70 years pma (post mortem auctoris or "after the author's death")
  • If author is unknown or corporate authorship, the earlier of 95 years after first publication or 120 years after creation
Template:PD-US contains a summary of the public domain logic, but the page is specifically talking about images. The logic could be pulled out of that into a central page that is then referenced for images and text as applicable. -- JLaTondre (talk) 08:10, 25 April 2024 (EDT)
I'm not certain as to the scope being proposed here. While I agree with avoiding links to Luminist and other dodgy sites, as proposed, this seems to also be prohibiting links to the Internet Archive. I would object to that. While the Internet Archive lost a lawsuit over their "National Emergency Library" (where they dropped the one viewer at a time restrictions during the COVID pandemic), their regular library so far has survived challenges. I do not see a reason to ban links to their texts. -- JLaTondre (talk) 08:10, 25 April 2024 (EDT)
Thanks for the reminder about Archive.org and its lending program. I agree that we don't want to disallow links to archive.org. As I wrote in the discussion linked above:
  • It looks like the consensus is that archive.org links are OK to add. By default, archive.org only lets you access copyrighted books' metadata, cover images and the first few pages of the text, which is similar to what Amazon's Look Inside does. You have to join their "Lending Library" program in order to be able to "check out" books. The legality of the LL program is currently under review by the courts and the last brief that I know of was filed on 2023-12-15. As long as archive.org remains a legitimate organization and complies with relevant court orders, linking to its Web pages shouldn't be an issue for us.
We may want to clarify the proposed Help language to make sure that we don't accidentally exclude legally operating "internet-based lending libraries" like archive.org. Ahasuerus (talk) 13:41, 25 April 2024 (EDT)
Luminist is not a dodgy site; they have no torrents, membership fees for faster downloads, or anything like Anna's Archive or the defunct Ocean of PDF or any similar sites have and the person running the site even has a board on our site under the name of Luminist. They offer individual PDFs (years ago they used other formats but converted most to PDF a while back) that are linked offline at, I assume, a free or cheap site to save money. They've been around for many years and are perfectly legitimate. Anna's Archive, in particular, is a very useful site to research from for sites like ISFDB because they have a huge number of books not on The Internet Archive (an almost full run of Pulphouse: The Hardback Magazine, for example) but links are not appropriate because they offer those shady things I mentioned above. I've added the bulk of the 300+ Luminist links here over the last few years and I don't recall anyone ever leaving a message asking for any of them to be removed. What should be done is the same thing as the Archive does; if any individual or publisher asks for something to be taken down, do it; if not, leave it. --Username (talk) 09:06, 25 April 2024 (EDT)
Based on what I have seen on the internet, there is a spectrum of questionable -- or "dodgy" -- Web sites that make copyrighted material publicly available without copyright holders' permission. On the one end of the spectrum we have sites that openly state that they do not respect copyright and do everything in their power to evade law enforcement.
On the other end of the spectrum we have sites that come up with legal theories justifying their actions. In the Luminist case the justification is 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.” By accessing files linked to this site you are agreeing to abide by these restrictions. If you do not agree, do not download. If any copyright owner objects to our inclusion of their material on this web site, please do not harass our hosting providers; just contact us with the pertinent information. We will remove contested content promptly upon receipt of legitimate requests.
As I pointed out in January 2024, 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.
To what extent this and similar interpretations of the copyright law would be upheld by the courts is not something that we can guess at with any degree of certainty. Even if a Web site cites the correct "fair use" language of Section 107 of the Copyright Act, we are not (and shouldn't be) in a position to decide which copyrighted texts fall under "fair use" and which ones do not. "Fair use" cases are settled by the courts on a case by case basis; we don't have the knowledge or the resources needed to deal with its complexities. That's why the current proposal would exclude them.
Another thing to note is that the Luminist site makes 3 different types of texts available:
  • All pre-1929 (pre-1930 come January 1, 2025 etc) texts that it hosts are always OK to link to because they are out of copyright.
  • Under the current proposal, a 1929-1963 text would be also OK to link to unless it happens to be one of the rare (under 15% based on some estimates that I have seen) cases where copyright has been renewed and we know about it, e.g. H. Beam Piper's Space Viking (1963).
  • All 1964-2024 texts are under copyright and links to them would be disallowed under the proposed Help language.
Ahasuerus (talk) 15:43, 25 April 2024 (EDT)
If nobody here is clear then nobody elsewhere is, either, because nobody to my knowledge has ever asked for anything from Luminist to be taken down. If they do for individual works because they think they can squeeze more money out of a new edition, take those links down; if not, leave them. They're here for research on this site to fix or add information, not for profit. Any real collector would want a physical copy, not some scanned electronic version. --Username (talk) 16:13, 25 April 2024 (EDT)
Sorry, I've tried to post this 3 or 4 times and keep getting edit conflicts. It may no longer make sense, but I'm out of time.
I'd suggest broadening the proposed bullet to be something like
  • works under copyright protection and made available in contravention of applicable copyright law
We need to worry about more than pirated texts, and "without the copyright owner's permission" is both narrow and difficult to determine. Plus, of course, there is the problem of which laws govern the work in question.
The rest of the discussion seems awfully complicated to me, and I still worry about anything that can be construed as ISFDB's having responsibility to determine whether something is "legally" posted. I feel we are over-thinking it. I do not like the suggested date-specific criteria at all (US-specific, doesn't allow for truly legal postings of copyrighted material, for two things). I think ISFDB's position should be that if a court with appropriate jurisdiction: (a) decides the availability is ok, then it is "legal" and (b) decides the availability is not ok, then it is "illegal". That could be widened a bit at both ends to allow for temporary orders in one direction or the other (e.g., an injunction blocking specific posting(s) or an entire site's content provision, or the denial of a motion to grant such an injunction, until an open case is decided). Beyond that, I believe the ISFDB should have a policy of allowing links, with some process for case-by-case exceptions (e.g., the copyright holder or their representative asks to have the link removed) and perhaps one bigger exception where if multiple copyright holders make the same complaint/request involving the same source the ISFDB might decide to prohibit links to anything hosted by the same entity. IANAL, but we're not producing, hosting, or delivering the content, merely documenting where it is available on the internet. If ISFDB's having a link constitutes copyright infringement or abetting copyright infringement, then that would be the case for every search engine as well. --MartyD (talk) 16:24, 25 April 2024 (EDT)

Accounting for Different Jurisdictions

Different jurisdictions have different copyright laws, which often means that a text may be under copyright in one jurisdiction and out of copyright in another jurisdiction. The US/Australian versions of Project Gutenberg are probably the best known example of the resulting divergence. For this reason I would like to change the proposed Help language from:

  • texts known to be under copyright protection and made available without the copyright owner's permission

to:

  • texts known to be under copyright protection in the jurisdiction where the third party Web site is hosted if it is made available without the copyright owner's permission

(This change doesn't account for the "internet lending libraries" exception discussed above since it will need additional consideration.) Ahasuerus (talk) 16:04, 25 April 2024 (EDT)

Pub Series homonymy

i'm submitting the titles for an italian Pub Series, named Fantascienza and published by Armenia. Here is an example. Now I found that in the database there are already two more series with the same name. I guess that upon approval my titles would be inserted automatically together with the others, which would be wrong. Actually, the three titles already in the db are part of two series published by different publishers, and should be separated because these series are totally unrelated to each other. I cannot submit this change because, as stated in the help page "you can change the name, note contents, and links, but not the list of titles included in the series" - but I think a moderator could do that. Now the issue is how to tell them (all three of them) from one another. The same help page says that for Regular Title Series the separation can be done by adding a parenthetical qualifier... should we do the same also for Pub Series? For example renaming the three series as "Fantascienza (Publisher)"? The canonical name is just "Fantascienza" for all of them. Please advise - Luca --Fantagufo (talk) 15:20, 29 April 2024 (EDT)

That's exactly right. We disambiguate publication series names, publisher names, author names, etc using parentheses. The ISFDB software is aware of this convention and displays links to similarly named records on each related page. For example, consider G-8 and His Battle Aces (Adventure House) and G-8 and His Battle Aces (Berkley). Each Web page links to the other Web page. Or, to use an extreme example, consider Andrew Smith (VII), which says:
  • Note: There are other authors with the same name: Andrew Smith, Andrew Smith (artist), Andrew Smith (I), Andrew Smith (II), Andrew Smith (III), Andrew Smith (IV), Andrew Smith (V), Andrew Smith (VI), Andrew Smith (VIII)
Help:How to separate two authors with the same name has more details.
Ahasuerus (talk) 17:40, 29 April 2024 (EDT)
Thank you! I self-rejected my submissions for that pub series and submitted them again with "disambiguated" names. I also submitted the new series names for the 3 books of different, homonymous series already in the db. --Fantagufo (talk) 07:49, 30 April 2024 (EDT)

Invalid ISBNs

In the Help, first bullet point states: "You may enter the ISBN with or without hyphens. If the ISBN is valid then the ISFDB software will display it with hyphens.". There's nothing there, that I can see, about (intentionally) entering invalid ISBNs. I see there have been many discussions on this topic, but it's difficult to see what the current state of play is.

I'm cloning this publication for my 11th impression (with the same ISBN), noting that the Catalog ID was used in the first instance. I submitted (then cancelled) my edit with the ISBN in the ISBN field and, as well as the 'Invalid ISBN format' yellow flag, the ISBN shows without any hyphens - and, of course, had I left it submitted, it would have been rejected - the moderator knowing it would show up in the 'Invalid ISBN Formats' cleanup report.

Clarity would be appreciated. Thanks, Kev. --BanjoKev (talk) 06:39, 2 May 2024 (EDT)

The first thing to note is that there are two types of invalid ISBNs and two separate cleanup reports that address them.
The first type of "invalid ISBNs" covers normal 10- or 13-digit ISBNs that follow the standard format, e.g. 13-digit ISBNs start with "978" or "979", but their "check digit", i.e. the last digit (or "X" for ISBN-10s), doesn't match the rest of the ISBN. These ISBNs are entered as they appear in the publication with a comment in the Notes field explaining that the check digit doesn't match the ISBN. The software displays the ISBN in red with the words "(Bad Checksum)" displayed next to the ISBN, e.g. see this publication record. The cleanup report "Publications with Invalid ISBN Checksums" looks for this type of bad ISBNs and lets moderators "ignore" the discrepancy if the issue with the ISBN is documented in Notes.
The second type of "invalid ISBNs" covers ISBNs with invalid formats. For example, this publication record states that the ISBN is "2940000163078", which starts with "294", not a valid prefix for ISBN-13s since ISBN-13 prefixes are limited to 978 and 979. It's 13 digits, so it's presumably an "International Article Number" or "European Article Number" ("EAN"). These are not allowed in the ISBN field and need to be moved to Notes or, if they are External IDs (like Barnes & Noble IDs), to the External ID field. We have a cleanup report for these types of problematic values, Publications with Invalid ISBN Formats, and it doesn't support the ability to ignore discrepancies.
Does this makes sense? If it does, then we may want to update Help with this information.
I am also thinking that we should modify the software to display the words "Invalid ISBN format" (instead of "Bad Checksum") for ISBN-13s which do not start with 978 or 979. Ahasuerus (talk) 10:48, 2 May 2024 (EDT)
I like this last idea of changing the wording. It might be good to also have the new wording link to here. ···日本穣 · 投稿 · Talk to Nihonjoe 17:35, 2 May 2024 (EDT)
I've submitted my edit here. Ahasuerus, if I'd seen those two paragraphs in the Help, I wouldn't have raised this post :) They seem to cover everything, and I like the inclusion of examples - the submitter won't be startled when they see their approved edit with lots of red! I suggest you include your paragraphs under a subheading "Invalid ISBN" after the present last bullet point.
I agree with Joe, displaying "Invalid ISBN format" is much more elegant. Linking that to the Help so that the submitter can follow up would be the icing on the cake. Thanks both! Kev. --BanjoKev (talk) 19:16, 2 May 2024 (EDT)
OK, the software has been updated to display correct error messages/yellow warnings. It turns out that we already had scenario where the software displayed "Bad format" for ISBN values, so I used that wording to keep things consistent. I like "Invalid ISBN format" more than "Bad format", so if other editors agree, I can change it easily.
Re: Help, I'll wait a few days to see if other editors have additional ideas and update the Help template as per Kev's suggestion. Ahasuerus (talk) 19:53, 2 May 2024 (EDT)
Excellent! Thank you! Kev. --BanjoKev (talk) 03:36, 3 May 2024 (EDT)

Outcome -- Template:PublicationFields:ISBN updated

Template:PublicationFields:ISBN and Rules and standards changelog have been updated. Template:PublicationFields:ISBN has been further streamlined to explain how the ISFDB software handles embedded hyphens, spaces, etc. Ahasuerus (talk) 16:47, 6 May 2024 (EDT)

Thanks for your work on that - it looks well covered now and very readable! Kev. --BanjoKev (talk) 15:31, 7 May 2024 (EDT)
Great! Ahasuerus (talk) 15:34, 7 May 2024 (EDT)