Rules and standards discussions/Archive/Archive16

This is an archive page for the Rules and standards discussions page. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from January - December 2018.

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

Special Genre Issues of Non-Genre Issues

Italics copied from ISFDB:Help desk#magazine issue on genre topic. See that discussion for context:

I guess yet another paragraph should be added to Help:Entering non-genre magazines. Something on the order of Headline: "Special issues" Text: "Sometimes, a non-genre magazine will devote an entire issue to speculative fiction and/or articles about it. This can be regarded as a genre publication and catalogued in its entirety, even though other issues of the magazine aren't." --Vasha 12:43, 13 January 2018 (EST)

I'm all for it. Is there any argument against this? Stonecreek 03:58, 14 January 2018 (EST)

Agreed. If the whole issue is genre, then that issue is a genre publication. -- JLaTondre (talk) 08:13, 14 January 2018 (EST)

Agreed. Jens Hitspacebar 08:55, 14 January 2018 (EST)

It seems reasonable, but it may be prudent to clarify what we mean by "the whole issue". Do "special issues", which concentrate on SF but include other things, count? If they do, do we want to include the non-SF content? Ahasuerus 09:51, 14 January 2018 (EST)

Good point. I would recommend only including genre content. How about "Sometimes, a non-genre magazine will devote an entire issue to speculative fiction and/or articles about it. This can be regarded as a genre publication and genre non-fiction should be cataloged along with the fiction (even though we do not normally catalog non-fiction from non-genre magazines)." -- JLaTondre (talk) 10:08, 14 January 2018 (EST)

I like that. --MartyD 13:37, 14 January 2018 (EST)

Me too. Christian Stonecreek 14:17, 14 January 2018 (EST)

I like the wording. Annie 22:55, 14 January 2018 (EST)

Sounds good. --Vasha 11:07, 15 January 2018 (EST)

I also like the proposed wording. ···日本穣 · 投稿 · Talk to Nihonjoe 14:01, 15 January 2018 (EST)

I have updated Help:Entering non-genre magazines to add the new section using the agreed upon wording. -- JLaTondre (talk) 09:13, 17 January 2018 (EST)

Revisiting standardization of author names

I've recently been told by multiple moderators that there's a policy to always write an author's first initials with periods after them, even if the author themselves and the publication don't. However, I find that Help:Screen:NewPub#Author(s) states "In the very rare case where an author prefers two (or more) initials as if they were a name (such as "TG Theodore"), without period or space, and is so credited, we follow the author's preference." What's the current thinking? I can see arguments both in favor of and against changing author names to a standardized form—I just want all moderators to agree on one policy, whichever it is, and have the help screen changed if necessary. --Vasha 08:56, 24 January 2018 (EST)

The original standard was to always use periods after initials. The reason for it was that some older magazines and anthologies had a habit of randomly dropping periods and capital letters. We didn't want to create variants and pseudonyms every time an editor chose to use "r a lafferty" instead of "R. A. Lafferty" or "r a heinlein" instead of "R. A. Heinlein".

Then Steven H Silver stopped by and pointed out that "H" was his middle name and not an initial. It started a discussion which, after reviewing the evidence, resulted in the current Help wording. I believe it still reflects our policies. Ahasuerus 16:06, 27 January 2018 (EST)

(BTW it really isn't a very rare case.) --Vasha 08:56, 24 January 2018 (EST)

It looks like it's become more common lately. We may want to delete the word "very" from the clause cited above. Ahasuerus 16:06, 27 January 2018 (EST)

(unindent) Let me get this straight: is the following correct? There should be two variants, one with periods and one without, if the author is credited both ways in publications. But that only applies IF the variant without periods is the author's preferred spelling. If the author usually writes their name with periods, then it should be recorded with them in the record for a magazine that prints the name without? --Vasha 16:19, 27 January 2018 (EST)

Pretty much. Let's use Steven H Silver as an example. Most pubs credit him without a period (his preference) and our canonical name doesn't have a period. The few times when he was credited using a period, we recorded the data as it was entered in the pub ("Steven H. Silver") and created variants.

On the other hand, take R. A. Lafferty. All of his "r a lafferty" stories (I recall coming across a few many years ago) have been entered as by "R. A. Lafferty". Ahasuerus 22:57, 27 January 2018 (EST)

And just to be clear, again, the form without periods should be canonical? Because it is the author's preference and in most cases it will be the commonest in publications. --Vasha 10:58, 28 January 2018 (EST)

Hmmm. Something's not consistent here. If our normal policy is to normalize the as-given credit to conform to our capitalization, spacing, and period-after-each-initial standard (ignoring any special exceptions), specifically so as to avoid creating variants, then it seems we should apply that same policy in similar situations where we're omitting punctuation or spacing or are down-casing. E.g., if all of these credits -- RA Lafferty, R.A. Lafferty, R. A. Lafferty, r a lafferty, etc. -- would be recorded as "R. A. Lafferty", then it seems any occurrence of Steven H. Silver should be recorded as "Steven H Silver", with a note about removal of the period, and no variant made.

Much as I lament our various deviations from the record-it-as-it-appears rule of thumb, it is even worse to vary what we do in specific situations. We should either normalize capitalization/punctuation/spacing of author names in all cases, or we should not ever do it. Note that "normalize" can nonetheless allow for different formatting according to circumstance. --MartyD 12:29, 28 January 2018 (EST)

I agree that it's a concern, but let's consider the following scenario. Suppose an ISFDB user who is aware of our "period and space after an initial" rule comes across "Steven H. Silver" elsewhere, e.g. in a publication or on Silver's Fancyclopedia page, which uses the "H." form of his name. The way we have things set up right now, the user does a search on "Steven H. Silver" and discovers that it's a pseudonym used by "Steven H Silver".

Suppose we were to change the data entry rules and update all of our "Steven H. Silver" titles to use "Steven H Silver". There would be no easy way for an ISFDB user to find his bibliography in the database. Ahasuerus 13:31, 28 January 2018 (EST)

Yet another case where some sort of fuzzy searching would be useful. It is also a problem that we have to create variant titles for trivial differences in.punctuation (British "Mr" vs. US "Mr." and so on). Is there any possibility of ignoring punctuation when searching? --Vasha 14:50, 28 January 2018 (EST)

I have run a few tests on the development server and it looks like we can modify the regular Author search to ignore punctuation without doing too much violence to performance. Titles would be tougher to do: a sample search went from 0.8 sec to 1.3 sec. Ahasuerus 16:40, 28 January 2018 (EST)

Well, it seems that cases like this one (SHS) were one of the reasons why the use of abbreviated names without punctuation was restricted to very few authors. If users won't find Steven H Silver the next step would be to search for Steven H. Silver, whereas the other way wouldn't occur to many users. Stonecreek 16:10, 28 January 2018 (EST)

Impossible to restrict this to very few cases if we're going to follow author preferences at all. In that case once we find out that an author uses a non-standard punctuation, we have to adopt it, and there are literally hundreds of them. (I just yesterday found four more that I hadn't known about.) --Vasha 16:33, 28 January 2018 (EST)

(unindent) My first inclination is to enter an author's name as he wishes, but in many cases it is as the country of origin wants it. Using the Steven H Silver example, in some countries it's mandatory to print the author's name with a period, while in others it is automatically left out. The same is true with authors like Andrew J. Offutt who often published with his name in lower case. While it would be nice to do what the author wants, there does seem to need to be as system to go by, or everything will just end up being just a mess. Listing an author's name with such variations as pseudonyms doesn't seem to be an option either. Might I suggest that the standardization as is continue, but we have a page in which writers names are listed as they are published, but then these names have a link to the author's standardized name? I.e., somebody types in KJ Kabza and they go to a listing of similar names, and they find KJ Kabza's name with a link to this author's official page. This would save the site the trouble of constantly creating pseudonym pages for authors who want an unorthodox variation of their names on their writings. MLB 19:22, 28 January 2018 (EST)

I agree with you that this is awfully messy. The more I think about this, the less I like any version of "make a variant for the name as it appears in the publication, but only if." It is inconsistent and user-unfriendly that way; it requires people who are entering new records to look up whether the author is one who can be varianted or not every single time. Wouldn't it be better to simply have one form of the name, even if that means making lots,of notes about what's in the publication (a field for that?) That single form could be the author's own or a standardized one. I can't decide which I prefer and there are arguments in favor of both.

Using a standardized form would be relatively simple in concept, but putting it into practice would involve a lot of editing of records. These are the parts of the system needed: 1. A single author page with their name in standardized form (K. J. Kabza). 2. A note on that page stating that the author himself spells it KJ. 3. A note on every publication and/or title record stating which spelling is used in which publication. 4. A search engine that will pull up K. J. Kabza's page if someone types in "KJ Kabza." --Vasha 06:19, 29 January 2018 (EST)

If we'd get an author request on a certain spelling, we should go with it, but then it still may be better to make an unpunctuated version of a name into a variant of the punctuated one, since we have to weigh in the needs of the users, too; one aspect seems to have gone out of consideration: that an author is published with an unusual spelling doesn't necessarily mean that it's the author's intention. There are other parties (editors, publishers) involved. Your submissions, Vasha, in many cases were done without giving proof of the author's request or preference, and often without asking primary verifiers. Stonecreek 06:23, 29 January 2018 (EST)

"doesn't necessarily mean that it's the author's intention" is why I say that if we are going to use the author's own spelling, we should consult their personal web pages and social media for evidence. It's true that if we only made variants for authors who personally requested us to, there'd be so few of them that the inconsistency wouldn't be very noticeable--we'd still on the whole be using a system with only one form of the name. But it seems unfriendly to the authors to force them to do that. We should have some way of indicating what the author's spelling is, taken from their web presence rather than only by request. And either create variants for any nonstandard spelling (which I don't really want to do) or for none--we can accommodate author requests some other way than by making variants. --Vasha 06:45, 29 January 2018 (EST)

Record-it-as-you-see-it and making variants based on that seems to be the most straightforward solution. Due to limitations of the system, we can't do that for capitalization variances, but we could for punctuation and spacing. It won't solve the "KJ" versus "K. J." search issues, but it will at least lead to being able to find a credit that one is looking at. What difference does it make if we have a lot of variants? Doing it that way then just isolates the how-to-normalize-it issue to choice of which name is the canonical. And that decision can be documented in one place. --MartyD 07:29, 29 January 2018 (EST)

Maybe we should all get together and just teach authors how to spell their own names. MLB 11:58, 29 January 2018 (EST)

Any more thinking on this? We need to get it settled. The latest problem case is G.Kay Bishop, who wants to spell their name without a space between the "G." and the "Kay." --Vasha 02:14, 3 March 2018 (EST)

Being as there isn't a resolution yet, here's what I'm going to do unless and until there's a consensus to do otherwise. The Help currently states "In the very rare case where an author prefers two (or more) initials as if they were a name... without period or space, and is so credited, we follow the author's preference." So therefore:

1a. I will use the author's preferred spelling (if known, e.g. from their website) as the canonical form of their name. 1b. If I can't figure out the author's preference, I will use a non-standard form of the name as canonical if and only if it is a lot more common than the standard form.

2. In all publications, I will enter the name exactly as it is printed, varianting if necessary. (Note that in at least one case which I've seen, the author has spelled their name with and without periods at different times, but is currently using the period. In such cases I will enter the publications exactly as printed even though the canonical name is not non-standard.)

3. Point (2) only applies if I have seen a preview or asked a verifier; in existing publications where I don't know what's in the book, I will leave the name as it currently is, varianting it to a new canonical form if necessary.

4a. Although the Help only refers to "two or more initals... without period or space," I will use these same guidelines for all non-standard author preferences for punctuation or spacing (e. g. "G.Kay Bishop"). However, initals with periods after them must always have spaces between them. 4b. Since it is not possible to have multiple different capitalizations as pseudonyms, I will use a non-standard capitalization as canonical if and only if there is clear evidence that the author wants it that way, and clarify publications with notes if necessary.

If anyone doesn't like this solution, speak up! --Vasha 13:08, 10 March 2018 (EST)

As long as there is no consensus, one should stick to the rules, as it seems there is no basis for just another self-defined rule. Stonecreek 14:02, 10 March 2018 (EST)

Exactly--sticking to the rules is precisely what I am doing. All that I did with those four points was to explain in detail how I am interpreting the guidelines in the Help. Can you show me a written rule that contradicts anything I said? --Vasha 14:12, 10 March 2018 (EST)

Now, that would be a new development! As with misusing the title tags, in changing names I have seen numerous examples where you seemingly followed your own preferences, not necessarily the author's one. At least there were numerous examples where you proposed changes based on the spelling on some publications (not all, and some against PVed ones), and some where you supplied no proof at all. Stonecreek 14:39, 10 March 2018 (EST)

What sort of documentation would make you happy? Shall I post a link to the author's website in the moderator notes of every single record that I add or edit, if there is something non-standard about the name? Shall I count how many publications use one spelling vs. the other and post that number in the moderator notes of every single edit? --Vasha 14:55, 10 March 2018 (EST)

Yes, something along these lines would be very much appreciated. And with your proposed point 2. you are absolutely NOT sticking to the rules as your recent submission for a publication edited by one 'Robert N Stephenson' made clear. Either you haven't understood the rules or are willingly messing things up. Stonecreek 15:10, 10 March 2018 (EST)

Yes, I did follow the rules. Robert N Stephenson uses that spelling AND it is spelled that way on the title page of the book, so it should be used in the database, according to Help.

You are the one who's not following the written guidelines. You can't just say "I've been doing it X way for years;" either rewrite the guidelines or follow them. --Vasha 15:13, 10 March 2018 (EST)

So, what is there that you don't understand in the sentence Initials should normally be entered followed by a period and a space as "Gordon R. Dickson" or "K. D. Wentworth", even if period or space is omitted in the publication.? Stonecreek 15:32, 10 March 2018 (EST)

I quote Ahasuerus above: "Steven H Silver stopped by and pointed out that "H" was his middle name and not an initial. It started a discussion which, after reviewing the evidence, resulted in the current Help wording. I believe it still reflects our policies." Thus, the part about following author's preference applies to middle intials also, not just two initials. --Vasha 15:46, 10 March 2018 (EST)

I am speaking of Robert N. Stephenson here, for whom you tried to apply your self-invented rule #2. Please apply the valid guidelines! Stonecreek 04:18, 11 March 2018 (EDT)

According to the written guidelines and the fact that, as Ahsduerus indicated, those guidelines apply to middle initials, Robert N Stephenson should be in the database as Robert N Stephenson. The fact that he is not only means that the the current form of his name is incorrect (by the written principles). But see Annie's comment below for a sensible proposal to make things easier. --Vasha 18:08, 11 March 2018 (EDT)

(unindent) When I started editing here, one of the easy rules on the books was about the author names - if it has an initial, add a dot and a space and ignore whatever the book decided to use; capital letters for the first letter in each name; small for the rest. It was refreshingly uncomplicated. If I was adding a magazine or an anthology, I did not need to go and check each name to see if that specific author is not saved in the DB under some fancy capitalization or with some fancy spaces and dots somewhere. The few exceptions were exactly that - well known exceptions.

Don't get me wrong, I am all for what the authors want but this is a DB and even if search can be tweaked to connect the dots (no pun intended), using arbitrary rules based on preferences makes the already tedious process of adding anthologies and magazines much worse. Add to this the fact that then the moderator needs to do the same checks (because almost none of the editors will explain why they used that spelling in a moderator note) and the thing just becomes silly. Not to mention just how confusing it is for someone that is just starting. We are a database and databases need to have strict standards. Adding a new field into the author and using that as a "display name" makes a lot more sense than having a standard that is essentially "if the editor finds a reference that the author really do not like dots, don't have dots". Either we go with "whatever is on the book" or we go with "here is our strict standard". Everything else is just unnecessarily confusing and complicating things without adding any value. Annie 17:33, 11 March 2018 (EDT)

I have to disagree with the "without adding value" part--authors have a right to decide how they spell their name, and apparently Steven H Silver managed to convince everyone of that when he posted--but everything else you say makes sense. Editors entering new publications should not have to know what the author's preferred spelling is. As things currently are, entering exactly what's in the book will create lots of duplicate names that aren't varianted to each other. So yeah, it would be very nice for people to be able to just enter a standardized form. Your idea of adding a field to the author record to indicate their preference is an excellent one. There's just one problem--how do we deal with the problem of recording exactly what's in the book? Some people will make notes if they know to, but not all the time. Are you willing to lose having that detail recorded? Personally I think it isn't important and yeah, it doesn't add much value to know exactly what's in the book, but I want to hear from other people whether they want the exactness. --Vasha 17:52, 11 March 2018 (EDT)

(unindent) I love Annie's suggestion of having an extra field in the author record that would contain the author's preferred spelling and cause their name to be displayed that way everywhere, while the main Name field would contain a standardized form. That makes it easy to enter publications, just use the standardized form, no need to look up what the author usually uses. A note in the publication record would be optional--if you know several different spellings are in use, note which one is used in that publication record. This wouldn't be too technically difficult, right?

And if we don't go with that solution, the Help needs to be clarified. Thanks a lot... --Vasha 08:32, 6 April 2018 (EDT)

Sorry, but there's no need to make this urgent, since we do have valid rules, AND we have such a desired field: it's called the note field. Any additional information regarding an author might be put into this nice item. Stonecreek 13:32, 6 April 2018 (EDT)

Christian, you have said many times that you disagree with my interpretation of the current rules, but I haven't heard from anyone else. That's why I'm begging and pleading for an update to the help section that will clarify matters. This can't be just between you and me. -- Vasha 13:54, 6 April 2018 (EDT)

Oh, it isn't. I've watched you espouse your views for quite a while, mostly ignoring anything that contradicted what you want. The entire database can't possibly be changed just so you don't have to think [in other words keep your edits to under 7 seconds, about the limit of your attention span]. You never interpret, you just don't want to do it that way. Waste of cyber-skin. --~ Bill, Bluesman 22:09, 9 April 2018 (EDT)

(unindent) I agree that our current lack of consistency re: author initials can be frustrating. I myself have run into this issue on a number of occasions.

I think the underlying problem is that common usage has changed over the last few decades. A hundred years ago, even 50 years ago, it was understood that having a period after an author initial was the standard. Occurrences of author initials without a period were due to typographical conventions (notably at some digest magazines) and not to be taken seriously by bibliographers. The few cases when an author insisted on a "naked" initial -- famously Forrest J Ackerman -- were seen as affectation at best. Over time common usage changed while our data entry standards haven't. It's not an uncommon scenario in the bibliographic field: we have repeatedly changed our data entry standards and the software to support e-books, audio books, links to 3rd party Web pages and so on.

That said, every time the world changes, it takes some time for our standards to catch up. Being too slow is bad, but being too hasty can be bad as well. For example, back when Wikipedia became big (2005-2006) Al and I kind of assumed that it would ultimately become the repository of all knowledge. Our plan was to work closely with it and to move some of our data, especially biographical data, to Wikipedia pages. Within a couple of years we realized that Wikipedia was not necessarily the panacea that we had thought it would be, e.g. their notability policies meant that they would never have Web pages for most of our authors. We also realized that our decision to create a single "Wikipedia URL" for author records had been wrong. We ended up rewriting the software to support multiple 3rd party URLs per author and, later on, per title, per publisher, per award type, etc.

And so, even though I understand the frustration, I think it's important to come up with a comprehensive consensual solution to make sure that we don't have to backtrack and re-do things in the future. Annie's suggestion is an interesting one, but there are other potential approaches. For examples, we could adopt a combined approach:

• state that due to recent changes in the publishing world we would start entering author names as printed on title pages and create pseudonyms/VTs when needed, and
• create a new "standardized author name" field in the author record; it would always include periods after initials and would be used by the search logic so that users could find author names regardless of the way they are "initialied" in books/magazines

Please don't be discouraged by the fact that the last Rules and Standards discussion didn't yield any immediate results. Sometimes it takes a number of iterations before we find a solution that is both feasible and addresses everyone's needs. Then we have to prioritize it, but that's a whole different issue. Ahasuerus 13:58, 6 April 2018 (EDT)

I had been on the record with a similar position as Christian as well - trying to bend the rules causes more problems than it solves and makes both adding books and moderating records much harder. I think that we should just stop trying to follow every fad of US publishing and just follow our rules and use the Notes (and/or new fields) to ride them. I dread working on magazines and small publishers anthologies these days because of creative naming conventions for the authors and us not following our own rules. Annie 14:27, 6 April 2018 (EDT)

That is certainly true, Annie, it is a big messy mess! I'm just trying to figure out what I should do right now, while things are unsettled (and while current rules allow for exceptions but are poorly worded as far as explaining exactly which exceptions are allowed).

I've been compiling quite a long list of authors who use non-standard initials/capitalization/spacing, along with which publications various spellings are used in. I guess I should put that on a Wiki page so that we can all refer to it. --Vasha 14:56, 6 April 2018 (EDT)

How about an innovative approach? Follow the rules. Initial, dot, space. Initial, dot, space. Notes in the author name and in the publications with creative naming conventions and we have all the information that we need. These are fixable downstream if needed anf if we change the rules but in the meantime they do not cause triple work for everyone.

I know you mean well and that you are trying to advocate for the authors but this is a DB and consistency is important. We need a system identification for the author - this is what our canonical name is for the most part - we default on the most used name but... how initials are different from suffixes - or are we going to decide to break that pattern as well? And if we start following this for the US market and keep changing the rules on the fly, what happens when I start asking the same for the Bulgarian ones for example? Let's not go there, shall we? :)

Plus the DB is a scary place for new editors anyway - making it almost impossible to figure put how to enter a name does not help matters. If anything, I think we should go back and clean the mess we had already created by not following our own rules. Annie 15:03, 6 April 2018 (EDT)

Or abandon any pretenses that we are actually following rules and start recording names exactly as they are written in the publication. Which I am even less of a fan of. Annie 15:05, 6 April 2018 (EDT)

I don't disagree with you; standardization would be user-friendly. But there was a decision, years ago, to allow exceptions. And that's still in the rules. So strict standardization isn't following the rules. That's one of the things this inconclusive discussion has been about--should we overturn the former decision that exceptions are allowed? --Vasha 15:08, 6 April 2018 (EDT)

There is a point of no return on all of this. Are we a bibliographic website or just another Index??? Once a decision is made to just STANDARDIZE [I really wish there was a way to make this the obscenity it is ....] there's really no way to go back. At best there has been an attempt to make the 'best-of-both-worlds' work. Sometimes it does, sometimes there's some twit who can't stand that there's a 'difference' in use and usage and the 'help/rules' get lambasted; and they're NEVER satisfied until it's changed to their way. They're flawed, so what. We grow. Bibliography is messy, it's what keeps us thinking [at least more than 7 seconds per edit] and why there are always going to be adaptations. Changing the perception that our Help listings are rules instead of guidelines would make most of these discussions much easier. We already have ALL the tools we need to deal with [supposed] aberrations in authors names/story titles/etc/etc that we need. By simply recording data as it is in the publications we list, and then dealing with the 'fallout', ........... if there's an editor or mod who can't deal with that, or just wishes to iron-clad/stomp/jackboot any variance into submission ..... they should leave. I'll help the twits pack ........ --~ Bill, Bluesman 22:35, 9 April 2018 (EDT)

I think it is - the exceptions are there for special cases but I think we stretch this a bit too much. I would argue that any exception needs to be discussed on the community pages - if it is not done or there is no agreement, then strict standardization is the way. Otherwise you end up with different editors having different ideas on what is "conclusive" and what should constitute an exception. Or even if we do not stretch it that far, using a non-standard initials handling should ALWAYS be recorded on the author page with links to the sources showing the author preference. Does not help with the adding of the works of that author but saves everyone a lot of time in finding out why... Annie 15:50, 6 April 2018 (EDT)

Yeah, that is all very cautious and wise... No problem, for now I will just be keeping a list of all the proposed exceptions. --Vasha 16:30, 6 April 2018 (EDT)

(unindent) Here is a list I've been keeping of unusual names. Note that these are just ones I've jotted down over the last couple months, the tip of the iceberg... --Vasha 15:14, 7 April 2018 (EDT)

Follow the rules and all is good. I've the feeling that some authors don't know what name they have --Wolfram.winkler 05:19, 16 July 2018 (EDT)

How to handle author credit of traditional stories (again)

I think we should decide on a rule regarding traditional, re-told tales and clarify the help accordingly. I'm coming up with this because we currently have an author page like that of Ludwig Bechstein, which has several titles where the variants are from other authors ("Grimm" and "Kletke"). Moreover, "Grimm" has also been made into a "pseudonym" of Ludwig Bechstein, which is not correct because the Grimms weren't. According to Ludwig Bechstein's Wikipedia page it looks like the Grimms and Ludwig Bechstein were different authors, each collecting tales independently. I therefore propose the following:

• If a data source doesn't clearly state that a tale is either a retelling by the credited author, or that the author has re-used the text from another author, we assume that it's a different work and use the credited author.
• Only if a data source clearly states that the credited author has simply re-used the text from another author do we use the orginal author (adding a note about the re-use by the credited author).
• Title records, which are retellings by different authors, must not be varianted. Example: "The Three Dogs" by "Grimm" must not be a variant of "Die drei Hunde" by Ludwig Bechstein.
• Different authors who retold tales must not be made pseudonyms.

See also the previous discussion How to handle author credit of traditional stories above.

Jens Hitspacebar 07:35, 10 February 2018 (EST)

I agree. However, there is an additional complication when a source uses the wrong author as a credit (this happens outside of traditional stories as well). I've seen it handled by 1) using the credit as supplied and varianting without a pseudonym; and 2) modifying the credit (example Ludwig Tieck (in error) and varianting with a pseudonym. I recommend we standardize on an approach there as well. For all these cases, notes should be liberally used. Links within the notes can show the relationships for those wanting to capture them. -- JLaTondre (talk) 09:08, 10 February 2018 (EST)

I feel it is ok to use a variant, with no pseudonym, for the wrong-credit situation, and have a note in the wrongly credited title with an explanation. This approach is consistent with our handling of works with no credit; we do record it as (not) credited, but we don't make that "uncredited" a pseudonym. --MartyD 10:19, 10 February 2018 (EST)

I'd like to thank --JLaTondre for inviting me to this discussion. While doing the Andrew Lang Fairy Books, I've come across this situation a few times & would really appreciate how to handle credit mistakes. --MartyD & JLaTondre both know the trouble I've had with The Enchanted Watch credited to Deulin. I always try to verify a story before I variant or link to a parent title(which takes me a long time when I can't check the original collections!) When I was doing The Green Fairy Book I noticed The War of the Wolf and the Fox didn't match any of the Grimm Brothers' stories, even tho it was mentioned on the wikipedia page for their tales. I found it in Hermann Kletke's collection. I was just going to list it under Kletke when it was put under Bechstein & I'm not sure what to do. I'd would rather break the variant to Bechstein & have it & the 2 other ones currently linked to Bechstein under Kletke. I know Kletke mentions Bechstein in regards to these stories but I think it just makes things confusing. I still haven't found the original of The Golden Mermaid credited to Grimm(& I know from doing this work titles & even character names were sometimes changed by Andrew Lang's translators). In regards to the Grimm as pseudonym, I was wondering of making those that haven't been verified under something like Grimm-x, with a note stating this is to be used for that purpose only? Sort of like that version of Maupassant's The Horla that is used when the translator is unknown. So far there's only a handful of stories in this situation & wonder if this could be a solution-any thoughts?Loviatar 02:02, 15 February 2018 (EST)

Entering translator

What is the rule on entering the translator? It's entered on the Title Page Note, of course, in the Tr template form, but can't it be entered additionally, as so many people have been doing, on each relevant Publication Page's Notes? It seems more user-friendly to have it there where a link may land you, and not have to navigate elsewhere. Likewise it seems helpful to me to mention the language in the Notes, although I can see that entering the original title, as I've been doing, can be redundant (although a first-time user might not scroll down when their laptop screen puts Contents below the fold). So--is there a rule on what can and can't go in the Notes as additional info? --Martin. MOHearn 12:00, 20 February 2018 (EST)

There are two things to consider here - the data in the DB and the users of the DB. For the first one, we do not need the translator in the publication - we all know to look at the title level and adding in the publication is just repeating known information. But for external users that are not so well versed in the way the data works here, having the translator inside of the publication record of novels is actually useful. I would not go out of my way to add it to other editors' work (I make sure it is in the title level when I am moderating) but when I am adding my own publications, I am adding to both (in a lot of cases it is because it is a Bulgarian publication and I need to make some notes on how the translation is marked - translation from a specific edition or an uncommon language, different from the original one for example)) and I would just leave it there if an editor adds it, especially if they are PV-ing. Just my 2 cents Annie 14:29, 20 February 2018 (EST)

I'm not aware of a rule that forbids to additionally state the translator in the note if a publication. Moreover, the software has no translator support (yet) therefore I think it's not a bad idea to do so. Language, however, is a different thing: it shouldn't go into the note redundantly (except maybe for some complex cases), because the software has language support and should itself be able, after some changes, to show the title's language automatically on the publication page, if such a feature is desired. I actually think it's a good idea if this feature would be added. Jens Hitspacebar 11:43, 21 February 2018 (EST)

It would be easy to change the Publication page to display the language of the container title. If you want to post this proposal on the Community Portal, we could then discuss the implementation details. Ahasuerus 17:30, 21 February 2018 (EST)

What Jens said is the same point of view that I can subscribe to. Christian Stonecreek 12:21, 21 February 2018 (EST)

So we all are on the same page here. Annie 12:33, 21 February 2018 (EST)

In my messages Hauck has said that the rule is that the translator data goes into the Title Notes and is not to be duplicated in the Publication Notes (and explained that it helps keep the data simpler to manage). After looking around the site for the rule I can see that it's implied in the Publication Notes help page, but it's merely an implication. Could the help page be made explicit one way or the other? If it specified that the translator data either is or is not allowed to be duplicated in the Pub Notes, then there would be a bona fide rule. As it is, it felt like the rules changed under me when I'd been adding the translator in the publication all this time without being told it was a problem.

Repeating the language entry from the container title via the software does sound like a very good idea. MOHearn 22:18, 21 February 2018 (EST)

OMNIBUS Definition Inconsistency

The OMNIBUS definition in Template:PublicationFields:PubType doesn't match the database. It states "Conversely, if a book is originally published as multiple volumes, and republished as a single volume, the latter is a novel unless the presentation within the single volume makes it clear that the works are presented as separate novels. For example, "The Lord of the Rings", by J.R.R. Tolkien, was originally published in three volumes; the single volume edition is to be categorized as a novel." However, the single volume The Lord of the Rings is entered as an OMNIBUS and not a NOVEL. Since it can be argued that some of these versions do present the individual books a separate novels, is this just a bad example that needs to be replaced? Or has practice changed & the whole quote should be struck? -- JLaTondre (talk) 17:15, 6 March 2018 (EST)

This seems like a similar situation to a fixup novel, or a novel-in-stories. If the final result is presented as a novel, that's the type it should be. LOTR seems like a bad example to choose to demonstrate the principle since there are some editions that are more omnibus-like and some that are more novel-like. --Vasha 17:27, 6 March 2018 (EST)

It really seems to be a bad example, which should be replaced. The initial novels were published independently, and, though they tell a chronological epic, the OMNIBUS should stay an OMNIBUS. We really have many examples of ongoing stories published in one volume that could be viewed as novels. Especially it would be a NOVEL that includes other NOVELs and that we already had ruled out, I think. Stonecreek 04:11, 7 March 2018 (EST)

The Date field for Titles that first appeared in a Magazine

I'm a new contributor. I read the rules about Date field in Magazine records. The two things that struck me were

'and also because there is no good way to determine actual publication date'

and

'For magazine cover dates that cannot be assigned to a specific month, use the year only: "Spring 1943" is just entered as "1943-00-00"'

For U.S. magazines there is a good way to determine actual publication date. Most of them have copyright registrations which list the date of publication for each issue. These are available online - for convenience I've saved on my hard drive PDF files of all annual Periodical Registrations books from 1891 through 1977.

I wish the second point would be reconsidered. I can see that there is a common consensus (i.e. not just ISFDB but elsewhere) to simply use the month and year for monthly periodicals. I don't believe there is common consensus to use "00-00" for quarterlies, annuals, and any other non-monthly periodicals. Why not put in accurate dates where known?

My real question, though, is would it be against the rules to make the Date field of individual titles be the actual publication date? E.g. "The Derelict of Space" by William Thurmond & Ray Cummings. This was first published September 1, 1931, in Wonder Stories Quarterly Fall 1931 which has a Date field value of 1931-00-00. The Title Record for the story also has a Date of 1931-00-00. Would it be OK to put 1931-09-01 in the Date field of "The Derelict of Space"'s Title Record? PatConolly 02:56, 16 March 2018 (EDT)

Yes, it would be against the rules as they are now stated. However, for quarterly publications, I see no reason to not state an actual month of publication. A publication for 'Spring' of a given year could be determined to have the first month of the timespan in question assigned (March), just as a publication for the first quarter of a year would be assigned YYYY-01-00. Stonecreek 03:20, 16 March 2018 (EDT)

I'm against it, particularly in this case as, to the best of my knowledge, September 1, 1931 is not IN Fall 1931 (for us europeans it's in Summer). It's for the same reasons that we do not date monthly magazines by the availability date. I've also strong reservations about the fact that the date registered by the copyright office is the real publication date (or what we mean by this term). In France books and magazines are also registered (it's the Dépôt Légal) and this date is rarely the same as our "publication date". I'm not sure that a copyright office's data has any relevance as a bibliographic reference (if it has, why not use the printer's records or the printing date that's on this issue).

Hervé, you're absolutely right on the exact date; the date of registration of an official national library has some more relevance than the date of availability as stated with some online bookshops, but not that much more. But in your stated case we would file the date of publication as September 1931 (1931-09-00), which would be an analogue to the three month span for any other quarterly published magazine. Stonecreek 06:30, 16 March 2018 (EDT)

In the U.S. copyright registrations, Publication Date for a magazine meant the same as Publication Date for a book - namely the date that the publication was on sale. The publisher submits that date with his registration. I would say it's as relevant as a statement printed in the magazine saying "next issue on sale (some date)" or in ads saying "this other magazine on sale (some date)". And if you check you will find that these publication dates stated in magazines match what the registrations show - at least it did for many decades (I don't really look at current magazines). I'm not sure why printing date would be preferable to on-sale sate. E.g. Asimov's first novel was published Jan. 19, 1950 and that's what ISFDB Date shows. If it was determined that the printing had occurred in December, would it be desirable to change that Date field to 1949-12-00? PatConolly 12:54, 16 March 2018 (EDT)

What information is the Date field intended to capture? If it is merely the date printed on a magazine as a means of keeping them separate (except for quarterlies which would NOT be separate per current rule), then it's redundant, since the Title field already says something like "Weird Tales, May 1930". Why not use it for the date the Title was actually published? This may be a personal quirk, but if Jan 1939 Amazing Stories came out 7 weeks before Jan 1939 Astounding Science Fiction (which it did) I like to know that. I like to know where Quarterlies fit in with the monthly magazines. I like to know whether Winter 1943 means the beginning of 1943 (Jan) or the end of 1943 (Dec). Well, basically I just like to know when things were published in the sense of "made available to the public". Hope this wasn't annoyingly long. PatConolly 12:54, 16 March 2018 (EDT)

My remarks had nothing to do with stating additional information in the notes, sorry if it seemed so. But as for publication dates, we go with the official ones as stated by the publisher, and that means for most (monthly) magazines the month stated with the title. Stonecreek 12:58, 16 March 2018 (EDT)

Yes I knew we are only discussing the specific Date field. I have been submitting the dates in the Notes field. And my main objection is not about the Date field for monthly magazines, but the Date field for non-monthly magazines. 1928-00-00 seems objectively less meaningful and informative than 1928-10-20. By the way, after you read the above I removed a mildly inflammatory statement about "accurate data" and replaced it with "Why not use it for the date the Title was actually published?" (following my point about it redundantly restating what is said in the Title field).PatConolly 13:42, 16 March 2018 (EDT)

Title Convention for Magazines

Hello. I have a question, please: How important is it to follow the convention for magazine titles? It is stated this way:

For the title of a magazine, the best source is the information (often below the table of contents) about the publisher, giving the address; this often says something like "IF is published monthly by . . . ." If this is not present, the magazine cover and the heading on the contents page are about equal in priority; again take a good guess.

I'm asking because I've noticed that some issues of Analog do not follow the convention. For example, the title for the October 1960 issue is recorded as Analog Science Fact -> Fiction, which matches the large heading on the contents page (which is the secondary choice for a title). However, it does not match the information paragraph, which says "Analog Science Fact & Fiction is published monthly by . . . ."

As another example, the information paragraphs in the current issues of Analog read: "Analog Science Fiction and Fact (Astounding) . . . published bimonthly by . . . .", but they are listed as only Analog Science Fiction and Fact in the ISFDB. In this case, the information paragraph is on the editorial page rather than the table of contents page. This alternate title is mentioned in the ISFDB publication notes for the first two issues of 2018.

I'm happy to submit these changes and similar ones from my Analog collection if they are wanted, or perhaps someone would like to propose a change to the convention. Or maybe it can all be left as is.

Please advise and accept my thanks for your answers. --Main 19:44, 18 March 2018 (EDT)

Looks like your query slipped through the cracks! Your link above doesn't work for me but this one has the title info you referenced. It looks like you're exactly right. Someone with higher seniority than I have will hopefully shed some light but I notice that there are conventions that are used that don't stick by the rules here that become sort of de facto rules that are only learned by using the site a lot. It may be a matter of it looks less cluttered in the display listings without the added "(Astounding)". I'll ask MLB and Hauck who do a lot of Analog entries for their take on this and see if they'll join. Doug / Vornoff 15:10, 23 April 2018 (EDT)

John W. Campbell reportedly hated the Astounding name, and if so, I can only wonder what he would think of this continued discussion. However, be that is it may, the cover states Analog Science Fiction and Fact, this is repeated on the title/contents page, while the headers just simplify everything as Analog. The (Astounding) part is only listed in the indicia. Not sure what the Statement of Ownership says, but personally, despite this being a good question, I would just keep it as it is. Somebody who has access to their whole collection could check out if Dell Magazines always listed, in their indicia, Analog as it is listed now. Again, I'm for keeping it as is. Still, I'm just a working bee here, the ultimate decision will be somebody else's. MLB 22:13, 23 April 2018 (EDT)

(unindent) What do you think the name of Fiyah magazine should be? They don't have a header at the top of their contents page; at the bottom it says ©Fiyah Literary Magazine. On the cover it says "Fiyah Magazine of Black Speculative Fiction" and they also start off every editorial by saying "Welcome to Issue (number) of Fiyah Magazine of Black Speculative Fiction." Should we just ask the magazine staff which is their name? --Vasha 23:53, 23 April 2018 (EDT)

Mark, I have a gap in my Analogs from 84 to 97 so the earliest Dell Analog I have is October 97 and the title over the contents as well as the cover both say "Analog Science Fiction and Fact" and the indicia has, again, "Analog Science Fiction and Fact (Astounding)". There seems to be some kind of leeway in the rules and with the amount of people that have verified these issues it seems the consensus would be to keep it as is. I like it that way as well - it's a lot cleaner.

And Vasha, I'd vote for "Fiyah Magazine of Black Speculative Fiction". There is no real indicia or contents page header apparently, so the next logical thing would be the cover, plus it gives you a good idea of what the mag is about, rather than just "Fiyah". Doug / Vornoff 00:52, 24 April 2018 (EDT)

Format pb vs. tp - interim solution for German publications

Continuing the discussion from above about Rules_and_standards_discussions#Format_pb_vs._tp which was left open with an unresolved proposal, and because "'pb' or not, that's the question" is a recurring problem with submissions for German publications, I suggest to add Chrstian's proposal from above as an exception rule for German publications to Template:PublicationFields:Format until a more generic solution has been found. Here it is again (I amended to values in inch to exactly fit the values in cm and added "or smaller"):

Since the histories and dimensions for the mass market paperbacks are slightly different for different countries, there are the following exceptions:

German paperbacks ("Taschenbuch" in German) / trade paperbacks ("Paperback" in German):

1. For books at least as tall as 19.01 cm (7.48") and at least as wide / deep as 11.5 cm (4.53") use "tp".
2. For books as tall as 19 cm or smaller, generally "pb" should be used. Exception:

• For books as tall as 19 cm or smaller, but with a width equal or higher than the height, "tp" should be used.

Jens Hitspacebar 06:03, 24 March 2018 (EDT)

I knew I was forgetting something! As per our last discussion, I changed the way publication formats are handled internally a few months ago. I was going to post to the effect that the enhancement should make it easy to add new formats going forward, but then it slipped my mind. Sorry about that! We can add "taschenbuch" to the standard drop-down list if desired. Ahasuerus 11:14, 24 March 2018 (EDT)

Oh, very cool! And no worries, we all forget things :) "Taschenbuch" would be a great addition. As per discussion in Rules_and_standards_discussions#Format.2Fbinding_-_proposed_software_change) we should on one hand indeed leave out "German" in order to include other German-speaking countries in this term. On the other hand, here comes the tricky part: we'd also need something similar for "trade paperback", but they are called "Paperback" (sic!) in German, which will certainly lead to confusion with the standard "pb". Therefore it's probably necessary to call it "Paperback (German)" for disambiguation with "pb". With "(German)" at the end we also imply other German-speaking countries (opposed to a term like "German Paperback", which rather implies paperbacks in Germany only). If we decide to use "Paperback (German)" we should probably also use "Taschenbuch (German)" for consistency. Jens Hitspacebar 05:38, 25 March 2018 (EDT)

I am afraid I don't know enough about these formats to have an opinion, so I'll defer to the experts. (We may want to wait a week for Christian to come back from hiatus.) Ahasuerus 11:08, 25 March 2018 (EDT)

Ok. I leave a message on his talk page. Jens Hitspacebar 11:32, 25 March 2018 (EDT)

Hello Jens! I can only say that I completely agree with Herve's point of view [1].If someone thinks that common sense leads to chaos, then a solution is barely achievable. It will always be searched to find new pitfalls. A solution-oriented discussion is no longer happening. I am in favour that everything should stay as it is. Rudolf 15:02, 26 March 2018 (EDT)

I fully understand Hervé's point regarding complexity, and I'm also ok with not adding new formats, as long as we somehow get rid of the recurring discussions about "tp or pb?" for German publications. If we leave everything as it is they will never stop and will be wasting time again and again. I currently don't see any other solution than to at least add a special rule (without adding new formats) for German sizes, as initially proposed above. If we don't, especially most of newer German paperbacks ("Taschenbuch") will all have to be changed to "tp" - which they are definitely not (that's what I meant with "common sense" on Wolfram's page). If there's another, better solution - count me in! :) Jens Hitspacebar 15:24, 26 March 2018 (EDT)

I am fond of the labels "Paperback (German)" & "Taschenbuch (German)", which really seem to be the most consistent. Thanks for your efforts, Christian Stonecreek 00:51, 3 April 2018 (EDT)

Before this peters out unresolved again and to sum it up it seems there are the following different solutions/opinions:

1. Leave everything as it is (please no).
2. Only add a rule exception for the German formats to the help, but don't add a new entry in the selectable list of formats. That's actually my initial proposal above. This needs no software change except for the the mouseover help which would become language-specific, and it would be sufficient to solve the problem. However, as for the sizes, I just dug around a bit and it looks like the current "official" German definition of "trade paperback" is quite different from my original proposal: according to these sources, a "trade paperback" for the German market must be at least as tall as 20.5 cm and must have inner flips on front and back cover. That's it. Everything else is just "paperback", no matter how tall or wide. Sources (all in German, sorry): buchreport.de, vlb.de, boersenblatt.de.
3. Add new formats for the German market (which, admittedly, adds complexity which may not be necessary considering options #2 and #4, and would also require to correct the format of all German pub records with softcovers).
4. Only add one new format: "sc" for "softcover". Wolfram and I think Annie also proposed something similar as well recently. With additional fields for width and height (and maybe an option to choose "inch" or "cm") it would provide all we need for cases which don't match the current pub format rules. Personally I'd prefer to get rid of the distinction between "pb" and "tp" altogether and only keep "sc", but I highly doubt that this is going to happen :) So, this would be "pb", "tp" and "sc" for softcovers to choose from, and you'd use "sc" if you can't make a clear decision regarding "pb" or "tp".

I'm in favour of #2, though #3 or #4 should probably be considered in the long run if more and more non-English pubs are added from markets which have other pub formats and can't be handled with just "tp" and "pb". Jens Hitspacebar 16:15, 9 April 2018 (EDT)

Hello Jens, meanwhile, I have the opinion, that we must include all countryspecific formats (worldwide).--Wolfram.winkler 13:58, 19 April 2018 (EDT)

It seems to be too complicated, therefore only softcover/hardcover and the dimensions height/width/thickness...--Wolfram.winkler 05:46, 27 June 2018 (EDT)

I agree. Come on, please, let us vote, so that there is finally a result.--Wolfram.winkler 08:07, 29 June 2018 (EDT)

I am not sure how we would go about it since we have 4 options on the table and 4 editors with different takes on it. I have left a message on Jens's Talk page to see if he would like to revisit this issue. Ahasuerus 16:23, 29 June 2018 (EDT)

Rewriting "Exception for works which have illustrations preceding their title pages"

Forgive me if my links and formatting are not ideal. I'm a newb.

This rule is not well written and should be changed. Template:PubContentFields:Page reads in part: Exception for works which have illustrations preceding their title pages - If a magazine presents artwork for a story or essay preceding the piece's title page, and it is apparent that the art accompanies the text, the starting page of the story or essay should be the page number of the artwork which illustrates it. If you're creating content records for both the work and its illustration, they would have the same starting page. (See "Sorting" below for multiple works appearing on the same page.) If there is no indication that the artwork is related to the text on the succeeding pages, and no indication in the table of contents that it illustrates the work, then do not count it as the first page of the work.

The initial use of the word "works" causes confusion. Let us consider any Easton Press published novel. Nearly all Easton Press novels include an illustration on page i or ii, preceding the first page of the story text by several pages. As written, the rule above says first page of the story should be listed as i or ii, because the novel is a "work," the artwork is for the story, and the artwork precedes the "piece's title page." One potential literal interpretation of the current rule leads to nonsense. A book with length [viii+232] could have a story length of 238 with the first page an illustration on iii, and 5 of those 238 pages being either blank or containing the colophon and other front matter.

For any work which has an introductory illustration, it should be included in the story's page count only if it immediately precedes the text of the story. Otherwise a preceding illustration listed as the first page could create a situation in which the illustration, 3 pages of ads, 2 blank pages, and an introductory essay included in a story's length. This very situation has just occurred on data I entered, with a moderator citing the rule as it is currently written (above), insisting on including an illustration and a blank page in a novel's length. (That is not how I read the current rule, but I'm not a moderator.)

The rule does make sense for magazines or collections that often include introductory art for a story.

Suggested changes: "magazine" to "publication" and "preceding" to "immediately preceding".

The rule would then read: Exception for works which have illustrations preceding their title pages - If a publication presents artwork for a story or essay immediately preceding the piece's title page, and it is apparent that the art accompanies the text, the starting page of the story or essay should be the page number of the artwork which illustrates it.

This is my first suggested change to the editing rules and standards. If my organization of this suggestion is off-base, forgive me.MartinStever

Hi, and welcome. I like your proposed rewording. --MartyD 06:58, 15 April 2018 (EDT)

It's been my understanding that this exception only applies to magazines which is certainly where this situation occurs most often. How have other editors interpreted it? --Ron ~ Rtrace^Talk 08:50, 15 April 2018 (EDT)

This wording makes sense for illustrated stories in anthologies and collections, but not so much for novels. We already have rules for how to deal with illustrations on unnumbered pages in novels. So instead of "publication," how about "magazine, anthology, or collection"? --Vasha 12:49, 15 April 2018 (EDT)

I am a bit rusty in this area, so perhaps it's a dumb question, but better safe than sorry. What is the rule for illustrations on unnumbered pages in novels, please? Ahasuerus 12:20, 21 April 2018 (EDT)

I was just wondering the same. I cannot think of a specific rule off the top of my head at the moment... Annie 19:38, 23 April 2018 (EDT)

I was just thinking of the rules in that same Regular Titles section which say to derive a page number and put it in brackets or use designations like fp, bp. Not actually any help in the current discussion. --Vasha 19:53, 23 April 2018 (EDT)

It's irrelevant here - this specific case here is for using the page number of one element to denote another one (the art page number to be used as a story/essay page number). Deriving numbers is clear (so if the art is there and not numbered, count and bracket it). I wonder if Martin is not conflating the two things as well a bit (because of the Easton Press example). The rule does not say "for example, if a a magazine". It specifically talks about magazines (and it is a lot to do with the way American and British magazines use artwork if you ask me). Anyway - back to the question at hand. Annie 20:01, 23 April 2018 (EDT)

Yes, that seems to fit most. Stonecreek 14:02, 15 April 2018 (EDT)

I like Vasha's suggested change to my suggested change.MartinStever

This is my first time participating in a rules and standards discussion. How are they concluded? MartinStever

Some discussions linger on, never coming to a conclusion.... In a case like this, where there's no apparent opposition, I suggest you "move the question": Post an out-dented summary statement of what was agreed to above and provide a full quote of your final proposed wording, plus a deadline for any objections (e.g., a few days or a week). When your deadline passes and no objections have been raised, edit the help text and update the change log. --MartyD 08:13, 21 April 2018 (EDT)

(unindent) I think clarifying that this exception only applies to interior art pieces which immediately precede the works that they illustrate would be useful. I am a bit leery about enumerating the publication types to which this exception applies since it would be likely to make things more complicated. For example, does the word "magazines" also cover fanzines in this case? What about non-fiction? Etc. Ahasuerus 12:25, 21 April 2018 (EDT)

An issue that has maybe been overlooked is that the passage MartinStever highlighted is in the section on how to enter Regular Titles. So some of this discussion (including some of the things I said) is moot. Novel titles in publications with type NOVEL are not usually treated as the kind of regular titles that need page numbers, but they can be--for example, if there are also introductions or afterwords. So in that case, it is necessary to decide which page to say that the novel begins on. I don't know why the exception was written to apply to only magazines. Martin's proposed change actually works, without enumerating which publication types it applies to, if we understand that it applies to novels only if we need to decide which page the novel text begins on for some reason.

I don't see any guidelines stating what to do if a story in an anthology/collection has a separate title page. The usual practice is to use the title page number, right? Is that stated somewhere I overlooked? That should be added, maybe as a section before "Pages without a printed number." Say, "Content items with a title page immediately before the text -- The item should be given the page number of the title page." Then, the exception will make sense in the form Martin proposed. --Vasha 16:27, 21 April 2018 (EDT)

It is not just fanzines - what happens when we have omnibuses that reprint (or rebind) magazines/fanzines? Do we change the number of pages compared to when they are single entities? What about stories added to a novel or collections that are essentially one novel and multiple stories. And then we have the not so uncommon anthology -> magazine conversion and vice versa - some of the publications we have are clearly one or the other; some are not and something conversion may be needed based on additional research. Having different rules for magazines and anthologies made this conversion especially weird if one applied the rule just for magazines).

How about we look at that from a different angle - instead of trying to define the containers, define the TYPES which first page changes based on a preceding interior art - short fiction, essay, serial, poem (don't forget the poems - they do not get art that often but they do now and then). That's it, isn't it? Covers all the cases and I think is what MartinStever was going for. Annie 18:24, 23 April 2018 (EDT)

Actually no, we have to decide what page a novel, anthology or collection begins on if it's in an omnibus. So it's all regular titles that have page numbers. --Vasha 19:17, 23 April 2018 (EDT)

Then add novel and non-fiction if so you wish - the same applies. Don't list the containers, list the types for which this applies. Or we are bound to find a corner case again or end up in the same situation.

Although we also need to be very careful - we have quite a lot of publications already and moving this rule from "magazines" to "any/most publication(a)" will invalidate the page numbers in a lot of them - because strictly speaking the rule specifically specifies magazines at the moment (not even fanzines). Interpretation had differed I am sure and I usually would stretch it to fanzines but the rest of the types? Tricky and a bit too stretchy (despite my grumbling for being a bit illogical above). It may be logical to change it but such a change is not trivial and changing such rules in the middle of a project are bound to backfire. And quite honestly I would rather have consistency in the DB than try to "make it correct". Maybe a few of the people with larger collections may chime in on how this had been interpreted by looking at some real books (collections and anthologies) and their records here and see just how different the page numbers will be. As for the Easton Press - I would not count the start that front piece as the start of the text... Just my 2 cents. Annie 19:30, 23 April 2018 (EDT)

After sleeping on it, I think I share Annie's concerns.

As an aside, it occurs to me that there is a noticeable difference between the way we enter interior art in books as opposed to magazines. When entering magazines, we generally enter each illustration as a separate INTERIORART record and assign pages to each one. On the other hand, when entering books, we often enter all illustrations as a single INTERIORART title. That's how it was done in the publication which prompted this discussion: according to the Note field, there are seven illustrations, but we have only one INTERIORART title. Help supports both ways of entering illustrations, but in practice the former way is more common on the magazine side of the database. It may be one of the reasons why this Help exception was originally created. Ahasuerus 21:26, 24 April 2018 (EDT)

Sure, that makes sense. But we still do have to add a line to that paragraph about whether or not to give the page number of a story (etc.) in an anthology as that of the preceding half-title page if there is one. Note that sometimes the title may be printed on the half-title page and at the top of the story text. --Vasha 22:24, 24 April 2018 (EDT)

The rule now says "magazines". Which is different from anthologies so it does not apply for anthologies. I am not sure why should we add anything more if the rule remains only for magazines... Or am I missing a place where it is unclear - because if you read the exclusion rule, it is very clear: "If a magazine" - not "for example if a magazine"... If you are advocating for adding a new exception for other types - then this is a separate topic but my concern above about existing data applies to it... Annie 22:39, 24 April 2018 (EDT)

My apologies for branching off to a different topic-- this is something that I noticed while reading the help section looking for info about the illustration question. I will start a different thread for the half-title issue. --Vasha 22:51, 24 April 2018 (EDT)

I'm jumping back to address Annie's comment of 24 April 2018 (I was on an 11-day road trip, no isfdb time while traveling). My reason for suggesting this rule change was based on an edit I submitted that was rejected. An interior illustration of a novel appearing two pages before the start of the story was listed as the starting page. I don't see that as the initial page. The moderator in question claimed this rule we're discussing applies to novels as well as magazines. That interpretation does not make sense to me. In order to correct the situation it seems that either a) the moderator should be informed the rule only applies to magazines and the rule is tweaked to make "magazine only" more clear; b)the rule is clarified to say "immediately preceding," which makes sense to me as I can picture many magazines and anthologies with illustrations including a title to lead off a story (which we called a splash page when I worked in publishing); c) the rule be left alone, which as written means Easton Press novels start on page i or ii.MartinStever 19:03, 7 May 2018 (EDT)

page numbers section of Help:Screen:NewPub -- half-title pages

Currently, the help screen talking about how to find the page number of content titles doesn't mention what to do if there is a half-title page (with the story title on it) preceding a story or essay or a novel in an omnibus, etc. We should probably add a sentence like "If the title is on a separate page from the beginning of the text, use the page number of the title." And we should also make clear what to do in the uncommon case that there is a title at the top of the text and a half-title page preceding it. I do not know, myself, what the decision would be in that case -- although using the number of the half-title page makes sense. In that case, it would be "If the title is on a separate page from the beginning of the text, use the page number of the title page, even if the title is also printed at the top of the text." --Vasha 22:57, 24 April 2018 (EDT)

Editor vs author data for OMNIBUS/ANTHOLOGY?

As I thought it of general interest, I've moved discussion from Willem's talk page to here for broader discussion. There's actually two topics here, so may want/need to split up:

• What to choose: editor vs author data in editor/author field of an OMNIBUS or ANTHOLOGY record, and
• Ambiguity in field name depending on edit mode

Hello Willem. I suggest to remove the editors (Scott Card & Tanith Lee) from this publication since the actual editor is not known of this (two-story-long) anthology, and to add an explanatory note to clarify why. This way I feel it's more correct. Can you agree? Regards. MagicUnk 18:50, 26 April 2018 (EDT)

Sorry, but I don't agree. The helptext has no specific rule for this case, but I think the rule for a multi-author OMNIBUS that has no editor credit (and no secondary source for the editor credit) comes closest, so it should be credited to the authors. --Willem 16:40, 27 April 2018 (EDT)

I think an omnibus customarily requires/has an editor, and since there isn't one identified it is still better to enter ‘uncredited’ because the author(s) are not necessarily the editor, and if you do enter authors in the editor field it is in any case redundant information as author information is already available from the 'Regular Titles' section.

That said, when I took a closer look at the data I noticed that its pub type is actually ANTHOLOGY (which is correct as per Pub Type definition page as far as I can tell). But regardless, same argument holds for ANTHOLOGY I'd think: editor=uncredited, authors identified in the 'Regular Titles' section so as not to end up with redundant information, but actually ending up with more information instead .

And, while looking into the records some more I noticed something else that struck me as odd; when in editing mode the Tweesprook 2 Title Record says 'authors' while in editing mode, and displays 'editors' when not in editing mode. Same observation holds for Tweesprook 2 Publication Record : ‘editors’ when displayed and ‘authors’ in its Publication Metadata section when editing the record. There’s even more: 'editor' is used as field name when adding a new ANTHOLOGY record in its title section. Question for Ahasuerus perhaps?
Thoughts? MagicUnk 10:20, 29 April 2018 (EDT)

Trying to make a single rule for two vastly different types is not very useful here.

For omnibuses - the authors of the individual books collected into the omnibus are the authors of the omnibus. Don't forget that we use omnibus for all kinds of box sets and other multi-books and multi-works publications as well. But it is always about books that had been published before Leaving something uncredited or attributing to some "editor" does not make sense - add that to the notes if you want to but unless the editor is very prominent on the title page, the authors of the collected books are the authors of an omnibus (the corner case of an omnibus where you have 2 novels and 10 stories is a bit different but if they are from the same author, I would rather use the author name). We may have somewhat confusing UI in places for that but leaving an omnibus uncredited or crediting to a house editor is a bit against any logic of how this DB works.

Anthologies are quite different - there you always have an editor and if it is not credited, it needs to be set to uncredited.

So let's not try to discuss omnibuses and anthologies as if they are the same thing? Think of the omnibuses the same way you think of novels - these have editors as well but you are not advocating to make them the authors, right? Annie 17:03, 2 May 2018 (EDT)

I do think that fiction OMNIBUSes are very similar to anthologies if they collect works by different authors (just as they are similar to COLLECTIONs when all the significant works are by the same author), they just have NOVELs instead of SHORTFICTION, and it in my opinion it does look more odd to have authors who may be already dead in place of getting together to edit it (because edited they were, and we should mirror that fact). Stonecreek 23:39, 2 May 2018 (EDT)

Books like this have been in the database for ages, no questions asked. Is it strange that I treat this one the same way? --Willem 14:53, 3 May 2018 (EDT)

No, it ain't. But maybe we should stick to the good old rule as to index what is stated on the title page. There are some who only state the titles of the container titles, and others (like this), who credit an active editor other than the authors of the novels. Stonecreek 15:28, 3 May 2018 (EDT)

That's what the rules say. If an editor had been credited for Tweesprook, his/her name would be in the author field. But there isn't. --Willem 15:31, 3 May 2018 (EDT)

That's exactly the point. The authors are in the editor field of the anthology example. Card and Lee haven't edited this anthology, so the current entry is 'wrong'. Sticking to anthology for a moment, I think you can agree having an editor makes sense so has to go in the editor field-uncredited if not known, and the authors mentioned with their respective contributions.

Rereading what I wrote earlier, I agree with Annie that omnibuses can be/are to be treated separately. And for omnibusers there's no editor field, but an author field instead (leaving aside the strange observation on field names when editing or creating a new record I mentioned earlier, which needs to be fixed eventually) . So in short editor for anthologies, authors for omnibuses, as that's what the field labels say too. MagicUnk

I think this is a side effect of the fact that these doubles are anthologies by name only (aka - these two are "short novels" and if they were longer they would have been omnibuses. The one that started this whole discussion (I just looked at it again) is exactly like this in my mind and I would much rather keep the two authors up in the field than go for uncredited (if there was one that was credited on the title page, that would be different). There is a big difference between an anthology like Dozois's Year's Best and this kind of doubles (which are anthologies in name only). Just thinking aloud. :) I am leaning towards leaving the authors as editors in this case if no other editor is mentioned anywhere... Annie 16:42, 3 May 2018 (EDT)

But why would we want to put authors in an editor field if we have author information in the contents section already? (apart from the anthology-in-name-only, that is :-). As I see it, using the editor field strictly for editor(s) when there's one, conveys more information than what you would get by repeating the authors MagicUnk 11:27, 4 May 2018 (EDT)

For proper anthologies? We would not - we put the editor in for these, uncredited if none is specified. And in your case we were not even discussing an editor - it was going to uncredited. Now - what kind of information does this convey? That the publisher cannot be bothered to name their editor? :) These anthologies in name only seem to be the main problem here, right? So let's just do the usual - if an editor is prominent on the title page, use that; if only the two author names are there, use that. It matches both our practice and the rules... Annie 12:09, 4 May 2018 (EDT)

Yes exactly: uncredited = publisher couldn't be bothered :-) Still not convinced it's the right thing to do, but as it looks like there's no consensus to the proposed change, I'll go with the current practice (btw, I couldn't find the relevant rule you're referring to, can you get me a pointer?) MagicUnk 16:48, 4 May 2018 (EDT)

That's what the notes are for :) The rule that we go with the title page? :) That's the one I am referring to (and the only rule we have around what names and titles to use) - sorry if it was not very clear what I meant. I can understand where you are coming from - it is just that our anthology type is a bit of a catch all which makes it a bit weird sometimes. Annie 16:53, 4 May 2018 (EDT)

A novell devided into multiple titles.

What is the procedure with this? I have Frank Herbert's Dune divided into three books. Debolestis 15:58, 2 May 2018 (EDT)

Add the books one by one as regular novels and then variant all 3 into the full novel. The French novels suffer from that a lot :) Annie 16:05, 2 May 2018 (EDT)

PS: Here is an example: The Snow Queen. See how the Portuguese titles are added? Annie 16:07, 2 May 2018 (EDT)

Does this relate to this discussion? ../Doug H 16:25, 2 May 2018 (EDT)

No - the one you linked is about how to date the original when it was published in pieces. This question is asking how to enter a translation that had been published in parts - the original already has a date and each of the parts here will get its own date (as it will be a separate title). So not related at all. Annie 16:52, 2 May 2018 (EDT)

OK, I now have these 3 titles, http://www.isfdb.org/cgi-bin/pl.cgi?663964, http://www.isfdb.org/cgi-bin/pl.cgi?663965, http://www.isfdb.org/cgi-bin/pl.cgi?663966. Do I add them all as variants of Dune? Do I need to merge them later? Debolestis 14:30, 3 May 2018 (EDT)

Yes, you make them variants of Dune (I actually would change the Title record to contain information that it is volume 1 of 3 for example (not the publication title but the title itself)). No, you do not merge them - they stay separate - the same way the 2 Portuguese ones in my example are. Annie 16:45, 3 May 2018 (EDT)

Author or Authors

On the title page appears "Author" on the publication page "Authors". Better is "Author[s]" for both pages.--Wolfram.winkler 05:44, 7 May 2018 (EDT)

No comment?--Wolfram.winkler 15:17, 18 May 2018 (EDT)

It is in the wrong forum - so people kinda missed it I suspect. Yes, that can be standardized via the usual way - a FR and then implementation :) Annie 15:29, 18 May 2018 (EDT)

Requests for changes to the user interface should be posted to the Community Portal. I am moving this there now. --Vasha 16:58, 18 May 2018 (EDT)

My mistake, but again a new administrator?--Wolfram.winkler 11:02, 20 May 2018 (EDT)

I assume you are referring to your earlier comment about our decision-making process:

• ... we need an administrator, who says: yes, we can do it. If no one makes a decision, the discussion will never ends.

At the time I wrote that:

• Most of our decisions are made after reaching consensus. The downside of this approach to decision-making is that there are times when we can't reach consensus, so things remain hanging for a long time. The upside is that we are less likely to lock ourselves into a solution which may prove unworkable or counterproductive in the long run. I find our current approach generally superior to the alternatives which we tried in the past.

Would you suggest abandoning the consensus approach and appointing a single person who would be responsible for making decisions about our data entry policies, project scope, software design, etc? Ahasuerus 11:57, 20 May 2018 (EDT)

I prefer a consensus approach, if it's the consensus of the majority. A few times in the past it only took one, angry, shouting person to derail the process.--Rkihara 15:58, 20 May 2018 (EDT)

We usually define "consensus" as "significant majority". 4-3 wouldn't be considered consensus, but 8-2 would. In my experience, a slim majority often indicates that the proposed solution may not be ideal and that a better solution may be possible. Ahasuerus 16:16, 24 May 2018 (EDT)

I meant that someday somebody should make a decision if there is no democratic majority. And this can only be an administrator. Wolfram.winkler 07:48, 29 June 2018 (EDT)

Here is how the process currently works:

• An issue re: data entry rules or project scope is posted on the Rules and Standards page
• Proposals are formulated and debated
• If a clear consensus emerges:
  • Someone (either a bureaucrat or a moderator) posts a note announcing that he or she is going to change Help if there are no further objections

• After a day or two Help is changed as per the outcome of the discussion

• If no consensus emerges:
  • The issue is dropped for the time being and Help is not changed
  • Another round of the same discussion may be started at a later point if new proposals emerge, we accumulate more experience with the issue or the bibliographic world changes
  • Sometimes it takes multiple iterations to come up with a consensual solution

Ahasuerus 14:42, 29 June 2018 (EDT)

In my post above it must mean: ...new moderator... and that related to Vasha. The moving of a post can only make a moderator. I didn't know that Vasha is a new moderator.--Wolfram.winkler 07:48, 29 June 2018 (EDT)

Ah, I see. Vasha is not a moderator, but she didn't have to be in order to move a feature request to the right Web page. Annie is a moderator and she made the same point at the time -- see above. The proposed feature was disscused and then implemented on 2018-06-02. Ahasuerus 14:42, 29 June 2018 (EDT)

Directory Entry for an American Von

This question was raised by Dirk P. Broer with regard to Eddy Von Mueller. This author is American and his last name, legally in the US, is "Von Mueller." Dirk argued that his directory entry should be "Mueller" and wrote, "The directory entry is the place where you look for the author in a library or store. I go to the ’M’ shelf for von Mueller." But I'm not sure about that. Shouldn't the directory entry be the actual last name?

Note that the same problem arises for "De" names. The "De" is included in the last name in some countries, e.g. US & Italy, and not others, e.g. France. --Vasha 06:34, 11 May 2018 (EDT)

I believe we had this discussion a couple of years ago. The US standard is to include the prefix, i.e. "Von Mueller" rather than "Mueller", in indices and directories. Other countries' standards vary, sometimes even within one country. For example, here is what the International Federation of Library Associations and Institutions says in Names of Persons: National Usages for Entry in Catalogs about "van":

• If the person is Dutch, "van Beukering" should be sorted under B
• If he or she is Belgian, sort it under V (but note the small print that says Belgian libraries aren't consistent across the country)
• If they're from the US, sort it under V

Ahasuerus 09:29, 11 May 2018 (EDT)

* If the person is Dutch, "van Beukering" should be sorted under B.

That is right, and the directory entry ("Beukering") takes care of that. The family name is "van Beukering" and the legal name field takes care of that.--Dirk P Broer 19:08, 11 May 2018 (EDT)

Odd one (Bruce Sterling)

Sterling has a story that was likely written in English first, then first published in Japanese, and latter published in English and other languages. Right now, the English title is varianted to the Japanese title, which creates a mess with the author pages. Thoughts on how to list this? ···日本穣 · 投稿 · Talk to Nihonjoe 20:02, 25 May 2018 (EDT)

I think it's covered by the following clause in Help:How to enter translations:

• If a work was written in one language, but a foreign language translation was published first, then the original language title should be considered the canonical title and the translated title should be considered variant title. The year of the canonical (i.e. parent) title should be set to publication year of the canonical title, not to the year of the translation (though the latter one was released earlier).

Ahasuerus 20:13, 25 May 2018 (EDT)

I just fixed this one (showed up on a report) as per our standard - the original is the parent even if it is published later (or never). Annie 01:54, 26 May 2018 (EDT)

Awesome, thanks. I'd never run into this situation before. (^_^; ···日本穣 · 投稿 · Talk to Nihonjoe 10:31, 26 May 2018 (EDT)

Record Numbers from Secondary Sources redux

There was a discussion a few years ago where I noted that I have been adding catalog numbers from secondary sources (e.g. Reginald, the Bleiler indexes) in the notes section of publication records (a recent example). At the time there was one editor who objected to my doing this. A compromise was suggested that such numbers be put below a {{BREAK}} tag, which I did adopt (once I learned what the tag was and how it is used) and a FR was created to add such numbers much as to a dedicated field as we now do with External IDs (perhaps it was the same FR). Regardless, there was no consensus at the time, or I believe since, that the addition of such numbers to the notes was prohibited. Today, I discovered that Reginald and Bleiler Early Years numbers that I added to this publication have been deleted. If I am mistaken, and this data is prohibited, I'll stop doing this. If it is not, what is our policy on the deletion of allowed data that an editor does not like? If editors should not be deleting allowed data, what is the recourse when they do so contrary to policy and etiquette. I am aware of this instance, but I wonder whether other data has been deleted without establishing a policy first. --Ron ~ Rtrace^Talk 22:03, 3 June 2018 (EDT)

While I was moving OCLC, LCCN and other external numbers out from notes, I made a point of keeping these where they were and cleaning the br's and what's not around them (and even pinging the verifier if one of the numbers was missing while it looked like it was supposed to be there). I think that they are useful information and if someone is deleting them, then we are losing information. Just my 2 cent.

Plus if one day we DO have a new field, finding the data again is much harder than just moving it. Annie 22:21, 3 June 2018 (EDT)

The Bleiler indexes do not have catalog #s, all that's being added is a page #, more than useless. We already have verification spots for all the Bleilers and Reginalds, none of which have any online source nor are they likely to. Adding Reginald's numbers or a page # adds nothing, what's next, adding the line from Locus? The ONLY verification source which also has an external ID spot is OCLC, as they can have multiple records even for a single edition [some worthwhile, many not] and that was the only reason for putting their record numbers in the notes in the first place. I thought the purpose for having the external ID slots was to remove links/numbers/clutter from the notes. If a particular datum has been gleaned from one or more of these external sources, by all means that should remain in the notes, but the number/link should still be removed [what's the point of having it there twice?]. The original discussion noted above had four participants, three of whom thought these 'identifiers' to be nothing but clutter, the fourth that any secondary sources shouldn't be present if the record is primary verified [like that's likely to be implemented ...]. They're still just clutter. --~ Bill, Bluesman 12:55, 13 June 2018 (EDT)

the suffix Jr.

Do we standardize the suffix, no matter how it is written in the book (if it appears as "John Jones jr." do we enter "John Jones, Jr.")? --Vasha 13:47, 5 June 2018 (EDT)

Yes. See "Ranks, suffixes, prefixes" in Template:PublicationFields:Author (last bullet). -- JLaTondre (talk) 16:44, 5 June 2018 (EDT)

[Copied from here --Vasha] Yes, but I understand that this is an English / American regulation that doesn't necessarily apply for other language authors (though this is not made clear in the help). So it seems that James Tiptree jr. (as published with many titles) is regularized because we want to avoid unnecessary pseudonyms, but other authors should not be, especially when there's no English involved. Christian Stonecreek 03:05, 6 June 2018 (EDT)

Are there examples of non-English names that use "Jr.", etc.? ···日本穣 · 投稿 · Talk to Nihonjoe 12:44, 6 June 2018 (EDT)

a few :) Annie 12:47, 6 June 2018 (EDT)

Keep in mind that we use the same rule for other suffixes like "III" even though it's not a standard English practice.

If I recall correctly (I am not 100% sure since it's been about 10 years), the original reason for this blanket rule was the way our software behaves when parsing new author names: it considers anything after the last comma to be a "suffix" and populates the Directory Entry field accordingly.

In retrospect, it may not have been a very good reason. We shouldn't be creating data entry standards based on the way the software behaves -- it should be the other way around. Ahasuerus 13:11, 6 June 2018 (EDT)

Novels published ONLY as serials

As per the date rules for serials, a novel takes its date from its first publishing in book form. What do we do when the book never made it to a book form? 0000-00-00 or 8888-00-00? I can see it both ways but wanted to ask (and we need this rule updated for that). Thanks! Annie 17:48, 11 June 2018 (EDT)

I think the majority of our "phantom" NOVEL titles for unreprinted SERIALs use the date of the first SERIAL publication. (Parenthetically, I still think that there is got to be a better way to handle the SERIAL/NOVEL date mess in the software, but it's not a high priority at the moment.) Ahasuerus 18:27, 11 June 2018 (EDT)

So should we add a note to that effect in the help page? Annie 18:30, 11 June 2018 (EDT)

It may be prudent to ask some of our "pulp" editors who have been active lately to be sure. I haven't entered any pulps in a long time. Ahasuerus 18:44, 11 June 2018 (EDT)

Not a bad idea. I was moderating around this one today and it came up - I went for 8888-00-00 initially (Vasha went for first serial date which the help page explicitly says not to do), then I decided changed it to 0000-00-00 and then decided to come and ask because neither of the 3 looks correct. I had never edited or moderated such a case before - so I was reading the help page and working based on that. :) If noone shows up in a day or so, I will ping some directly on their pages. Annie 18:49, 11 June 2018 (EDT)

I tend to give these cases the date of the end of the serial, but with only year precision. Thus in the example you cite, I would have dated it "1934-00-00". When the serial spans years, I go with the later one. I'm not wedded to this approach. Of the two that you mention I would think that 8888-00-00 makes more sense insofar as the book publication was never published, rather than the date of that publication not being known. --Ron ~ Rtrace^Talk 19:52, 11 June 2018 (EDT)

I like this approach even if it goes against the help pages (plus that won't send the novel at the bottom of the author's list). Probably we should add a line in the help page about that specific case considering that we already have a big note about serialization of novels and dating around them. Annie 14:07, 12 June 2018 (EDT)

If there are no objections, I will change Help later this week. Ahasuerus 10:03, 21 June 2018 (EDT)

(unindent) Is it still appropriate that the mouseover text for 8888-00-00 is "announced but never published"? By now, it is being used not just for that case, but quite often to record the original title of something that's only been published in translation. And here we have yet a third kind of unpublished title. --Vasha 09:23, 12 June 2018 (EDT)

It would be easy to change the mouseover text, but we would want to agree on the data entry rules and update Help first in order to keep everything in sync. Ahasuerus 13:59, 12 June 2018 (EDT)

Outcome

I have made a number of changes to Help:Use of the SERIAL type‎, Help:How to connect serials to titles‎ and Template:TitleFields:TitleType to reflect the outcome of this discussion and to eliminate all vestiges of the infamous "lexical match" logic. Back when we changed the software, I reviewed our Help pages and deleted all explicit references to "lexical match", but apparently I missed a number of indirect or implied references. Without the original context, they muddled the waters and made the affected Help pages hard to follow. Sorry about that!

Hopefully the updated Help makes sense. Ahasuerus 17:13, 23 June 2018 (EDT)

Alternate (as opposed to sub) titles.

The current rule explains how to enter a title that has a subtitle. Would the same rule apply to alternate titles? An example is "Five Weeks in a Balloon; or, Journeys and Discoveries in Africa by Three Englishmen." (in 4 variations of font style and size).

If so, would the "; or," be replaced by a ":"? Could the consensus be included in the help entry? ../ Doug H 15:08, 18 June 2018 (EDT)

We actually have quite a lot of these (they were kinda common in the past) and the current practice is to threat the whole thing as a title - not as a title/subtitle. That's a search for Title contains "; or,". Annie 15:22, 18 June 2018 (EDT)

What's the thinking on how to punctuate it if the title page isn't clear on that? Like, if it's written one above the other (Five Weeks in a Balloon / or / Journeys and Discoveries in Africa). I know the old way was " ... in a Balloon; or, Journeys ..." and I know that has fallen out of fashion. Nowadays most people use either "Balloon, or, Journeys" or "Balloon, or Journeys." Do we have a preferred choice here? --Vasha (cazadora de tildes) 16:04, 18 June 2018 (EDT)

I don't think there is a standard; I have seen it done all kinds of ways. Ahasuerus 10:02, 21 June 2018 (EDT)

Online publications that are not downloadable

A recent submission which I currently have on hold (the link is moderator-only) has raised the issue of non-downloadable online works yet again. At this time, ISFDB:Policy#Rules of Acquisition are as follows:

• downloadable e-zines are in
• web-based publications (webzines) available only as an HTML readable file are out except when:
  • published by a market which makes the author eligible for SFWA membership
  • shortlisted for a major award

Under the current rules, the submitted anthology is apparently ineligible even though its contributors include Gregory Benford, Brenda Cooper, Bruce Sterling, Nancy Kress, James Morrow, Margaret Atwood, Karl Schroeder, Paolo Bacigalupi, Mike Resnick, Kevin J. Anderson, Catherine Asaro, and Peter Watts, to name only some of the better-known authors. They probably have a few hundred awards and nominations between them.

We have discussed this topic a number of times over the last few years. There seems to be general agreement that the policy is outdated because:

• it's too US-centric ("eligible for SFWA membership")
• the SFWA membership criteria change from year to year
• more and more major SF works appear online as seen in this case

At the same time there has also been reluctance to start accepting all online publications. As far as I can tell, the reasons are as follows:

• there are literally tens of thousands of stories getting self-published online every year; there are well over 100,000 Twilight fanfics on FanFiction.net alone
• HTML-based online publications can be unstable as authors can tweak them whenever they want to (hence the "downloadable" rule that we currently have in place); they can also be taken down abruptly for copyright and other reasons, leaving no trace except, perhaps, at the Internet Archive
• many online works appear as Web serials with dozens and hundreds of installments, which our software is poorly equipped to handle -- see the ever-popular Mother of Learning, Worm, Everybody Loves Large Chests, etc
• the serialization problem is further complicated by the recent proliferation of translated serializations -- see Novel Updates, Wuxia World, etc. Different translators and groups of translators may be translating different parts of certain serials, sometimes with a considerable amount of overlap. (Some are machine translations with minimal human editing involved.)

As far as I can tell, there are two ways we could approach this issue.

The first approach would be to try and further relax out policy to make it less US-centric and more accepting of online works published by notable authors while trying to keep fanfiction and Web serials out. If we decide to do it, then I think "stability" may be one of the more viable (if admittedly subjective) criteria. Author Web sites, publisher Web sites and projects like the one that prompted this discussion are generally fairly stable.

The second approach would be to throw in the towel and accept -- in theory -- all online publications. It may not be as scary as it sounds. Back when we were considering relaxing our e-book inclusion criteria, we were concerned about a potential flood of ASIN-only ebooks published on Amazon. As it turned out, it hasn't been a significant problem. Just because some works are eligible for inclusion doesn't necessarily mean that there are dozens of people dying to submit them.

So, what do you think? Ahasuerus 23:29, 26 June 2018 (EDT)

Discussion

The current rules also give smaller languages a lot more leeway - there is no webzine in Bulgarian that publishes fiction that had not been nominated at least once for one of the big awards for example... Even without relaxing the rules, they would be eligible. Which is not true for the major languages. On the other hand as much as I love Bewildering Stories, I am not sure that we should be indexing it. Or maybe if someone cares enough, we should...

One of the major issues in our rules is the "downloadable" part - it is trivial to create a file for an ebook reader with a browser plugin these days and most of the webzines would rather leave that to the reader and not force the full content (including good old 'Strange Horizons' which makes it in the DB because of its awards). It made sense 10 years ago I suspect, these days - not so much. Annie 00:02, 27 June 2018 (EDT)

Yet another reason why it'd be good to have broader criteria for inclusion is the magazines that used to have ebook-editions and then went web-only. It just adds a degree of complexity to have to explain why only part of a magazine's run is in the DB. And then there's the time you enter a webzine issue because the editors say "we'll be releasing the ebook any day now" and you look back years later and find that either the ebook never appeared or it's hard to tell whether it did or not (e.g. there's a download link which is dead--was it ever live?)... Yeah, that would be solved by only cataloguing with the ebook in hand, but that isn't always possible. My point is that allowing more webzines would simplify some mess. --Vasha (cazadora de tildes) 16:11, 27 June 2018 (EDT)

Fan Fiction and Web Serials

I'm in favor of a middle ground - FanFiction.net and similar username based publications are out; Annie 00:02, 27 June 2018 (EDT)

An interesting idea. "Username-based sites and forums" is a good fit for what I have been looking for. It would automatically exclude a great deal of fan fiction and Web serials. We'll just need to define "username-based" carefully. For example, the author of "Mother of Learning" is listed as "nobody103", but his user page also links to his Patreon account, which is set up under "Domagoj Kurmaic", his legal name. Ahasuerus 14:11, 27 June 2018 (EDT)

If someone links to their real author page, they are allowed in - using pseudonyms is not new in our genre (one of those fan fictions even changes the names of the characters and is pretty popular these days). :) And then common sense will apply. If Stephen King decides to post some fan-fiction over on a fan fiction site tomorrow (I know, implausible but stay with me) and links his username page to his real author page, we will catalog it. But that won't make the whole site eligible. So for these, it will be down to "check carefully. Annie 14:29, 27 June 2018 (EDT)

The more I think about it, the thornier the issue looks. The big problem that I think we have with Web serials is not (so much) the number of potentially eligible works out there, but their format. For example, the author of Worm and other Web serials, J. C. McCrae, makes a living writing Web serials, which is more than many traditionally published SF novelists can say. There are hundreds (!) of Worm fanfics, some of them quite long and quite popular. He has had multiple TV offers, etc. I would be all for adding these serials to the database if not for the fact that they were published in -- quite literally -- hundreds of installments. Cataloging the output of just a few of the more successful Web serial authors would quickly overwhelm us if we were to use our usual standards.

After typing the paragraph immediately above, it occurs to me that perhaps the way to address this issue is not to exclude Web serials outright but to change our data entry rules a bit. If we were to allow the inclusion of Web serials as standalone "container" publications -- perhaps with a new publication type, "web serial"? -- it would alleviate most of the issues that I have with them. They may require somewhat more extensive Notes than other publication types, but nothing outrageous.

Of course, if this change in policy resulted in tens of thousands of Web serial records getting submitted, we would still have a bandwidth problems. However, our experience with allowing e-books without ISBNs suggests that there is a big difference between what is potentially eligible and what actually gets submitted. We were concerned that we would be overwhelmed by thousands of ISBN-less e-books, but the reality has been different. Ahasuerus 16:15, 27 June 2018 (EDT)

We do not need to solve all the issues in the current policy in one single change - we can always leave the web-serials out (for now) so we can see how relaxing the policy for webzines and what's not work and then tweak it again. Or we can add a new type (which sounds better and better the more I think about it). I just really do not want us to get boggled into this kind of issues - and never move anywhere (I know that a comprehensive change is always better but... sometimes divide and conquer is a much better strategy). Plus who knows what will be the new fad tomorrow or in a year anyway - digital publishing is changing daily. Annie 16:25, 27 June 2018 (EDT)

I agree that we don't need to address all forms of Web-published fiction at the same time. Web serials present unique challenges and may need to simmer a bit longer before we can decide how to handle them. If we were to create a new publication type for them, it would alleviate some issues but it may also raise others. For example, what should we use as the publication date of Mother of Learning, which was started in 2011 and is still in progress? Should we create a new title type to handle monstrosities like Worm, which is over 1.6 million words long? How do we handle Web serials which have been partially published in book form (Everybody Loves Large Chests)? Etc.

That's why I created a separate subsection for fan fiction and Web serials -- hopefully this preliminary discussion won't prevent us from reaching consensus re: webzines and possibly other forms of Web-published fiction. Ahasuerus 18:31, 27 June 2018 (EDT)

I am actually going to argue that we need to call FanFiction.net by name Annie 14:29, 27 June 2018 (EDT)

Keep in mind that FanFiction.net is only one of a number of sites hosting fan fiction, crossover fiction and related works. One of its cousins, Archive of Our Own, hosts 3,942,000 [sic] titles. Ahasuerus 15:52, 27 June 2018 (EDT)

True - and yes, "FanFiction.net, Archive of Our Own and other predominantly fan-fiction sites" is probably a better wording. If an author cannot tell a Annie 16:05, 27 June 2018 (EDT)

and just exclude it - it's fun and so on but... do we want to deal with all that when an author gets in a fit and decides that all the fan fiction using her characters should be removed? Annie 14:29, 27 June 2018 (EDT)

Well, we already have to deal with withdrawn/replaced e-books, so Web-published fiction wouldn't be qualitatively different. However, the volume of changes may be a concern. I have seen a number of Royal Road authors take down their serialized novels prior to e-book publication.

Ultimately, it shouldn't matter much as long as we document our sources and the time frame carefully. Changing/removing links to third party sites may be time-consuming, though. Ahasuerus 15:49, 27 June 2018 (EDT)

This is why I am not keen on adding author sites (and even publisher ones) as eligible - it is a lot less likely for someone publishing a webzine to take down a story than for a publisher/author to do it because they are about to publish it in an e-book/vanity press. Middle ground and common sense should be foundable. Annie 16:05, 27 June 2018 (EDT)

Different Types of Sites

... publisher and personal author sites are out unless the author is already in the DB (this will keep the online only poets and authors out until they actually publish something in a place that is eligible - although most of them probably are going the kindle route anyway) - I know that they are stable but... do we want the index of every wanna-be writer and poet in the world? Webzines, special projects like the one that started this and so on are eligible provided that for a webzine, it had published 3 issues or at least 10 poems/stories in the first year of its existence (numbers are open to debate) and it had published more than one author. That should keep some of the one-day webzines out... Annie 00:02, 27 June 2018 (EDT)

Yes, Annie's point of view is the one I also subscribe to. Christian

Me three. There are some venerable, major webzines (Kaleidotrope comes to mind, but there are plenty of others). It would sure be nice to have a systematic record of their contents integrated with our other records. And the well-run ones are a lot less ephemeral than some ebooks (try tracking down info about the contents of an anthology that was on sale for a month four years ago and you're not sure anyone even bought a copy!)

So, as I see it, there are three classes of websites that need to be discussed: webzines; sites that publish fiction by multiple authors but don't call themselves magazines; and single-author websites.

I don't really have an opinion on whether or not to allow the single-author sites, but I'm skeptical about them. [snip some things I changed my mind about]

Webzines & publishing sites, though, that should be doable. Annie's criteria are good. Also think about specifying that they must publish predominantly genre. But currently, when including pro or award-winning web publications, we draw a distinction between those that have regular issues & are indexed completely (Strange Horizons) and those that are blog-like or general-purpose websites and only have fiction indexed (Tor.com). Can we systematize that distinction to make it easy to classify new websites one way or the other? --Vasha (cazadora de tildes) 01:04, 27 June 2018 (EDT)

I would disagree on the "predominantly genre" a bit. Take for an example Words Without Borders which is non-genre for the most part. However they have Special themed issues that fit us better than some of our genre sites (it is not eligible now because it is not downloadable...) and they also have a speculative story or 2 in pretty much any country-based issue (and the magazine usually translates authors that are popular in their own countries but unknown or less known in the English speaking world). Then there are the web-versions of the standard journals and reviews (Kenyon Review for example had expanded their site in the last few years from being just a supporting platform, posting a story or 2 from the printed versions into a full blown online magazine to complement the printed one). None of those would be called genre publications but restricting them when we allow stories from printed non-genre magazines makes little sense. And while early in the webzines development they were almost segregated by genre, I see more and more of them using more of a "let me include different genres so I get more readers" kind of approach. So I would say that we should treat a non-genre webzine/web-magazine the same way we treat a printed one.

PS: My 3 issues or 10 stories above was exactly because of the two types of sites we have out there (the structured webzines vs the blog/non-issues specific ones) :) Annie 13:49, 27 June 2018 (EDT)

It doesn't entirely make sense to me to omit non-fiction just because the contents aren't organized into issues. Some of the articles on Tor.com have won awards, for example. And there's a degree of arbitrariness to the distinction: Fireside Magazine went from posting issues monthly, to a continuous stream of posts, and then quickly started organizing their content into monthly issues again but having the contents appear bit-by-bit over time, as many webzines do. I actually did leave out the nonfiction for those few in-between months but that feels wrong. Consider that if we want to only index fiction from sites that don't have "issues," then a site could post the exact same things and we'd treat it differently if they said "new content will appear on the 1st, 10th, 18th, and 24th of March" or if they said "the contents of the March issue will appear on the 1st, 10th, 18th, and 24th." --Vasha (cazadora de tildes) 14:39, 27 June 2018 (EDT)

I was just talking about fiction above :) As for non-fiction - indexing the whole tor.com (or locus's online site) does not make much sense either - although indexing some of their reviews may be a good thing, that is a bit outside of our main focus. My view is that we are predominantly a fiction DB and the non-fiction we index is mostly ancillary (aka - it just happened to be in the same place where the fiction was so we started cataloging it). Quite frankly, I would not have indexed the reviews in "Strange Horizons" either but that ship had sailed. But it is a very thin line between "we allow some of it in" and "we get overwhelmed by reviews and essays". And then some awards start having categories for "short non-fiction" that allows blog posts and the like so we may be stuck indexing these. On the other hand (again), we are a fiction DB, not a "webzines/magazines/blogs" db.

One possible option is whitelisting of webzines and sites - we keep a list of "fully indexed ones" (led by Strange Horizons), if someone wants a new one to be added, we discuss and add it eventually (and maybe have some automatic rules such as "nominated in a major award as a whole"); if one changes, we can put an "up to date" date on it. If it is not on the list, only fiction is eligible for inclusion. That will remove some of the differences. Annie 15:42, 27 June 2018 (EDT)

I mentioned something similar before in another discussion: I like the idea of a publication's being in or out. We could apply that thought to webzines/blogs/whatever. If we were to take Annie's idea that a web-only work is "in" if written by an author having other, non-web-only "in" works recorded in the database and extend it a bit to then let us categorize the site -- i.e., it's a "genre" site and everything is "in" because some of the things are "in", or it's a "non-genre" site and only the "in" works are "in" -- that might be a workable compromise that would also be fairly easy to remember and to apply. I do believe we need some way to be more permissive with regard to this medium. --MartyD 19:32, 27 June 2018 (EDT)

(unindent) I still think that we should make a difference between web-only in webzines/online magazines (which we should be treating in a way similar to printed ones and not as the red-headed step children we really do not want to hear from) and web-only somewhere on someone's site, in a blog and so on. Annie 19:44, 27 June 2018 (EDT)

I think the most relevant distinction is not between whether a site calls itself a magazine or not, but whether it publishes multiple authors, or is one person's work (even if with occasional guest appearances by others). This is a pretty easy distinction to make, whereas sites that publish multiple authors' fiction come in many forms and fall on a continuum that makes it hard to say what is really a magazine. --Vasha (cazadora de tildes) 20:14, 27 June 2018 (EDT)

That's down to how we define a webzine. As for the single author - untangling pseudonyms may turn a 10-authors magazine into a 1 author one overnight - would you then delete it because it is now not a real webzine? And if one of big authors of the genre decides to start publishing a webzine with their new stories, it will be excluded if we just look at how many authors participate in it. Annie 20:55, 27 June 2018 (EDT)

OK then, how would you define "webzine"? I might try "A site intended for a public audience, which adds new content of fiction or feature articles somewhat regularly;" there's a lot of undefined terms in that though. --Vasha (cazadora de tildes) 23:18, 27 June 2018 (EDT)

Your description will cover an author's blog (for an author that publishes fiction now and then) :) . And regularity is not really needed for a magazine (look at some of the schedules of the smaller printed ones). Webzines, ezines and so on are still "-zines" - which had always implied issues for me - just posting stories or articles online makes you a blog (and yeah, I am looking at tor.com) - Daily Science Fiction, Escape Pod or Strange Horizons are online magazines. Annie 00:00, 28 June 2018 (EDT)

I meant "intended for a public audience" to exclude blogs; too much informality and personal chitchat makes them the equivalent of some more personal fanzines, not magazines. And Tor.com seems blog-like because it has so many Tor publishing announcements, talk about their authors, words from editors, as well as "filler" posts provided by Tor.com staff that aren't substantial articles. As for the updating part, a site which posts a whole bunch of stories and then never updates again isn't a magazine, even if we might perhaps want to index it. How could I express this better?

BTW we don't draw a distinction between all-fiction webzines that have "issues" and ones that don't (like Daily Science Fiction). Is DSF a blog that just happens to post lots of fiction? --Vasha (cazadora de tildes) 00:29, 28 June 2018 (EDT)

How we deal with web publications now and how we want to deal with them is the whole point of this discussion, isn't it? If it is online, it is intended for the public audience - otherwise you keep it at a place where only your friends can read it. DSF posts a monthly schedule - so it is a magazine in my book - in a funny way the way we are recording it here is how I see their "issues". Especially because this is all they publish. :) Why would a site that posts 20 stories and then stops be ineligible - that would exclude all the special publications like the one that started this whole discussion - and which should obviously be eligible? Printed magazines with a single issue or 2 are eligible. Plus, DSF is a special case - together with sites like 365 tomorrows for example (which I am not planning on indexing - just pointing out another daily story site) - it just happens to be paying (better). I am not sure why we need to define webzine, site and what's not - what matters is the fiction, right? We are not discussing just making webzines eligible - if a site does not fall under that category, it can fall under another eligible one. Annie 00:40, 28 June 2018 (EDT)

Right, taking a step back here, the main issue is finding what online fiction we want to index. Until we find a site that publishes fanfiction in monthly issues, we should (I hope) be able to define exclusions narrowly enough that we can assume everything else is OK; it'd probly be better to try defining what is out rather than what is in. Talking about what else we want to do with the sites we include (like maybe indexing their full contents) can wait until we've settled what to include. We can keep the ones that are full-contents now under a "grandfather clause," especially since it's a very short list: just Strange Horizons, Fireside, and Abyss & Apex at present, and maybe a few more in the past like the recently-defunct Sci Phi Journal after it stopped being a print zine. --Vasha (cazadora de tildes) 01:13, 28 June 2018 (EDT)

Proposed Subcategories

After sleeping on it, I came up with some subcategories:

• The Genre web publications
  • 1. The webzines - anything that publishes issues weekly/monthly/irregularly - things like Strange Horizons, Bewildering Stories, Kaleidotrope, all the zines that stopped publishing paper versions (and sometimes even e-books) but kept the creation of the issues - the stuff that would have been on paper once upon a time and already has at least 3 issues.
  • 2. The sites and blogs that publish a lot of fiction from different authors: Daily Science Fiction can go in either this or the previous category, tor.com is here (minimum of 10 stories and predominantly publishing genre fiction for a site to be eligible)
  • 3. Special publications - like the one that started this round of conversations - anthologies and one-time sites that have SF, are not likely to get updated but obviously belong here. Maybe this needs to be merged down with 1a?
  • 4. Personal author sites
  • 5. Publisher sites
  • 6. Random sites (aka - I am not an author but I wrote a story)
  • 7. Random stories in random weird non-fiction related places (the town of Nowhere asked a genre writer to create a story for them and then published it on their site.
  • 8. Fan fiction sites
• The non-genre ones
  • 1a. Special speculative fiction issues of non-fiction webzines (like the one from Words Without Borders I linked somewhere above)
  • 1b. Single stories inside of issues non-fiction webzines (that have issues)
  • 1c. Single stories in sites that publish stories but are not exclusively genre (Everyday Fiction comes to mind).

Based on that:

• Eligible: 1, 2, 3, 1a (we would catalog a speculative fiction issue of a printed journal)
• Conditionally eligible: 4, 5, 7 (when we already have the author in the DB)
• Never eligible: 6, 8
• Not sure: 1b and 1c. I want to say "yes" but then we will get into "what is speculative fiction" so to be on the safe site, I;m more inclined to go for the conditionally eligible for these as well...

A few more additional thoughts:

• When cataloging a webzine, adding a link to the issue AND a link to archive.org that shows the issues should be strongly recommended. That will give us both proper dating (the archive links contain a timestamp) and even if the site dies, we catalog. Links for each story/poem in their title records may be annoying and/or an overkill but will protect us from having too much ephemeral content. The smaller (and newer) the zine is, the more we want this.
• Same rules for all individual stories that are eligible under the rest of the categories.
• Do we want to discuss a new type (e-chapbook) for these kind of stories? Alternatively, we can standardize the Daily Science Fiction style convention and have monthly "webzine" entry for those? This way the webzines that publish issues are catalog with their issues numbers and so on; everything else goes under the "monthly" plans. The more I think about it, the more I like the monthly idea thingie...

So... any obvious gaps? Anything that sounds weird? Any other thoughts? Annie 18:37, 28 June 2018 (EDT)

Webzines and single issue anthologies

The good news is that this discussion uncovered a lot of permutations which we (or at least I) hadn't considered. The bad news is... well, the same as the good news -- there are a lot of permutations.

I wonder if we could take a piece (or maybe two pieces) of this puzzle and present it as a concise proposal, probably in a new section? Something like:

• Webzines, which are defined as online periodicals with distinct issues, are "in"

perhaps? We could add more qualifications if desired, although in my experience quantitative thresholds ("at least X issues", "at least Y authors", "at least Z stories", etc) do not work too well. We always seem to end up with too many exceptions and unexpected scenarios which take longer to sort out than to simply enter into the database. Ahasuerus 21:41, 1 July 2018 (EDT)

Sounds like a plan. And I agree - the number of issues will probably not going to help much - I would prefer to wait until a second issue appear but that's not really that important. I kinda want to be able to include the publication we started with (and the Words Without borders special issue) in that first pass (because a lot of the big names will show up in these project and not in the proper genre webzines). Let me think on how the two can be defined in a single line as with the webzines. Maybe:

• Special speculative fiction issues of non-genre webzines and one time web-anthologies Annie 22:07, 1 July 2018 (EDT)

If I am reading this correctly, your proposal would be as follows:

• Remove the words "and downloadable e-zines" from the "Included" section of ISFDB:Policy#Rules_of_Acquisition
• Add the following lines to the "Included" section:
  • Speculative fiction webzines, which are defined as online periodicals with distinct issues
  • Special speculative fiction issues of non-genre webzines
  • One time speculative fiction anthologies published on the Web

Is this about right? Ahasuerus 13:56, 2 July 2018 (EDT)

I do not think we should be removing the "downloadable" part of that specific rule - these are still eligible based on their e-book versions (the ability to be downloaded) even if they have no issues - e-zines and webzines are different animals so I would leave the e-zines rule alone and just add a webzine one(s). Annie 14:14, 2 July 2018 (EDT)

Oh, I see. Makes sense, although it may be beneficial to clarify what we mean by "downloadable e-zines". For example, does a downloadable single issue anthology count? Ahasuerus 14:33, 2 July 2018 (EDT)

Between the e-zines rule and the last line in the Includes section, we have the downloadable content pretty well covered I would think. The single issues anthology if downloadable will be eligible under "Internet-based publications which are downloadable as electronic files in any number of ebook formats (ePub, Mobi, PDF, etc)" even today... I was actually thinking that we should split the Included section into 3 sections: printed, downloadable e-books and web-only publications and work based on that. There will be some repetition but it will make it clearer. Or we can make two separate sections (printed and downloadable e-books)- which is what we have now and a new "web-only publications" where we start qualifying them. Annie 14:47, 2 July 2018 (EDT)

OK, let's leave the current "downloadable e-zines" language alone for now. As you said, one thing at a time. Ahasuerus 16:22, 2 July 2018 (EDT)

I would leave the Included section as is now, add the 3 lines above at the bottom and rework the first line of the Excluded section by adding a third line under 1:

• Webzines and online anthologies that are included as per the Included section.

Annie 14:14, 2 July 2018 (EDT)

Presumably we'll also want to change the sentence which starts with "Works published in a web-based publication (webzine) and available only as an HTML readable file are not eligible for inclusion". At the very least we will want to delete "(webzine)". Ahasuerus 14:36, 2 July 2018 (EDT)

Well, this sentence finishes with "with the following exceptions" and we are just adding more exceptions - so no need to change that sentence (yet). Although deleting the word webzines makes sense - I think it is a hold over from the time when these were the only sites that actually published fiction. Annie 14:47, 2 July 2018 (EDT)

We could have a third category besides included and excluded, and sites w/out issues and Non-genre webzines would belong to it: "Exercise caution: only catalog a story from these sites if there is a clear justification for including it in the database (by a genre author, reprinted in a genre anthology, nominated for a genre award, etc.)" That way, it wouldn't be impossible to include these stories if you had a reason to, you would just have to have a reason. It would be similar to our patchy coverage of non genre print sources but more cautious. --Vasha (cazadora de tildes) 15:14, 2 July 2018 (EDT)

I'd leave this alone for now - let's take baby steps. There are a lot of stories we do not cover yet (even with the changes for the webzines) but trying to cover all of them in a single change will end up in doing nothing. So let's start working on the different types one by one - now the webzines, next step will be other genre stories... We can always start a wiki page for these and keep an eye on how many we find and where - and then use that during the next discussion on what else to allow in. The more I think about it, the more that wiki idea sound like something we should do. Annie 15:19, 2 July 2018 (EDT)

Page created, with sample entry under M. --Vasha (cazadora de tildes) 15:59, 2 July 2018 (EDT)

How about adding one more column: type of site - author, publisher, blog, non-genre webzine, (whatever else exists)? The idea is to see what kind of stories we would like to add but are just online and having them properly categorized will help in some data mining without clicking on all links. And then leave the notes for any other notes :) Annie 16:25, 2 July 2018 (EDT)

Done --Vasha (cazadora de tildes) 16:47, 2 July 2018 (EDT)

Then, if accepted, depending on the results once we start seeing what is being added, we can start looking at the rest of the possibilities.

The other question is how we want to treat non-genre fiction and both genre and non-genre non-fiction in these webzines - as we treat them in a genre magazine (which would make sense) or we draw the line there so that we do not end up cataloging 100 reviews that accompany the 2 stories in the issue. I am not sure what I am leaning towards quite honestly - cataloging the complete issue will be consistent with the e- and printed magazines so we probably should try that and keep an eye on what is added. Annie 22:07, 1 July 2018 (EDT)

One thing I would like to see is for former print/ezines to continue being treated the same if they become web-only; that'll simplify life. They will presumably continue having issues online, any rare exception can be dealt with individually. So there's one class of webzines that should automatically go on a "whitelist" if we have one. I think the list still wouldn't be very long--a couple dozen at most? --Vasha (cazadora de tildes) 22:38, 1 July 2018 (EDT)

Special rules for "print magazines-turned-webzines" would only be necessary if we were to decide to treat print magazines and webzines differently, right? If so, then I think it's another argument in favor of using the same rules for print magazines, downloadable magazines and webzines.

Re: Annie's concern that "reviewzines" with a couple of fiction pieces tacked on may overwhelm us, do we know how many webzines use this format? I have seem some sites/blogs that post mostly reviews with a smattering of stories. However, most of them seemed to be "not webzines" in that they did not have distinct issues. Was my sample not representative? Ahasuerus 14:21, 2 July 2018 (EDT)

No, if we stay with the "have issues" for now (aka actual webzines), we should be fine - I would say to just say/assume that webzine, e-zines and printed zines are treated the same (so fully cataloged as long as they are considered genre). The special case that remains is the case where a printed/e-zine switches to web-only and drops the issues (so it becomes just a blog with stories). In this case we just declare the 'zine dead, record the URL and if one day we decide to catalog the non-zine sites (aka blogs and so on), we will catalog then.

The two big sites that we might have needed to do special provisions for are tor.com and the Daily Science Fiction (as they are not webzines). But they are eligible now under the awards rule so we are covered. If another shows up, we can revisit. Annie 15:02, 2 July 2018 (EDT)

(unindent) Are we ready to create a new R&S section and post a proposal encompassing the three new areas of inclusion (genre webzines, genre issues of non-genre webzine, single issue anthologies)? Is there anything that we may be forgetting? Anything unintended that the new language may be covering? Should we take a step back and wait until tomorrow morning in case anything pops up overnight? Ahasuerus 16:26, 2 July 2018 (EDT)

Well, the only new thing today is the language itself so I'd say to make a new section and see if anyone else is interested (because I can understand why people do not want to read the maze of a discussion here). We probably should allow for partial success as well - if there is strong opposition against one of the three new categories (not sure why but it is possible) :) Annie 16:48, 2 July 2018 (EDT)

Author-specific sites

I am somewhat burned out after sorting out thousands of Fixer's ISBNs at the moment. I'll probably post a more coherent response tomorrow, but for now I'll comment re: author-run sites.

The main issue that I have repeatedly run into while working on Fixer's submissions has to do with urban fantasy and paranormal romance series. A number of UF/PR authors make certain stories "online exclusives", which means that you can't find then anywhere else. They seem to be "freebies" given by their authors to loyal fans or something close to it. When I come across them, there is no easy way to catalog them short of creating a "naked" title record.

Of course, other authors make some of their stories available on their sites as well, but most of the time they are not exclusive -- e.g., see Greg Egan's site -- which makes it easier to process them.

I hope that whatever standards we come up with will address this issue. Ahasuerus 21:29, 28 June 2018 (EDT)

That's an interesting point - I am not sure I was aware how many of those are out there. I wonder if some modification of my "monthly webzines records" idea (or even yearly records) won't work here as well - it will collect them into somewhat of a series and won't litter the page with the multiple chapbooks or fill the DB with pub-less titles... I need to think a bit on that. Annie 21:48, 28 June 2018 (EDT)

Space before ellipsis

We have a rule for the space after an ellipsis if it is in the middle of a title but we do not have a specific rule for a space (or lack of one) before it. In my book, an ellipsis at the end of a title is the same as a dot or a question mark (so no space is needed); an ellipsis in the middle of a title serves the same purpose as a comma so it also does not require a space before it but the DB has both styles (in non-trivial amounts). Should we standardize this a bit? Or leave each editor to do whatever they prefer about that? Annie 19:41, 27 June 2018 (EDT)

It should be simple to search-and-replace correct everything to whichever one we choose; I can't think of a single case where this would go wrong, can you? Then we could just do that periodically rather than expecting users to be consistent when they enter things. --Vasha (cazadora de tildes) 20:24, 27 June 2018 (EDT)

Having a backend process changing data in the DB is not as easy as you seem to think and I will be very careful in proposing such unsupervised changes - such mass replacements tend to backfire when a special case that noone thought of appears. Plus it is really easy for the editors to just pay attention and do it properly to start with - we somehow have most people get the capitalization and the price formats right. That's not the question though - the question is "do we want to standardize and if so, it what direction - or in short: to space or not to space" :) Annie 20:50, 27 June 2018 (EDT)

Both Chicago Manual of Style and MLA Style Guide (also Harvard University and quite a few others) recommend "word . . . word" No help there. I don't think that style looks good in most computer fonts. Here are the other possibilities and sources I could find recommending them in a quick search.

* Word... word Newcity Network

* Word...word - University of Oxford

* Word ... word AP Style Guide; Florida Virtual School; Google Developer Documentation

I close with a quote from Wikipedia: "According to Robert Bringhurst's Elements of Typographic Style, the details of typesetting ellipses depend on the character and size of the font being set and the typographer's preference. Bringhurst writes that a full space between each dot is "another Victorian eccentricity. In most contexts, the Chicago ellipsis is much too wide"—he recommends using flush dots, or thin-spaced dots (up to one-fifth of an em), or the prefabricated ellipsis character ( … ). Bringhurst suggests that normally an ellipsis should be spaced fore-and-aft to separate it from the text, but when it combines with other punctuation, the leading space disappears and the other punctuation follows." --Vasha (cazadora de tildes) 21:19, 27 June 2018 (EDT)

(unindent) The big problem that we ran into when we last reworked out ellipsis standards was that, as we discovered, different languages support different ellipsis conventions. Some allow style A and B, but not C. Some allow B and C, but not A. And so on. The current, somewhat incomplete, standard was the best compromise that we could come up with without stepping on any language's toes (do languages have toes?) while still supporting language-independent searching.

If we are going to revisit this issue, we'll want to be careful not to upset the delicate balance that we currently have in place. Ahasuerus 21:32, 27 June 2018 (EDT)

"Supporting language-independent searching," that's the hundred-dollar issue. Have you made any progress in investigating the possibility of searches ignoring punctuation? --Vasha (cazadora de tildes) 21:34, 27 June 2018 (EDT)

I am afraid not. It's a bigger can of worms than I originally realized. There are some third party software packages that may be worth looking into, but I haven't done anything on that front yet. Ahasuerus 22:15, 27 June 2018 (EDT)

Well, my point is that we do not have a standard at the moment for the space before the ellipsis - and everyone seems to have their own idea about it :) So if there are search issues out there, we already have a problem. Annie 21:39, 27 June 2018 (EDT)

People should not have to know what our standard is in order to search, right? That's an issue with all kinds of punctuation (like commas). --Vasha (cazadora de tildes) 21:43, 27 June 2018 (EDT)

I think all of our supported languages follow the "word, word" standard for commas. Colons, OTOH, are a different matter: Alien: Resurrection and Alien : La résurrection have to remain that way in order to accommodate language standards. We'll need to make the search logic smarter if we are to make our searches more user-friendly. Ahasuerus 22:09, 27 June 2018 (EDT)

Sure, and how not having a standard at all helps with the search now? Instead of people knowing the standard to search, they need a crystal ball to know how to search for the title. :) I realize that we have an issue with the search when punctuation is involved but this specific question was about a case of a missing standard and the fact that we are inconsistent in the DB - and if anything, using the same standard will make the search a bit more consistent. Annie 22:52, 27 June 2018 (EDT)

(unindent) So do we want to have a standard about it or should we just supply a crystal ball to everyone that tries to search for a title (so they know how to look for it) and/or just tell them to use google to search us? Annie 15:29, 2 July 2018 (EDT)

Poll of contributors? My vote would be to follow the AP style guide and go with "word ... word." --Vasha (cazadora de tildes) 15:58, 2 July 2018 (EDT)

Could the search manage to insert optional blanks before and after the special characters like colon and ellipsis? ../Doug H 19:46, 2 July 2018 (EDT)

Sorry, I missed this question when it was asked 12 days ago.

At this time the search process uses the same logic as the data entry process. What this means is that if you enter ". . ." anywhere in your submission, the software will replace it with "...". The same thing happens when searching: ". . ." is automatically replaced with "...". Similarly, we replace multiple adjacent spaces like " " with a single space. It's all part of the "filter" that we use to massage the entered data before submissions are created.

We could use the same "filtering" logic if we can agree on what the data entry rules should be. If we decide that we should always have a space between words and ellipses, we could modify the software to insert them as needed. I think it would be a better solution than making the search logic and the data entry logic diverge. Ahasuerus 13:22, 14 July 2018 (EDT)

Inclusion of webzines and one-time anthologies

The discussion of non-downloadable online publications above has identified a significant number of subcategories. They all have their quirks and caveats, so it's probably best to handle one or two subcategories at a time. Based on what we have discussed so far, the following proposal has been formulated:

1. Add the following lines to the "Included" section of ISFDB:Policy#Rules_of_Acquisition:
   • Speculative fiction webzines, which are defined as online periodicals with distinct issues
   • Special speculative fiction issues of non-genre webzines
   • One time speculative fiction anthologies published on the Web
2. Remove the word "(webzine)" from the "Excluded section" (after "Works published in a web-based publication".)
3. After the bullet point "It has been shortlisted for a major award" add another bullet point which reads:
   • Webzines and online anthologies as per the Included section

Please note that this proposal was designed to exclude blogs, author-run sites, fan fiction, Web serials and other online oddities. We may revisit all or some of them at a later point, but the proposed language was crafted to make this iteration of scope expansion more manageable. Ahasuerus 19:49, 2 July 2018 (EDT)

For the Exclude section changes, instead of adding that bullet point, how about just changing the main bullet text along the lines of:

Works published in web-based publications not designated as included -- such as blogs, author-run sites, fan fiction, web serials, and other online oddities -- and available only as an HTML readable file ...?

I think the examples will help clarify the framers' intent for posterity. --MartyD 22:38, 2 July 2018 (EDT)

I am fine with either of the two ways to word the change in the Excluded section. Annie 22:53, 2 July 2018 (EDT)

Yes, both would be fine, I think. Stonecreek 00:05, 3 July 2018 (EDT)

(unindent) It looks like there is substantive agreement re: including genre webzines and single issue anthologies.

Re: the best way to organize the Rules of Acquisition, it occurs to me that it may be possible to change the structure to make it more user-friendly. Currently the structure is:

• Inclusion list
• Exclusion list with the following exceptions:
  • SFWA-qualifying markets
  • Shortlisted for a major award

"Exceptions to exceptions" are hard to parse. How about we move them to the Inclusion list? That way the Exclusion list will simply say "Works published in all other types of Web-based publications available only as HTML readable files that are not listed in the Included section" and then we can optionally add "such as blogs, author-run sites, fan fiction, web serials, and other online oddities". Ahasuerus 12:33, 3 July 2018 (EDT)

There are lots of online publications that call themselves magazines even though they don't have distinct issues and thus aren't webzines by the definition here. How can we help people to not overlook this distinction? Add "magazines not organized into issues" to the list of examples of exclusions? How does it sound to say "... such as websites publishing without distinct issues (even if they call themselves magazines), blogs, author-run websites..." There must be a less-wordy way. --Vasha (cazadora de tildes) 12:50, 3 July 2018 (EDT)

I think it would be better to keep everything webzines-related in one place. Perhaps add something like "(and only with distinct issues)" to the end of the bullet which currently says "Speculative fiction webzines, which are defined as online periodicals with distinct issues"? Ahasuerus 13:01, 3 July 2018 (EDT)

Maybe after the sentence you originally wrote, "If a publication calls itself a magazine but does not have distinct issues, it is not included." --Vasha (cazadora de tildes) 13:16, 3 July 2018 (EDT)

Planned Changes to Project Scope

I'll leave notes on the Talk pages of some of the more active editors to make sure that we are not missing anything. Unless substantive objections are raised in the next couple of days, I plan to change the Rules of Acquisition tomorrow night. Ahasuerus 11:08, 4 July 2018 (EDT)

I'm in favour of more material even if it stretches boundaries. My concern is dilution of effort. There are a lot of projects on the go and on the books. The reason for the rules is to prevent scope creep. If allowing this expansion means a few people enter things that moderators can approve rather than argue about, I'm good. If it means days of software updates, more moderator discussions to set boundaries or fewer editors entering material for the existing bounds, I'm not so good. The exact definition of the boundary is far less relevant. ../Doug H 11:29, 4 July 2018 (EDT)

I don't expect this particular change to require software changes. We already have "webzine" listed as a valid publication format and I can't think of anything else that we would need to change on the software side. Some other scope changes discussed earlier (Web serials) may require software changes, but they are not included in this proposal. Ahasuerus 12:00, 4 July 2018 (EDT)

On the other hand, we may get more editors - ones that come to enter webzines and stay for other things and that would have never started helping if they did not start with their favorite webzine. I don't really expect people to rush in adding them but on the other hand, we already allow any e-zine and downloadable webzines - so this just closes the circle. And a magazine changing to e-zine changing to webzine won't cause head scratching on what is eligible and why we stop cataloging it... :) Annie 15:27, 4 July 2018 (EDT)

I'm OK with proposed expansion as both useful, and consistent with our charter. That said, I doubt we enthusiasts will be able to keep up with the wave front of qualifying online publications. But it's a worthy goal, and could attract new, web-first enthusiast editors. Markwood 11:49, 4 July 2018 (EDT

I can agree on the intent of the rules change, however I'm unsure on the format: sometimes I guess it's going to be difficult to discern a periodical as such, especially if they're loosely formatted and displayed. Are we going to leave it to the editors to decide whether it's to be entered as a webzine or one-ofs? MagicUnk 13:59, 4 July 2018 (EDT)

If they are loosely formatted and the issues are not set as such, it won't be eligible under the current changed rules - we will get it on the next change (eventually). The idea is not to catch all periodicals but to slowly start allowing them - the reality of genre publishing are changing and there are quite a few webzines out there that had survived longer than some print and e-zines :) A one-of is something like this - a non-periodical that just happens to be on the internet and not in an e-book or in print. Do you have an example of what may be hard to categorize? It's a first step - not a "let's get all periodicals and web-stories in the DB". Annie 15:27, 4 July 2018 (EDT)

I have no problems with this change. I personally have little interest in webzines, but I can see the value of including them. --Willem 15:11, 4 July 2018 (EDT)

Sadly, we're gonna lose the smell of books, but that's the future. It's okay with me.Rudam 01:38, 5 July 2018 (EDT)

I support the increase in scope. I would also support going further, but this is a good initial step. --Ron ~ Rtrace^Talk 07:13, 5 July 2018 (EDT)

List of Webzines

FYI: I have been keeping notes on current webzines, so I've put them on a Wiki page: Webzine List. It contains all the current ones I know of, plus a very few random examples of the hundreds of former ones. Feel free to add to this list, or edit it to be a more formal directory (I don't feel like spainding time on that). --Vasha (cazadora de tildes) 16:02, 4 July 2018 (EDT)

A nice list! What about non-English non-downloadable webzine or web-anthology? sanfeng 07:35, 6 July 2018 (EDT)

As long as they produce distinct issues, any webzine is eligible, regardless of the language. Same for the one off web-anthologies - as long as it is an anthology and not someone's blog where they post stories all the time for example. Annie 10:10, 6 July 2018 (EDT)

Outcome of the webzine discussion

Consensus has been reached; ISFDB:Policy#Rules_of_Acquisition has been updated. As per earlier discussion, "exceptions to exceptions" have been moved to the "Included" section. The old note about excluding author-run sites has been incorporated into the language proposed by Marty ("such as blogs, author-run sites, fan fiction, web serials, etc".) A note re: unstructured online periodicals has been added to the webzine definition as per Vasha's suggestion.

Hopefully the new language is reasonably clear and reflects the consensus reached above. Ahasuerus 19:39, 5 July 2018 (EDT)

You may want to update the note at the bottom of the section that serves as an "update log". Annie 19:44, 5 July 2018 (EDT)

We have something even better these days -- Rules and standards changelog :-) Ahasuerus 19:58, 5 July 2018 (EDT)

But we should do something for that last like in the ROA section - because someone not knowing about the new changelog will think that these rules had not changed since 2014. Annie 20:00, 5 July 2018 (EDT)

Hm, that's a valid point. For now, I have added a link to the changelog to the bottom of the section.

Ideally, we would want to review the last 12 years worth of discussions on the R&S page and add all agreed-upon Policy changes to the changelog, but that could take quite a bit of time. Ahasuerus 20:07, 5 July 2018 (EDT)

Page number for novel in an omnibus

The rules are not clear on whether a novel in an omnibus is taken from the title page or first page of content. Which should be used? Oddly, in this volume some entries have a title page and others do not. ../Doug H 21:34, 13 July 2018 (EDT)

Yep, good question. The same goes for anthologies and collections which have a title page before the start of the story. It might even happen that they have such a page (technically called a "half-title page") and then the story title is repeated again at the top of the text. --Vasha (cazadora de tildes) 22:16, 13 July 2018 (EDT)

I would go with where the content begins over what the table of contents says. It's the most accurate. ···日本穣 · 投稿 · Talk to Nihonjoe 12:18, 14 July 2018 (EDT)

The rules are clear about using the actual page rather than table of contents. The question is whether to use a title page as the start or the actual text. The title page provides a pretty clear beginning, but we use text for single content novels. I guess you could consider the title pages in the omnibus (anthology / collection) as more of a half-title page. But it is what we should be using for the official title (as we should not be relying on the table of contents for that either unless necessary. Consistency is a reasonable goal and more detailed rules to cover this instance would help. ../Doug H 12:40, 14 July 2018 (EDT)

Prepositions to add to the list of uncapitalized words

Although most changes proposed in last year's capitalization discussion didn't find favor, one thing that people did seem to like was the idea of adding a few more one-, two- and three-letter prepositions to the list of uncapitalized words at Template:PublicationFields:Title. As is the most significant one, followed by o’ (short for "of"); the others would be ca. (abbreviation of "circa"), per, re, ’til, via, and vs. Is everyone here OK with all of those? --Vasha (cazadora de tildes) 14:05, 17 July 2018 (EDT)

I am against it on general principle - we already have a list that everyone is used to - just adding some more to it because someone thought of them or cannot be bothered to remember or check the list is just replacing one list with another and the longer the list is, the harder it is to remember.

We can discuss making ALL prepositions and conjunction words (based on length for example) uncapitalized but just extending the list with another set of prepositions and conjunctions ("as" is not always a preposition - is your proposal to change it only when it is a preposition) is counterproductive in my opinion. Annie 14:53, 18 July 2018 (EDT)

Have I stepped into mirror world? :-) Last year I proposed uncapitalizing all one-, two-, and three- letter prepositions and conjuctions, and you said it was too hard for people to understand a rule and argued in favor of having a list instead! I definitely would prefer a rule to a list, so if you and other people want to work on one, just say the word. --Vasha (cazadora de tildes) 15:00, 18 July 2018 (EDT)

Just different context and more exposure to the DB :) I still do not like the idea of making up a rule that requires a degree in English to decipher. On the other hand making the list very long ensures that noone will remember it and we will spend more time fixing this and monitoring this than anything else. You want the short ones on a list - that's fine - but how is "as" different from "so" or "and" from "but" for example? Just adding a few more random ones into the current list does not solve any of the issues of this list. Annie 15:09, 18 July 2018 (EDT)

(unindent) OK, so you do want to keep the list? It does not have all the common prepositions on it, though; at the very least as and o’ need to be added. However, the main issue is that the situations where preposition-like words are lowercase are fewer than the ones where they are uppercase, so it would be simpler to define that. After all, you also capitalize a preposition when it has been separated from its object ("What Are You Thinking Of, Joe?") and the exceptions go on and on. So, maybe a rule that says to lowercase the preposition in a preposition phrase (Lost in the Woods, Believe in Me, None of That, Never on Monday, etc.) and in comparisons (It's as Simple as That), would get us close. Someone with a shaky understanding of grammar doesn't have to know what a phrasal verb is in order to think about "Turn On the Light" and realize that there is no preposition phrase "on the light" in there. —The preceding unsigned comment was added by Vasha (talk • contribs) .

After moving this to the correct discussion (as the list has NOTHING to do with the other discussion)...

No. We need simple rules. Not complicated ones that require an English degree. Annie 16:37, 18 July 2018 (EDT)

OK, what's your simple rule? --Vasha (cazadora de tildes) 16:38, 18 July 2018 (EDT)

The one we have now suits me just fine - a flat short list - it may not be perfect but it is consistent. If we decide to make an exception for phrasal verbs (the same way as we do for first words in a title and/or sentence), that's a separate conversation. But something that contains "preposition phrase" and "comparisons" (how about "like" - how come it does not get the same treatment in this case)? And if someone does not know what a phrasal verb is, they won't know what a "preposition phrase" is either. Just saying. :) Annie 16:42, 18 July 2018 (EDT)

I am just saying that IF we decide that the list approach is insufficient (and I really think it is--if we applied it consistently, our data would look awful) we should say that the default is to capitalize all words and come up with a rule for the rare cases when they are not. Definitely no exceptions-to-exceptions, no "put only these few words in lowercase except when."

Defining a few simple grammatical terms is actually the simplest way to talk about these things. That's why grammatical terminology exists. If, instead of a list to memorize we say 1. One-, two-, and three-letter prepositions (What is a preposition?) plus from and with, only when they form a phrase ([ What is a preposition phrase?]). 2. The as of comparisons. 3. The conjunctions and, or, and but. 4. The articles the, a, and an. -- In short, we trust that people can understand, and they will do a better job if they understand rather than trying to keep an arbitrary-seeming list in mind. --Vasha (cazadora de tildes) 17:11, 18 July 2018 (EDT)

Capitalization in English and special cases

We have two special cases where our capitalization rules are just ignored in practice:

• Poetry - everyone seems to these as they are in the magazines/collections until we need to merge two of them and then one title prevails. Which is the whole reason to have the rules to start with - the ability to not have myriad of variants and invalid data because an editor chose to specify a title in a specific way
• Phrasal verbs - the way the current rules are written mean that "on" for example is not capitalized even when it is a part of a phrasal verb. I like that interpretation but seems like most editors do the opposite. Plus that makes it easier for non-native speakers (and for ones that have no idea what a phrasal verb is) to know how to add the title.

Can we either agree on changing the rules for these two cases OR alternatively, we should start enforcing the rules as written... And while we are at that, we really need to insert a "in English" to this rule - otherwise pretty much any Spanish, Bulgarian and so on title is not following the rule :) Annie 14:58, 18 July 2018 (EDT)

I just don't see any way to get around the fact that capitalization in English is tricky and requires understanding a little grammar in order to tell the difference between phrasal verbs and prepositions. Could we have a non-native speaker help desk or something? One thing I definitely do not want to see is a rule applied to the data in this database that results in capitalization different from standard publishing practices -- we should try to look good, and that may require a little extra effort. --Vasha (cazadora de tildes) 15:06, 18 July 2018 (EDT)

We make the rules of THIS DB and we follow them as written as opposed to trying to follow rules of other DBs and standards? The rules say that "Go on" is capitalized this way and not as "Go On" regardless of the fact that it is a phrasal verb. That makes is easy to understand and implement for anyone.

Which publishing practices? Can you show a capitalization standard that is valid for all possible standards? Because the publisher standard depends on the house style - and depending on which one they follow, it differs between publishers. So as soon as I find a standard that all English language publishers (USA, UK, Australia, South Africa, Singapore and so on (add all other countries where English is used enough to publish books into)) follow and I will support you in changing to that.

If we want to separate the phrasal verbs as separate rule for capitalization and then work with non-native editors, then fine - but we need a rule change for that. We don't just decide to ignore the rules just because :) Annie 15:15, 18 July 2018 (EDT)

Publishing standards vary, but they don't vary infinitely. There are some things that are common to all of them. I don't think we are yet widely enough known to be able to use practices that no one else does and have them follow us instead of the other way around. :-) We should be copying either big publishers or big academic bibliographies & library catalogs. Being as the bibliographies and catalogs tend to use sentence-case, we need to figure out what the practices of major English-language publishers are. They have a lot in common with each other. Alas, leaving phrasal verb "prepositions" uncapitalized is not something anyone else does, so I don't think we have the freedom to do so ourselves. --Vasha (cazadora de tildes) 15:35, 18 July 2018 (EDT)

Then we need to make that an official rule :) Or an editor that actually follows the rules will be told that they actually need to do something else. At the end of the day, as long as we are consistent in our DB, we will be fine. When we are not, we look as if we are just making up things as we go (and each editor does it differently). Which is why I started the conversation - because as they stand now, we are NOT capitalizing these (despite a lot of people doing it and some moderators allowing them to go through). :) Annie 15:40, 18 July 2018 (EDT)

In that case, I actually agree with you. The list approach has the problem that it does not allow for context and does not admit that the words on the list may have multiple uses, some of them capitalized and some not. We should not continue the current situation where in theory, the list reigns over all, but in practice, a lot of people are entering standard-publishing-capitalization and the moderators allow it. It's true that this is far from the only instance here where practice diverges from principle, but that's just an argument in favor of updating the rules when they don't match what people actually do (rules changes are perpetually lagging behind). --Vasha (cazadora de tildes) 15:55, 18 July 2018 (EDT)

Two separate issues... let's try to work on them one by one? This is for the phrasal verbs ONLY - the list does not matter at all - the list is already irrelevant if it is the first word in a sentence/title; the phrasal verbs exception will go on the same level :) The conversation about the list is elsewhere and connecting them again ends up in a long convoluted and fruitless discussion... Annie 16:02, 18 July 2018 (EDT)

(unindent)According to "The Practical Speller" (p. 97)

• 10. The titles of books, the headings of chapters, etc. take capitals for the notional words and small letters for the connecting words, that is, the articles, prepositions, and conjunctions.
• 11. Wherever there is any doubt about the use of a capital letter it is safe to use a small letter. Something is left to the judgement of the writer in regard to very important words, but excess of capitals is to be avoided."

Even if this is the 20th Century Edition, Things may have changed since 1901. ;-) ../Doug H 16:51, 16 October 2018 (EDT)

:-) Things haven't really changed, if the number of titles I see with words like "is," "it," and "my" uncapitalized is any indication. The only difference is that now "something is left to the judgement" of the computer more often than to that of the writer. --Vasha (cazadora de tildes) 17:04, 16 October 2018 (EDT)

Convention for name of series published in multiple languages?

Now that we are becoming so multilanguage, we are getting more series that have different series-names in their original language and English translation. Should we establish a standard convention for handling that -- for example, "Original / English" or "English / Original" (I would favor the latter)? --Vasha (cazadora de tildes) 20:23, 18 July 2018 (EDT)

And what happens if the series already has a / in the name? Or if we first enter the books in German while the original is Bulgarian? I think that we need to start talking about "series name per language" instead of how to do a "Frankenstein" name for a series so that people can see the main DB record (in English), the original DB record (in whatever language) and the "language version" when attached to the book in that language (or any combination of the above). This way you don't end up with a "German / English" series on the display of a Russian book (with the Russian name of the series nowhere to be seen unless you want to put it in the notes (and that does not work very well for publications with multiple stories from multiple series). Annie 20:35, 18 July 2018 (EDT)

It would certainly be nice to have the series name displayed on a title record match the language of that title, but implementing it sounds tricky.

One possibility is that the series could have multiple names, each with a language assigned to them. That way, the title page could pull up the appropriate one for the language. And when entering a new publication, you could put any of the names into the Series field and the software would find it. There'd have to be a way of combining with an established series if you entered a publication with a language that hadn't been used before. Disadvantage: doesn't deal with the problem of series having been published under multiple names in the same language (you'd have to keep dealing with that the same way we do now, with slash-separated alternate names and use of notes). --Vasha (cazadora de tildes) 21:01, 18 July 2018 (EDT)

We can even make it easier - when adding a new publication, if it is a new series, mark it with whatever language it is in and make that default for this series. Then make series mergeable as titles are. When showing a series for a title, use the language of the title or the default language if the proper one does not exist (I do not know why a Russian magazine series need to have an English name in the same record or as a record at all... but that's a different question:)) There are a lot of ways to implement it - as long as we agree to at least put it on the development's plan list at some point. :)

PS: Are you saying that we should not have any non-English series without an English name in the name of it? Or is that only for the ones that are not in the Latin alphabet (for this we need transliteration - and we have a discussion about it). Because we have a lot of German and French series out there. :) Annie 21:36, 18 July 2018 (EDT)

I was only thinking about the display on the author's summary page, and only in those cases where the series has been published in English. I suppose it could be an option like the "display all translations" option: display series names in default language only, or in default language + English if both exist, or English only! Yeah, I'm just tossing out some thoughts that might be implemented at some point --Vasha (cazadora de tildes) 21:47, 18 July 2018 (EDT)

I am thinking about the publications pages for collections for example where the series names are on the lines with the content they belong to. The author pages are a different kettle of fish altogether :) Annie 21:52, 18 July 2018 (EDT)

Location of ISFDB biographies: Policy/FAQ updates

Please note that ISFDB:Policy#Biography Policy, Help:Contents/Purpose#Biographies, and ISFDB:FAQ#Can I enter author biographies in the ISFDB? have been updated to reflect the fact that author bibliographies are now entered directly into author notes as opposed to using Wiki-based "Bio" pages. Ahasuerus 13:09, 26 July 2018 (EDT)

Rules of Acquisition -- unpublished books

ISFDB:Policy#Debatable currently includes the following item:

• Unpublished works by established authors, e.g. John Taine's manuscripts? Or do we just mention them in their respective Wikipedia articles? On their ISFDB Wiki page?

This was something that I wrote back in May 2006 when we were trying to define the scope of the project before opening the beta phase of ISFDB 2.0. The John Taine example was a reference to his manuscripts which are known to exist but remain unpublished -- see the Eric Temple Bell Papers stored at University of California at Santa Cruz. Over the last few years this has become a common occurrence as more and more archives make their catalogs available online.

On the other hand, Template:PublicationFields:Date, which reflects our current data entry standards, says:

• 8888-00-00 is used to date publication records for books that were announced but never published. In most cases these publications turn out to be vaporware, but there is good reason to keep these records in the database: it prevents Fixer (along with other automated bots) and editors using secondary sources from submitting records with duplicate ISBNs as well as establishing a bibliographic record of important titles like Last Dangerous Visions.

I suggest that we try to reconcile the Rules of Acquisition and Help. How about adding the following line to ISFDB:Policy#Included:

• Announced but never published works of speculative fiction may be entered as "unpublished"

Similarly, ISFDB:Policy#Excluded can have the following line added:

• Works of speculative fiction which have been neither announced nor published. This includes manuscripts, unpublished drafts, etc.

? (We may also want to clarify what "published" means, but that's a separate topic.) Ahasuerus 11:27, 27 July 2018 (EDT)

Yes, good idea. However, 8888-00-00 is also quite frequently used to record unpublished original-language titles of translated works. These have never been "announced" but they ought to stay anyway. --Vasha (cazadora de tildes) 11:49, 27 July 2018 (EDT)

That's a good point. The language should probably cover Fixer's forthcoming books as well. How about:

• [Included] Unpublished works of speculative fiction which have been:
  • announced as forthcoming within the next 90 days
  • announced but never published (entered as "unpublished")
  • published only in translation (the original should be entered as "unpublished")
• [Excluded] Unpublished works of speculative fiction unless they fall within one of the categories explicitly listed in the "Included" section

? Ahasuerus 13:48, 27 July 2018 (EDT)

I like the wording. Annie 14:41, 27 July 2018 (EDT)

I have inserted "only" between "published" and "in translation". Ahasuerus 15:19, 27 July 2018 (EDT)

Looks good. It was implied already considering that it is in an "unpublished" section but adding it does not harm.

I think that we do need to revisit the "unpublished" when something is published only on translation (we had a similar discussion for when something it only published in serialization) - the "unpublished" sends the book at the end of one's list which is annoying. Maybe a change in the way the author page orders books is how we can handle that - use the earliest date even if it is an a variant. But this is a separate topic and I need to think a bit more on that. But we will see more and more of these as we add more and more international works. :) Annie 15:30, 27 July 2018 (EDT)

Outcome -- Unpublished books

ISFDB:Policy#Rules_of_Acquisition has been updated with the proposed language. Ahasuerus 11:26, 31 July 2018 (EDT)

Unpublished books -- Mouseover text

We probably should drop the mouseover text on "(unpublished)" now. It is not true that that indicates "announced but never published," and I think rather than wordily explaining both uses of 8888, "unpublished" is informative enough. --Vasha (cazadora de tildes) 13:28, 31 July 2018 (EDT)

True, the current wording can be misleading. I'll remove the mouseover bubble until we can think of a better way to do it. Ahasuerus 15:06, 31 July 2018 (EDT)

Done. Ahasuerus 15:12, 31 July 2018 (EDT)

Does the rule for dating magazine issues need to be reworded?

I know that magazine issues should be dated according to their cover date rather than their publication date. However, the corresponding section of Help:Screen:NewPub doesn't say so in so many words:

For magazines, the month on the cover is rarely the month of actual publication. However, since this is a well-known fact about magazine publication schedules, and also because there is no good way to determine actual publication date, both month and year should be given where possible. For bimonthly magazine dates, use the earlier month: "January-February 1957" should be entered as "1957-01-00", for example. If a bimonthly magazine only quotes a single month in the title, use that month. E.g. the March 1959 issue of Fantastic Universe was preceded by January 1959 and succeeded by May 1959; it should be entered as "1959-03-00". If a bimonthly issue spans a year boundary, such as a December-January 1960 bimonthly issue, use the earlier year and month: "1959-12-00". For magazine cover dates that cannot be assigned to a specific month, use the year only: "Spring 1943" is just entered as "1943-00-00".

That is really pretty unclear. I think the section should start with the words "Magazine issues should be entered using their cover date as the date" or something to that effect. --Vasha (cazadora de tildes) 22:12, 1 August 2018 (EDT)

I agree on both counts. I also think that much of the verbiage could be consolidated. How about something like:

• When entering the date of a magazine/fanzine issue, use the year and month which appear on the cover. If more than one month appears on the cover, use the earliest year and month, e.g. "December 1959/January 1960" should be entered as "1959-12-00". For magazine cover dates which cannot be assigned to a specific month, use the year only, e.g. "Spring 1943" should be entered as "1943-00-00". (Note that the ISFDB uses the cover month even though magazine issues typically become available to the public earlier.)

I think that should cover everything. Ahasuerus 17:35, 13 August 2018 (EDT)

I have cleaned up the proposed text a bit and added a note about the difference between cover dates and publication dates. Ahasuerus 10:23, 14 August 2018 (EDT)

Yes, that's a great improvement. However, sometimes magazines don't have a date on the cover, but do have one on the masthead. How about, "If there is no date on the cover, use the date on the masthead; if there is none there, use the year only. Do not use a publication date." --Vasha (cazadora de tildes) 18:47, 13 August 2018 (EDT)

A couple of thoughts:

1. If the publication date is known/stated and is consistent with the cover/masthead date, why not use it? I don't see why we'd want to record a less exact date.
2. For magazines with no cover/mastehead data, if the publication date is known it should be used as-is (not in the title, though).

--MartyD 08:03, 14 August 2018 (EDT)

It looks like there are two separate issues here. The first one is rewording the current Help text to make it clear and concise. The second one is changing the substance of this Help paragraph: cover dates vs. masthead dates vs. actual publication dates. I suggest that we tackle the first issue in this section and then create a separate section to discuss proposed substantive changes. Ahasuerus 10:20, 14 August 2018 (EDT)

Template:PublicationFields:Date has been updated with the proposed language. I will copy the other part of the discussion to a new section. Ahasuerus 00:47, 18 August 2018 (EDT)

Serial essays

Can we either create a title type SERIALESSAY or treat essays the same as SERIAL when they are combined into a single volume. In this case, there are a bunch of essays that differ only in a number at the end, and they were all collected into a single volume. Basically, these are a serial, but non-fiction. Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 20:10, 10 August 2018 (EDT)

Would it be feasible to put them in a series? Essay series are supported by the software. Ahasuerus 20:24, 10 August 2018 (EDT)

I don't yet know how many there are, so I don't know if adding them all as titles to the collected volume is feasible. ···日本穣 · 投稿 · Talk to Nihonjoe 20:12, 10 August 2018 (EDT)

After doing a little digging, there are at least 133 of them. That would be cumbersome to add to one volume, to say the least. Can that many titles be in one publication? ···日本穣 · 投稿 · Talk to Nihonjoe 20:15, 10 August 2018 (EDT)

At this time the software allows up to 1,999 titles per publication. However, once you get into the 500+ range, validation and filing become very sluggish. In addition, submissions can time out or error out, so I don't recommend it. Ahasuerus 20:23, 10 August 2018 (EDT)

Opinions wanted -- how absurd does a satire have to be to be included here?

I am puzzling over the line between comedy/satire and fabulation/fantasy with regards to Machado de Assis's "The Alienist." In this story, a doctor who has a new theory of insanity sets up a madhouse in a Brazilian city and begins interning in it people who are vain, covetous, duplicitous, etc. eventually locking up four-fifths of the city ... events escalate from there. Now, if a story like that was set in a country with an imaginary name, I think we'd all include it here. But if it's set in Brazil, and involves a lot of recognizable local detail, is it fantasy? (PS I have found this story in an anthology of fantastic stories, so I guess it is in.) --Vasha (cazadora de tildes) 02:03, 11 August 2018 (EDT)

"A doctor locks up 80% of the city population in a madhouse" sounds like magical realism, so I would be inclined to include it as a genre story. Admittedly, magical realism can straddle the fence between fantasy and reality, but in this case I'd say that it favors the speculative side. If it was only 8%, it wouldn't call it speculative. Ahasuerus 17:24, 13 August 2018 (EDT)

Canonical name for an author with lots of non-genre and a few genre works only

From the Glossary: "Canonical name. (...) For authors who publish under multiple names, the canonical name is the most recognized name for that author."

This is not clear enough and it should be added if "most recognized name" is based on the genre works of an author only, or regardless of genre. Imagine an author with a vast amount of non-genre works published under his real name and only a few genre works under a pseudonym. We'd include the genre works only in the ISFDB then, and based on the help text an editor may believe that "most recognized name" means that the author's pseudonym should be used as canonical name (because we are a SF database). I'm actually not sure right now if there is an unwritten rule / standard practice for this already (is there?), but I think the help should be more precise. It should be similar to one of these (italics are my additions):

Either

"For authors who publish under multiple names, the canonical name is the most recognized name for that author, regardless of whether the author's work is mainly genre or non-genre."

or

"For authors who publish under multiple names, the canonical name is the most recognized name for that author, based on the author's genre works only."

Jens Hitspacebar 14:59, 13 August 2018 (EDT)

I think Help should say something like "the canonical name is the author's most recognized name within the genre". William Fitzgerald Jenkins, who published most of his SF as Murray Leinster, comes to mind. He was very prolific (hundreds of stories) in multiple genres and I wouldn't be surprised if it turned out that he had as many stories published as "Will [F.] Jenkins" as "Murray Leinster" if you counted all of his mysteries, westerns, etc. That shouldn't affect our choice of "Murray Leinster" as his canonical name. Ahasuerus 15:28, 13 August 2018 (EDT)

Agreed. -- JLaTondre (talk) 18:35, 14 August 2018 (EDT)

"Within the genre" has been added. Ahasuerus 09:19, 18 August 2018 (EDT)

SERIAL in Help templates

At this time the "(Complete Novel)" rule for novel length works published in a single magazine issue is given in Template:TitleFields:Date:

• Novel length works (40,000+ words) printed as a single installment in a magazine are treated as serials and given the date of the issue in which they appear; the Title Type is "Serial" and the text "(Complete Novel)", preceded by a space, is appended to the title.

Template:TitleFields:TitleType says:

• SERIAL. Use for a title that would otherwise be either SHORTFICTION or NOVEL, but which is being serialized. Include "(Part 1 of 3)" etc in the title of the work. Occasionally this will lead to odd situations. For example, Frederik Pohl's novel "Jem" was serialized in Galaxy; it was initially intended to be a four part serialization, so the first two parts were listed as "Part 1 of 4" and "Part 2 of 4". By the time the third part was printed, it had been changed to a five part serialization, so the titles of the next two parts were "Part 3 of 5" and "Part 4 of 5". In this case it makes sense to go back and change the title of the first two parts; a note should be added to the note field of the publication to make it clear what has been done and why, if such a situation arises. Note that all new SERIAL titles need to be turned into variants upon approval -- see Help:How to connect serials to titles for instructions.

I am thinking that we may want to move the "Complete Novel" and the "Part 1 of 3" instructions to Template:TitleFields:Title. Also, the part which says:

• Novel length works (40,000+ words) printed as a single installment in a magazine are treated as serials

should be moved from Template:TitleFields:Date to Template:TitleFields:TitleType. Ahasuerus 00:33, 18 August 2018 (EDT)

Here is the proposed "before vs. after" language.

Current version of Template:TitleFields:TitleType:

• SERIAL. Use for a title that would otherwise be either SHORTFICTION or NOVEL, but which is being serialized. Include "(Part 1 of 3)" etc in the title of the work. Occasionally this will lead to odd situations. For example, Frederik Pohl's novel "Jem" was serialized in Galaxy; it was initially intended to be a four part serialization, so the first two parts were listed as "Part 1 of 4" and "Part 2 of 4". By the time the third part was printed, it had been changed to a five part serialization, so the titles of the next two parts were "Part 3 of 5" and "Part 4 of 5". In this case it makes sense to go back and change the title of the first two parts; a note should be added to the note field of the publication to make it clear what has been done and why, if such a situation arises. Note that all new SERIAL titles need to be turned into variants upon approval -- see Help:How to connect serials to titles for instructions.

Proposed version of Template:TitleFields:TitleType:

• SERIAL. Use for a title that would otherwise be either SHORTFICTION or NOVEL, but which is being serialized in a magazine or fanzine. The SERIAL title type is also used for novel length works (40,000+ words) printed as a single installment in a magazine or fanzine. Note that all newly added SERIAL titles need to be turned into variants once the original submission has been approved -- see Help:How to connect serials to titles for instructions.

Note: The "Jem" example will be moved to Help:Use of the SERIAL type.

Current version of Template:TitleFields:Date:

• Magazines: Serial installments of a work are always given the date of the magazine in which they appear even if the work has been published previously in book or serial form. Novel length works (40,000+ words) printed as a single installment in a magazine are treated as serials and given the date of the issue in which they appear; the Title Type is "Serial" and the text "(Complete Novel)", preceded by a space, is appended to the title.

Proposed version of Template:TitleFields:Date:

• Magazines and fanzines: Serial installments of a work, including novel length works (40,000+ words) printed as a single installment, are always given the date of the issue in which they appear even if the work has been published previously in book or serial form.

There is nothing in Template:TitleFields:Title re: SERIALs at this time. Here is the proposed language which includes what used to be in Template:TitleFields:TitleType, Template:TitleFields:Date, and Help:Use of the SERIAL type:

• SERIALs. If the title of a SERIAL installment is unique, e.g. "Butterflies in the Kremlin, Part Eight: As the Bear Turns" or "Ciężki bój (cz. 1)", then use the full form of the title. If, on the other hand, the title is shared by all SERIAL installments of the work, append a space and a parenthetical statement such as "(Part 1 of 3)" to the title. For novel length works (40,000+ words) printed as a single installment in a magazine or fanzine, append a space and "(Complete Novel)" to the title.

Please note that this proposal is an attempt to reorganize and clarify existing Help templates without altering existing data entry rules. I plan to revisit the issue of SERIALs in CHAPBOOKs as a separate issue. Ahasuerus 15:17, 24 August 2018 (EDT)

I recently ran into a serial where each installment had a different title. You can see it here. I entered it that way because it's a serialized story, but the name of the overall story is different than each entry. ···日本穣 · 投稿 · Talk to Nihonjoe 16:30, 24 August 2018 (EDT)

I think it should be covered by the proposed first sentence of Template:TitleFields:Title which says "If the title of a SERIAL installment is unique , e.g. "Butterflies in the Kremlin, Part Eight: As the Bear Turns" or "Ciężki bój (cz. 1)", then use the full form of the title". Ahasuerus 16:37, 24 August 2018 (EDT)

How exact should we be in following the title in the magazine? If it is printed "Vandals of Varda--Part Two" or "Vandals of Varda: Chapter 2" should we enter that as "Vanadals of Varda (Part 2 of 8)"? --Vasha (cazadora de tildes) 17:22, 24 August 2018 (EDT)

That's a good question. This exception was added to Help:Use of the SERIAL type a few years ago. If I recall correctly, we didn't discus where we wanted to draw the line at the time. It may be worth having a separate discussion of this issue once we get the Help templates reorganized. Ahasuerus 17:34, 24 August 2018 (EDT)

One note on the wording: "the title is shared by all SERIAL installments of the work" - change "all" to "more than one of". That will cover cases where the first and last installment say: "Story: The Beginning" and "Story: The End" but the middle ones are called just "Story" and cases where some of the installments have separate names and some do not (because the magazine started naming in one way and finished in another for example. Other from that, the proposal looks good. Annie 22:14, 24 August 2018 (EDT)

I agree with Annie, and it looks good to me overall, too. For the "exception" case being discussed above, the intent behind what's there was to avoid appending "(Part m of n)" needlessly. So for Vasha's example, if "...--Part Two" or "...: Chapter 2" were unique to the installment, i.e., other installments are titled "...--Part One" or "...: Chapter 3", then just use the titles as-is. The "(Part m of n)" only needs to be appended when different installments use the same title. --MartyD 08:39, 25 August 2018 (EDT)

Something else comes to mind. Suppose a novel has been serialized multiple times. Further suppose that some serialization installments had unique titles and some didn't. Should the "uniqueness" test apply:

• across all serializations, or
• to the set of SERIAL titles that the current title belongs to

? Ahasuerus 20:11, 25 August 2018 (EDT)

Supposing that one serialization put numbers in the title and one didn't, it would actually make the display more clear rather than less to have one set of installments in the database as "Story (Part m of n)" and the other as "Story: Part N." The important thing in order to reduce confusion would be to have each set of installments be consistently labelled.

And for that reason, if there are numbers in the titles but they are inconsistent, I'm in favor of dropping them and going with m of n instead. I just recently did so in a case where the two parts of a story were titled "Story" and "Story, Part II" -- I figured it would be better in terms of clarity to enter those as "Story (Part 1 of 2)" and "Story (Part 2 of 2)" while recording the actual titles in notes. --Vasha (cazadora de tildes) 21:21, 25 August 2018 (EDT)

SERIAL in Help templates - Outcome

Given the consensus re: the bulk of the proposed changes, I have updated the affected Help templates with the posted language. I have also changed "shared by all SERIAL installments of the work" to "shared by at least one other SERIAL installment of the work" as per Annie's suggestion. Next I will create a new section for the discussion of title disambiguation. Ahasuerus 16:44, 26 August 2018 (EDT)

Magazine dates -- cover vs. masthead vs. actual publication date

[Re: Template:PublicationFields:Date ] Sometimes magazines don't have a date on the cover, but do have one on the masthead. How about, "If there is no date on the cover, use the date on the masthead; if there is none there, use the year only. Do not use a publication date." --Vasha (cazadora de tildes) 18:47, 13 August 2018 (EDT)

A couple of thoughts:

1. If the publication date is known/stated and is consistent with the cover/masthead date, why not use it? I don't see why we'd want to record a less exact date.
2. For magazines with no cover/mastehead data, if the publication date is known it should be used as-is (not in the title, though).

--MartyD 08:03, 14 August 2018 (EDT)

Who/what is a cover artist?

I can't find it in searching, but I have a memory of someone remarking that the COVERART record is only intended to record the author of "art" and therefore we wouldn't enter things like NASA credited for a photo of the moon, an agency credited for a landscape photo, etc. Is this wholly a matter of opinion? Personally, I don't enter either photo agencies or the photographers of stock photos, but I'd like to hear other opinions. --Vasha (cazadora de tildes) 19:07, 24 August 2018 (EDT)

Not sure about the NASA photos at the moment, but as for stock photos: the increased use of them by cover designers (from Shutterstock and the like) for book covers are a problem, I think. It's often debatable if these photos are really art and not just snapshots (of landscape, a house, the blue sky, etc.). We maybe already have some "cover artists" in the database who only provided a snapshot photo for a cover but got the credit. Moreover, if such a photo is photoshopped or whatevered into a book cover by a cover designer, who is the cover artist? I currently tend to decide this on a case-by-case basis, because the cases are too diverse. Examples from my verified books:

• Terror uses a photo from a stock image provider by an (it seems) established war photographer, therefore I credited him even if it's a "just" a photo (I also dimly remember that this photo got a prize, but that's not stated in the book and I don't know anymore where I got that information from).
• Venus siegt uses a photo from a user at a stock image provider, which looks there like it's a work of art itself (probably not a photo at all but a very realistic digital drawing), but the cover designer of the book added own artwork to it, therefore it's "combined artwork" and I credited both (see note there).
• Blade Runner: the cover designer used images from a user at Shutterstock, but the user doesn't exist there anymore, so it's hard to tell who provided which part of the cover. Therefore I only added the infos to the note.

To sum it up I'd say that the creators of photos from stock image providers should not be credited unless the photo is either by a well-known artist or the photo itself is well-known. But even this can be debatable, because even if a photo would not be eligible according to such a rule, some photos might be really skilful works of art...

On a more general note I think if an image from a stock image provider is worthy to be credited we should only credit the user name/artist name(s) from there, but never credit the stock image company itself. We should add this to the rules and remove these companies from all cover artist records in the database. Jens Hitspacebar 11:50, 25 August 2018 (EDT)

As photography is considered an art form, why would we exclude it? If we're going to make value judgments on their artistic merits, are we going to make the same value judgments for other art forms (some self-published covers for example)? Whether the photo is via a stock image provider or not shouldn't be a determinant for whether it gets credited. As for how photos from a stock image provider are credited, that is a different matter. If the photo is credited as "name / stock provider", I credit based as "name" only (the same goes for artwork via an art stock provider). If it is only credited to the stock provider, I have mixed opinions and would be fine excluding them. We are supposed to credit per the pub, but the stock provider is clearly not the one who took the picture. Your comment on designers does raise an issue that occurs even outside of photos: the designer can make significant changes to the artwork that it is significantly different than the original (more than cropping, coloring, etc.). At the moment, we are supposed to exclude them (though there are many designers credited in the database) and I tend to think this is the best approach across the board vs. trying to make judgments on when enough changes have occurred to be credited in their own right. -- JLaTondre (talk) 10:44, 26 August 2018 (EDT)

I agree. Photographs are are just as much art as drawings are. I don't object to/remove anyone's use of a provider if they choose, but to me they're uncredited if there is no identification of the photographer(s). As for cover designers, I think of them as akin to editors. While they certainly contribute to the final published work, perhaps significantly, they did not create the work, and our author/artist fields record the creators. --MartyD 13:54, 26 August 2018 (EDT)

Disambiguating SERIAL titles

(copied from here)

How exact should we be in following the [SERIAL] title in the magazine? If it is printed "Vandals of Varda--Part Two" or "Vandals of Varda: Chapter 2" should we enter that as "Vanadals of Varda (Part 2 of 8)"? --Vasha (cazadora de tildes) 17:22, 24 August 2018 (EDT)

That's a good question. This exception was added to Help:Use of the SERIAL type a few years ago. If I recall correctly, we didn't discus where we wanted to draw the line at the time. [...] Ahasuerus 17:34, 24 August 2018 (EDT)

[...]

[...] the intent behind what's there was to avoid appending "(Part m of n)" needlessly. So for Vasha's example, if "...--Part Two" or "...: Chapter 2" were unique to the installment, i.e., other installments are titled "...--Part One" or "...: Chapter 3", then just use the titles as-is. The "(Part m of n)" only needs to be appended when different installments use the same title. --MartyD 08:39, 25 August 2018 (EDT)

Something else comes to mind. Suppose a novel has been serialized multiple times. Further suppose that some serialization installments had unique titles and some didn't. Should the "uniqueness" test apply:

• across all serializations, or
• to the set of SERIAL titles that the current title belongs to

? Ahasuerus 20:11, 25 August 2018 (EDT)

Supposing that one serialization put numbers in the title and one didn't, it would actually make the display more clear rather than less to have one set of installments in the database as "Story (Part m of n)" and the other as "Story: Part N." The important thing in order to reduce confusion would be to have each set of installments be consistently labelled.

And for that reason, if there are numbers in the titles but they are inconsistent, I'm in favor of dropping them and going with m of n instead. I just recently did so in a case where the two parts of a story were titled "Story" and "Story, Part II" -- I figured it would be better in terms of clarity to enter those as "Story (Part 1 of 2)" and "Story (Part 2 of 2)" while recording the actual titles in notes. --Vasha (cazadora de tildes) 21:21, 25 August 2018 (EDT)

(unindent) Please note that I have posted about the larger issue of title disambiguation on the the Community Portal. Ahasuerus 17:19, 27 August 2018 (EDT)

Newspapers

ISFDB:Policy says that "newspaper publications" are "debatable".

We have a number of newspaper issues on file, including The Chicago Times-Herald and <The> New York Times. Based on current usage, I propose that we move newspapers from the "Debatable" section to the "Included" section. Ahasuerus 11:03, 25 September 2018 (EDT)

Sure! I'm all for it. Also, otherwise we would lose some valuable information, like this. Stonecreek 12:04, 25 September 2018 (EDT)

Yep, we ought to keep the publications for all the classic fiction that first appeared in daily newspapers. --Vasha (cazadora de tildes) 12:42, 25 September 2018 (EDT)

Agreed. It should probably be emphasized that these are entered as non-genre magazines and that the editors are usually entered as "Editors of XY" (unless the editors are also publishing within the genre). Jens Hitspacebar 12:46, 25 September 2018 (EDT)

If the proposal is accepted, we'll presumably want to change Help:Entering non-genre magazines to Help:Entering non-genre magazines and newspapers and adjust the language accordingly. I suppose "genre newspapers" are another possibility, but I can't think of any paper-based ones (daily SF periodicals do exist, e.g. see Daily Science Fiction.) Ahasuerus 13:04, 25 September 2018 (EDT)

(unindent) It does raise an interesting question about our current decision to restrict webzines to those that publish "issues" rather than simply having dated posts. In what ways would it differ from a newspaper if we made a record for "Weird Fiction Review, October 29, 2015" and included in that publication the story and accompanying interview they published that day? Something to think about. --Vasha (cazadora de tildes) 12:42, 25 September 2018 (EDT)

Newspapers - Outcome

Seeing no objection, I have moved newspapers from "Debatable" to "Included" on the ISFDB:Policy page. Next we'll need to review and tweak our Help pages to make sure that the language covers both magazines and newspapers. Perhaps "periodicals" would be a more concise term, but we need to make sure that we don't inadvertently/implicitly expand the scope. Ahasuerus 00:42, 1 October 2018 (EDT)

It turned out that Help:Entering non-genre magazines already covered newspapers and other non-genre publications. I have changed the page name to Help:Entering non-genre periodicals and replaced "magazines" with "periodicals" throughout. I have also cleaned up the text a bit, but it needs a lot more work since many instructions are outdated and/or poorly organized. Ahasuerus 15:00, 1 October 2018 (EDT)

Canonical names for transgender authors

Hello! It recently came to my attention that ISFDB treats the names of transgender authors who have transitioned as a pseudonym of their birth names. After some discussion with Ahasuerus I understand the reasons for this, including software limitations and vague use of the word "pseudonym", but want to argue for a policy change with regards to how the canonical name for transgender authors is determined.

In short, listing a transgender author's birth name as canonical is not only insulting, but has the potential to open that author up to increased harassment and abuse online. Although I understand that ISFDB wants to keep a clear record of authorship and linkage between author names, there are ways to do that which are more respectful than the current policy. The current policy can also lead to incorrect information being propagated elsewhere, as people who pull citations from ISFDB may not understand how ISFDB is treating author names in these cases.

For example, a recent post to File770 which listed Hugo-eligible work published in 2018 listed a story by Anya Johanna Deniro as by Alan Deniro, a name the author has not used since 2015. It seems likely they got their information from ISFDB, which credits the story that way on the entry for issue of Shimmer the story was published in. That entry includes an interview with Anya, also credited as with Alan. This is confusing, and surely less clear than if Anya's current name were her canonical one.

Although I have been led to understand that some of the reasons for this are technical, it seems that others are just a matter of policy. (Which policy makes perfect sense in other situations! Just not in this one.)

With that in mind, I would like to propose a simple addition to current policy on determining canonical names:

When an author is known to be transgender, the canonical name should be their current name.

This would be clearer to outside users, and would also be more in line with policies on transgender individuals from other organizations. For example, policy on referring to transgender people in articles is to "Give precedence to self-designation as reported in the most up-to-date reliable sources, even when it doesn't match what's most common in reliable sources". policy for citing books published by transgender authors under a previous name is to use the current name and pronouns, with a "see also" note which cross-references the name the author was using at the time of publication.

Note that this would not affect authors who are not transgender but use a pseudonym of the opposite sex. (e.g. James Tiptree Jr. was not a transgender author; she just wrote under a male pseudonym.)

Would be happy to discuss this further if anything I've suggested isn't clear. Thanks! Stewartcbaker 13:25, 26 September 2018 (EDT)

I wholeheartedly agree with this. I do think that the importance of such matters to transgender people goes somewhat beyond mere ordinary courtesy. I like your proposed addition to policy. Thanks for bringing this up, Stewart. --Vasha (cazadora de tildes) 13:34, 26 September 2018 (EDT)

This makes sense to me.--Michaelc 14:25, 26 September 2018 (EDT)

(unindent) Digging through the archives, I see that I didn't have a strong preference when the issue was last raised in late 2016. I was willing to go along with changing Heather Brewer's canonical name to Zac Brewer as long as the person remained active within the genre under the new name. Darrah Chavey made a related comment:

• We don't always follow the preferences of an author, such as those that have been asked to be removed, or to have a particular item removed, or those who've asked to have birthdays removed (when they are public data) <2018 note: this policy was changed in 2017>. Gendered names are a bit different though, and I think we should normally follow an author's preference when it comes to such a name. But I think it's also appropriate to wait until that name has "announced" to the SF community, e.g. through a genre book with their new name. For example, Amos Salmonson was *living* as Jessica for about a year before making a formal announcement of her change -- which happened in "Mom's Home Made Apple Fanzine" in 1974 when she published a particularly feminine picture of the person that "Amos" had become. Most such announcements aren't quite as easy to pin down to a particular date, but publishing two genre books under his new name would certainly do that. Chavey 17:27, 30 December 2016 (UTC)

I also had the following general concern:

• I think the biggest problem with following the author's preferences in a certain subset of cases is that there are other, equally compelling, subsets of cases. For example, take an author who has gone through a very painful divorce and wants to have as little to do with her old married name as possible. Or take an author who has been forced to use a pseudonym because the books published under the original name didn't sell well and whose resurrected career is still hanging in the balance. She may well see anything that promotes the pseudonym as helping her career to survive and vice versa.
• It's all understandable and justifiable, but if we start making exceptions for "worthy causes", it will be very hard to preserve any kind of objective standard. Ahasuerus 18:22, 31 December 2016 (UTC)

I'm unsure whether I'm responding to a concern from 2016 or what is still a current concern, but it may well make sense to wait until a transgender author has a relevant publication under their new name before making that new name canonical. Unless I'm misunderstanding ISFDB's process, there would be no reason to actually enter in that new name without a new publication anyway, correct? I would recommend making the requirement a new publication, rather than a book specifically, as many authors are prolific short storyists who may only publish the occasional book. I definitely think the current policy of waiting until the new name becomes the "most recognized" one is unreasonable in this case, though.

I guess one option if you wanted to wait until a publication might be "When an author is known to be transgender, the canonical name should be the name under which they have most recently published."

Regarding objective standards, I must confess I have a hard time seeing how matching up an author's canonical name with their current name would result in a loss of objectivity. If anything, since transgender authors are highly unlikely to go back to their birth name (I cannot think of a single case where this has happened) a policy which makes the current name of a transgender author their canonical name would more likely match objective reality. Stewartcbaker

The current policy -- which uses the "most recognized name for that author within the genre" rule -- is objective because anyone can review our Summary data, count the number of genre works published under each name and compare the results. That's what we use to decide that it's time to do a canonical name/pseudonym swap.

The originally posted form of the proposed policy -- "When an author is known to be transgender, the canonical name should be their current name" -- would only be objective if all authors used the same name after the gender change. However, that's not a given even outside of the publishing world: people get married, divorced, etc. In addition, a lot of authors use multiple forms of their name and/or pseudonyms either sequentially or simultaneously and we need to account for that. Typical scenarios include:

• There is another author who is using the same name. That was the reason why Lawrence Evans became "Lawrence Watt-Evans" when he began his career 40 years ago.
• Certain genres require or at least encourage gender-specific pseudonyms. That was why Mack Reynolds published his gothic novels as Maxine Reynolds. Ditto Frank Belknap Long/Lyda Belknap Long, Charles Platt/Charlotte Prentiss, etc.
• In traditional publishing, if a new author fails to live up to the distributors' algorithms' (sic!) expectations, he or she has the option of relaunching his/her career under a different name.
• Many authors prefer to use different names for different sub-genres even if it has nothing to do with gender. For example, many urban fantasy/paranormal romance authors use at least two different names.
• Some authors use different names for lighthearted-vs-grimdark works. There is a good reason for it. For example, when J. A. Sutherland, the author of the "Alexis Carew" series of space operas, published a much darker book (Of Dubious Intent) under his own name, reader reviews were negative because they had been expecting more stuff in the Alexis Carew vein. Within a couple of months he pulled the book and reissued it as by "Richard Grantham", a new pseudonym.

The list goes on and on. An author may have a pseudonym for the adult market, a pseudonym for the juvenile market, perhaps another one for the YA market, etc.

This may not be obvious when you look at it from the outside, but our robot processes up to 30,000+ new books every months, so we get to see a lot of permutations. Any policy changes should account for this variety, which is one reason why I always hesitate to make exceptions for special cases.

To go back to the last form of the proposal, "the canonical name should be the name under which they have most recently published" may work for authors who publish under one name post-gender change, but it wouldn't work for authors who use multiple names because "most recently published [name]" would be constantly changing. Ahasuerus 19:31, 26 September 2018 (EDT)

The current policy seems to admit that it is already somewhat subjective in any case. For instance, it notes that an argument could be made either way for Robin Hobb / Megan Lindholm. Stewartcbaker

The Robin Hobb / Megan Lindholm case was actually fairly straightforward. Back when the project started (1995) the vast majority of her output had been as by Lindholm. The first Robin Hobb book came out the same year that we started, so Lindholm was the obvious choice at the time. In addition, "Robin Hobb" was an undisclosed pseudonym at first, so we didn't have the two names linked in the system. Some of us knew her identity as individuals, but we only enter publicly available data into the database. (There are some "semi-disclosed" pseudonyms like Wrigley Stuart, but that's a whole different headache.)

In any event, in due course of time the pseudonym was publicly disclosed and we linked the two names. (The details were different because we were using very different software back then.) Eventually she became much better known within the genre as "Robin Hobb" -- the threshold as per the current policy -- and we swapped the canonical name.

That said, there are times when it's not clear which "working name" should become the canonical name. Sometimes an author is publishing under multiple names and there is no clear favorite. Other times an author may have a dozen stories published as by "X" and 2 novels as by "Y", so it's not clear which name is better known. In cases like that we usually wait for the dust to settle. Ahasuerus 19:52, 26 September 2018 (EDT)

I can certainly appreciate the slippery slope argument, and the worry that allowing for one exception will invite others. However, I would argue that the current one-size-fits-all approach, while it might be easier to maintain, is not necessarily the most accurate bibliographic approach regardless. (As you mentioned on my talk page, there's a feature request to differentiate pseudonym types. Stewartcbaker

I agree. Actually, I have been thinking of ways to address this issue. One simple way to improve things that wouldn't take much time would be to revisit what we display on "Pseudonym" pages. At the moment William Fitzgerald Jenkins says "Pseudonym. See: Murray Leinster (or view all titles by this pseudonym)". Then, when you follow the link to the Murray Leinster page, you are told that "William Fitzgerald Jenkins" was his legal name, which is confusing. Similarly, Владимир Набоков tells you that it's a "pseudonym" even though it's merely the Russian form of Vladimir Nabokov's name.

We can't easily change the way the software handles pseudonym associations throughout the system, but we can easily change the way things are displayed on Summary pages. For example, we could say something like "Canonical name: Murray Leinster (or view all titles published as by William Fitzgerald Jenkins)". In the case of Anya Johanna DeNiro, it would say "Canonical name: Alan DeNiro (or view all titles published as by Anya Johanna DeNiro)". It seems like it would make the software behavior less confusing across the board. Ahasuerus 20:33, 26 September 2018 (EDT)

The linked discussion has other relevant comments. Ahasuerus 15:55, 26 September 2018 (EDT)

Right. Re Zac Brewer, it is more than high time to change his canonical name; he now has several nonspec books under that name. (I will do it this evening). --Vasha (cazadora de tildes) 16:24, 26 September 2018 (EDT)

Have his SF books been republished as by Zac Brewer? All I can find is 2 new books published in 2016-2017, apparently not SF. I suggest we wait until this issue has been discussed. Ahasuerus 16:39, 26 September 2018 (EDT)

I will refer you to Stewart's argument that it is important to accept a trans author's name change promptly and automatically so as not to add to the storm of voices challenging their validity. I'm sure Stewart can say it more eloquently. --Vasha (cazadora de tildes) 16:45, 26 September 2018 (EDT)

That's part of his proposal. Until a new policy is adopted, the current one remains in effect. Ahasuerus 16:50, 26 September 2018 (EDT)

Some things in that 2016 discussion don't reflect very well on the ability and willingness of the ISFDB (as a consensus-based collective entity) to respond promptly and without friction to the needs of trans authors. --Vasha (cazadora de tildes) 16:24, 26 September 2018 (EDT)

Sometimes proposed policy changes are adopted or rejected quickly, but in many cases there are multiple iterations of the same discussion until consensus can be reached. There are also times when a proposed policy change is addressed in the software (e.g. transliterations) or some other changes happen that make the issue moot. It's all part of the process. Ahasuerus 16:48, 26 September 2018 (EDT)

That's why it would be very valuable to have a simple explicit statement in policy which would obviate the need to debate each individual case. --Vasha (cazadora de tildes) 16:24, 26 September 2018 (EDT)

New policies are typically formulated after dealing with multiple similar cases. Once a commonality is established and the side effects of different approaches become clear, policy proposal are made and debated. As I mentioned earlier, it tends to be an iterative process. Ahasuerus 17:00, 26 September 2018 (EDT)

Is there a specific way policy proposals are decided, by the way? Or is it purely consensus-based, and based on discussion on this page? Just curious. :) Stewartcbaker 17:11, 26 September 2018 (EDT)

Policy changes are made once consensus has been reached on this page. Ahasuerus 17:18, 26 September 2018 (EDT)

(And I do not overlook how much it adds to the stress of transition for someone to have it debated by us whether their change of name is "yet" ready to be fully implemented ...) --Vasha (cazadora de tildes) 16:24, 26 September 2018 (EDT)

I see no issues with changing their canonical name once something has been published under the new name and we have some indication (a post, article, etc.) that is their preferred name now. Unless we rewrite the software quite a bit, we must first have in the database an entry using the desired canonical name. Once we have that, it's just the matter of taking the time needed to revariant everything accordingly. Depending on the number of entries by the person that are already in the database, this can be a quick process or a very slow process since we don't have an automated way to change everything. ···日本穣 · 投稿 · Talk to Nihonjoe 18:05, 26 September 2018 (EDT)

To clarify, these are the steps I see needing to happen to change the canonical name:

1. Have an existing entry in the database using the new name.
2. Have a post by the person, an article about the person, or some other reliable source that says this is the new name (just like we currently need to variant a pseudonym to a canonical name).
3. Variant the existing entries to the new name.

This makes the process fairly easy, and it doesn't require debating every instance. If they meet the criteria, that's how it's handled. Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 18:10, 26 September 2018 (EDT)

We would definitely need at least one title record associated with the new form of the name. Otherwise there would be no author record to swap the canonical name with. Ahasuerus 20:10, 26 September 2018 (EDT)

Maybe we need to look at other names for "canonical"? Where on the site do we use the word "canonical"? And "pseudonyms" are already listed as "used these alternate names", so I don't see an issue there. ···日本穣 · 投稿 · Talk to Nihonjoe 20:34, 26 September 2018 (EDT)

"Canonical name" is used in Advanced Search and when editing author records. The latter links to Template:AuthorFields:CanonicalName. As an aside, that template needs to be adjusted. It contains a general discussion of canonical name/pseudonym issues even though it is linked from the Edit Author page. It should cover two other areas: explain that the field can only be edited by moderators and warn that changing the canonical name will change all titles and publications that use that name. Ahasuerus 20:55, 26 September 2018 (EDT)

Softwarewise, is it possible to have title records for an author with no publications using that name -- I mean, as long as the titles have titles-with-publications varianted to them? --Vasha (cazadora de tildes) 20:32, 26 September 2018 (EDT)

There must be at least one title record using the name. Each title must have at least one publication that matches it. ···日本穣 · 投稿 · Talk to Nihonjoe 20:35, 26 September 2018 (EDT)

It's true that each publication has to have a "reference" title whose authors match the publication authors. However, the opposite is not necessarily true. You can have a title without any publications which match that title's authors. For example, consider collective pseudonyms like Kimberly Kinrade. The canonical names are {A|Dmytry Karpov-Kinrade}} and Lux Karpov-Kinrade. If you run an Advanced Publication Search looking for publications associated with either of the canonical names, the search won't find anything. Similarly, we have a lot of Japanese author records without Japanese publications.

So from the technical perspective we could variant an author's titles to a name that otherwise doesn't exist in the database and everything would still work. Whether it's desirable from the policy perspective is a different matter. Ahasuerus 22:43, 26 September 2018 (EDT)

However, please note the issue of authors with multiple working names which I discuss above. Ahasuerus 20:10, 26 September 2018 (EDT)

It is true that the proposed wording, "the canonical name should be their current one," cannot be followed to the letter when trying to decide whether to choose Alex Acks or Alex Wells as that author's canonical name. But it helps that this is being proposed as a relatively narrow policy addition to help out trans authors. We therefore can state "should be a current one" (if they are currently using several). We do not need to deal with the entire complexity of the issues you have mentioned in order to be courteous to authors in this one respect. --Vasha (cazadora de tildes) 20:49, 26 September 2018 (EDT)

We certainly try to be courteous when dealing with authors. If nothing else, it's a mutually beneficial relationship: we get bibliographic data while they get exposure. However, our primary objective is creating accurate and comprehensive bibliographies. For this reason we have rejected multiple requests made by authors to delete or change bibliographic data which was known to be correct.

As a general rule, we do not let author preferences dictate our policies. As I indicated earlier, one of my concerns is the creation of a precedent which can lead to a slippery slope. In addition, special policy provisions for certain exceptions have proved problematic in the past. In my experience, policy statements should be formulated as broad rules to the extent possible with exceptions being the last resort.

Re: the following wording:

• [the canonical name] should be a current one" (if they are currently using several)

what would be the method for determining the canonical name? The most commonly used name since the gender change? Ahasuerus 21:27, 26 September 2018 (EDT)

While scanning our data looking for examples that may clarify various permutations, I came across Poppy Z. Brite's bibliography. Born Melissa Ann Brite, this author now uses the name Billy Martin on LiveJournal. Wikipedia and some other sites also mention hormonal therapy ca. 2010. (I have been unable to determine whether the legal name has been changed.) At the same time there is no indication that he has used any names other than our canonical name "Poppy Z. Brite" in print since changing gender -- not that there is much post-2010 material since he apparently retired around the same time.

What would be this author's canonical name under the current proposal? Ahasuerus 23:08, 26 September 2018 (EDT)

P.S. The author's Web site is http://www.poppyzbrite.com/ . It says "Poppy Z. Brite Official Site" and doesn't mention "Billy Martin". The Wikipedia article is under "Poppy Z. Brite", but the first line says:

• Billy Martin (born May 25, 1967), known professionally as Poppy Z. Brite

Wikipedia also has a page for Billy Martin (author) which redirects to "Poppy Z. Brite". So they use the author's professional name as the name of the main article, which is their equivalent of the ISFDB canonical name. Ahasuerus 09:55, 27 September 2018 (EDT)

Legal and Other names

It sounds like, even if a policy change is made, that policy change will only affect authors who have published works under their new name, because otherwise a new variant(?) isn't created for that author anyway. Is that a correct understanding? In that case, I don't think that author's name change would affect their canonical name, since there's no new publications under the new name.

I don't know what ISFDB's policy on including legal names is in author entries, or when/why they're included, but I do think the author's current name should be added to the bibliography. If the author is now Billy Martin, and has been for almost 8 years, referring to them by their birth name is not only insulting and potentially harmful, but it's obviously factually incorrect and has been for close to a decade, apparently. Unless the preference is to include factually incorrect information in the database, of course...? An "Used these alternate names" for the birth name may still be appropriate if that information is in some way relevant.

I will say that in terms of what's best for transgender people it isn't preferred to wait for a legal name change, as some people may not have that option depending on where they live, or may not be able to actually afford that option. In many places, getting a legal name change is expensive, especially for low-income individuals. Stewartcbaker

It is not factually incorrect as all the information in the database is entered under the name used when the publication was released. That they now use a different name is irrelevant to ISFDB unless they publish one or more qualifying works (meaning they qualify to be listed here) under the new name AND the connection between the two names is publicly known. In the "Poppy Z. Brite" case (given the links provided by Ahasuerus, above), the new name is publicly known, so that's one criterion met. There are, however, no works in the database under "Billy Martin". Therefore, until and unless such a work is added to the database, the entry will continue to be under the only known name under which the works were published.

Arguments about whether it's insulting or potentially harmful are irrelevant to this discussion. ISFDB presents information exactly as it was published, including typos and all. If someone finds that insulting or potentially harmful, that is not ISFDB's problem.

That said, I have no problem with updating their legal name (there is a field for that in the database already) as long as the change is backed up by solid and public sources. ···日本穣 · 投稿 · Talk to Nihonjoe 15:00, 27 September 2018 (EDT)

One thing that is important to keep in mind and, more often than not, is not intuitive when new editors join the project, is that we deal with multiple different types of names:

• Names as they appear in publications. These are always entered into the database as they were published, including pseudonyms, typos and so on, except for punctuation and case regularization (and interviewee names due to software limitations.)
• Canonical names, i.e. names used to organize authors' Summary bibliographies on one page. Currently they are defined as "the most recognized name for that author within the genre". If an ISFDB bibliography is incomplete -- which is common when dealing with non-English authors -- the canonical name may not be the same as the name with the most publications in the database. Ditto collective pseudonyms since they can't be canonical names.
• Legal names. They are currently defined as "the most recent legal name for the author" and are captured in author records. They are not used for anything else at this time.
• Names used by the author socially. They are currently not captured in the database.

In the case of Billy Martin/Poppy Z. Brite, all of the author's works have appeared as by "Poppy Z. Brite" and are captured as such in the database. The canonical name is also "Poppy Z. Brite". The legal name was originally "Melissa Ann Brite"; it's not known if it has been changed. The primary "social" name as used on the author's LiveJournal site and apparently elsewhere, is Billy Martin. This can be confusing at first, but, if you think about it, people can have multiple names in real life as well. Depending on circumstances, people go by their full legal names, first names, last names, nicknames, variants of first names ("Leonard" - "Len" - "Lenny") and so on. Ahasuerus 15:54, 27 September 2018 (EDT)

One possibility for instances like Brite: we can add a note to the Author page stating something like "This author has changed their legal name to Billy Martin, but no works have been published under that name." That would at least put the information out there for people to see. ···日本穣 · 投稿 · Talk to Nihonjoe 17:59, 28 September 2018 (EDT)

In this case we don't know if the legal name has been changed. Even it has, the new legal name may not be the same as the "social name" used on LiveJournal and elsewhere, e.g. "William Martin" vs. "Billy Martin". Of course, if we determine that it has been changed, we'll adjust the record accordingly. For example, see the Simon Hawke record, whose Note field reads "Born Nicholas Valentin Yermakov, but had his name legally changed to Simon Hawke, after using that as a pseudonym for some time."

Since we are not sure that the legal name is still what our record states, we should probably remove "Brite, Melissa Ann" from the "Legal Name" field and add something like:

• Born "Melissa Ann Brite"; publishes speculative fiction as "Poppy Z. Brite". Changed the name [self-identification?] to "Billy Martin" ca. 2011; current legal name is unknown.

to the Note field. Ahasuerus 18:29, 28 September 2018 (EDT)

That works for me until more solid information sources can be found. Speaking of sources (and this may need moving to another section for proper visibility), should we add a way to enter sources? I think it might be good to have a separate sources table, and have a way to attach a single source to more than one author, title, or publication. ···日本穣 · 投稿 · Talk to Nihonjoe 18:44, 28 September 2018 (EDT)

This approach does a good job of balancing the legal name (which you all want for ISFDB purposes) with what's likely to be accurate in the real world. For language, maybe:

Legal name unknown. Currently known as Billy Martin; publishes speculative fiction as "Poppy Z. Brite." Other names associated with this author: Melissa Ann Brite.

That way the language might be more consistent in other cases where an author's name is likely to have changed for whatever reason, and this can be applied more broadly to various authors. (In an ideal world, the birth name would just be deleted for trans authors as no longer relevant, but I get the impression that will not fly here and this is a decent compromise.) Stewartcbaker 21:46, 28 September 2018 (EDT)

Actually, ISFDB:Policy makes a distinction between bibliographic and biographical information. For bibliographic data the policy is that we do not "delete publications/titles that are known to have been published". This includes the way authors were credited in publications. However, for biographical data, the policy is that:

• If a living author (or their authorized representative) requests that the ISFDB remove the author's detailed biographical information, the ISFDB will comply after confirming the requester's identity. The ISFDB will remove as much biographical data as needed in order to accommodate legitimate privacy concerns while preserving, to the extent possible, the work of the editors who have compiled the data. A note will be added to the author's record explaining what type of information has been removed and why.

If Billy Martin or another author (or their authorized representative) requests that we remove the current (or past) legal name, we will honor the request as per the current policy. Ahasuerus 23:25, 28 September 2018 (EDT)

Hi folks. Is there anything else I can do at this point to provide clarity on why this name change would be helpful to users, or how it might work in specific unique cases? If not, I'll keep checking in until I notice there's been consensus reached. I'd be happy to work with you all to come up with a more comprehensive proposal that makes sense for all authors, rather than an exception, too, if you like. Stewartcbaker 17:42, 28 September 2018 (EDT)

Sorry Stewart, I haven't forgotten. I will write more soon --Vasha (cazadora de tildes) 16:54, 3 October 2018 (EDT)

No worries! Just wanted to make sure I wasn't misunderstanding how this process works. :) Stewartcbaker 15:41, 4 October 2018 (EDT)

There are a few things that we could do on the software side to clarify things. For example, the word "Pseudonym", which we use on Summary pages, could be replaced with something more generic -- like "Alternative name" -- which would help minimize confusion when our "pseudonym" is actually the person's legal name (see the Murray Leinster/Simon Hawke examples above.)

I think this would be a great step to take that would immediately make a lot of things clearer. Stewartcbaker 15:41, 4 October 2018 (EDT)

We could also try to rearrange the order of titles and names in the Contents section of the Publication page. As I wrote during the original discussion, the current order, which has evolved over time, is not particularly intuitive and can confuse users. For example, here is the order of translated titles in the Contents section:

• <translated title> <canonical author name> <canonical title of the story> <pseudonym/transliterated author name>

I am pretty sure we should be able to do better than that. However, we'll need to be careful because the Contents section handles a dozen+ different permutations: serializations, translations, different vs. identical dates/titles/authors. A redesign would have to account for all of them. Ahasuerus 20:31, 3 October 2018 (EDT)

As a librarian (my day job), our preference in cataloguing books and other materials is always to match the cover, and note variant names elsewhere. I don't know if that kind of model would add more complexity to ISFDB or be totally unworkable, but from an end-user perspective it would certainly be less confusing. Since the point of the canonical name is (I believe?) largely to aid in allowing people to connect variant author names, what do you think of that approach? Or would it not be well-suited for ISFDB? Stewartcbaker 15:41, 4 October 2018 (EDT)

It looks like there are a few separate issues here.

The first one is what we want to capture in the main "Title" field. This was something that we needed to standardize at the very beginning of the project. After considering our options, we settled on the form of the title displayed on the title page. This rule applies to books as well as to Contents titles. The main reason why we decided against using covers was that, in our experience, book covers often include promotional information which can be hard to distinguish from subtitles. In addition, some verification copies may be missing their original dust jackets.

The second issue is what we capture in additional title-related fields. At this time we have only one additional field: "Transliterated Title", which supports multiple Romanized forms of the title. MARC21 has a number of other related fields and sub-fields; we have occasionally discussed adding some of them, but it's still very tenuous.

The third issue is how we want to display the captured data on various bibliographic pages (Summary, Title, Publication, Series, etc.) The display logic behind the Contents section of Publication pages is particularly complex because it supports so many different permutations. It's this part that I think we can improve by tweaking the software. Ahasuerus 11:44, 5 October 2018 (EDT)

Canonical names for transgender authors - Consensus?

Hi again. Sorry if this is clearer to regular ISFDB folks, but I'm having a hard time telling if a decision was ever made about this. There seems to have been broad consensus, generally speaking, that it was an issue worth tackling, if not necessarily agreement on a way to do that.

If more discussion is needed: What are the pinch points? How can we move forward? Thanks. Stewartcbaker 16:44, 3 October 2018 (EDT)

I don't think that at this time a consensus to have a transgender author change his/her canonical name upon request can be reached. The problem imo is that as far as I can see, there's no easy implementation that does not muddy the biobliographic contents of isfdb. While I understand that a person that has changed genders would like to be known with his/her changed name, and that it has a significant emotional meaning, history can't be erased (well, it shouldn't); the fact remains that as long as the author has more publications/is better known with his/her old name, the current rule has it that that will be the canonical name. Now, that said, I'm sure the community is willing to accommodate in other areas where changes are easier and possible to effect. Just my 2cents. MagicUnk 17:00, 6 October 2018 (EDT)

(unindent) As per this announcement, the term "pseudonym" has been changed to "alternate name" throughout. Ahasuerus 18:08, 17 October 2018 (EDT)

Magazine issue dates

Dirk just brought to my attention that there's an omission in the rules for dating magazine/fanzine publications. In both Help:Screen:NewPub and Template:PublicationFields:Date, which use the same wording, it does not say what to do with a magazine that does not have a cover date. The section ends with the words, "For magazine cover dates which cannot be assigned to a specific month, use the year only, e.g. 'Spring 1943' should be entered as '1943-00-00'."

Clearly, we need to get some info about the date from the publication date if there is no cover date. But isn't it the current practice to use just the year in such cases? If so, I propose adding another sentence: "If the magazine has no cover date, find out the publication date and enter the year only." --Vasha (cazadora de tildes) 20:28, 30 September 2018 (EDT)

Well, the problem with using the publication date, i.e. the date when the magazine issue was made available to the public, is that magazine publication dates traditionally do not match the cover date. A January issue would be mailed to subscribers in late November of the previous year, ditto February-December, etc. So one could plausibly argue that a magazine issue without a cover date published in December 2017 should have its date set to 2018-00-00.

Also see this discussion which was started in August but didn't get much traction. Ahasuerus 21:16, 30 September 2018 (EDT)

What bibliographic guides did you consult when setting up these rules? --Vasha (cazadora de tildes) 21:30, 30 September 2018 (EDT)

The magazine data entry rules were primarily defined by the editors who were working on magazines back in 2007-2009. We could try to find the discussions which occurred then, but, as I recall, the rules were not based on third party bibliographic standards any more than the rest of our data entry rules were. As I wrote a few weeks ago:

• Back when we were getting ready to resume accepting submissions (2006), we needed to agree on data entry standards, so we reviewed the standards used by OCLC and other bibliographic projects. We thought that their rules were excessively complex and said "Surely we don't need that level of complexity! We can come up with simple common sense standards that anyone would be able to understand!" And so we proceeded to create our own standards, simple and logical.

• Within a couple of years we realized that we had been wrong. The OCLC and other data entry standards were complex because they were intended to reflect real world complexities. As the scope of the ISFDB project expanded, we had to create more and more rules to account for more and more permutation and borderline cases. Our rules grew more and more complex. The rules for magazines are different than the data entry rules for books. The rules for non-genre magazines are different from the rules for genre magazines. Novel-length works published in magazines, whether serialized or not, present unique challenges and we have special rules for them -- see Help:Use of the SERIAL type. And so on and so forth.

I'm surprised that this rather major matter has been overlooked so long. --Vasha (cazadora de tildes) 21:30, 30 September 2018 (EDT)

I am not sure what most of our editors currently do if there is no cover date. I have mostly entered magazines from my pulp and digest collection and they rarely have this problem. We should probably ask the editors who have been working on magazines for years, e.g. User:Rkihara and User:Rtrace. Ahasuerus 21:56, 30 September 2018 (EDT)

I think that in the case of 'Theaker's Quarterly Fiction' (as well as in other, similar cases) the magazine does imply in its title that there's a well-defined chronological order (four times a year, each time covering three months), and the issue grid as well as the yearly order should reflect that. Stonecreek 00:05, 1 October 2018 (EDT)

As some of you may have recognized, I do much work on the (in)famous Perry Rhodan series which is published weekly, and not a single one of the issues has a cover date. Instead we use the schedule and other available information to determine the day of publication. With 52 or 53 publications in one year, the issue grid would become a chaos, if we'd enter only the year.

I'd propose that if we do have available dates (if not from the cover date then preferably from the publisher's / magazine's web site, or from amazon) we should use this for modern magazines (for older ones where no website exists and amazon is unreliable, I think we should stick to just the year), with the exception of magazines published quarterly, which either should officially be published in January, April, July & October or - when it is a seasonally published magazine - in other months.

That said, there are also multiple examples of magazines without a cover date, but well-defined, -known and -indexed dates of publication, for example Italian ones (Urania etc.), or magazines inside a publication series. Stonecreek 02:06, 1 October 2018 (EDT)

Is anyone but me disturbed by using sometimes cover dates, sometimes publication dates? Consider it from the point of view of someone reading the data. It'd be nice if they could look at the date and know, "That is the cover date. If it is just a year, there is no cover date, or no exact one." Whereas if editors are all entering dates according to their own ideas, it'll be really hard to know what they meant by it when you look at it afterward.

If there is a cover date, we use that, but we do discuss the case(s) where there is none. Stonecreek 15:36, 1 October 2018 (EDT)

My point is that if you saw an issue dated 1977-02-00, how would you know if that is an issue with a cover date of February, or one that's been assigned a date according to some individual rule? Maybe the way around that is to get better about always putting the cover date in the title of the publication. For instance, what Beneath Ceaseless Skies displays at the top of the issue is something like "Issue #258, August 16, 2018." Currently we would enter that in the database as "Beneath Ceaseless Skies, #258," and date it 2018-08-16. But if we put the date in the title, it makes clear that 2018-08-16 is a cover date not a publication date. That way, people could always know the cover date by looking at the title, and the rules for assigning a date could be as elaborate and individualized as we want. --Vasha (cazadora de tildes) 15:57, 1 October 2018 (EDT)

I personally think the magazine's issue identification should be captured as the title, with some disabiguation guidelines for when the magazine does nothing to distinguish one issue from another, and the publication date should be the date of publication, as best we are able to determine (defaulting to cover date, falling back on assumed date for things like "quarterly"). If there's no date in the magazine's identification, we shouldn't add one; if there's volume and issue information in the identification, we should use it. The grid might have to be tweaked to look at both the title date and the publication date and to prefer one over the other. Notice in the Theaker's Quarterly example, the issues are placed by the publication date, even though there is no cover date. Would it bother me if a "January" issue of some magazine were placed in the prior year's December box? Probably. But what's the purpose of the grid, to show a sort of chronology of issues or to give a visual display of the actual dates of publication? --MartyD 07:22, 2 October 2018 (EDT)

Note that all of the major awards stipulate that the year of a work's eligibility is according to the cover date, not the publication date. Diverging from this potentially increases confusion instead of clarifying: do you want to have a bunch of stories with notes like "Although dated December 21, 1992, this story was nominated for a 1993 Hugo Award because of having been published in a magazine with a cover date of January 1993"? Cover dates are also used by our standard secondary sources. All in all, it would be a bad idea to abandon this and start doing our own thing. --Vasha (cazadora de tildes) 09:41, 2 October 2018 (EDT)

Christian has a point that undated issues don't look good in the issue grid. But maybe that is better addressed at the display end. --Vasha (cazadora de tildes) 15:15, 1 October 2018 (EDT)

It is not just bad-looking in the issue grid, but on a given author's page as well: Consider an issue of a seasonally published magazine, say the Winter 1977/1978 one (published in December 1977). All entries would appear on the author's page before the very most of the author's other 1977 titles in the same title type category - even before the ones published in January of the same year. There are cases when we son't have trustworthy information (be it as a cover date or, in absence of that, a publication date), and then it has to be that way, but not in the cases in question. Stonecreek 15:36, 1 October 2018 (EDT)

(unindent) Let me just copy-and-paste two proposals from the August discussion so that everything would be available in the same place:

• Sometimes magazines don't have a date on the cover, but do have one on the masthead. How about, "If there is no date on the cover, use the date on the masthead; if there is none there, use the year only. Do not use a publication date." --Vasha (cazadora de tildes) 18:47, 13 August 2018 (EDT)
• A couple of thoughts:
  1. If the publication date is known/stated and is consistent with the cover/masthead date, why not use it? I don't see why we'd want to record a less exact date.
  2. For magazines with no cover/masthead data, if the publication date is known it should be used as-is (not in the title, though). --MartyD 08:03, 14 August 2018 (EDT)

Ahasuerus 11:01, 1 October 2018 (EDT)

I don't think I've ever encountered a magazine without a cover date. Using the masthead date seems like a good alternative. Lacking that, I would suggest using the publication data which is always printed near the magazine index. It usually gives the volume, issue number, magazine date, or date of printing or distribution.--Rkihara 12:48, 1 October 2018 (EDT)

Well, there do exist magazines that have no such thing: Urania & Heyne Science Fiction Magazin are such examples, and Theaker's Quarterly Fiction mentioned above seems to be another one. Stonecreek 13:48, 1 October 2018 (EDT)

Issue numbers in title

A bit of an aside: I believe the current rules/guidelines call for the date in that Beneath Ceaseless Skies example to be included in the title in this instance. There is a strong preference to have dates included in titles. Why, I do not know. --MartyD 07:22, 2 October 2018 (EDT)

As I read the guidelines, The issue of BCS above should be titled simply "Beneath Ceaseless Skies, August 16, 2018" This template (under Missing or variant dates) states that the date is preferable. It does go on to suggest an exception for Interzone, but further states that only the publication number should be used (I realize Interzone has both). I'm not comfortable with using issue numbers in the title when a date exists if only for the reason that it has never been defined when to do so vs. when to use the currently documented method. Nor has it been documented on how to reflect the issue number (pound sign, "No.", "Number", volume and number). If we're going to have three different standards on how to title magazines (date, number, both), we really need to decide when to use each one. We should also decide whether we want to reflect issue numbers in a standard way. --Ron ~ Rtrace^Talk 21:16, 2 October 2018 (EDT)

I like having both a number and a date in the title. For one thing, the issues are so often referred to by number that it needs to be easy to find them that way. For another thing, if a magazine has been published irregularly, you have to check the numbers to see whether any issues are missing. I don't see why issue numbers have to be standardized, though; just quote whatever the magazine says. --Vasha (cazadora de tildes) 21:41, 2 October 2018 (EDT)

Having the standard as whatever appears in the magazine still raises questions. Do we go with what is on the cover, masthead or spine? What do we enter if we don't have a primary source? A particularly long example would be "The Baum Bugle, Volume 61, Number 3 (Whole Number 180), Winter 2017". That's from the masthead. The cover omits the whole number. How would it parse in the issue grid? I fear that a lack of standard would result in many different formats within a single magazine's issue grid. For example, the first issue of Interzone would be "Interzone, Vol 1 No 1, Spring 1982" going by what's on the cover. The latest issue would be "Interzone 274, March-April 2018". This assumes that we continue to standardize the dates. Regardless, adding issue numbers in addition to dates, is going to necessitate a very large number of edits to bring everything to the new standard. --Ron ~ Rtrace^Talk 22:23, 2 October 2018 (EDT)

If you are expecting everyone here to actually agree on a standard way of entering issue numbers, don't hold your breath while you're waiting :-) We couldn't even decide whether or not to put spaces in front of an ellipsis last summer. A step forward would be to explicitly allow issue numbers (certainly, we shouldn't remove any that are already there) and poll current editors about their preferences and practices, hopefully leading to some spontaneous unification. --Vasha (cazadora de tildes) 23:03, 2 October 2018 (EDT)

Sounds like a plan to me. Christian Stonecreek 00:18, 3 October 2018 (EDT)

I'd be happy to see is consistency within a magazine. And the format / style could depend on the nature of the magazine's covers, masthead, issue schedule etc. If you want to make a 'rule', have it reflect what the software can handle. ../Doug H 08:32, 3 October 2018 (EDT)

We already have two documented standards: (A) Date Only, which is stated as preferred, and (B) Issue Number Only, which is used when there is no date, and in two other ill-defined cases (i.e. I can't determine when the format is supposed to apply). What is being proposed is is a third standard of (C) both Date and Number, which has been used in the project, but again, no one seems willing to define when C should be used instead of A or B. Further, for those formats that include issue number, there is no standard, stated or otherwise, on how the number should be reflected ([2], [3], [4], [5], [6], [7] all follow different conventions and do not all match what is seen on the publication). I am not opposed to developing a new standard, but we need to have a clear definition on when to use each of the standards. We also need to have a clear definition of how issue numbers should be reflected, whether that be standardized as we do with dates, or otherwise. If we do end up with a date plus issue number standard, we need to know which of those comes first, and how to separate each of the three pieces of data (spaces, commas?) in the title field. I strongly disagree that we should formally allow a third standard without getting a consensus on when and how to use it. I know that getting a consensus can take some time, but that really isn't a reason to abandon the process. I wouldn't delete issue numbers outright, and I certainly wouldn't alter a verified publication without a discussion with the primary verifiers. However, moving the issue number from the title to the notes is merely bringing an publication in line with the stated standards. No data is lost. I've certainly done this to unverified zines, though I generally will only do so if I have some other reason to edit the zine. --Ron ~ Rtrace^Talk 20:00, 3 October 2018 (EDT)

I agree with Ron that it's a fairly big can of worms. For example, back in the day the copyright section of Stupendous Stories, March 1952 might have said something like "Vol. 12, No. 3, whole number 147". We'd need to decide which (if any) components to move to the title field.

My current thinking is that Stupendous Stories, March 1952, vol. 12, No. 3 would not be better than what we have now. Ahasuerus 10:37, 4 October 2018 (EDT)

(unindent) Re: the points that Vasha raised earlier, i.e. "issues are so often referred to by number that it needs to be easy to find them that way. ... if a magazine has been published irregularly, you have to check the numbers to see whether any issues are missing."

If you think about it, "issue number" is a separate data element just like "price" or "page count". The best way to handle a separate data elements would be to have a separate field for it. Once it's been added, we could modify the issue grid to display its value under the issue title.

If we were to decide to add a new field, we would need to address a few issues, including:

1. What do we do when the issue number is also a part of the title -- see the Interzone example
2. What do we enter in the new field for issues like the one that Ron mentioned earlier: "The Baum Bugle, Volume 61, Number 3 (Whole Number 180), Winter 2017"

Ahasuerus 11:15, 4 October 2018 (EDT)

I do not see the repetition as a problem (better to have the same information in two places than not at all - plus the new field is useful for searches (we are a DB after all). Some magazines just use their numbers for identification, that does not mean we do not need the number in a separately searchable number (same way as we do keep a date even if the title has it).

As for the special cases as above, it will need to be either 61.3 or 180 - whatever a reader of the magazine would expect (we have similar problem for Analog for example or any of out other major magazines). Or maybe make it a multi-field allowing both values. Annie 12:41, 4 October 2018 (EDT)

Upon further reflection, a magazine issue can have up to 3 separate data elements: "volume number", "issue number", "whole issue number". From the design perspective, it may be better to capture them in three separate fields and then display them intelligently.

On the data entry side, we could have a single line for capturing this information, e.g.:

• Volume: [field] Issue: [field] Whole Issue: [field]

In the New Publication data entry form we would gray out these three fields except for MAGAZINEs and FANZINEs just like we gray out the two Series fields for variant titles. In Add/Clone Pub, we would add pop-up validation.

Capturing this information in a structured manner would make it possible to create a cleanup report that would look for gaps. The report would be similar to Series with Numbering Gaps. Ahasuerus 12:57, 4 October 2018 (EDT)

That works for me :) Annie 13:10, 4 October 2018 (EDT)

I really like this idea.

I also don't see a problem with keeping magazine titles the way they are if they already have a number in them even when implementing the new field. That would be good for magazines that use an individualistic numbering system, like The Future Fire with their "... 2017.42, 2017.43, 2018.44 ..." We would put that in the title in case anyone searches for the issues that way, but also put "... 42, 43, 44 ..." in the "Whole Number" field. --Vasha (cazadora de tildes) 18:35, 5 October 2018 (EDT)

I like this idea, too, and I agree with Vasha about the titles, which I still think should reflect how a magazine's issue identifies itself. --MartyD 08:06, 6 October 2018 (EDT)

OK, I have created FR 1202 for the new fields. We'll probably want to discuss the implementation details before they are added. Ahasuerus 09:53, 10 October 2018 (EDT)

(Unindent) You could take your idea one step further and have a field for Cover Date while you're at it, which would allow us to capture both the date on the cover (which might not be a standard date, like Winter or Michaelmas or a range of months) and the actual date of publication. --MartyD 08:06, 6 October 2018 (EDT)

One problem with this is that all the contents would then inherit the publication date rather than the cover date, and that would completely mess up trying to find things that are eligible for awards in a given year. But if the magazine's overall date continued to be the cover date, the publication date could be a separate field. (Also, in the case of older magazines, we are more likely to know the cover date than the publication date; thirdly, all existing issues were theoretically supposed to be entered according to the cover date so it'd be a lot of work to convert them.)

Thoughts on division of labor between fields

How about this: the cover date goes in the title, word-for-word as it's stated (which is more-or-less the current practice); publication date goes in its own field; the overall date is derived from the cover date according to the current rules; and if there is no cover date ... well, that's what started this discussion.

The more I think about it, the more it makes sense to divide up information this way. The title field is for recording exactly how the issue describes itself (in both numbering and dating), and the other fields are for breaking that down into database-worthy data.

Having a separate field for the publication date helps with the problem of grid display without changing the current rules for how to derive a date for the issue: the current rules leave us with a lot of year-only dates which don't display well, but the grid could take months from the publication date. Plus, the current rule for what text to put in the grid is (I think) "everything between the first comma and the date" which can lead to pretty bulky text -- if we adjusted that so it would capture only the cover date (between the last comma and the year, say) the grid could display issue number (taken from issue number field) and cover date (taken from title) which would be nice and tidy. --Vasha (cazadora de tildes) 09:19, 6 October 2018 (EDT)

Credits not in the book

Just checking, with reference to the Nemonymous anthologies. All of them have the stories uncredited in the actual book, with the authors being revealed several months after publication. The lists of actual authors are available in various places online, such as in the book's Goodreads entry. Currently the whole series is entered with the stories credited to their authors in the publication record. But that isn't right, am I correct? We should be handling it the way we do known-but-not-credited editors: enter the basic title record as "uncredited" and then variant it to the real name. --Vasha (cazadora de tildes) 13:58, 9 October 2018 (EDT)

Yes, that's correct: we credit the authors as stated in the publication. Stonecreek 14:05, 9 October 2018 (EDT)

Agreed. We credit per the pub. If the pub is uncredited, they are entered as uncredited and varianted to actual author. -- JLaTondre (talk) 17:02, 9 October 2018 (EDT)

Omnibus vs. Novel

In the help instructions for a new pub, the rules for selecting Novel as a type specifically point out that "For example, "The Lord of the Rings", by J.R.R. Tolkien, was originally published in three volumes; the single volume edition is to be categorized as a novel". However, examination of the title record for The Lord of the Rings shows it to be an omnibus. I'm trying to apply the rules to several Jules Verne novels (omnibuses?) and would appreciate any discussion to help clarify the intention. For example, omnibus requires prior publication of the parts - but does a serialized publication count, given that it has not resulted in a publication entry under the title(s)? ../Doug H 12:59, 13 October 2018 (EDT)

This discussion is relevant. There was agreement Lord of the Rings was a bad example, but not much to help whether or how it should be applied in other cases. However, there was little participation in the discussion. Since this has come up again, I'm going to remove the use of Lord of the Rings as an example. -- JLaTondre (talk) 13:12, 13 October 2018 (EDT)

Editors of non-genre periodicals

When entering non-genre periodicals, we usually credit their editors as Editors of Die Zeit, Editors of The Anglo-Portuguese News, Editors of Знание - Сила, etc. However, Help:Entering_non-genre_periodicals says:

• Enter in the "Editor" field "Editors of PERIODICAL NAME". If the actual editor is known with reasonable certainty, you may optionally enter the name(s) of the editor(s) as co-editors, but leave "Editors of PERIODICAL NAME" as one of the editors in any case. [...]
• If the actual name of the editor (or editors) of a non-genre periodical is entered (instead of using the "Editors of Periodical Name" format), it may be desirable to make such a name an alternate name of "Editors of Periodical Name". [...] Alternatively, the actual editor may be listed as a second editor (co-editor) along with "Editors of Periodical Name".

I see the following issues with these instructions:

• Turning a real person's name into an alternate name of "Editors of PERIODICAL NAME" doesn't seem to make sense.
• Crediting non-genre people as co-editors of non-genre publications ("if known") doesn't appear to be desirable. We could probably determine the names of the editors of most major magazines and newspapers at various points in time, but that would defeat the purpose of the original rule about "Editors of PERIODICAL NAME"
• On the other hand, crediting genre people who happened to be editing a non-genre periodical at a time when it was publishing SF may be a good thing. Unfortunately, I am not sure what the best way to distinguish between "genre people" and "non-genre people" for the purposes of this discussion would be. The best I can think of is "already present in the ISFDB database", but it's not perfect.

For now I propose that we delete the section that reads:

• If the actual name of the editor (or editors) of a non-genre periodical is entered (instead of using the "Editors of Periodical Name" format), it may be desirable to make such a name an alternate name of "Editors of Periodical Name".

and discuss the rest. Ahasuerus 17:42, 17 October 2018 (EDT)

I concur with removing the sentence for now. Annie 17:54, 17 October 2018 (EDT)

As parliamentarians say, "Without objection, so ordered." Ahasuerus 14:38, 21 October 2018 (EDT)

(Unindent) Another foggy issue is when a periodical has one special genre issue; should the real name of the editor be used there? If so, then only there? (That is how Tin House Magazine is currently done.) --Vasha (cazadora de tildes) 17:16, 21 October 2018 (EDT)

I am a bit on the fence about that one - I lean towards using the real editors quite honestly. Using the editors as they are written shows clearly that it is actually a genre issue (and we used that as an argument to include these issues when they are in online only magazines) so you can spot it easily in the list of issues. For all intents and purposes, this is a genre issue, not a non-genre one. On the other hand the usual editors are usually part of the team in that issue so it defeats the purpose of not using their names. I wonder if a middle ground is not a better idea here - name the special guest editor(s) and use Editors of for the regular crowd. This way we still have the visual separation, our author(s) get(s) their credit and we do not add a new non-genre editor... On the other hand, we come back to "who is considered one of our writers/editors?" and it is a very easy question. Food for thought. As it is, I would rather list the actual editors than use the Editors of for the special genre issues. Annie 17:43, 22 October 2018 (EDT)

The consequence of putting the editor's name on only the genre issue is that the non-genre issues don't appear on their page. This is actually consistent with the policy that non-genre work shouldn't be on an author's page unless they're "above the threshold" (although in practice, a lot of authors have non-genre work on their page because it was in genre magazines ... that contradictory consequence of two policies is another issue, which has been discussed here but not resolved). So I like doing it this way, I guess. It makes sense that Rob Spillman and Win McCormack have a page in the DB because they edited one genre issue of Tin House, and it doesn't have all their non-genre work (other issues) because they aren't primarily genre authors. --Vasha (cazadora de tildes) 22:40, 22 October 2018 (EDT)

Graphic Novel status?

Hi, I've been reading up on discussions around inclusion/exclusion of Graphic Novels. For what I understand, Graphic Novels are 'somewhat tolerated', especially if they are 'graphified' novels. Has there since the last discussion (2012 I believe) be any evolution in our position? I.e. would Twilight be allowed? Or rather not? Thanks! MagicUnk 15:12, 22 October 2018 (EDT)

If I were to decide, I would not add them (and it is not because I do not like the medium, I quite like it - I just find them out of scope). But as you had gathered, it is a bit of a gray line. Annie 00:27, 23 October 2018 (EDT)

ISFDB:Policy defines speculative fiction to exclude graphic novels. They are currently treated like other non-genre works, i.e.:

• Works (both fiction and non-fiction) which are not related to speculative fiction, but were produced by authors who have otherwise published works either of or about speculative fiction over a certain threshold [are included]

This means that only graphic novels whose credited authors are "above the threshold" should be included. On the other hand, graphic novels which are based on works by "above the threshold" authors but not written by them should be excluded. At least that's the theory -- there are borderline cases where the extent of the original author(s)'s involvement is unclear.

FWIW, Fixer ignores almost all graphic novels. It's a huge can of worms, so he leaves it to human editors. Ahasuerus 10:55, 23 October 2018 (EDT)

Thanks for posting that reference :) We had been accepting a lot more GNs than the policy implies though but that's just how these things work. Under the policy, the one that started this conversation is clearly excluded. And I would like to leave this particular door closed - there had been a lot of "here is that novel/story as a GN" lately and I expect to see more and more of it. And we are not well set for recording those (illustrators matter as much as authors) or have the capacity to cover them (so things like this is in (above threshold author credited as an author) (although I think we are missing a co-author - need to talk to the PV); this is out - (adaptation from a non-above threshold author in graphic format. This Twilight is like the second case - so out it goes. Annie 12:50, 23 October 2018 (EDT)

It's quite clear:). Thank you both for the clarification. Regards, MagicUnk 19:17, 23 October 2018 (EDT)

The "Juvenile" flag and YA titles

It was my understanding that Help didn't explicitly require using the "juvenile" title flag for young adult titles. However, it turns out that Template:TitleFields:Juvenile says:

• Juvenile - Check this check-box if this work is targeted at the juvenile or Young Adult market.

After running a number of queries against the database, I see the following usage pattern.

First, definitions:

• A "YA tag" is a tag that starts with "teen " or contains "young adult"/"young-adult"
• A "juvenile tag" is a tag that starts with "child "/"children" or contains "juvenile"

Note that some titles are on the border between "juvenile" and "YA", so they have both "juvenile" and "YA" tags. With that in mind:

• Total titles with YA tags: 12,047
• Total titles with YA tags and no juvenile tags: 11,030
• Titles with YA tags which have the "Juvenile" flag set: 3,731
• Titles with YA tags and no juvenile tags which have the "Juvenile" flag set: 2,732

This means that of the 11,030 "pure" YA titles only 2,732 -- or 24.77% -- have the "Juvenile" flag set. Granted, I am our most active tagger, which means that my erroneous understanding presumably skews the data.

Basically there is a significant discrepancy between what Help says about the use of the "juvenile" flag for YA titles and the current practice. Possible solutions include:

1. Reaffirm the decision to set the flag for YA titles. Identify YA titles without the flag and set it.
2. Decide not to use the "juvenile" flag for YA title. Change Help accordingly, identify YA with the flag and remove it.
3. Change the "juvenile" flag to a drop-down list, which we could call something like "Audience Age" or "Target audience". Supported values could include "Juvenile", "Young Adult" and possibly more. See field 521 of the MARC 21 bibliographic standard for some examples of what librarians use.
4. Delete the "juvenile" flag.

A new cleanup report may be required to support some of these options.

My personal preference would be 2 (not to use the "juvenile" flag for YA books) or 3 (create a drop-down list). I would be against deleting the "juvenile" flag since it would wipe out the work of many editors over the course of a number of years. Ahasuerus 13:23, 28 October 2018 (EDT)

Relevant here is MARC 21's list of definitions of target audience terms. They divide up non-adult audiences into 4 groups but also have a catch-all "juvenile" term which they define as "intended for use by children and young people, approximate ages 0-15."

I'd be in favor of adopting MARC 21's definition and changing Help accordingly. Note that their "juvenile" (0-15) doesn't exactly correspond to the boundary between their "pre-adolescent" (9-13) and "adolescent" (14-17), which is understandable for blurry categories; target ages 14-15 are a judgment call, and probably make up a significant portion of the books we have tagged both "juvenile" and "young-adult."

This is an instance where it's very useful to have an external standard developed by experienced librarians—different ISFDB editors have very different ideas of what "juvenile" means, so this external standard provides both a good compromise and needed clarity. (My personal "juvenile" would only cover MARC 21's "pre-adolescent," or "middle-grade" books, but in fact for tagging and flagging in this database I have been using MARC 21's "juvenile" definition for the past two years.)

As for why we should not use the juvenile flag both for books tagged "juvenile" and for books tagged "young-adult"—well, that's down to the issue of what the purpose of the flag is. I can only say that "young-adult" books are also popular among adults whereas this is significantly less true of juvenile books (with a few significant exceptions like The Hobbit and the earlier Harry Potter volumes), so it seems sensible to have the flag marking this difference in potential audiences. The flag makes it easy to exclude juvenile books from a search. Seems to me (subjectively) that someone searching for "fantasy" would not mind having YA books mixed in but juveniles wouldn't interest them. --Vasha (cazadora de tildes) 14:17, 28 October 2018 (EDT)

(after sleeping on it) I think what we are saying here is that a title can appeal to multiple target audiences: juvenile/YA or YA/adult. If so, then what we are really dealing with are two separate data flags: "juvenile" and "young adult".

Instead of trying to decide what the "primary audience" is -- compare MARC 21's "When ... appropriate for more than one target audience, the code is recorded for the primary target audience" -- we could use the old adage (inevitably attributed to Yogi Berra): "When you come to a fork in the road, take it" and create two separate title flags, one for "juvenile" audiences (younger than teenagers) and one for "young adult" audiences (teenagers.) That way our users could search for young adult and/or juvenile titles in Advance Search while editor wouldn't have to try to determine what the intended "primary audience" is.

Of course, we could take it even further and create separate flags for "chapter books", "new adult books", etc, but I don't think it would be the best way to spend our resources. Ahasuerus 15:11, 28 October 2018 (EDT)

I like the idea of having a YA flag. The introduction of the recent Hugo category indicates there's significant interest in this as a thing of its own. If we do have a YA flag that's another reason not to have the juvenile flag cover the whole range of teen years.

I think we should call on MARC 21 for a definition of "juvenile", but they don't (at least on the pages I saw) have a definition of "young adult" and I think we should likewise leave it undefined because how it's used in practice, and in the Hugos, is very slippery. Basically, "If anyone thinks it's YA, it's in." --Vasha (cazadora de tildes) 15:32, 28 October 2018 (EDT)

The MARC 21 definitions which you linked earlier are as follows:

• a Preschool Children, approximate ages 0-5 years.
• b Primary Children, approximate ages 6-8 years.
• c Pre-adolescent Young people, approximate ages 9-13.
• d Adolescent Young people, approximate ages 14-17.
• e Adult Item is intended for adults.
• j Juvenile [age not specified]

In my years of working with Fixer's submissions, I have seen a lot of different ranges specified by publishers. "8-12" is probably at least as common as "9-13". "10-14" is common. "12-18" is surprisingly common, but I have also seen a lot of "13-18" and even "11-18".

Given the overlap between these ranges, my preference would be to use "Juvenile" for books aimed at people who haven't reached their teens yet (I don't want to call them "preteens" because the term is typically used to refer to 10-12 year-olds) and "YA" for teenagers. And if a book is aimed at 10-14-year-olds, well, we can simply use the two flags at the same time. Ahasuerus 16:56, 28 October 2018 (EDT)

In other words, publishers' practices differ from MARC, not too surprisingly. And as it is easier to find publishers' data than MARC classifications for new books, using them would not be a bad idea. Now, from the age ranges you cite, publishers have a blurry dividing line between younger & older books at about 11-13 years. MARC 21 says "pre-adolescent" is up to about 13, close enough that we could use that to correspond to the publishers' division instead of their suggested "juvenile" of up to 15. So maybe, if we're working from ranges, we should say that "juvenile" is under 11 and also 11-13 if that is the upper end of the range; i.e. 9-13 is juvenile but 12-18 is not. And if we only have a MARC classification to go on, "pre-adolescent" is juvenile, "adolescent" is not. --Vasha (cazadora de tildes)

I'm in favour of the original use for our juvenile flag, ie include YA. Why try to make a fine-grained subdivision if it surely will spur discussions such as to which category a pub belongs anyway?:) MagicUnk 16:31, 28 October 2018 (EDT)

Well, the whole idea behind having a separate flag for "YA" in addition to "Juvenile" is that we won't have to choose one or the other -- we will be able to use both at the same time. A book aimed at 11-15 year olds? Click the two checkboxes and you are done! Ahasuerus 16:45, 28 October 2018 (EDT)

On the other hand I can see the appeal to follow an external standard. But if we do, it will be a pain to rectify what's already in the database. I for one, have not been using the YA tag, but did flag the books that qualify as juvenile. MagicUnk 16:31, 28 October 2018 (EDT)

(unindent) I think we're missing the point. Does anyone plan on locating as source of MARC 21 categorizations of titles and consulting it? Do you want to use their definition (not currently provided) for levelling the book? Figuring out the grading and interest level in a book is non-trivial and there are many systems for grading books - assuming we can agree on the cut-off between juvenile and YA and YA and adult. We should go back to the original definition "... work is targetted ...". If you want to have flags for either or both of these, please provide criteria for evaluating the targeting rather than the content. As an example, I sort donated books for a charity and some criteria we use is looking at reviewers, imprint/publishers (e.g. Scholastic) and web-sites (e.g. Harperteens). I've got an abridged copy of 20,000 Leagues Under the Sea that is part of the PageMaster Series that's clearly juvenile, and The Annotated Jules Verne Twenty Thousand Leagues under the Sea that is clearly targeted to an adult audience. And early Heinlein is pretty juvenile but was never marketed that way. ../Doug H 17:09, 28 October 2018 (EDT)

Yes, you have a good point. It is non-trivial to figure out classifications. That's precisely why I like Ahasuerus's proposal of having a new YA flag which could be used simultaneously with the Juvenile flag. As I see it, the YA flag could be used in a very loose manner, with nobody being called wrong in applying it. Some lists of YA books on the internet include Frances Hardinge's books with their 13-year-old protagonists, and some include Pierce Brown's "Red Rising" series which is definitely post-adolescent and is sometimes classed as adult. --Vasha (cazadora de tildes) 17:43, 28 October 2018 (EDT)

So let's put a flag there that noone can define and can be used at editors discretion based on whatever they find to be YA? Yep, that's not a recipe for a disaster at all. If we cannot define it, we cannot have a flag for it - especially because then we end up with two flags that step on each other toes. Tags are different - they are not moderated and they are user data; if we are going to make something part of the full record, we need to be consistent. Otherwise if Editor 1 and Editor 2 try to save the flag the opposite way, who makes a decision on which one is correct? Last one to be approved? The moderator that approves? A discussion for every title?Annie 17:58, 28 October 2018 (EDT)

Good point. But it's going to be difficult to define "young adult" even though it is an important category which does not correspond to "juvenile" (see the considerable progress being made toward defining "juvenile" above). One starting point would be to search the discussions around creating the "young adult" Hugo Award, since I am sure many people made an attempt at definitions. What other discussions or articles on YA spec fic include definitions? --Vasha (cazadora de tildes) 18:10, 28 October 2018 (EDT)

To the point that Ahasuerus made above concerning wide variations in age categorization, I believe we'll never be able to come up with an acceptable definition of what juvenile/YA/etc. really is, let alone be able to label the books in these narrow categories. I'm with Annie that it'll likely result in discussions back and forth. Our efforts are better spent elsewhere... However, what I think the community can come to agree upon is that books are either targeted to ages below 18(21) = Juvenile, or above = non-juvenile. Incidentally the current definition of our Juvenile flag. You'll still risk similar discussions whether a book is intended for adults, or not, or both, but it should be much less.

On the usage of tags (iso flagging): it's bibliographic usefulness is very limited, I believe, since tags provides a means to an editor to convey unmoderated information. MagicUnk 08:33, 29 October 2018 (EDT)

There is one area where tags are very useful. There has been a marked increase in the number of books which are published outside of our regular genres yet contain speculative elements. In many cases it's some kind of minor psychic ability possessed by the protagonist and hidden from the world at large. If I had a penny for every thriller with a borderline psychic criminal profiler that I had to sort out... well, I would have a really nice collection of pennies. Tagging these books with "psychic ability" makes it easier for our users to understand why a speculative fiction database has a bunch of seemingly unrelated thrillers (and, to a lesser extent, books from other genres) on file. Ahasuerus 10:32, 29 October 2018 (EDT)

Sticking to a broad category also allows us to rely on publisher's classification without having to worry unduly whether their definition deviates or not from the standard. Take this example (click on Kind & Jeugd) from a Dutch publisher; everything in all of these listings would then classify as 'juvenile'.

And we may want to change the name of the flag to remove ambiguity & make it clearer what it really means if we decide to reinforce the current definition (Youthful reader, perhaps ? :-D)... MagicUnk 08:33, 29 October 2018 (EDT)

I agree that if we keep the current policy, if will be advantageous to change the name of the flag from "Juvenile" to something that would clearly include YA works. Unfortunately, I can't think of a good term. "Youthful" is typically used to refer to looks, appearances and so on. The technical term for people under the age of majority is "minor", but it's not used in the publishing world. Ahasuerus 10:23, 29 October 2018 (EDT)

I knew it you could tell English is not my mother tongue! :) MagicUnk 20:01, 29 October 2018 (EDT)

"Children & YA" maybe? Why come up with a fancy name when the descriptive one is short enough. :) Plus we won't need to explain what we mean with the field name or usage. Annie 12:30, 29 October 2018 (EDT)

YA -- the big picture

Let me take a step back and consider the big picture.

A database or a catalog is a reflection of the world at large or at least of some part of the world. The data that it captures and the way that data is organized reflect the way the relevant parts of the world are structured. For example, we distinguish between novels and short fiction not because of some inherent distinction between them, but because readers and writers (and, by extension, publishers, librarians, award committees, etc) do.

Once reader preferences change, the publishing/cataloging/award world follows, sometimes reluctantly. To use an obvious example, only 10% of all SF novels published in 1900 were part of a series. Now it's a bit over 75%. Lo and behold, a "Best Series" category has been recently added to the Hugo awards. (BTW, we may want to create an FR to support series awards.)

Similarly, "young adult" as a distinct marketing category with its own imprints and now awards is a relatively new development. I am not sure if it's a worldwide phenomenon, but the rise of Japanese light novels parallels and perhaps predates the rise of YA books in English-speaking countries.

Of course, the world is a huge and enormously complicated place. No matter how faithfully we try to mirror its structure, there will always be exceptions and borderline cases. Still, it's been my experience that, in the long run, our best course of action is to try to reflect real world patterns as closely as possibly. We don't need to be hasty about it since some trends peter out quickly, but long term trends have a way of dragging us along, whether we like it or not. Ahasuerus 11:20, 29 October 2018 (EDT)

Well, it may be new in the English language genre publishing but the novels for the YA population had always been there technically (back when I was in school, all the adventure novels and a lot of propaganda novels were marked that way for example). If we decide to add a tag, we need to define it - while I am not expecting YA to go away anytime soon, it is still used as a marketing gimmick more often than I wish - especially outside of the English language market (where a YA label may make a book impossible to sell). I am just saying not to rush into adding something without definition - when does a children book become a YA is as an important question as is "where is the border between YA and adult". I know that we cannot find exact borders but we should at least try. Alternatively, we can just start using the publisher's marketing categories.

I read a lot of YA novels - and when I read them, they are easily recognizable as YA. But unless I know the author or the publisher, I cannot figure that out from a description for the most part. Is Uprooted YA? Would this categorization be different if the writer was not Novik or if another publisher had published it? Annie 12:27, 29 October 2018 (EDT)

On the same note: many people speak of all-age-literature, and I do think they mean just the YA label with that denomination (or a huge partition of it). Where would we put those titles, even when they are marketed as for 'Young Adults', how shall we draw the line? Stonecreek 14:21, 29 October 2018 (EDT)

Is the point to determine if a book is a) truly appropriate for juvenile/YA b) an editor thinks it is juvenile/YA or c) whether the publisher identifies the book as targetted to juvenile/YA readers? I'd prefer the third and focus on how they tell us and how we'd record it - flags, tags, series, templates or notes. ../ Doug H 14:35, 29 October 2018 (EDT)

If we go for the third - if it is marketed as YA in USA but as non-YA in Bulgaria, do we have variants with mismatched flags? Or do we give preference to the original (in which case clear YA books in English may need to be marked as non-YA because of how they were originally marketed when originally published)? Or do we give preference to the English language designation? Annie 16:00, 29 October 2018 (EDT)

If info is recorded on the pub level, and is intended to convey the intent of the publisher, then there's no problem having mismatched flags between original and translation. MagicUnk 15:04, 30 October 2018 (EDT)

Yeah but when this comment was made, we were still discussing title level only :) Annie 15:06, 30 October 2018 (EDT)

Here's yet another reason why I don't like the idea of having the "juvenile" flag cover both children and teens. I mean, it's pretty clear that there exists such a thing as works written for children who are not yet able to read adult literature. But above a certain age (perhaps varying with era and culture) teenagers are inevitably going to be reading some adult literature along with books aimed specifically at them, and this is why the YA/adult distinction, although often obvious when looking at products of a certain subset like the Anglophone literature of the past few decades, is extremely dependent on era and culture and not at all clear in a global sense. I mean, there was a time when the default audience for Anglophone science fiction was assumed to be youthful and if you were selling your book specifically to adults you had to make that clear somehow; now it's the other way around, and yet a lot of "classic" science fiction has been so seamlessly absorbed into the new conception of science fiction that the difference of target audiences is not noticed. But books for younger children, that has always been a different beast. I think we have a good chance of coming up with a definition of children's books that will make the juvenile flag useful for them. But a way of recognizing YA books that's unambiguous enough for a flag? Not likely. --Vasha (cazadora de tildes) 17:43, 29 October 2018 (EDT)

Alternate approach

Here's a suggested alternate approach: Abandon the subjective checkbox and replace it with a recommended reading age field. Record only what the publisher lists. If the publisher doesn't list anything, leave it out. Take it back to a bibliographic concept (record what the publisher says) instead of leaving it something that invites people to make their own decision on whether the checkbox applies or not. -- JLaTondre (talk) 18:41, 29 October 2018 (EDT)

I like this idea. It will mean losing a lot of the work done through the years but... sometimes it is worth going that way. I wonder if we should not move that on the PUBLICATION level instead of the TITLE level this eliminating the problem of the same book having different designation when published by two different publishers (especially in different years) or across languages. Or if it stays on the title level, it will need to allow multiple values I guess. Annie 19:09, 29 October 2018 (EDT)

I also like this idea. I think putting it on the PUBLICATION level would work best for the reasons Annie mentions. ···日本穣 · 投稿 · Talk to Nihonjoe 19:54, 29 October 2018 (EDT)

Agree! MagicUnk 20:10, 29 October 2018 (EDT)

Yes, not a bad idea at all.

I wonder how many hours it would take to go through all the existing juvenile-flagged items and tag them with one or more of the standard tags "juvenile fantasy," "young-adult sf," etc? There are currently 26,514 juvenile titles, of which 12,780 are already tagged. Chapbooks don't particularly need tags; cutting them out gives us 22,515 items of which 12,340 are tagged. Reduce that further by however many of those titles are variants, and that's what, maybe 8,000 to tag? An exhausting but not impossible prospect. --Vasha (cazadora de tildes) 20:42, 29 October 2018 (EDT)

A few thoughts:

• Sometimes publishers specializing in YA books do not give an explicit recommended age range. I suppose their reasoning is that it already says "Teen" or "Young Adult" or "Young Readers" in the name of the imprint, so why bother repeating it? Besides, if you are familiar with the field, their plot summaries and blurbs already make it clear that the book is aimed at teenagers. What should we do with books that are clearly aimed at a certain audience by virtue of the name of the imprint, blurb, cover art, etc, but doesn't make it explicit?
• Popular juvenile books like The Hobbit and Narnia can have dozens, if not hundreds, of editions and printings. For this reason I don't think it would be feasible to capture the "intended audience" information at the publication level. We could use Publication level notes to document any deviations, e.g. "According to the publisher, this edition of Harry Potter has an "adult" cover", instead.
• We could either create a new title field for this information or just add it to Notes. The main advantage of having a separate field is that it would facilitate Advanced Searches. Perhaps we could add "not empty" to the list of supported operators on the Advanced Search page.
• There are a few ways to handle the conversion process without losing any data, e.g. we could create a cleanup report that would look for titles with the Juvenile flag.

Ahasuerus 21:25, 29 October 2018 (EDT)

(unindent) On further reflection, I don't think JLaTondre's proposed field is an adequate substitute for the juvenile flag & I'd rather not get rid of the flag. Here are some reasons: 1. The difficulty of finding the publisher's own data means the field would rarely have data in it. 2. Proposing replacing a simple flag with a complexity of data means altering its purposes and uses. Many factors point to a book being for children and no one of them is as reliable as a combination; the flag provides a summary of all the indications in a form that can be used in searches and lists. Consider the nightmare of having to design a search that covers the complexities I'm about to list, as opposed to having the person who entered the book look at all the indications of audience and summarize them in one checkbox.

• The publisher may provide age info in the form of an age range, a grade range, a phrase, etc. We could try to convert all this to a standardized form but, quite apart from this being not straightforward, it defeats JLaTondre's aim of an exact bibliographic record. So we record it verbatim, and searching becomes difficult.
• If you know that a particular publisher, a particular imprint, or a particular publication series is aimed at a certain age of readers, you put this information in the notes to the publisher or series, where it is not seen when looking at the publication record.
• There's a separation between target audience (subject matter & style) and reading level (complexity of language). A book written for 17-year-olds who have difficulty reading, full of fast cars and hot romance, is clearly not a book for 9-year-olds though it has a 4th grade reading level. Subject matter is of more interest to this database than reading level. However, suppose the publisher only states a reading level, should you enter it? Should we have a separate RL field?

So, even if it'd be nice to have all these kinds of data and we could add more fields, they don't substitute for the juvenile flag which is a summary based on all of them plus the synopsis, blurb, etc. etc. I don't know if I'm the only person here who regularly uses the juvenile flag in searches, but I for one would badly miss having it. --Vasha (cazadora de tildes) 06:19, 30 October 2018 (EDT)

I guess the first question that we need to tackle is whether information about the "intended audience" is inherently associated with:

• title records
• publication records
• both title records and publication records

The way I see it, if the term "intended audience" reflects the author 's intent, then it's a title attribute. If, on the other hand, it reflect the publisher 's intent, then it's a publication attribute: the same text may be packaged differently when marketed to different audiences like in the Harry Potter example above.

After thinking about this issue, my tentative take on it is that it's both a title attribute and a publication attribute. For example, take a classic text like The Wonderful Wizard of Oz. A recent edition by Project Gutenberg or a bare bones reprint by one of many minor public domain publishers may not have any information about the publisher's intent: no blurb, no age information, no illustrations. On the other hand, a reprint by a major publisher may have lost of touches that indicate the intended audience and even state the intended age range explicitly. The raw text and the audience intended by the author remains the same, so the title attribute remains the same, but the publication attribute is different because the publisher's intent is not stated in the first case.

It's similar to the way our novelization flag works. It exists at the title level and is set regardless of whether the publications associated with the title state that they are novelizations. If they do, we also try to record the details in publication notes to help with distinguishing between editions, e.g. "Based on the motion picture screenplay by David Peoples and Janet Peoples on the cover". In other words, we capture summary information at the title level and detailed/supporting information at the publication level, which is similar to what Vasha talked about earlier. Ahasuerus 15:08, 30 October 2018 (EDT)

(looks like you've posted similar thoughts while preparing my own reply -- copied below :))

I guess it depends on one's point of view. Doug H pointed it out above; do we want to make a distinction between a) truly appropriate for juvenile/YA b) an editor thinks it is juvenile/YA or c) whether the publisher identifies the book as targeted to juvenile/YA readers?

• If you look at it from the point of view of recording actual bibliographic information as provided by the publisher (option c), then recording that information on the book (or pub) level is the way to go. It allows for capturing cases such as a publication of a work targeted at juvenile/YA audiences, and another pub of that same work targeted at an adult audience instead (see the Harry Potter example above). You can't make that fine-grained distinction when recording on title level. The argument that the intended target audience field(s) would be rarely populated when relying on publisher-provided info should not be used as an argument in favor of title-level recording of the info; it is not because info is sparse, that it should therefore be recorded elsewhere (in consolidated form or not).
  Likewise, it is not because there can be tens or hundreds of different pubs of a single work that we should shy away from providing the information on pub level if it’s available.
  Neither should the fact that recording information on publisher, imprint, series,... prevent you from entering similar information on pub level; in fact, allowing pub-level target audience data entry will allow for the inevitable exceptions where e.g. a mostly YA imprint publishes a book targeted at an adult audience.
• Alternatively, if we want to record the intent of the author of the work (option a), then that information should be recorded on title level instead. Unfortunately, this has the drawback that 9 out of 10 we can't know the intent of the author. Often, only indirect information is available to derive the author's intent. So that's not ideal either. So not in favor of recording author's intent either.
• I am neither in favor of option b) where the editor collates all available information and makes a judgement call as to whether the work (or pub for that matter) classifies as juvenile or not (even though this may have been exactly what have been done in the past when populating the juvenile flag, and is what your preference is if I have well understood). After all, who is to decide who is right, and who is wrong if there’s a difference of opinion? Subjective data is to be avoided where we can - see Annie's related remark above). So, for what I am concerned, option b) is to be avoided too going forward.

To summarize: I agree that moving info from title to pub level does change its meaning. But is that a Bad Thing™? Doug H’s question still remains to be answered: do we a) record the author's intent, or c) record the publisher's targeted audience? For me it is clear: out of the available options, only option c) is the one that records data as is provided - i.e. it records the info as provided by the publisher, nothing more.

On the data complexity. When choosing to enter the info on pub level (as opposed to consolidate on title level as you propose), I guess only 3 fields are required to cover all cases: label (drop down if we choose to limit choices), from age, till age. None would be mandatory. It would cover cases where [a publisher] labels their pub as juvenile, YA, ... -- this would go in the label field; or if only age range is available, then in this case only from, till fields would be populated, or, why not, a combination: Adolescent, ages 14 till 17? See also the MARC 21 definitions provided by Ahasuerus above. (granted, may need some more thinking to make sure we don't end up with weird cases we haven't considered, but still...)

On reading level. As far as I am concerned, this is not relevant in the current discussion, unless you take this kind of info into consideration when assessing whether a title (or pub) is to be classified as juvenile or not - but then, see 3rd bullet above.

On the distinction between juvenile & YA. If we would choose to put age info on the title level, then I'm against two flags, unless there's an accurate enough definition of what juvenile, and young adult, means exactly. Can we come up with an accurate enough definition? Personally I don’t think so. Then - if we decide to continue to record target audience info on title level - I am in favour of reconfirming current definition, and perhaps, for clarity's sake, rename into 'Children & YA' as suggested elsewhere. Thoughts? MagicUnk 16:24, 30 October 2018 (EDT)

Capitalization and international titles

We have our rules for English (let's not derail this discussion of them) but the rules do not specifically say they are for English (in effect they are saying that any non-English title should have all its words starting with a capital letter). Which is neither how we want to use the rule, nor how we use it in practice. So I propose two changes to the help pages:

• Insert "In English" after the word Titles at tche start of the rule.
• Add a link at the end of the section to a wiki page where we can set the rules per language (Bulgarian in sentence case and so on) (with a "The Rules for other languages can be found here:"). This way we do not need to keep changing new language and making the help page too big but we do have the proper rules added.

The case for regularization is valid in all languages, making it easier for new editors to determine how we deal with titles in their language will make it easier for everyone to add book. And even more experienced editors that do not work in a specific language a lot will find it easier to figure out how to add a specific title. Please note that this is just adding to the help a description of what we are doing anyway and not a change in policy or usage. Annie 12:32, 31 October 2018 (EDT)

I agree in principle. Specifically, we could change:

• "Titles should have case regularized ..." to "Titles should have case regularized according to language-specific rules ...", and
• "Regularized case means ..." to "In English, regularized case means ..."

Once we have a Wiki page with case regularization rules for other languages, we could link the main Help template to it. Ahasuerus 12:43, 31 October 2018 (EDT)

Works for me. :) It was a first swing at wording - as long as we figure out a way that makes it explicit, I am fine with it. :) And I can start the help page now if that is all that is blocking us. Annie 13:04, 31 October 2018 (EDT)

Thank you Annie. Most appreciate you picked this up and went ahead taking initiative to clarify :) MagicUnk 15:40, 31 October 2018 (EDT)

Yeah, your note on your page reminded me that I had been planning to post about this for a long time - we all know what we are doing, trying to explain it to a new editor is... challenging. :) Annie 15:53, 31 October 2018 (EDT)

Thank you very much. I was hoping to get some clarity on Italian and Portuguese. I already found a discussion that satisfactorily clarifies matters for Italian but I wish we had some native Portuguese speakers here so they could argue about the matter among themselves. --Vasha (cazadora de tildes) 20:36, 31 October 2018 (EDT)

So was that a vote for fixing the help page for that rule? :)

Italian is sentence case. So is Spanish, right? And French. And Russian. I have a feeling that the answer will be that all languages that are using alphabets and are not English use sentence case :) Annie 20:49, 31 October 2018 (EDT)

There are different language-specific conventions, e.g. in German all nouns are capitalized. Well, except when they aren't, of course... but only if the text was published prior to the 1996 reform. Most of the time, that is :)

Then again, there are multiple different English conventions, we just opted to use a certain flavor of "title case". Ahasuerus 21:07, 31 October 2018 (EDT)

The capitalized nouns in German are still proper sentence case though - that's how you type sentences - it just requires a definition of sentence case per language. But that is why I proposed a special rule set per language :) But ok = "most of the languages that use alphabets and are not English" :) Annie 21:46, 31 October 2018 (EDT)

Yep, if you can capitalize sentences in German you can capitalize titles. But that "if" hides some difficulties. Compare "Ich fange an zu schreiben" (I am beginning to write) with "Ich brauche eine Feder zum Schreiben" (I need a pen to write with). Or, "Wir haben nichts" (We have nothing) versus "Eine Sprung ins Nichts" (A leap into nothingness). I do not envy the people who are going to be writing the guide for foreigners ... in fact, if we had a betting pool for when such a guide will be written, "never" would be the safest bet. --Vasha (cazadora de tildes) 23:47, 31 October 2018 (EDT)

That becomes VERY important in the non-Latin languages where А/а are different letters so searching for Ана and ана produces different results. But we will see. Don't start changing titles yet - the data is already ratty anyway - we may be able to find a way to do something a bit less work intensive... a semi-automatic cleaning the way we did title languages maybe. But step by step - first change the formal rule. Second - build the help page slowly. Third - figure out how to deal with the current titles. :) Annie 20:49, 31 October 2018 (EDT)

Alas, it is not so easy in Portuguese. In many people's opinion, Portuguese orthography is still governed by the Acordo Ortográfico de 1945, which prescribed that like in English, Portuguese titles would have some words capitalized and others lowercase (somewhat different words than in English); see Base XLIV on this page for the wording of the rules in the Acordo. Naturally there is widespread uncertainty over which exact words to capitalize; the rules are only 3 sentences long, but not entirely easy to understand. However, there is also a sizeable portion of Portuguese-users who write titles in sentence case -- like any sensible language :-). Here is one forum commenter who summarizes some differences in usage -- in oversimplified form. --Vasha (cazadora de tildes) 21:27, 31 October 2018 (EDT)

And then there is Brazilian vs Continental Portuguese as well - although last time I checked, they were looking at standardizing at least that. I never said it will be easy for all languages. At the end, we will need to decide how we standardize - or let ratty data in. But again - one problem at a time. :) Annie 21:46, 31 October 2018 (EDT)

I agree that it would be best to minimize the scope of the proposed Help change. There are all kinds of related issues that we could conceivably address, but it could paralyze the discussion. For example, pronoun capitalization is a big can of worms in some languages ("Du"/"du" in German, "Vy/vy" in Russian, etc), but it's best to let sleeping dogs lie for now. Ahasuerus 22:02, 31 October 2018 (EDT)

Oh, absolutely. I didn't actually mean to start a discussion. I was just giving a reason why I wanted to thank you for establishing a page where the results of such discussions can be recorded if they happen. --Vasha (cazadora de tildes) 22:12, 31 October 2018 (EDT)

That's ISFDB - everything starts a discussion! :) Annie 22:14, 31 October 2018 (EDT)

Yep. And then we can attack the languages one by one - deal with the easy ones; then deal with the rest. It is sie (they)/Sie(the polite you) in German btw, du is always du. Annie 22:14, 31 October 2018 (EDT)

It's a bit more complicated than that. As the German Wikipedia says, "Im Regelfall wird das Du (Dich, Dein, Dir) kleingeschrieben, die Großschreib-Regel der Substantivierung trifft auch auf das Du zu. Mit der Einführung der Rechtschreibreform von 1996 wurde die zuvor übliche Großschreibung zum Ausdruck der Höflichkeit z. B. in Briefen zuerst gestrichen, bei der Änderungen der Rechtschreibung im Jahr 2006 aber wieder erlaubt: In Briefen können die Anredepronomen „du“ und „ihr“ mit ihren Possessivpronomen auch großgeschrieben werden (§ 66). Eine bis heute nicht abgeklungene Verwirrung erzeugte dabei der während der Zwischenphase, im Dezember 1998, gefasste Beschluss der deutschen Nachrichtenagenturen zur Umsetzung der Rechtschreibreform [1], in dem es heißt: „Die vertraulichen Anredepronomina du und ihr werden ebenso wie die zugehörigen Possessivpronomen dein, euer usw. weiterhin großgeschrieben“, womit eine konsequente Großschreibung von du und ihr nicht nur praktiziert, sondern sogar fälschlich als Teil der alten Rechtschreibung ausgegeben wurde."

In other words, a royal mess :-) Ahasuerus 22:50, 31 October 2018 (EDT)

It is indeed a royal mess, but fortunately irrelevant to this database, since that's only talking about the capitalization of Du as a polite form of address in letters; maybe there'll be a title which is supposed to be politely addressed to a "Du" someday, but we can worry about it then. --Vasha (cazadora de tildes) 23:08, 31 October 2018 (EDT)

Ah, that case. Doubt that we will see it in a title but if we do, we will go and ask Christian what to do :) We can work around that. Eventually. :) But that is for a new discussion later on. And quite honestly, getting a single case wrong is much much better than having the mess we have now in most languages - half are based on the English rules, half based on local rules. We may not get it to be perfect but didn't you have a quote about perfection somewhere that applies here? Annie 23:20, 31 October 2018 (EDT)

Yes, Voltaire once wrote "The best is the enemy of the good." There is also Shakespeare's take on it: "Striving to better, oft we mar what's well."

I think we should start by simply spelling out that we use language-specific capitalization rules. It is our current policy, after all, we just need to put it in writing. Ahasuerus 23:56, 31 October 2018 (EDT)

The third sie (she) is easier to recognize because it requites a verb in a different conjugated form :) Although these are problematic for an editor that does not speak the language - one that speaks the language won't have the issue figuring it out - and not too many titles will have the issue. So we may find a way to define it. But not on this change - this one is just to clarify that the current rule is English and to get us on the path of building a help page (which is not considered rules until we agree as a community - till then it is business as usual.) Annie 22:14, 31 October 2018 (EDT)

Actually, the "title case" rule for German is pretty simple: "Use the German grammar (that was valid when the pub was published)" Well, ok, there's a bit more: a few extra remarks should be added, especially because titles are often printed in ALL UPPERCASE in German pubs, leaving it to the submitter to convert it into regular German. One problem because of this are proper names, which are written in word case in German. Example Der Eiserne Rat) would be "Der eiserne Rat" in regular German, and is stated as "DER EISERNE RAT" on the title page. But since the "Iron Council" is a proper name in this work it is word case in German. Another thing, as already mentioned above, are pronouns, which can be uppercase depending on context. Example: In Ihrer Freizeit is using the uppercase "Ihrer" (your) instead of "ihrer" (their). These must be detected properly by the submitter.

There's no use-case I can think of where a title should be capitalized differently here than in regular German grammar (except if the author intentionally used different or wrong grammar). This, of course, means that moderators who don't know German grammar either have to trust the submitter of a German title, or have to leave moderation to the ones who know German grammar, but this has been the current practice anyway so far and has worked pretty good, I think. I we tried to devise title case rules for German which were different from regular German grammar but understandable for all moderators we'd probably open the door to rule hell :) Also, chances are you change the meaning of a title or create nonsense meanings. Example: "Der Gefangene floh" means "The prisoner fled", whereas "Der gefangene Floh" means "The imprisoned flea" (only the case of the last two words has changed). Changing it to all case "Der Gefangene Floh" (noun and noun, literally "The prisoner flea") wouldn't make much sense in German, except when "Floh" would be the name of a prisoner (then meaning "The prisoner Flea"). Simple, isn't it? :) Jens Hitspacebar 14:20, 1 November 2018 (EDT)

Or in short - German uses "sentence case" as "title case" (technically title case does not exist is a better way to say it but that is good enough) - you write titles the same way you write regular sentences. :) Annie 14:47, 1 November 2018 (EDT)

(unindent) So we are converging on the following change (in the usual ISFDB way of going on tangents :)):

Change the help pages for the title regularization as follows:

• Change "Titles should have case regularized ..." to "Titles should have case regularized according to language-specific rules ...", and
• Change "Regularized case means ..." to "In English, regularized case means ..."

Anyone opposing? Anyone with alternative wording? Annie 11:53, 1 November 2018 (EDT)

Sounds good. Jens Hitspacebar 14:20, 1 November 2018 (EDT)

Sure, but can we have a 'one-line' reminder for (some?) other languages too? Would be nice for foreigners to know that eg Dutch uses "sentence case", with a reminder that eg names of persons/place should be capitalized. No? MagicUnk 17:27, 1 November 2018 (EDT)

Yes, we can set up the linked page with some of the easier languages right away. --Vasha (cazadora de tildes) 17:35, 1 November 2018 (EDT)

Or we can do it step by step by changing this first (as it technically does not change anything - it just codifies what we are doing) and then starting a new discussion on "language-specific" starting with the easy Dutch if you want - and that discussion includes us formulating a plan for fixing what we have in the DB - manually or not. What do you think? :) Otherwise we are expanding the scope again... Annie 19:15, 1 November 2018 (EDT)

(unindent) Based on the consensus above, I have created Template:AllFields:Case. I then leveraged it in Template:PublicationFields:Title and Template:TitleFields:Title to ensure that we no longer have to worry about title- and publication-specific templates diverging.

Also, I wonder if we want to mention that cartoons are an exception and always use the original case. The rule is currently documented in Template:TitleFields:TitleType, but it may be better to mention it in Template:AllFields:Case as well. Ahasuerus 21:42, 1 November 2018 (EDT)

Good catch about the cartoons. I think we should indeed move it in the new template. Annie 22:33, 1 November 2018 (EDT)

Done. Ahasuerus 09:56, 9 November 2018 (EST)

(unindent)Discussion Page - with the update done, let's go and start working on the languages. Let's start with Dutch (because apparently it is easy), Russian and Bulgarian (because I know what I am doing there) and Spanish (as it seems easy). If you want to add another language, add a discussion topic. Once we agree for a language, we can bring the proposal(s) back to this board and then add it to the proper help page. This way all discussions will stay together so if someone disagrees with some rules, they can see how we made them up. If you do not like the format of the discussion, we can always change it... Annie 01:51, 2 November 2018 (EDT)

Just a thought, I presume transliterations to English will follow English rules?--Rkihara 11:45, 2 November 2018 (EDT)

It had not so far in my experience - it usually follows the rule of the language it transliterates - so the capitalization matches the capitalization in the title itself. But because in English the search ignores capitalization and we do not have the problem with variants in case of differences, it really does not matter as long as it uses the correct letters. Is there a side effect of not making it clearly defined? Because seeing non-English text starting with caps all over the place makes it weird (especially if it is an almost Latin language) Annie 11:54, 2 November 2018 (EDT)

PS: And the more I think about it - those are not transliterations to English - that would be a translation. We just transliterate to the Latin alphabet, staying in the original language - and the rules are per language, not per alphabet. So sticking to what the language uses makes more sense to me. Annie 12:34, 2 November 2018 (EDT)

Yes, that makes sense--Vasha (cazadora de tildes) 12:41, 2 November 2018 (EDT)

Agreed. Ahasuerus 13:21, 2 November 2018 (EDT)

(unindent) Should we follow the practices of the title's language as to whether or not to capitalize the first letter of the subtitle? (Spanish does not capitalize it) --Vasha (cazadora de tildes) 15:10, 2 November 2018 (EDT)

Also, sometime (not right now) we need to discuss punctuation. The essential issue is whether or not to follow the practices of the title's language. --Vasha (cazadora de tildes) 15:10, 2 November 2018 (EDT)

Well, I made the new page say Regularization and not Capitalization (after a few renames...) so that may also be part of it. Maybe start a new thread here to first decide if we want to make these rules (one by one) language-specific and then if we decide to make that language-dependent, we just add it to the other discussions. If we try to get everything in a single discussion, we go nowhere so next question for a new thread here: "Should the capitalization of the first letter in a subtitle follow language-based rules or do we want to make it consistent across all languages and enforce capital letter (aka treat it as if it is a complete title?"). Or something to that effect. :) Annie 15:45, 2 November 2018 (EDT)

Ok, new thread --Vasha (cazadora de tildes) 16:02, 2 November 2018 (EDT)

Capitalization of subtitles

We have agreed to follow language-specific rules for how to capitalize within a title. However, nothing has been said about another capitalization rule, the principle that (in English) subtitles after a colon should start with a capital letter.

Firstly, this rule is not mentioned in Help:Screen:NewPub (where it just says to separate with a colon).

Secondly, we need to decide whether to make this rule language-specific (Spanish, for one, does not capitalize the subtitle after a colon). I'd be in favor of language-specific practices; it'd look a little odd to follow them in all other respects but deviate here. --Vasha (cazadora de tildes) 16:02, 2 November 2018 (EDT)

(Copied from Talk:Title Regularization) The more language-specific rules we have, the harder it becomes for people to help with languages they do not at least read and that is not necessarily a good thing - it will reduce the number of international titles... So we need to thread carefully I think -- treating subtitles as full titles (which is the current rule) does have some sense even if some languages do something different. We need to find a middle ground - there is such a thing as being too correct thus becoming impractical. :) Annie 15:49, 2 November 2018 (EDT)

My take on that:

• Add the rule for the capitalization of the first word of the subtitles to the new template
• Follow a general ISFDB convention of always treating a subtitle as a new title regardless of language

I am not entirely convinced about the second one and I am still thinking on it but it will make our lives easier without making the titles too foreign looking... Annie 17:02, 2 November 2018 (EDT)

I agree with both points in Annie's take. --MartyD 10:47, 3 November 2018 (EDT)

I think both suggestions are a sensible rule to follow. Agree MagicUnk 10:16, 9 November 2018 (EST)

Yes, I also do agree. Stonecreek 11:46, 9 November 2018 (EST)

OK, everyone thinks the first letter of a subtitle should be capitalized. Yes? --Vasha (cazadora de tildes) 12:26, 9 November 2018 (EST)

Yup. Who's doing the honours of updating the relevant rules ? And update the rules changelog accordingly if needed? MagicUnk 17:05, 9 November 2018 (EST)

It sounds reasonable, but I wonder if there may be languages which do not allow subtitle capitalization. I have left a note on Linguist's Talk page to see if he may be aware of any. Ahasuerus 09:01, 10 November 2018 (EST)

[unindent] Just a little note to say that capitalizing a subtitle in not systematic in French titles ; especially, but not only, if the subtitle starts with a conjunction like ou “or" : in this case, the capital is usually on the second word (like here, for instance). But there are many other examples of subtitles without a capital. In fact, there is no strict rule about this. Linguist 09:15, 10 November 2018 (EST).

The reason I asked in the first place is that Spanish does not capitalize subtitles. But I think we have four votes above for overriding language-specific practices (I might prefer to follow them, but I don't feel strongly about it.) --Vasha (cazadora de tildes) 13:27, 10 November 2018 (EST)

Come to think of it, Linguist's comment re: subtitles that start with conjunctions like "ou" ("or") applies to other languages as well. If you run a regular "Fiction Search" on "; or", the majority of matching fiction titles do not capitalize the word "or". The word that follows "or" is capitalized instead, just like in the French example above.

Something else that comes to mind is that it's not always easy to tell whether a title contains a subtitle or whether it's a composite title. Requiring that subtitles should be capitalized across languages may create issues for languages that generally do not capitalize words in titles. Ahasuerus 14:10, 10 November 2018 (EST)

So are we back to deciding the capitalization of subtitles based on the language? If so, then let's codify the English rule (making sure that we mention it is only for English) and work on the other languages together with their regular capitalization rules. Bulgarian does not usually capitalize subtitles either but then for English our rules are not really standard either :)

It comes down to policies - as long as we are consistent in a language, either works. I would have preferred less "language specific" rules but if the capitalization of subtitles needed to be done one of the different things, I can live with that. Annie 14:46, 10 November 2018 (EST)

How much detail do we need to go into to clarify the English rules? "Or" titles are a can of worms that we have not sorted out yet. However, there are other uncertain cases too. Haven't you ever seen a title like "John Smith, an Appreciation" alongside the more common "John Smith: An Appreciation"? If I saw "JOHN SMITH—AN APPRECIATION" in a magazine I would not be sure whether or not to capitalize the "an" when I entered it. When em-dashes appear in titles, they usually don't indicate subtitles. I would not capitalize the "and" in The Planet Xorx—and What We Did There. There are so many possible variations that the best practice, in cases that don't obviously fit the "Title: Subtitle" model, is probably to do what's in the source (consulting the table of contents if it's printed in all-caps at the head of the story, etc.)

That said, we should add a simple rule to policies. "In English, separate subtitles from titles with a colon and capitalize the first word of the subtitle. See [[Page]] for discussion of alternate titles beginning with "or" and other complex cases." --Vasha (cazadora de tildes) 14:55, 10 November 2018 (EST)

To Ahasuerus' point, isn't it so that if there's an intervening "or" it means that the subtitle is actually to be considered part of the title proper (ie composite titel)? It would mean that in that case there would be no need for additional rules but the two Annie proposed? And yes, it would mean a departure from following language-specific rules for subtitles, but as discussed above these are not well-defined anyway. So, for subtitles I feel that simplicity should trump (no pun intended) language-specificity, so I would go with applying the rule for English to other languages too... MagicUnk 15:06, 10 November 2018 (EST)

Not true that "or" titles behave like ordinary titles. The usual practice is to capitalize the word after the "or" even if it's something like "the" (Amalia; or, The Queen of the Tigers). I don't want the matter of "or" titles to stop us from stating a rule that fits the majority of cases though. Are we ready to start another discussion and hash out what to do about "or" titles so that we can add a rule about them to the help also? --Vasha (cazadora de tildes) 15:47, 10 November 2018 (EST)

Personally, I am not too happy with systematic normalization, and forcing round, language-specific pegs into square, standardized holes. But then, this db has so many other quirky rules… The one about French titles, for instance, which is not the standard French one, used by most publishers and reference books, but a rarer — and indeed easier and more systematic — variant. It has always annoyed me, but I have complied with it, and even edited spellings which, to my eyes, were more correct. What is being discussed here is whether we want sleek, uniform formats and easy rules, or a truer reflection of reality. Linguist 05:26, 11 November 2018 (EST).

I also am not happy about imposing rules that make data in certain languages look weird to people who are familiar with that language. Who is the data for, anyway? The largest group of users is those who read and don't contribute, who don't care if it's easy to enter data but would like to have it readable. As for contributors, in all languages but English you have more contributions by native/fluent contributors (entering large quantities of data) than you do by outsiders (entering an occasional original-of-a-translation, for example). Why impose rules that are supposed to make things easier for this last group which is the smallest? When they can just copy from their source instead of using rules? --Vasha (cazadora de tildes) 09:35, 11 November 2018 (EST)

As one of the latter, I would like to know the rules, as the titles are often in upper case. I'm not fussy about what they are, I'd just like to enter them so that no one has to change them to be 'correct'. ../Doug H 09:52, 11 November 2018 (EST)

Sure, I like having a reference guide to the rules also. My argument was in favor of using language-specific rules, and ones which aren't artificially simplified to the point that they produce results that look strange to native speakers. --Vasha (cazadora de tildes) 10:12, 11 November 2018 (EST)

After years of struggling with these issues, I am inclined to agree with Linguist and Vasha. Whenever we adopt data entry rules that apply to all languages, we almost invariably end up with scenarios that look odd or just plain wrong to the native speakers of some languages. Ahasuerus 11:33, 11 November 2018 (EST)

OK. I can agree with that. So, can we get to something alike having a 'default' rule that is to be followed (colon, capitalization of subtitle), unless overruled by a language-specific one? Or is that opening a can of worms in that ppl will be confused as to what rule to apply: the default one, or language-specific one? If the community feels it is too confusing, could we go with a rule for subtiles akin the one for titles? ie 'colon, follow language-specific rules for subtitles? MagicUnk 12:46, 11 November 2018 (EST)

Keep in mind that not all languages use colons. For example, Japanese doesn't use colons or semicolons. In addition, the Japanese punctuation characters are different, e.g. the quotation marks are「 and 」. Also, some Japanese titles incorporate Latin characters, so you end up with odd combinations like "バッカーノ!2002「B side」Blood Sabbath" and "バッカーノ!1933〈下〉THE SLASH チノアメハ、ハレ". Ahasuerus 14:38, 11 November 2018 (EST)

They do not have capital letters either though. Annie 15:10, 11 November 2018 (EST)

(unindent)Let's step back a bit and remember why we have the discussion.

• Without capitalization rules, we may end up with a lot of variants that differ by a capital/small letter

That was the rule that started it all. Since then, and with the international titles in the picture, we have two additional items to think of:

• Editors that do not speak the language they add books in (and moderators approving them).
• The search for non-Latin1 characters - without standards for Russian and Bulgarian, search becomes useless.

With all that in mind, having SOME rules makes sense - it may not be perfect but you do not end up with someone trying to find out how to add a Serbian (or German, or Hungarian) title when they are working on adding translations needing to find the rules on the internet. And while for some languages we do have editors to consult, we do not for all of them. And we cannot say: "Let's do the French as the French do but who cares about all the West Slavic languages - let's just do what we want there". Or continue without even adding something about all that in the rules.

So - I would love to just have everything as a native speaker will expect it BUT we do not do that for English either and it is not very practical. But if that is the consensus, then fine... Do we then just let everyone add any way they want and just variant - even when it is the same translation for example? Annie 15:18, 11 November 2018 (EST)

This is an area where software improvements could help a lot (sorry, Ahausuerus).

• Firstly, adding equivalency of capital/small letters in Cyrillic and other alphabets, so variation is not as important (and while you're at it, improved handling of non-Latin-1 characters in the Roman alphabet too).
• Secondly, adding the ability to ignore punctuation when searching and checking for duplicates.
• Thirdly (perhaps as a stopgap in place of the other two suggestions), following up on a suggestion which was made some while ago, to add a field for alternate titles which differ only in punctuation or accents; the alternate titles in this field wouldn't be displayed (they wouldn't be mouseover text like transliterations are) but would help with searching.

These improvements would make variation less of a practical problem and shift it to a purely stylistic one, and allow more tolerance for variation. We'd still have to have ways of resolving matters like "editor A thinks this title should look like so, and editor B thinks it should look like so," but some of the pressure would be off. --Vasha (cazadora de tildes) 15:49, 11 November 2018 (EST)

Re: "adding equivalency of capital/small letters in Cyrillic and other alphabets", it's certainly desirable and important. However, it's not really "adding functionality", it's "upgrading/replacing the database engine". Think of it as replacing a gasoline engine with a diesel engine: doable but time-consuming and but has all kinds of ramifications for the vehicle. Given my other current responsibilities (Fixer, maintenance, e-mail communications with 3rd parties like SFE3 and MIT, Wiki, etc) and my less than stellar health/energy/ability to learn new things these days, big projects like that are a steep hill to climb. We could really use another developer or two. During the summer I helped a volunteer set up a separate development server, but I haven't heard back from him in a few months. I have other prospects as well, but nothing definite at the moment. Of course, a number of our active editors have development backgrounds, but for some reason they seem to be unwilling to quit their day jobs and concentrate on ISFDB development. Most peculiar! (Fixer tells me that it may have something to do with the rumor that humans like to eat.)

Re: "the ability to ignore punctuation when searching and checking for duplicates", the main Duplicate Finder has three modes: Exact, Similar and Aggressive. The last two ignore punctuation. On the searching side, I experimented with different approaches a few months ago, but it turned out to be significantly more complicated that I had expected. Apparently, the guys at Google are well paid for a reason... Ahasuerus 16:30, 12 November 2018 (EST)

Yes, I remember you've said before that these things aren't really feasible without some change in circumstances (like additional developers). Sorry for bringing them up again. But since real improvements are far off, what do people think of adding a field for this sort of variant titles? We are currently using the transliteration field for the purpose, but it's not the best practice to mix two kinds of data in the same field. --Vasha (cazadora de tildes) 16:49, 12 November 2018 (EST)

Agree re: "it's not the best practice to mix two kinds of data in the same field". The issue has been raised in the past, but no FR has been created (that I recall.) The proposed functionality, as I understand it, would be a subset of the functionality supported by Field 246, "Varying Form of Title", which exists to record "alternative form[s] of the title" in the MARC 21 standard. The linked Web page includes various examples. If you would like to pursue this line of thought, we may want to move the discussion to the Community Portal. Ahasuerus 20:39, 12 November 2018 (EST)

Copied --Vasha (cazadora de tildes) 21:47, 12 November 2018 (EST)

The non-Latin1 characters need an overhaul of the DB. Doable but not in the short term. We need a solution in the short term.

Searching for duplicates is not the problem here. Or searching at all. Deciding WHICH one to become the valid one is. If we have "O Dio" and "O dio", who will decide how they will be merged and what will be the correct one after the merge? That is not something you can solve with a software change. And that becomes a problem with your third proposal as well - who decides which one is the "proper one"? We can solve the search issues in various ways but they are secondary. The main one is: "When Someone opens a page, what title do they see?"Annie 15:59, 11 November 2018 (EST)

Your phrasing of the question suggests the answer :-) People should see what will seem right to them (our data should look good) and that means, above all, aiming to use practices that are right for native speakers of the language in question. After all, someone who doesn't know what it should be won't notice anything, right or wrong, so it's only native speakers who can be disturbed by getting it wrong. --Vasha (cazadora de tildes) 16:09, 11 November 2018 (EST)

And in the meantime? The duplicate finder finds "O Dio" and "O dio" and the translator matches so we need to merge. Which one becomes the name of the now merged record? If we leave that to the handling editor without any guidelines, you end up with an author page that looks like it is edited by a 3rd grader - half of the titles in the same language use English-style title case, the other ones with the sentence case one. Which looks worse than having a unified view - even when it is slightly off. :)

Don't get me wrong, I will love to follow the language rules (and make it look nice to native speakers) but we need something that can be implemented and is practical... Annie 16:16, 11 November 2018 (EST)

Didn't you start this whole conversation by proposing something which can aim at this goal in the short term? Clarify that existing rules are just for English, and set up a page for language-specific rules. I think that is an excellent start. The parenthesis in the discussion was just debating how much language-specificity we want (to which Linguist, Ahasuerus, and I answered "a lot"). --Vasha (cazadora de tildes) 16:29, 11 November 2018 (EST)

And effectively in the long run - once we have the rules in place, it is viable. But then the conversation decided to go into the "or" specifics (and not pull that in a separate discussion) and things got muddled :) But yes - this goes back where we started - language rules for all regularization including subtitles (as it looks like "making it simpler albeit a bit off" is not the preferred solution). So if we all agree on that, we can start working on that. Annie 16:40, 11 November 2018 (EST)

I think that your proposal does make an immediate and concrete improvement—people may have been confused before as to whether they should use English rules or the language-specific rules, and now at least they will know the answer to that, even if they won't yet have a guide to language-specific rules and will have to rely on their own knowledge or copying from a book. --Vasha (cazadora de tildes) 16:46, 11 November 2018 (EST)

I would like to get closure on this R&S discussion; the current consensus if I understand correctly is

• to have language-specific case rules for subtitles too - which is already covered by the new wording of Template:AllFields:Case, nothing needs updating there, I believe.
• the current wording still requires a colon. Are we keeping this rule, or also relaxing this one and make this the default rule UNLESS overruled by language-specific rules (cfr. Japanese example above)? - If yes, we'll need to update the wording of Template:PublicationFields:Title and Template:TitleFields:Title accordingly (create Template:AllFields:Subtitles perhaps?). MagicUnk 17:34, 14 November 2018 (EST)

Title case cleanup

For the Dutch titles - and likely for other languages too - almost all titles hitherto entered in the DB follow the English capitalization rules. Now that we have made it clear language-specific capitalization rules are to be followed, titles need to be updated. What would be the best approach? Discussion here. MagicUnk 17:34, 14 November 2018 (EST)

I would wait until we define the rules for the language in question before we start changing things. Considering that we have a few more Dutch(-speaking) editors (and a moderator), why don't you invite them over in the other discussion so we can try to close that language? BTW: Most Dutch titles are actually in the proper cases - we have languages with much worse ratios. Annie 17:55, 14 November 2018 (EST)

Done. Invited Dirk P. Broer, Willem H., and Wjmvanruth (I haven't seen W&W in a while, so don't know if they will contribute, but still). These are the only people I am aware of that are Dutch-speaking. Anyone else you know of I should contact? MagicUnk 08:11, 15 November 2018 (EST)

Punctuation: language-specificity? (and how much?)

While we are writing rules for capitalization in various languages, it would be nice to set standards for punctuation too, so that we can clean up titles in one pass. This applies both to matters specific to book titles (such as separation of subtitles) and more general punctuation standards. We already do have a degree of language-specificity: French titles use spaces before "heavy" punctuation like colons, and German titles are inconsistent as to whether they separate subtitles with a colon or a period. Do people want to follow language-specific rules in all respects? (My vote would be yes.)

I am going to put some remarks about Spanish in particular on Talk:Title Regularization but the short version is that unlike English, Spanish does have an obvious single international source of norms, the Ortografía de la lengua española (2010 edition). I would suggest following it exactly, so that there is no confusion or inconsistency. Some languages have such a single source, others don't. Finding and following an external authority, if one exists, will save a whole lot of time and brain cells. --Vasha (cazadora de tildes) 18:58, 15 November 2018 (EST)

Is that followed by all Spanish speaking countries (in the Americas for example)? Because with Portuguese, the two big countries speaking the language are making a mess of any idea to have unified rules (or so it looks from what I read...)

You know, this is one of the rare cases I am happy that my language is spoken in a single country :) Annie 20:06, 15 November 2018 (EST)

Yes, the OLE is a joint publication of the academies of 23 countries. I know that such internationalism is rare, but when it exists, take advantage of it! --Vasha (cazadora de tildes) 20:11, 15 November 2018 (EST)

SERIALs in webzines

Help:Use of the SERIAL type says:

• Note that current discussion suggests that a webzine may not be treated as a periodical publication, and so may not get the SERIAL type and the "(Complete Novel)" title form.

With the recent expansion of the scope of the ISFDB project to include webzines, I think we now treat webzines as all other periodicals. Should we delete this sentence from Help:Use of the SERIAL type? Ahasuerus 10:09, 9 November 2018 (EST)

Yes, I do think that would be meaningful. Stonecreek 11:47, 9 November 2018 (EST)

Yes, no reason for this sentence at all. --Vasha (cazadora de tildes) 12:20, 9 November 2018 (EST)

Yep, I did not even remember we still have this on the book - webzine is a web magazine, no reason to treat it differently. Annie 14:02, 9 November 2018 (EST)

SERIALs in webzines - outcome

The sentence has been deleted as per the consensus above. Ahasuerus 11:39, 11 November 2018 (EST)

While we are at it, can we add webzines to the other place we are listing that can have serials (the Title type help page? Annie 15:42, 21 November 2018 (EST)

Let's consider our options. The Help text currently says:

• SERIAL. Use for a title that would otherwise be either SHORTFICTION or NOVEL, but which is being serialized in a magazine or fanzine. [snip]

We could either:

1. add "webzine" to the list at end of the sentence, or
2. assume that "webzine" is just of many formats that magazines and fanzines can use, so it goes without saying

Since we are pretty much treating "webzine" as a format at this point, wouldn't it be safe to go with option 2? Ahasuerus 16:48, 21 November 2018 (EST)

I really need my coffee... You are right - I mixed up formats and types here for a second. Sorry :) However - we may want to add something about chapbooks (as in practice we do use serials inside of chapbooks). Annie 17:06, 21 November 2018 (EST)

Right, we have 437 CHAPBOOK publications with SERIAL titles. We began experimenting with this option a couple of years ago and it seems to be working well for authors like Donna Grant. Without SERIAL/CHAPBOOK combos her Summary pages would be in much worse shape.

I guess it's time to decide if we want to make it official. I will start a separate discussion in a minute.

P.S. I'd be careful with all that caffeine. It works great until it doesn't... Ahasuerus 17:36, 21 November 2018 (EST)

Caffeine and I are old friends. If it stops working, I know that I need to kick it off for awhile - then it decides to behave when I get back to it. One day it will stop working completely I suspect but until then... Although if I keep talking about pots of drip coffee, half of Europe is going to disown me for calling that thing coffee... :) Annie 18:18, 21 November 2018 (EST)

"Stops working" is one of caffeine's more benign failure modes. "Headaches" and "insomnia" are markedly less benign. They may not kill you, but they tend to make it hard to work on bibliographic projects and who wants that?! Ahasuerus 17:44, 22 November 2018 (EST)

On the other hand when insomnia hits, caffeine is the only way to get me anywhere close to a functioning human being. Which of course starts a vicious cycle. One cup of a caffeinated drink, two cups of clear water usually manages to offset the bad effects before they happen. Annie 18:30, 22 November 2018 (EST)

Anthology of Essays: "Nonfiction" or "Anthology" or "[something new]"

Regarding "The Analog Science Fact Reader" http://www.isfdb.org/cgi-bin/pl.cgi?269898, this is currently classified as "Nonfiction", which automatically causes the editor, Ben Bova, to be classified as its "Author". I tried changing it to "Anthology", but that edit was rejected by Stonecreek who noted, "This is a NONFICTION title, it has no SHORTFICTION in it". So, unless I'm missing something (a possibility of course), it would appear we do not currently have a publication category for an anthology of essays, unless we (a) create one, or (b) relax the definition of "Anthology" to include an anthology of essays, or (c) add the ability to change "Author" to "Editor" for "Nonfiction" publications. I've already placed my bet on "b", and lost. :) What say y'all? Thanks. Markwood 18:28, 13 November 2018 (EST)

There is a feature request, FR 40 ("Support for non-fiction anthologies"), to implement this functionality. It's been on the back burner since I created it in 2006.

For what it's worth, back when I looked into it, the first thing that I noticed was that it wasn't always easy to tell whether a non-fiction book was a monograph by a single author or a compendium of articles by multiple authors. As of late, it's gotten somewhat easier thanks to Amazon's Look Inside and Google Books. Ahasuerus 18:55, 13 November 2018 (EST)

Non-fiction is secondary for our DB anyway so I Like having just one type. Maybe change the label to say Authot/Editor for non-fiction and call it a day? Otherwise, if we pull away the "multiple authors" in a separate type, should we also separate the collection from single non-fiction ones (essays from the same author vs a monograph that does not contain separate essays)? Annie 19:03, 13 November 2018 (EST)

So how about just adding an "Editor" field to "Nonfiction" publications? Markwood 19:08, 13 November 2018 (EST)

Let me make sure that I understand correctly. Are you suggesting creating a separate "Editor" multi-field for title (and publication?) records? One that would be in addition to the current "author/editor" multi-fields? And that would only be populated for non-fiction works which contain multiple essays? If so, then it wouldn't be feasible. Ahasuerus 19:23, 13 November 2018 (EST)

I'll admit ignorance to the current plumbing, so please let me know what I'm missing here … but when I add an Anthology, the name in the "Author 1" field is displayed as "Editor", but when entering a Nonfiction, the name in the "Author 1" field is displayed as "Author". Markwood 10:15, 14 November 2018 (EST)

Well, when you are entering an anthology, a magazine or a fanzine, the software knows that the publication has an editor but no single "author", so uses "Editor" as the name of the field. In all other cases it uses "Author".

When you are entering a non-fiction book, the software has no way of knowing whether it's a single-author monograph or a compendium of articles by multiple authors. It defaults to "Author", but it's a coin toss really. The only way for the software to know for sure would be to split the NONFICTION title type into two types: "SINGLE AUTHOR NONFICTION" and "MULTI AUTHOR NONFICTION" as per FR 40. It's doable, but it's a much lower priority than a lot of other things, which is why it has been on the back burner since 2006. Ahasuerus 10:39, 14 November 2018 (EST)

I'm suggesting that a book might have an "Author" AND/OR an "Editor" (e.g. posthumous collections often have both) so it would make sense to have an "Editor" multi-field separate from, and in addition to, the "Author" multi-field. Markwood 10:15, 14 November 2018 (EST)

It's certainly true that some books have both authors and editors. Single author collections and, as of late, novels published by certain houses like Tor are good examples. In addition, there are many other types of contributors whose contributions are currently only captured in notes: cover designers, translators, adapters, etc. It would be definitely desirable to have a way to capture and display their names in a structured manner as opposed to dumping everything in notes. There is a feature request, FR 97 "Contributor roles", which covers all of these "roles". However, at this time we don't have a solid design for this functionality (we don't even have a solid design for translators yet.) Whatever we come up with will more than likely take a lot of man-hours to implement, so it's been on the back burner as well. Ahasuerus 10:48, 14 November 2018 (EST)

I can appreciate this is low priority for volunteer hours … so, back to my original issue (how to identify an editor for a nonfiction anthology) the two "free" options are (a) status quo (Nonfiction editors labeled 'Author', with Pub Notes to clarify), or (b) allow Anthologies with all nonfiction contents. Personally, I prefer (b), but I'm OK with whatever moderator(s) decide. Thanks. Markwood 15:15, 14 November 2018 (EST)

I would be in favor of keeping the ANTHOLOGY title/publication type reserved for fiction books. Ahasuerus 15:47, 14 November 2018 (EST)

Me too - anthology/collection requires fiction. Annie 17:56, 14 November 2018 (EST)

And me three: it seems only logical since ESSAYs seem to be the short forms of NONFICTION, so a bundled bunch of them (book-length) becomes nonfiction, I'd think. Stonecreek 23:41, 14 November 2018 (EST)

Me four. I've always entered essay collections as NONFICTION. ···日本穣 · 投稿 · Talk to Nihonjoe 02:34, 15 November 2018 (EST)

OK, that's settled. Thanks for the discussion and consideration. Markwood 10:23, 15 November 2018 (EST)

Science Fiction Book Club

The last update for the page for assisting with entry of Science Fiction Book Club books pre-dates the splitting of the field for ISBN / Catalog #. What would the rules for entering ISBN's be now? Interestingly, the New Novel entry still refers to it as a single field. ../Doug H 08:34, 20 November 2018 (EST)

Good catch. I have updated Help:Screen:NewNovel. Ahasuerus 09:13, 20 November 2018 (EST)

What about the SFBC page itself? There were 'rules' regarding when to include ISBNs and what to record in the combined field. Is there a new standard? ../Doug H 13:40, 20 November 2018 (EST)

SFBC IDs are supposed to be entered into the new "Catalog ID" field. There is also a cleanup report for "SFBC Publications with an ISBN and no Catalog ID".

I could do some limited cleanup of Help:How_to_enter_a_SFBC_publication, but it would be better if one of our SFBC experts chimed in and updated that page. There may be SFBC-specific subtleties that I may be missing. Ahasuerus 13:47, 20 November 2018 (EST)

And while the Publisher field references the SFBC page, the catalogue does not. ../Doug H 08:34, 20 November 2018 (EST)

SERIAL titles in CHAPBOOK publications

Template:TitleFields:TitleType currently says:

• SERIAL. Use for a title that would otherwise be either SHORTFICTION or NOVEL, but which is being serialized in a magazine or fanzine. [snip]

This means that the use of the SERIAL title type is restricted to MAGAZINE and FANZINE publications. At one point this restriction was enforced by the nightly cleanup reports. However, a couple of years ago we changed the cleanup reports to allow SERIAL titles in CHAPBOOK publications. It was an experiment to see if SERIAL/CHAPBOOK combinations may help organize certain authors' bibliographies in our age of e-publications. Authors like Donna Grant frequently publish their new novels as 3-4 standalone e-chapbooks first. Then they publish the completed novel as a regular book. (Some 20th century publishers had tried similar approaches prior to the Internet era, e.g. see R. A. Lafferty My Heart Leaps Up, but it was uncommon.)

At this point we have 437 CHAPBOOK publications with SERIAL titles and they seem to be a good match for authors like Donna Grant. I would like to propose that we make this combination official and adjust Help accordingly. Ahasuerus 17:58, 21 November 2018 (EST)

I support codifying the current practice. With all the e-readers and what's not, more and more works are issued serialized outside of what we consider magazines/fanzines before they are published as a whole (some of them in up to 12/13 parts...). Annie 18:07, 21 November 2018 (EST)

Another use of the type: this anthology has a novella scattered through it in three parts, each of which has a title. SERIAL seems the only sensible way to handle that. --Vasha (cazadora de tildes) 18:52, 21 November 2018 (EST)

The use of SERIAL titles within ANTHOLOGY and COLLECTION publications was hotly debated at one point. As I recall, no consensus was reached at the time. We may be ready to revisit the issue, but I'd like to limit the current discussion to CHAPBOOKs. One step at a time and all that :-) Ahasuerus 08:21, 22 November 2018 (EST)

(unindent) The US holidays are now over, so hopefully everyone with an interest in this issue has had a chance to review the discussion. Any objections to the proposed change? Ahasuerus 12:20, 27 November 2018 (EST)

SERIAL titles in CHAPBOOK publications -- Outcome

Template:TitleFields:TitleType and Rules and standards changelog have been updated to allow the use of the SERIAL title type in chapbook serializations. Ahasuerus 15:16, 29 November 2018 (EST)

One thing to keep an eye on is where we draw the line between serialization in chapbooks and splitting a book into volumes (which we record as the original type). Dune for example is often printed in 3 volumes - but it is long enough to be obvious that those are not chapbooks. For other novels it may not be as clear. Unless we decide to start treating these splits as serials as well (I am not starting this discussion - I am just thinking aloud - I think we should wait and see what we get for awhile - it is one of those things that are easier to recognize than define). Annie 15:36, 29 November 2018 (EST)

Date of excerpts

Template:TitleFields:Date doesn't answer this, and I don't recall it being discussed. What should the date of an excerpt be? That of the original work, or that of the publication of the particular excerpt? I think it should be the latter, since that is the date when it first appeared in that form. --Vasha (cazadora de tildes) 23:17, 17 December 2018 (EST)

I think we have been using the date of publication of the excerpted work. In a way, that makes sense: That is the first publication of the excerpt in its entirety. --MartyD 21:19, 20 December 2018 (EST)

My understanding is that we use the first publication date of the except. Ahasuerus 23:09, 20 December 2018 (EST)

OK, we should add that to the help template. --Vasha (cazadora de tildes) 00:07, 21 December 2018 (EST)

Added to the help template. Stonecreek 09:06, 21 December 2018 (EST)

There's some unclarity here. If I'm understanding what you wrote, Christian, you said to use the excerpt publication date (the later one), not the date of the text it's taken from? That's not what Marty said, and I think Ahasuerus said to use the date of the full text too, although now I look at his words again, I'm not sure. --Vasha (cazadora de tildes) 10:37, 21 December 2018 (EST)

Sorry if I misunderstood, but looking at it, it seems possible to excerpt both meanings from it. Please, Marty & Ahasuerus, could you explain what exactly is the excerpted work: the original one with more words, or the excerpt proper (?). Stonecreek 11:06, 21 December 2018 (EST)

I believe what Marty and I are saying is that the standard practice has been to use the date of "the first publication of the excerpt in its entirety". However, searching some popular author's pages for "(excerpt)" -- e.g. H. G. Wells -- finds a roughly 50-50 split. Ahasuerus 12:09, 21 December 2018 (EST)

Still not clear to me... is it the date of first appearance of the excerpt (ie typically the first few pages of a full work); or is it the date of the work itself the excerpt is from? MagicUnk 14:20, 21 December 2018 (EST)

It's the former, i. e. the date of the first appearance of the excerpt as a separate text. Ahasuerus 14:49, 21 December 2018 (EST)

To me, it should be the former, not the latter, because it is possible for an excerpt to be published BEFORE the actual work is published in full. MagicUnk 14:21, 21 December 2018 (EST)

What you are describing is pre-publication excerpts which have become popular over the last few decades as the number of series increased. The publisher may decide to print chapter 1 of Volume N+1 in the back of Volume N as an extended ad.

However, anthologies and collections can also include excerpts from previously published works, e.g. A Century of Science Fiction or, to use an extreme example, Keith Laumer's Chrestomathy. They are not necessarily limited to the first few pages of the full text. Ahasuerus 14:48, 21 December 2018 (EST)

(Unindent) Time for a poll: who uses which date? --Vasha (cazadora de tildes) 12:26, 21 December 2018 (EST)

I use the date of the first appearance of the "excerpt as a distinct text", not the date of the first appearance of the complete text. One of the reasons is that excerpts need not be in the same language as the original. For example, Sir Thomas More's Utopia was originally published in Latin, but we currently list excerpts in a number of languages. I expect that we will eventually list even more excerpts in a variety of languages. I don't think it makes sense to use 1516 as the publication date of a text that first appeared in 1950. Ahasuerus 14:56, 21 December 2018 (EST)

I use the date the excerpt was published on as an excerpt - the same way we do when we list serialization installments with their own dates and not with the original date even if they are after the full text). Annie 15:03, 21 December 2018 (EST)

I too use the date when the excerpt first appeared as that particular excerpt. I do not use the date for the work from which it was extracted. --Ron ~ Rtrace^Talk 15:58, 21 December 2018 (EST)

(Unindent) Now that misunderstandings are cleared up, it sounds like just about everyone agrees that we should be using the excerpt-date. How about this as wording for the help? "Use the date that the excerpt was first published as an excerpt, not the date of the larger work it's taken from." --Vasha (cazadora de tildes) 19:37, 21 December 2018 (EST)

I would suggest that we wait another couple of days in case other editors have additional thoughts, but otherwise it sounds about right. Ahasuerus 20:37, 21 December 2018 (EST)

My understanding was the opposite of this consensus (what I stated -- the date of the excerpted work, not the date of the excerpt), although that clearly breaks down for previews. I don't particularly care. I don't think excerpts or their dates are of much bibliographic importance. A challenge, of course, is that it's impractical for us to determine whether one excerpt of a work is the same as another. --MartyD 06:55, 22 December 2018 (EST)