ISFDB:Community Portal

From ISFDB

(Redirected from Community Portal)
Jump to: navigation, search


ISFDB Noticeboards
Before posting to this page, consider whether one of the other noticeboards might suit your needs better.
Help desk
Questions about doing a specific task, or how to correct information when the solution is not immediately obvious.
• New post • Archives
Verification requests
Help with bibliographic, image credit, and other questions which require a physical check of the work in question.
• New post • Archives
Rules and standards
Discussions about the rules and standards, as well as questions about interpretation and application of those rules.
• New post • Rules changelog • Archives
Community Portal
General discussion about anything not covered by the more specialized noticeboards to the left.
• New post • Archives
Moderator noticeboard
Get the attention of moderators regarding submission questions.
 
• New post • Archives • Cancel submission



Archive Quick Links
Archives of old discussions from the Community Portal.


1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 · 9 · 10 · 11 · 12 · 13 · 14 · 15 · 16 · 17 · 18 · 19 · 20 · 21 · 22 · 23 · 24 · 25 · 26 · 27 · 28 · 29 · 30 · 31 · 32 · 33 · 34 · 35 · 36 · 37 · 38 · 39 · 40 · 41 · 42 · 43



Contents

Roadmap 2017

As of late, I have been tweaking the software based on user feedback and in support of various cleanup projects. It was a good match while I was recovering from the most recent unpleasantness over the last couple of months.

Going forward I hope to get back into more heavy-duty development work. With that in mind I have compiled a list of software projects that are currently close to the top of my list of priorities for 2017. I'd like to post them and solicit feedback re: their desirability and relative priority. If a description seems too vague or incomplete, please don't hesitate to ask for a clarification. Detailed suggestions/feedback re: individual FRs may need to be spun off as separate Community Portal sections. Additional requests welcome!

Continuation and further development of currently active projects

  • Language cleanup:
    • Assign languages to all titles
    • Assign languages to all authors. Should we use the manual approach that we used with titles? Or can we automate some parts of the process based on the language of each author’s titles?
      I think that for the ones remaining (with 3 titles or less), we can just assign automatic English if all the titles they have are English (after all titles have languages). We will miss a few that are not English but we will also miss a lot of them if we go manually - they will require a lot of research at this point. If a non-English title ever appears, we can always change it. That won't be much different to what will happen now if someone adds a magazine in English and do not change the author language manually after the addition... Annie 20:38, 19 February 2017 (UTC)
      Or if someone is too worried about doing it, going the manual way will always work of course. Annie 05:38, 21 February 2017 (UTC)
      I guess the "low-hanging fruit" would be identifying language-challenged pseudonyms of language-enabled authors and auto-assigning the parent author's language to the pseudonym. Ahasuerus 16:07, 21 February 2017 (UTC)
    • Yes, I think we even talked somewhere about that. If those can be done, this will eliminate a pretty good chunk of what is remaining... Annie 16:25, 21 February 2017 (UTC)
  • Wiki cleanup
    • Move series/magazine-, publisher- and publication-specific etc Wiki pages to the database
    • Move Bio and Bibliographic data to the database
Globally OK.Hauck 20:00, 19 February 2017 (UTC)
Along with this (though it might not qualify as part of your currently active projects), I would like to add, it would be very good to try and untangle the main DB from the wiki DB (particularly with respect to login credentials and sessions, etc.; I have a few ideas on this). This would facilitate upgrading our MediaWiki software which we are sorely behind on (including many security/bug fixes as well as features). Uzume 20:41, 19 February 2017 (UTC)
Unfortunately, my knowledge of the MediaWiki software is very limited. Last we talked about development, Al was going to look into this issue, but he has been unavailable for a long time. I could try and get myself up to speed, but, unfortunately, learning new things gets harder as you get older. Ahasuerus 16:18, 20 February 2017 (UTC)
It is already untangled in terms of sessions - I get asked to login in wiki while my session is fine on the site and vice versa. So it is just the users that are the same. Annie 05:38, 21 February 2017 (UTC)

New projects

  • Create a separate queue for automated submissions, which will enable non-moderators to work on them (and let me concentrate on development). Rather time-consuming to implement. Discission moved here.
  • Internationalization:
    • Support multiple prices per publication.
    • Support multiple publishers per publication. May require more thought re: imprints.
      I wonder if that cannot be solved in a different way - allow books published from a joint venture to show on 3 publisher pages - a combined one and the two separate publishers ones. Or if we go for multiple publisher, we should have a way to see which books are from both publishers at the same time without the need to compare lists manually Annie 20:52, 19 February 2017 (UTC)
      An interesting point. I have never considered it, but the proposed behavior seems desirable. We'll have to think of the best way to implement it. Ahasuerus 02:18, 20 February 2017 (UTC)
      And here comes a live example that I and KarenHunt had been investigating: Братья по оружию. It is a joint venture between the Moscow based Астрель and the Minsk based Харвест with separate ISBNs for both (9785271439186 for Астрель and 9789851815919 for Харвест) and usually all copies of the books carry both publishers and both ISBNs (I've had a few similar ones). If you create a fictitious "Астрель / Харвест" (incorrect it is not an imprint) or "Астрель & Харвест", it will remove the book from the two individual publishers' pages. Adding two separate copies (one per ISBN) is not exactly correct either as it is the same book (ha - this is one of the other projects below - multiple ISBNs). The cleanest seem to be to go for one of them and note the other in the notes. None of those scenarios are very good though. Annie 01:09, 25 February 2017 (UTC)
      And just to illustrate why it is a bad idea to use the double name, this wonderful example of Russian thought has 4 valid ISBNs on the same publication - it is again a joint venture between Харвест and this time another of the publishing group АСТ's imprints/daughter companies - this time Хранитель instead of Астрель. But just to make it funnier, it also got ISBN for the parent as well and an ISBN for the pure Moscow "АСТ Москва" which was consolidating their list. So we will need yet another joint version for this one. As you can imagine, there are more of those. :) Annie 01:25, 25 February 2017 (UTC)
      We had a discussion of Russian publishers a while back, which is what pushed this FR up the list of priorities. There are other publishers who use multiple ISBNs, but Russian language publishers account for the majority of our "problem cases". Ahasuerus 01:34, 25 February 2017 (UTC)
      Well - I saw an example, decided to post it so it is in the thread. Sorry if it was repeating already known data - I do not remember stumbling on this discussion when I was reading archives. Annie 01:52, 25 February 2017 (UTC)
      Not a problem at all! I was just sharing additional background information. Part of my job description :-) Ahasuerus 02:03, 25 February 2017 (UTC)
    • ISBNs (mostly to support internationalization, but will also help in other cases):
      • Support multiple ISBNs per publication
      • Move catalog IDs to a separate field, which will help with pubs that have both an ISBN and a catalog ID, e.g. book club publications.
      • Add two checkboxes next to each ISBN: “derived” and “corrected”. The latter will let us capture two versions of invalid ISBNs, the stated one and the corrected one. Discussion moved here.
    • Add support for translators (as well as other roles like single-author collection editors?) Discussion moved here.
    • Move “Add Authors/Transliterated Titles/etc” buttons to the right to free up screen real estate. This will become more important with the addition of new multiply occurring fields (see immediately above.)
    • Change “Family Name” to “Directory Entry”. If we allow multiple values to support different alphabets/scripts, how will we sort by this field? Discussion moved here.
    • Separate translations from VTs on the Title page. Allow users to limit the list of publications to the ones associated with the canonical title.
      Won't that also allow us to group same language translations? If not, then can we have that as well? I would love to be able to see that there are 30 different language translations without seeing that there are 10 French ones until I really want this information :) Annie 20:52, 19 February 2017 (UTC)
      I don't recall this functionality mentioned in the past. I would suggest creating a separate section to explain what you would like to see and to gauge other editors' reaction. Ahasuerus 16:23, 20 February 2017 (UTC)
      I'd like to see this one. -- JLaTondre (talk) 01:22, 20 February 2017 (UTC)
  • Re-do verifications. This will let us have more than 5 primary verifications as well as multiple transient verifications per publication.
    I'd also like to see this one. -- JLaTondre (talk) 01:22, 20 February 2017 (UTC)
  • Create a history of changes to primary-verified publications by storing a snapshot of the way each verified pub looked like right before it was changed.
  • Support for third party IDs (OCLC, BLIC, LCCN, BNF, ASIN, Goodreads, etc). This will enable Fixer to submit ebooks without ISBNs, an increasingly common scenario. Discussion moved here.
  • Make the Quick Tags list user-definable.
  • Add support for additional title types like PLAY and FICTITIOUS ESSAY (requires discussion to come up with additional values.) Discussion moved here.
  • Cleanup reports:
    • Make additional cleanup reports available to non-moderators
    • Review old/incomplete FRs for cleanup reports, add missing title types
  • Add a “non-genre” field to author records; create a cleanup report to find non-genre titles by non-genre authors. Discussion moved here.
  • Add a disambiguator field to title records (requires discussion)
    Don't understand. Hauck 20:08, 19 February 2017 (UTC)
    Is that for the ability to add the reason for the variant (change in name, change in author, translation, abridgement)? If so, I'd love this one. If not - What are talking about? Annie 20:52, 19 February 2017 (UTC)
    The problem that we are trying to address here is that some authors have identically named titles. For example, H. P. Lovecraft's Summary page shows three "The Lurking Fear and Other Stories" collections published in 1947, 1964 and 1971. They are all different, but you can't tell until you drill down to the title level and check each title's Note field. Similarly, there are 2 version of the story "The Rats in the Walls" (1924 and 1956), 4 "[To Albert A. Sandusky]" poems and 9 (sic!) poems whose title is currently entered as "[To ?]". Analogously, Clark Ashton Smith wrote 4 different "Ennui" poems, 2 different "A Sunset" poems, 3 "sonnet" poems, etc. A "disambiguator" field would be used to capture a brief summary of what's unique about each title, e.g. "1964 version", or perhaps the first line if it's a poem. It would be displayed on the author's Summary page. In addition, we will need to decide whether to use this field to capture other information like "abridged", "revised" or "greatly expanded". There were different suggestions the last time we talked about it. Ahasuerus 16:53, 20 February 2017 (UTC)

Ahasuerus 19:50, 19 February 2017 (UTC)

Additional Proposed Projects

  • Add a "printing rank" field to order multi-reprinted titles without pub dates.Hauck 19:57, 19 February 2017 (UTC)
    Yes, please! And not only for dateless - even if we know the dates, a field for edition and printing will be awesome (so if I have 2nd edition, 3rd printing, I can look for 2.3 or something like that and find it without looking through 200 publications). Annie 20:52, 19 February 2017 (UTC)
    There is a Feature request to "Add a 'printing' field to publication records". It would be relatively easy to implement, but we need to decide what kind of data we want to capture. Unfortunately, "editions" and even "printings" do not always follow the standard numbering scheme, so "2.3" won't work. For example, this 1989 edition of Heinlein's "Have Space Suit—Will Travel" says "First Ballantine Books Edition: December 1977 Thirteenth Printing: October 1989". On the other hand, this 2005 edition simply says "First Pocket Books trade paperback edition [no printing number]". This is very common, so you can't expect "editions" to be sequentially numbered. We could limit this field to numeric printing numbers, but there can be any number of "first printings" from different publishers. It doesn't mean that we won't want to implement this field, but we need to understand its limitations. Ahasuerus 04:07, 20 February 2017 (UTC)
    The 2.3 was mostly an oversimplification for the multiple printing scenarios - not necessary requesting it to be numbers only. But we do need some guidelines for formatting in that field - or we will end up with unsearchable field. It will still be better than the current way to find where we are. Even if it is just printing in the field - the edition is kinda clear from the ISBN unless if it gets reused. Annie 04:13, 20 February 2017 (UTC)
    What!? I cannot put Chinese characters in the printing field? But I know of some publishers that mark their printing this way (OK, I am facetiously joking but it is true). Uzume 04:58, 20 February 2017 (UTC)
  • Add a new title type for excerpts, and one for plays/scripts. Neither of those fit nicely in SHORTFICTION. --Vasha 21:02, 19 February 2017 (UTC)
  • Another useful feature, mainly for links from external site to the ISFDB: stable IDs. Currently, if a title is merged into another its ID ceases to exist in the database. As a result, links from external sites to such a title record will become broken links. The software should instead keep track of merged IDs and redirect to the new ID (there was a discussion about this some time ago but I just couldn't find it). Jens Hitspacebar 21:10, 19 February 2017 (UTC)
  • [Discussion of third party IDs moved here.]
    1. The next item I would like to see is revamping verifications. I am not sure how valuable it is but it has plagued us for a while now. Uzume 21:36, 19 February 2017 (UTC)
    2. Finally, author and publisher directory updates seems like a good item. This could bed sooner but probably will be easier after more internationalization/localization is completed so doing it a bit later might prove less problematic. Uzume 21:36, 19 February 2017 (UTC)
  • There had been some discussion about reviews of magazines here. I think the discussion petered out rather than concluded, but did get into alternate ways of dealing with magazine series. I'm not pushing for it, just reminding. Doug H 21:59, 19 February 2017 (UTC)

Editor review of submissions: "automated" submissions sidebar

Create a separate queue for automated submissions, which will enable non-moderators to work on them (and let me concentrate on development). Rather time-consuming to implement. Ahasuerus 00:10, 20 February 2017 (UTC)

Yes, please :) I know that it is a time consuming change and so on but it should help in the long run... Annie 20:52, 19 February 2017 (UTC)
I am for editor review of submissions (by Fixer, etc.), however I question the need for a separate queue. Uzume 23:02, 19 February 2017 (UTC)

(unindent)I have some reservations about "automated submissions" (but that maybe based on my lack of understanding) referred to above in #Roadmap 2017. Why do we need another submission type like this? It seems to me this FR is just to allow editor reviews of submissions (like a voting system to help moderators close pre-reviewed submissions). Is there some reason some submissions should not allow editor review? Uzume 21:45, 19 February 2017 (UTC)

Because at the moment Ahasuerus is doing all of them. This is the process by which all newly published books are added to the system - so we do not wait for people that have them to add them. The idea is to allow the queues that the automated robot finds to be worked on by other people as well - which includes verifying that it is indeed eligible, cleaning the record based on our standard and so on. It is not about approval of entries from people, it is about adding the entries in the DB so the moderators can approve them. Annie 21:50, 19 February 2017 (UTC)
I do not see a difference with this. It is true bot generated submissions maybe of lower quality but then so are submissions from new users. I do not think we need a new field specifying the origin of the submission. I like the editor review of submissions concept (perhaps with comments and queued results to be approved by moderators). In the general sense, consistent high volume and high quality editor reviews of submissions could be a very good way of selecting prospective new moderators. Uzume 21:55, 19 February 2017 (UTC)
We do not have real bot submissions now - Fixer is manually operated by Ahasuerus, title by title, every month. That idea is to allow these queues to be processed by other people as well. Annie 22:03, 19 February 2017 (UTC)
I understand he is doing significant manual work on his side of bot before making submissions and to lighten his load, we want to be able to have his bot make lower quality submissions with better review from our community (instead of just him) and I agree with and would not argue on that point. However, it seems to me a generalized editor submission review process is a better solution than a one targeted just at this problem (which adds unneeded extra levels and fields for such). Uzume 22:10, 19 February 2017 (UTC)
And how is that different from making more people moderators if there are people that can do reviews like that? :) Plus - the process of accepting edits is already slow, do you really want to add more layers? A new editor will get discouraged fast. Plus - that won't solve the issue with Fixer and the sumissions - no matter who reviews, they need to get into the system one way or another. If you want all that it finds to be just added and someone reject it if it is not genre and/or not eligible, that would add a lot more work on the moderators... A Fixer, non-massaged entry is missing series information, have (or can have) a weird spelling on author and publisher, title (Amazon adds a lot of junk there) and it may need to be merged somewhere. No matter how many layers of whatever we have, if those just get dumped into the regular queues, the real updates will get delayed with days until the pure deluge of those is dealt with. Annie 22:18, 19 February 2017 (UTC)
I never suggested making editor review mandatory—just generalized. I do not understand why the normal queues would cause delay. Submissions are not required to be moderated in any particular order (I understand there are order issue with regard to moderation however). Uzume 22:36, 19 February 2017 (UTC)
No but delaying a new contributor edit while we are dealing with internal ones is a bad idea. And if you have 12000 Fixer submissions in the regular queue, how will a moderator find a new one. And even if they can, working on those ratty ones will be a longer and more involved process than on user-created ones. Thus the separate queues and the ability to do something before they hit the moderators. :) Annie 23:13, 19 February 2017 (UTC)
That sounds to me more like better management of the current queue is needed not another queue. Make the submission queue searchable and a moderator can for example just say "I do not currently want to see submissions by Fixer" (or say any user with a bot flag; this list is not very long Special:Listusers/bot). Uzume 23:40, 19 February 2017 (UTC)
The main difference is allowing editing before submitting - allowing the entry to be cleaned up before it goes into the proper queue for approval. This is not something moderators can do now - they can only accept and then edit after that. If you are saying that moderators should be able to directly edit ANY submission before acceptance, then yes, better management is all that is needed. But I am not sure we really need that. Annie 23:45, 19 February 2017 (UTC)
OK, so you are saying you want a separate queue and allow reviewing editors to edit/transmute the submission as part of the review process? That would be very powerful but also extremely difficult to code (since submissions are actually only semi-structured XML documents). Uzume 00:39, 20 February 2017 (UTC)
Thus the comment above that it will be very complicated :) If you read the FR, this is what it describes. Annie 00:40, 20 February 2017 (UTC)
Maybe Annie means a tiered submission process where editors can review automated submissions converting them to normal submissions or some such. But then in my mind "automated" just means "I want editor review first". Uzume 22:02, 19 February 2017 (UTC)
Not really. The current automated process is; Fixer finds the titles (wonderful), Ahasuerus submits them to the DB (from his name or from Fixer's name). Anything with low priority just stays in the queues until someone gets to them. Or forever. It is not about magazines, it is really about what the robot finds and needs to be processed. And you do not want to add to the moderator load all the "cleanup" that is needed - there is a LOT of work to be done on those submissions. And just dumping them into the DB without cleanup, even with a flag will dilute and make the DB useless fast.Annie 22:09, 19 February 2017 (UTC)
I am not arguing against editor review for Fixer submissions. I am arguing this is more elegantly solved with a generalized process of editor review not specific to Fixer submissions. Uzume 22:13, 19 February 2017 (UTC)
Baby steps. Plus Fixer (and any other robots that can find more entries). It is not Fixer specific really - it is the ability to handle those kinds of submissions in a slightly different way. And I still argue that if a person does the submission, you do not need it to be delayed further. :) Annie 22:20, 19 February 2017 (UTC)
So you are saying the "automated" flag adds another tier delaying such submissions. I do not think that is necessary (and if you read the FR a moderator can directly moderated such without editor review anyway). The point is we do not need the flag can just implement generalized editor review of submissions. Uzume 22:35, 19 February 2017 (UTC)
If someone new does not understand the page, they may click it, sending the submission in a long(er) queue. What we need is a double queue - one that only moderators can deal with (the ones that had been seen by a human and just need the second set of eyes) and one for any "ratty" submission where more people can help with. Which is what this FR is for - allow a new process for submissions. Annie 23:09, 19 February 2017 (UTC)
I am all for editor review of submissions to help with such "ratty" submissions. I do not think we need another queue (chosen at submission time), just better management of the current queue should suffice.

OK, after some discussion...if we are talking about a totally different process where someone or something provides suggested changes (I am intentionally avoiding the word "submission") which are reviewed and potentially generate submissions (a sort of reviewed and potentially modified transmutation from the original suggestion), that is a totally different beast and I am not at all sure it makes sense to leverage the current submission system for this sort of thing. I believe Fixer and other bot like "animals", normally do not make arbitrary submissions but only very specific types (mostly NewPub; someone tell me what other types it can make). As such, it might make much more sense to have separate NewPub (suggestion? submission?) queue. A reviewer could then load such NewPub items into the NewPub editor, potentially make changes, and make the submission (or mark the suggestion rejected). This would greatly simply such an implementation (which would still not be trivial). It might also make sense for bots to mark items in this NewPub queue as completed (or actually delete them) when it noticed the items appear in the main database that match the NewPub suggestion it previously made. For this purpose, there should probably be a way for bots to query this queue (with search requests like "give me all outstanding NewPub items I provided but I have not marked as completed/deleted"). Frankly, I see this as more of a remote bot NewPub queue than an automated submission queue. Uzume 01:24, 20 February 2017 (UTC)

AddPub as well when it manages to recognize where a new publication belong. Fixer founds a few thousand new ISBNs every month. They need to be added to our DB - but they do need a lot of cleanup. What is your proposal - how those say 6 0000 publications get added without the new process and without burning out all the moderators? If Fixer can just dump them in a new queue where someone can go, edit and then submit to the moderators, this is exactly what the FR is proposing - call it automated submission queue, call it "list of possible NewPub/AddPub", call it whatever you want :) So do you just disagree with the name? What you describe here is exactly what the FR is proposing... It is not about changing the current process, it is about adding a new process BEFORE the current one that will allow these ratty ones to be streamlined for the process. Annie 01:39, 20 February 2017 (UTC)
That is not how the FR is written "Change the submission table to support a new submission state, “A”utomated" implies modification of the current submission queue. I do not think that is appropriate if what is really needed is a remote bot queue with limited input types. Uzume 02:11, 20 February 2017 (UTC)
Look at point 10 in the FR - that is what ties it back into the standard process. Annie 02:13, 20 February 2017 (UTC)
I see that now, however, I still think that unduly complicates the submission system with little value. I believe we would be better off with a new different specialized queue than trying to shoehorn this functionality into the current submission queue. Uzume 02:20, 20 February 2017 (UTC)
The linked FR would effectively create a new specialized queue. Editors would then use "automated"/robotic entries in that queue to create regular submissions. Implementing various supporting mechanisms -- like putting the automated submission on hold as described in the FR -- would be somewhat challenging, but I think it should be doable. Ahasuerus 04:58, 20 February 2017 (UTC)
I imagine it would be challenging both on the API server side of things as well on the bot client side of that (what does/should Fixer do with such a remote queue item marked on hold?). Uzume 05:23, 20 February 2017 (UTC)
When Fixer sends a submission via the Web API, he marks the ISBN as "submitted" in his internal database. From that point on he ignores the ISBN.
When Fixer gets the next weekly backup, he updates his internal database with the data from the backup. At that point the status of moderator-approved ISBNs changes from "submitted" to "already in the ISFDB". Moderator-rejected ISBNs remain "submitted" forever. It doesn't really make a difference because Fixer ignores "already on file" ISBNs just like he ignores "submitted" ISBNs. Adding a new queue on the server side shouldn't change this logic on the client side. Ahasuerus 15:16, 20 February 2017 (UTC)

(unindent) If there's a separate queue, that queue will probably back up (esp. once the novelty wears off). If it does back up, we'll end up in a situation where more of its entries turn out to be duplicated by live-editor submissions that were created without consulting that queue. It might be better to get more moderators and let Fixer submissions flow through without any sort of pre-qualification/filtering that's being done now. We are not particularly aggressive about recruiting moderators. --MartyD 12:03, 20 February 2017 (UTC)

Unfortunately, the main problem is not that we don't have enough moderators. The main problem is that processing robotic submissions requires a significant amount of research and TLC. Some moderators have the time and the energy to do the necessary research. Others don't and simply click the "Approve" button if the submission looks OK. This is the main reason why I only submit on Fixer's behalf if:
  • I am 99% sure that the submission (usually an AddPub) is truly OK "as is", or
  • it's a relatively low priority ISBN and it's better to have it on file in a potentially suboptimal state than not to have it at all
Over the years, I have tried a number of strategies to improve the situation, all of them unsuccessful.
The "automated submissions queue" is the latest stratagem in a long list. The idea is that having human editors do the requisite research up front would result in higher quality submissions to be reviewed by moderators. Admittedly, if a moderator wants to pull up the new queue and simply approve everything in it, it won't stop it.
If there is a better solution to this problem, I'd love to hear it. The current situation is barely sustainable as it is and will continue to deteriorate as my productivity declines over time. (Not to mention the increased number of SF books being published.) Ahasuerus 19:15, 20 February 2017 (UTC)
That makes more sense and I prefer that idea. This is why I am sort of against the idea of another queue (particularly if it is shoehorned in the the existing one as the first part of the current FR describes) despite its possible value in being able to modify the entries at review time (it seems like much work where this is the only key value). That said, it might make sense to add generalized editor review for the main submission queue (it could be limited in some way if there are security concerns). Such reviews could help moderators more quickly decide when closing submissions. As Marty points out such things may turn into a novelty (but then same could be said of moderators; look at how many inactive or barely active ones we have at the moment), but it might ebb and flow with new or return editors, etc. Aside from the possible value of possible valuable reviews to speed in moderation, it seems like a better tool to use when hunting for moderator candidates (so though this too might be complex and end up perhaps with smaller value this part seems to me the most valuable part). Uzume 18:24, 20 February 2017 (UTC)
If we go this way, we really should have a prioritization of some type available so that live editor updates do not get backed up when Fixer dumps its submissions. It is very annoying to wait for a moderator to approve something so you can send the next update (especially when you are new) :) Annie 05:32, 21 February 2017 (UTC)

Adding support for translators and possibly other roles

Add support for translators (as well as other roles like single-author collection editors?) Ahasuerus 23:45, 19 February 2017 (UTC)

Ok for the former, no to the latter, why not editor of novels? Hauck 20:08, 19 February 2017 (UTC)
A collection editor usually chose the stories, a novel editor does not chose content in the same way. I think we should allow such editors be somehow recognized. Annie 20:52, 19 February 2017 (UTC)
I am with Hervé here. I would like to added more contribution types supported when we have no way to capture such now but I would be wary of taking what we already have and dividing it down (into different types of authors/editors or different types of artists, etc.). I definitely want to see translators (since our current rules have no fields to capture this short of ad hoc notes). I am curious why we cannot just add translators as authors of translation title records and just reach through the VT to find the original authors instead of adding a new contributor type (we do not need one for artists they just tend to have different types of works; a different language VT work is a different type of work). If we implement such, we could have a cleanup report for translations where the canonical authors (jumping through pseudonyms) match authors of the original work allowing us to convert these over time (translations without translators report). Uzume 23:20, 19 February 2017 (UTC)
So you are proposing the Bulgarian variant of Dune to have as an author the translator? How would you separate on the translator page the ones he did write from the ones he translated? And if a user does not understand out variant system, they can misunderstand who the writer is. Plus this way we are losing the ability to have the actual author name in the language of the variant... Artists are different because they are different works - the translation by definition has two authors - the original one and the translator. Annie 23:24, 19 February 2017 (UTC)
The software can display these differently by reaching through the translation variant and author psuedonyms to properly credit authors vs. translators. Uzume 23:33, 19 February 2017 (UTC)
How about an example. Dune by Frank Herbert is translated as Дюн by Франк Хърбърт, translated by Виолета Чушкова. Which name goes where in your idea? Annie 23:37, 19 February 2017 (UTC)
I am assuming we are referring to T1442269. I am suggesting we just add Виолета Чушкова to this record as an author and the software just credits the translators differently than the authors by chasing the authors of this record from A237335 to A30 (since A237335 is a pseudonym) and whatever number Виолета Чушкова ends up (since that is not currently in the DB) and comparing that to the authors from T2036 (which is the variant translation parent of T1442269) which is A30. The software can clearly detect that A30 is the original author and whatever number Виолета Чушкова ends up is not (and thus is a translator). In a similar vein, cleanup reports could be generated for translations without translators. It should be noted that all of the above data is already required to display and properly link these entries anyway. This proposal just suggests handling that data differently to allow for and credit translators (there is already an FR900 for displaying translations differently; I just want to add translator credit support to that and propagating it to more than just title pages). Uzume 00:10, 20 February 2017 (UTC)
None of your links work by the way - you may want to look at the templates again:) The example was from my library. I get where you are getting with that. But... what about the case where it is not just a translation but also an abridgement. This is when the second author gets added now. And that will make the translator author page a nightmare for the software to work out - in which cases the author is a translator, in which cases he/she is a co-author, in which case an author. Not to mention that the searches for translators will become overly complicated. Annie 00:14, 20 February 2017 (UTC)
Currently our rules rule out using the VT mechanism for any derived (where the contents is significantly different) works except translations so that should not be an issue. In partially translated works and the like there should be no VT parent and whether an author is an translator or not becomes muddled to begin with. I believe it would be best to list all contributors as authors and specify who did what in the notes (much as is done with certain types of authorship now). Uzume 00:26, 20 February 2017 (UTC)
Not true. Abridgements are variants - in a way a lot of translations are abridgements, especially the older ones. As are two parts of a split novel. Revised novels also get varianted... Annie 00:42, 20 February 2017 (UTC)
We did not used to allow VT title records except to change the title or author and then later allowed translations (i.e., changes in the language). If indeed we now allow other derived type VT title records, I would prefer a translation flag marking a VT title record as a translation over adding new contributor role flags (add a field to title records instead of to author records). The problem with using the VT mechanism for generalized derivatives is: what constitutes a derivative? Is something a derivative if it is barely inspired by something else? As you can see this gets rather sticky quickly and the resulting connections are more then the VT system can support (since that would actually require a generalized graph construct). Uzume 01:38, 20 February 2017 (UTC)
We are already using them for that (as much as I dislike it in some cases - split novels for example) - the question is how to make it a bit more manageable. And as for the flags - if a variant is an abridgement AND a translation from different people, how do we determine who is who? We can use additional authors but we will need a flag on the author as well... Annie 01:52, 20 February 2017 (UTC)
You cannot use VT to specify abridgement AND a translation by different people anyway. That would require more than a two title records to express (one would have to link the translation to the abridgement and the abridgement to the original; our VT mechanism specifically does not allow generalized graph structures). Uzume 02:08, 20 February 2017 (UTC)
You cannot have a variant of a variant - so if my Bulgarian book is an abridgement and a translation, it goes under the main work directly. You need one title to express - you just need to add notes. Annie 03:21, 20 February 2017 (UTC)
Exactly, and in the same notes that qualify such you can qualify the contributors as well. If you cannot have the titles linked perfectly (outside of notes) it makes little sense to try and credit people perfectly. Just as you have to pick the parent (even if it is not right) you can pick if it is a derived work or a translation (even if neither of those is entirely right either). If we ever support such a linking then the credits will be able to be right too. I see little reason to hack around credits to solve the real problem. Uzume 04:12, 20 February 2017 (UTC)
And here we ended up on a circle. I cannot see how your idea is easier than a new role (considering all the additional work that will be needed to populate summary pages. And if it is not separate, something will need to be done on the publication and title pages to identify them as translators/whatever they are anyway because new contributors will be confused. Annie 04:20, 20 February 2017 (UTC)
Easier solutions do not always make for better solutions (in fact the easy way often ends up being hard long term). I am not sure how new contributors would be any more confused than by which is the proper variant parent (which cannot be set properly too). Uzume 04:54, 20 February 2017 (UTC)

(unindent) The issue of translator support has been open for close to 5 years now, so I don't expect us to come up with a perfect solution in a few hours. There are various aspects to consider, e.g. the fact that there are awards given to translations rather than to titles.

The reason that I mentioned "roles" is that whatever design we come up with will ideally also support other "roles" like single-author collection editors. The exact roles that we will eventually choose to support are less important than the decision to make the design extensible -- or not, if we find that translators are sufficiently different from other "roles". Ahasuerus 05:11, 20 February 2017 (UTC)

Yes, I was not suggesting the translators issue would be solved anytime soon. I am against dividing our author records into roles however, and just wanted to state that there are potential solutions for the translators issue without depending on the introduction of divided contributor role tags. If we do end up supporting such role tags, I would recommend corollaries for publisher roles (or any other authority control item), etc. Uzume 05:16, 20 February 2017 (UTC)
I don't know how useful it is to researchers, but I like the idea of being able to capture/catalogue various roles. Part of the challenge right now is that the nature of the relationship is embedded in one of the two ends (mostly in the publication/title end, but sometimes in the author end). Thinking about the relationship in terms of a generic "contributor", I could see typed Contributor relationships instead -- the nature of the relationship stored on the relationship. There could be many of these, and a pre-defined set of types. One type would be "Author", another "Editor", another "Translator", and so on. The bibliographic summaries for people could be organized by roles, then title types. To keep things simple and avoid big mistakes, the software could provide some specialized handling for required roles (e.g., it could still present Author (for single works, collection, and omnibus) or Editor (for anthology and magazine)) and then "other contributors" or something like that for everything else (just like we get the Title record but then other Contents). --MartyD 17:42, 20 February 2017 (UTC)
I have been thinking along similar lines for a while now. However, redoing the author logic (and the variant title logic and a lot of other things) would be a daunting proposition. It may be less painful to declare authors and anthology/magazine editors a "special privileged case" and implement a generalized "contributor/role" mechanism for translators, other types of editors, etc, separately. Once the new mechanism is in place and everything is working smoothly, we can look into migrating "special case" authors/editors to the new system.
Granted, a two-step approach is not without its dangers -- I once worked on a multi-billion dollar project which took this route and never completed the second step -- but we may be better off even if we never complete the second step. Ahasuerus 23:59, 20 February 2017 (UTC)
Well, we need to solve it if we really want to claim that ISFDB is international. I do not want to add 3000 works just to have to edit them in a year to get the translators sorted... Just saying :) And I like Marty's idea. Annie 00:04, 21 February 2017 (UTC)

Non-Genre flag for authors

Add a “non-genre” field to author records; create a cleanup report to find non-genre titles by non-genre authors. Ahasuerus 00:04, 20 February 2017 (UTC)

This will open the gates to well-intentioned persons that will want to enter such authors (as with the "graphic" or "non-genre") as it will look possible. Bonus problem: how to define a non-genre author.Hauck 20:08, 19 February 2017 (UTC)
[Discussion of author birthdays and dates of death displayed on the front page moved here.]
I believe adding a genre field to authors should be fairly easy and provide good value (like moderator warnings on submissions when adding non-genre or graphic content to non-genre authors, etc.). @Hauck: all authors are non-genre by default and those "over the fold" get promoted to "genre" and we allow (nearly) all content for them. Uzume 21:36, 19 February 2017 (UTC)
Sure, and promoted by who? We're back to the remarkably precise definition of "over-the-threshold" (or "over-the-fold"). This may sound funny but trying to explain for the upteenth time to another new contributor all these perfectly nebular notions may prove an interesting expreience for those who do not practice it. Hauck 21:55, 19 February 2017 (UTC)
Promotion (and demotion) is via the moderated submission process (changing the author genre/non-genre checkbox). It could be restricted to moderators too (like author canonical renaming, etc.) though I am not sure that is necessary. Education is always useful and we already have the rule. The only difference is having a field to declare such. Uzume 22:58, 19 February 2017 (UTC)
I really do not like the idea. I do not think that we should have non-genre works even by the over the threshold authors but that ship had sailed. And how we will determine who is genre author? If someone has a story/essay here, they are an author with genre works. Anyone that has at least one non-genre work is again an author with genre works. Trying to draw the line in the middle is going to cause a lot of unnecessary edits.
I can think of a few Bulgarian authors that had written both SF and Crime works and are considered masters of both. I would not add their crime works here or call them non-genre because they just happen to also had written outside the genre. Even the ones that under the usual "rules" will be over the threshold. Annie 00:10, 20 February 2017 (UTC)
This is already specified in our Rules of Acquisition. I agree some cases are not clear cut. If you want to challenge the rule (or champion a specific author for promotion or demotion) that is a different discussion than implementing it in software. Uzume 00:20, 20 February 2017 (UTC)
It does not determine what a non-genre author is. Again - where would you like to draw the line? If you want it to be based on what an editor wants to nominate, it becomes meaningless... Annie 02:14, 20 February 2017 (UTC)
What you are saying is the current Rules of Acquisition are meaningless in this regard (and I agree that could be argued) but that is a different (though valuable) discussion. Uzume 02:38, 20 February 2017 (UTC)
No, what I am saying is that the rules determine which works are added but do not qualify authors as genre or non-genre ones (above the threshold is not a determination of that).Annie 02:52, 20 February 2017 (UTC)
This seems a pointless designation. If an author has written a genre story, they are by definition a genre author. If it is attempt to designate authors who write primarily genre stories, what is the point? If it is an attempt to designate authors who are 'above the threshold', that is not the same as authors who primarily write genre stories so the name is deceiving. And as Hauk has pointed out 'above the threshold' is a nebulous thing that is not well suited to a database field. -- JLaTondre (talk) 01:15, 20 February 2017 (UTC)
I am curious why you think this is not suited to a database field? I agree an author could be promoted or demoted to threshold levels over time (an author starts out in one genre and gravitates to another). Should instead every editor and moderator know and potentially reevaluate each author upon every submission? That would make for very complex submission criteria and potentially break the rules of acquisition as different people came to different conclusions at each point in time. Uzume 01:58, 20 February 2017 (UTC)
Everyone has a different opinion of 'above the threshold'. People already come to different conclusions. A database field is not going to change that nor is it going to change that different editors and moderators apply it differently. It would just be a field they can now edit war over. -- JLaTondre (talk) 02:17, 20 February 2017 (UTC)
Edit warring over that single field is considerably more noticeable than generalized edit warring over all the pubs and titles of an author which what we are already at. If there is an edit war over that field it will be noticed and brought to community discussion more easily and thereby a consensus can be reached. I agree we could also likely generate some sort of automated heuristic to flag if an author should be above the threshold (by counting title and/or pub records) but it would not take into account exceptions by consensus. But the proposed flag is meant to hold the current value of community consensus (which obviously can change). Uzume 02:32, 20 February 2017 (UTC)

Derived and corrected ISBNs

Add two checkboxes next to each ISBN: “derived” and “corrected”. The latter will let us capture two versions of invalid ISBNs, the stated one and the corrected one. Ahasuerus 00:12, 20 February 2017 (UTC)

Not frequent enough to clutter the screen, better use the note field instead.Hauck 20:08, 19 February 2017 (UTC)
I agree with Herve on that one. Annie 20:52, 19 February 2017 (UTC)
The proposed check boxes would only appear on Edit/Add/Clone Publication pages. Regular Publication pages would continue to display "(Bad Checksum)" for invalid ISBNs like they do now.
The advantage of having invalid ISBNs captured in the ISBN field vs. in the Notes field is that it enables searches. For example, if you search for the ISBN stated in Alpha 5 (0-345-24140-X), you won't find it because it's only available in Notes at this time. When dealing with invalid ISBNs, we have always had to choose whether to enter the stated version or the corrected version in the ISBN field. The other version was relegated to Notes where it became unsearchable. With the addition of support for multiple ISBNs, we will no longer have to choose -- we will be able to eat our cake and have it too! :-)
As far as derived ISBNs go, there are hundreds, if not thousands, of them in the database, mostly from the mid-1970s when the industry was transitioning from catalog IDs to SBNs and then to ISBNs. It's always made me uneasy that we claim, e.g., that Alpha 4's ISBN is 0-345-23564-9 even though it says "345-23564-9-125" in the book. If we add a check box to the edit page, the main Publication page will be able to indicate that the ISBN value was derived. Ahasuerus 03:15, 20 February 2017 (UTC)
I too agree and believe implementing ISBNs as identifiers (a different FR) will likely solve much of this problem anyway. The key is to differentiate identification (e.g., cover price, etc. which can be erroneous) from identifiers (which probably should not be erroneous). Uzume 00:23, 20 February 2017 (UTC)
I still believe most usages of this can be covered by treating valid ISBNs as identifiers (in the as yet unimplemented identifiers FR) and invalid ISBNs as catalog numbers. Here is some potentially pertinent information on this subject: MARC ISBN. Uzume 03:47, 20 February 2017 (UTC)
Back when I was working on the design of the "third party identifiers" FR, I considered making catalog IDs and ISBNs just another type of identifier. Eventually I decided against it for a number of reasons:
  • Unlike third party identifiers, catalog IDs and ISBNs are stated in publications
  • Also unlike third party identifiers, catalog IDs and ISBNs are displayed on various "Publication Listing" pages
  • ISBN searches are "privileged" in the sense that they are available via the main search box
  • ISBN searches use a special ISBN-10/ISBN-13 conversion algorithm so that a search on an ISBN-10 finds a matching ISBN-13 and vice versa
  • ISBNs can be "derived" or "corrected" as described above
In the end I decided that catalog IDs and especially ISBNs were sufficiently different animals to merit separate places in the database. Ahasuerus 04:17, 20 February 2017 (UTC)
Those are good points, however, an ISBN as an identifier can be converted to a 13 digit number in all cases (and left in original ISBN-10, SBN or other brokenness as stated in the publication in other cases like the catalog number field). When searching ISBN identifiers, ISBN-10s to search for can be converted to 13 digit numbers before comparisons (so there is no issue). It would not be so hard to have a combined search for these and the catalog field(s) (matching original input against catalog fields and any convertible ISBN-13 against the identifier fields). Though very common, all (and sometime any) ISBNs for an edition are not always stated in its printed publications (e.g., I have seen some that only have an GTIN-14 which contains the ISBN-13 and another digit). I am not sure it is worthwhile to implement ISBN as both catalog numbers and identifiers and to accomplish such (but it is something to think about). Another issue is broken punctuation in ISBN. I know sometimes an ISBN is stated in a publication with improperly placed hyphens (I have no idea how we handle this if at all; I am guessing we do not maintain this original statement). Uzume 04:49, 20 February 2017 (UTC)
There is a great deal (2427 lines to be precise) of ISBN formatting logic in this module. It's yet another way in which ISBNs are different from third party identifiers. Ahasuerus 15:20, 20 February 2017 (UTC)

Changing "Family Name" to "Directory Entry"

Change “Family Name” to “Directory Entry”. If we allow multiple values to support different alphabets/scripts, how will we sort by this field? Ahasuerus 02:39, 20 February 2017 (UTC)

Switching to unicode and then sorting by it will be cleanest. Until then I kinda like it as it is (as long at non-Latin entries actually have their transliterations properly and not getting defaulted to the name of the author in English. Annie 20:52, 19 February 2017 (UTC)
I have been looking forward to the DB moving to some Unicode encoding like UTF-8 for a long time (currently it is only pseudo-Unicode by way of using HTML entities). Multiple directory entries solves more than just multiple script/language issues (see ISFDB:Community Portal/Archive/Archive41#Sorting, author directory and nobiliary particles). Multiple scripts can be handled by multiple directories. Uzume 00:51, 20 February 2017 (UTC)
Multiple scripts/multiple directories becomes a bit of an issue with author names with letters from two separate scripts - Macedonian or Ukrainian names for example which combine Cyrillic (somewhat creative on some letters) and Latin letters. Which directory will have these forms? Or do we start a separate directory for each of them? Annie 02:44, 20 February 2017 (UTC)
This is not necessarily their real name anyway. I would suggest making entries for both (all Cyrillic and all Latin even if their real name is a mixture of the two). Uzume 03:03, 20 February 2017 (UTC)
For Ukrainian and Macedonian authors, it IS their real name. And we all hope we will get more international entries, right? Saying that they do not deserve their real names in a directory while we have everyone else's name there will be a bit... uninclusive. Annie 03:19, 20 February 2017 (UTC)
I do not see how that is an issue. The above mentioned "nobiliary particles" discussion proposed having multiple entries based on different cultural conventions in different places. Uzume 03:29, 20 February 2017 (UTC)
So separate directory for each language we support (I am just trying to make sure I understand the idea) Annie 03:37, 20 February 2017 (UTC)
No, not for every language. One per script (all Latin in one, all Cyrillic in another, etc.) implemented as time and demand allows. This means for example English language authors would be found in the same directory as German ones. Another example of directory entries that do not match real names would be Japanese (which would have kana entries) where I do not think we want to consider a kanji Chinese character directory (I have no clue how we would do such for actual Chinese names). Uzume 03:53, 20 February 2017 (UTC)

(unindent) Let's take a step back and examine the uses (current and proposed) of the "Family Name" (soon to become "Directory Name") field. At this time it is used to do 2 things:

Making this field a "multiply occurring field" will make it possible for an author record to appear on multiple Author Directory pages. For example, if we specify "de la Roche" as well as "Roche" as Mazo de la Roche's Directory Entries, his name will appear on two different Author Directory pages. Once we implement non-Latin Author Directories, it will also enable us to make the same author appear in multiple directories.

So far so good. However, once an author has multiple "Directory Entry" values, we will have no way of using them for sorting purposes. Unless we make one of them our "preferred sorting value", I guess. Ahasuerus 03:49, 20 February 2017 (UTC)

Agreed. And as such instead of having a "preferred" tag, why not treat this like transliterations. You have a Family name with "transliterated" Directory names on top. This would allow a mixed (potentially collating) name as a family name (we could have Cyrillic and Latin or kanji and kana in this field) and single script directory names on top. Uzume 03:57, 20 February 2017 (UTC)
I am not sure I understand what "on top", "mixed name" and "potentially collating name" mean in this context. It sounds like you may be suggesting that we keep the current "Family name" field and add a new multiply occurring field for "Additional Directory Entries". The latter would be used for names like "Roche" in the "Mazo de la Roche" example above as well as transliterations. The former would continue to be used for sorting purposes. Is that about right? Ahasuerus 15:45, 20 February 2017 (UTC)

Add support for additional title types

Add support for additional title types like PLAY and FICTITIOUS ESSAY (requires discussion to come up with additional values) Ahasuerus 02:41, 20 February 2017 (UTC)

Not enough candidates to warrant new types.Hauck 20:08, 19 February 2017 (UTC)
Maybe not for fictitious essays; --Vasha 00:52, 20 February 2017 (UTC)
"Fictitious essays" or "fictional essays" are basically "in-universe" essay. They are written as if the events of the work that they are associated with were true. They have prompted many discussions since it's not clear whether they are "essays" or "short fiction". Help currently says:
  • Some books contain fictional essays, purporting to written by a character in the book, as introductions or afterwords. There is no "FICTIONAL ESSAY" title type, so you have to choose whether the title is better described as SHORTFICTION or ESSAY.
There have been repeated request to implement a separate title type for them. Ahasuerus 02:48, 20 February 2017 (UTC)
Well then, if there have been repeated requests, clearly there are enough cases to justify adding the new type. And I will say it definitely sounds useful to me. --Vasha 03:37, 20 February 2017 (UTC)
but for plays, even if there aren't that many of them, the difference between them and SHORTFICTION is so obvious that they need their own title type. --Vasha 00:52, 20 February 2017 (UTC)
I would also like to see a separate title type for plays, but we will need to decide how to handle related cases first. For example, are movie scripts, TV scripts, etc separate types or are they all the same type? If they are the same, should we come up with a more generic term than "play"? Ahasuerus 02:51, 20 February 2017 (UTC)
"script" covers plays, audioplays, tv scripts, and even librettos I think. --Vasha 03:37, 20 February 2017 (UTC)
(And for the other one I would like to see a new type for, excerpts, there really are a lot.) --Vasha 00:52, 20 February 2017 (UTC)
The vast majority of excerpts are works of short fiction. Should we treat them as a separate title type or as a separate "length" value like novellas, short stories and novelettes? Ahasuerus 02:52, 20 February 2017 (UTC)
I'd like a separate length value for these - this will separate them from the real short fiction. Annie 02:54, 20 February 2017 (UTC)
Keep in mind that if we add "excerpt" as a new "length" value, it won't separate excerpts from other works of short fiction on Summary pages. Ahasuerus 03:18, 20 February 2017 (UTC)
I believe you mean it won't currently separate such (the software could be made to do such without huge issues though I am not convinced that is a good idea). I vote for an "excerpt" "length" value. Uzume 03:21, 20 February 2017 (UTC)
I do not like seeing novel excerpts mixed in with the actual short fiction on the summary pages. --Vasha 03:37, 20 February 2017 (UTC)
It does not do it now anyway (See Asimov - the third and the forth are excerpts... If we go for separation one day, we may start talking about separating all lengths from each other anyway. What I meant upstream was that it will allow the excerpts to be seen easier. And allow searching them - and searching without stumbling at them. Annie 03:40, 20 February 2017 (UTC)
Agreed. Though I am not convinced about separating such on author pages. One can create custom pages via search that show just what you want. Uzume 04:17, 20 February 2017 (UTC)
I would be strongly against separating all lengths on author pages. It's not of interest under most circumstances, plus every author has lots of unspecified lengths. --Vasha 04:21, 20 February 2017 (UTC)
I am not exactly vying for it but it might be interesting to consider user preferences allowing one to select which lengths one wants to see (much as language selection is possible this way). Uzume 04:25, 20 February 2017 (UTC)
I like that, it would speed up searching for titles on prolific author pages. Maybe checkboxes for all title types on the author page rather then setting it in user preferences.--Rkihara 18:23, 20 February 2017 (UTC)
The question wasn't choosing which title types to display, but rather displaying ss/novelette/novella separately. It's this that I think would not be good on summary pages. As a preference option, OK maybe. --Vasha 20:55, 20 February 2017 (UTC)
If the intent is to separate excerpts from other SHORTFICTION titles on Summary pages, then I believe they need to be turned into a separate title type. Using "length" values to create separate Summary page sections would go against the basic "one title type = one Summary page section" principle. Ahasuerus 23:43, 20 February 2017 (UTC)
Excerpts are short fiction -- it does not really make sense to have them as a separate type any more than novellas are separate type. At least I cannot see a difference... If we split them in a separate type, then why would not things like half a novel published in a volume go into that same new type? Where do we draw the line? At least as a part of the short fiction, that line is clear. Annie 17:05, 21 February 2017 (UTC)
Yes, sometimes excerpts are carefully selected so that they form a complete short story; more often they actually don't seem complete in themselves. --Vasha 17:39, 21 February 2017 (UTC)

(copied from a side discussion in another section) If excerpt got added to the mix [of length options] it would be hard to "sort" that with the others. Uzume 01:28, 25 February 2017 (UTC)

I think you may be the only one who likes the idea of "excerpt" being a length... everyone else was talking about it being a title type --Vasha 01:32, 25 February 2017 (UTC)
That was not how I read things (but perhaps I misunderstood). As title type seems that seems problematic. As a length, I could usefully apply it to essays and nonfiction (like an excerpt of a bibliographic index, etc.) as well. Uzume 01:43, 25 February 2017 (UTC)
I read it as a support for Length as well :) However to your latest point here: length may be showing when you are working on publications for all titles but if you read the documentation and look at editTitle for example, it is valid only for Short Fiction. Making it usable elsewhere will probably require a lot of other changes (even if it becomes a length). Annie 01:46, 25 February 2017 (UTC)
OK, I can sort of see excerpt as a length, in that I hesitate to assign a length to excerpts -- they don't strike me as "really" being a short story or novelette. But that is the very reason that they seem to me to be a separate kind of thing, not short fiction. Besides, I like the idea of having excerpts be a separate section on summary pages, for which they'd have to be a title type. But maybe I am the only one who wants that. --Vasha 02:15, 25 February 2017 (UTC)

Third party identifiers

Support for third party IDs (OCLC, BLIC, LCCN, BNF, ASIN, Goodreads, etc). This will enable Fixer to submit ebooks without ISBNs, an increasingly common scenario.

Good to have some way of identifying ebooks but most of those you listed are ephemeral. Particularly ASIN. (I don't put ASIN in the notes field of an ebook because there are probably numerous ones most of which will disappear soon.) Goodreads has the advantage that they keep records for out-of-print ebooks (it's their policy not to delete) but they're very unsystematic, their records can get merged with each other, etc. --Vasha 17:49, 20 February 2017 (UTC)
Library IDs (OCLC/WorldCat, BLIC, LCCN, BNF, etc) are generally stable, at least as stable as our IDs. We currently link to them in Notes, which requires crafting URLs manually, an error-prone and time-consuming process. In addition, manually built links can become defunct if/when the third party changes its URLs conventions, as we have seen over the last few years. ASINs may not be quite not as stable as, says, LCCNs, but in most cases they work even years after the data was captured. They are frequently the only way to link to the source of our data. Even more importantly, without ASIN support Fixer is unable to submit ebooks without ISBNs, which are increasingly common. Even major authors publish e-stories without ISBNs these days and Fixer can't do anything about them. Ahasuerus 18:10, 20 February 2017 (UTC)
I would certainly like to be able to use library IDs as identifiers for non-ISBN print books. Having support for that makes sense. But Worldcat doesn't list ebooks -- what about other libraries? --Vasha 18:30, 20 February 2017 (UTC)
Some do, but the intent of this FR is to support third party IDs regardless of whether they are for paper books or ebooks. Ahasuerus 18:46, 20 February 2017 (UTC)
As for automatically importing ebooks, that opens a can of worms. We had an inconclusive discussion a while back about what constitutes an "edition" of an ebook. Personally I would like to see them condensed not too many; the way Goodreads has so many is bewildering and ugly. So if you automatically import them there's have to be manual combining. --Vasha 18:30, 20 February 2017 (UTC)
Fixer has been generating ISBN-based ebook submissions for many years. Submitting ASIN-based ISBN-less ebooks should be no different than submitting ISBN-based ebooks once we add support for third party IDs. Where it will get challenging is if we start using another source of ISBN-less ebooks. Suppose Fixer creates a database record for an ISBN-less ebook whose ASIN is "ABC". 6 months later an editor (or another robot) wants to add a record for a Barnes & Noble ebook whose B&N ID is "XYZ". How can we tell whether it's same version of the ebook, especially if the publisher is not specified and the publication dates are close? OTOH, as long as we stick to ASINs, we should be in reasonably good shape since different ASINs are generally associated with different editions. Ahasuerus 18:56, 20 February 2017 (UTC)
Amazon actually doesn't have every single ebook. Graydon Saunders is a fairly notable example of an author who choose not to offer his books via Amazon because he objected to their terms. --Vasha 20:21, 20 February 2017 (UTC)
Oh, definitely. However, as long as we add these books manually, we are less likely to run into problems. It's the robotic submissions that worry me. I have seen many cases where an ebook was originally added by a human editor without an ISBN -- because Amazon doesn't display ISBNs for ebooks even if they exist -- and then Fixer came along and tried to add the same book using its ISBN. (This particular problem will be alleviated by adding support for third party identifiers.) If we were to start using other online sources to grab ISBN-less ebooks, the problem would become more severe. Ahasuerus 20:35, 20 February 2017 (UTC)

(unindent) I believe the next most interesting item is adding support for identifiers. This is a bit sticky but useful concept. The sticky part comes in where we apply these. I believe you were mostly considering at the pub record level, however, the identifiers in this area are themselves usually defined for the concept of an "edition" (which we have no real records for since our pub records most closely correspond to a printing, save for SFBC). This probably is not a major issue but something to consider. Uzume 21:36, 19 February 2017 (UTC)

True, different catalogs and Web sites create records at different levels. For example, an OCLC record may cover multiple ISBNs. Alternatively, multiple OCLC records may cover the same ISBN. I don't think there isn't much we can do about it. Adding support for third party IDs is the best I can think of. Ahasuerus 19:50, 20 February 2017 (UTC)

Next, there is the issue of untangling ISBNs. Currently, our pub record ISBN field represents an identifying mark of the publication (along with cover price, etc.) and we just sort of play games with it to allow is to link to other sites by using it as an identifier. Uzume 21:36, 19 February 2017 (UTC)

Well, ISBNs were originally supposed to be unique and they are still used as "mostly unique" by the industry, e.g. when ordering books. Of course, as we know, publishers can tweak things like cover art and price while keeping the same ISBN. And sometimes they mess up and reuse an ISBN which they didn't intend to reuse. Still, ISBNs are the only IDs that can be used cross-catalog and cross-site with a reasonably high chance of success. Our use of ISBNs to link to other sites is the way they were intended to be used. It's not perfect -- see the OCLC discussion above -- which is one of the reasons why adding support for third party identifiers is useful.
For example, suppose an editor wants to record that some of our data originally came from OCLC records. At this time the only choices are either to:
  • enter something like "Corrected page count from OCLC 1234567, artist from OCLC 987653", or
  • manually craft URLs pointing to the OCLC records
The second options provides more value but is error-prone and time-consuming, not to mention that URL structures can change over time. Once we add support for third party identifiers, editors will be able to enter OCLC record numbers directly. Ahasuerus 20:05, 20 February 2017 (UTC)

An easy way to untangle this is to run some code on our ISBN fields and move them to ISBN identifiers and keep the original field for verbatim publication catalog markings (including wrong/broken ISBNs, etc.). This way we change the current field to a verbatim publication identification field and move identifiers and external linking to use the newly introduced identifiers (including derived/corrected ISBNs etc.). Then we just have to update documentation to keep the fields separate (I recommend renaming the current field as catalog number and documenting it to apply to cover or copyright page catalog numbers, etc.; we could expand it to allow multiple values but frankly I would say that unneeded and can be covered in the note field). Finally, identifiers are also useful at other levels such as author and/or publisher authority control (e.g., VIAF, LCNAF, etc.), and for tags with regard to subject indexing terms (e.g., LCSH, FAST, etc.). Some identifiers also apply to the work (our title record) level (e.g., Open Library works, Wikidata, etc.). Uzume 21:36, 19 February 2017 (UTC)

Yes, author-based and title-based identifiers are something to consider. Currently we enter VIAF links as "Web sites". If and when their URL structure changes, we will have to change the URLs manually or write a conversion script. If we had author-based third party identifiers, the problem could be resolved in 10 minutes. I agree that they would be nice to have, but I see it as a much lower priority than publication-based identifiers. Ahasuerus 19:32, 20 February 2017 (UTC)
Maybe, but we could for instance move Wikipedia links to go through Wikidata identifiers with links like enwiki/Q982133, frwiki/Q982133 and jawiki/Q982133 for John Norman allowing the links to be translatable (without storing and maintaining all the various links) not to mention other possibilities like enwikiquote/Q982133 (provided the sitelink for such articles exist). Uzume 00:47, 25 February 2017 (UTC)

(unindent) OK, I have been thinking about this -- how about this for an addition to the edit-publication page. Below the ISBN field, the ability to add fields for an indefinite number of other identifiers (a button that says "Add Identifier" or whatever), and next to each of these fields a dropdown list to specify what the identifier is: ASIN, OCLC number, publisher's catalog code, however many options seems useful. (Naturally, the submission would be flagged if you try to submit a number without specifying its type -- and maybe, at least for some of them, could be flagged if it's not a valid format for the type specified.) Would that work? --Vasha 01:52, 2 March 2017 (UTC)

That's actually extremely close to the implementation that I sketched out last year :-) Ahasuerus 03:11, 2 March 2017 (UTC)

Last Activity Date

I would like to suggest that we tag the moderators listed on the Moderator Availability board with their “last activity date” as we do for the permanent verifications. If it’s not to difficult, perhaps we could tag all editors too.--Rkihara 13:50, 30 September 2017 (EDT)

We have limited control over the Wiki software, so I don't think we could get it to display "last activity" dates automatically. However, we could create a new ISFDB page to display an arbitrary editor's last activity date. We could then create a template to link to it from the Wiki side. The ISFDB URL would be something like "www.isfdb.org/cgi-bin/last_user_activity.cgi?Ahasuerus". Ahasuerus 15:06, 30 September 2017 (EDT)
Whatever works. I think this would allow us to more effectively direct queries regarding edits, and find active moderators.--Rkihara 12:33, 7 October 2017 (EDT)
Sounds good. I will work on it once I finish mitigating the changes to the Amazon API. Ahasuerus 14:15, 7 October 2017 (EDT)
P.S. FR 1097 has been created. Ahasuerus 14:18, 7 October 2017 (EDT)

Help with Tianxia Bachang

Delacorte Press is about to release The City of Sand, a book by Tianxia Bachang, a Chinese author. I assume that it's a translation of the first novel in his Gui Chui Deng series, but I am not positive. Do we have anyone whose Chinese is good enough to sort things out and to set up VTs/pseudonym?

Also, according to SFE3, the original text, which was self-published online, contained multiple fantasy elements:

  • ... risking censure for its possible allusion to Magic, regarded as an illegal superstition in the People's Republic (see Huang Yi). Consequently, there is a marked difference between the original online samizdat publications of the stories and their later appearance in print, which interpolated mitigating explanations for all supernatural phenomena, suggesting that any Zombie infestations, spirit hauntings and demonic assaults (see Gods and Demons) experienced by the characters were, for example, mere dreams or the result of poisoning by noxious tomb vapours.

It would be nice if we could determine whether subsequent Chinese editions and foreign translations contain the SF elements which were "mitigated" when the first volume was officially published. Ahasuerus 14:28, 30 September 2017 (EDT)

天下霸唱 (Tian Xia Ba Chang).--Dirk P Broer 16:54, 1 October 2017 (EDT)
Thanks, I have moved things around a bit to make everything consistent. Ahasuerus 15:35, 3 October 2017 (EDT)

How many Brian Millers are there?

Undoubtedly, not all the titles on the Brian Miller author page are by the same person. But I have no idea how to sort them out. Anyone here know any facts? --Vasha 23:05, 2 October 2017 (EDT)

I'd guess there are three: a pulp artist, a modern artist and an author (though the latter two may prove to be the same). Stonecreek 12:56, 3 October 2017 (EDT)

Murray Leinster's Long Ago, Far Away proved to be not a NOVEL

If there aren't any major objections this, so far a (Complete Novel), was found to be in fact a SHORTFICTION of novella length, and should be unvarianted and re-title-typed. The background is to be found here. Stonecreek 13:02, 3 October 2017 (EDT)

White Noise Press

The small press White Noise Press has a list of everything (?) that they have published since 2006 here. Most (and perhaps all) of the listed books look like SF by genre regulars. Calling for volunteers willing to work on reconciling it with our data. Ahasuerus 15:29, 4 October 2017 (EDT)

Server downtime - 2017-10-08 at 3:30pm

The ISFDB server will be unavailable between 3:30pm and 3:45pm server (US Eastern) time. Ahasuerus 14:57, 8 October 2017 (EDT)

The server is back up. Old versions of Wiki pages have been purged. "User Search", which shows "Last User Activity Date" has been added to Advanced Search. Next I will create a Wiki template to link to it. Ahasuerus 15:46, 8 October 2017 (EDT)
A new template, "UserActivity", has been created. If you enter {{UserActivity|<username>}}, it will link to the new User Search page for the specified user. For example, UserActivity Fixer links to the User Search page for Fixer. Ahasuerus 15:51, 8 October 2017 (EDT)

"Probe", as credited to Margaret Wander Bonanno

We list the Star Trek novel Probe as being by Margaret Wander Bonanno. Bonanno wrote a 17-page essay about this book, the first sentence of which is "This is the tale of PROBE: The Novel I Didn't Write". A summary of this essay is on Fanlore. The executive summary of this essay is: Bonanno wrote about 7% of that book; an initial rewriter (believed to be J.M. Dillard) wrote much of it; the final rewrite, and hence the final version, of the book was done by Gene DeWeese, whom Bonanno credits as the primary writer, and who received most of the royalties. Based on this essay, I propose to change the author of that book to "Margaret Wander Bonanno and Gene DeWeese", with a link to this Fanlore page. This affects 5 verifiers (who will be notified separately), but is an unusual enough change that I felt I should bring it here to see if there are any objections. Chavey 04:22, 11 October 2017 (EDT)

My first reaction was "How confident are we that the "initial rewriter [who] wrote much of it" was J. M. Dillard?" After reading this discussion, I think the answer is "there are some hints, but nothing official". I guess the "initial rewriter" will have to be relegated to Notes. Ahasuerus 16:07, 11 October 2017 (EDT)
I agree. Maybe some day the book will be re-issued and 'corrected' [like MZ Bradley/PE Zimmer for Hunters of the Red Moon]. Until then notes are fine. --~ Bill, Bluesman 12:17, 12 October 2017 (EDT)
The suggestion that J.M. Dillard is the initial rewriter doesn't seem strong enough to put that in the credits: I was only thinking about it as something someone would find if they followed the link to fanlore, but putting it in the notes seems reasonable. That Gene DeWeese is a co-author (or main author) seems strong enough to put it in the actual "author" slot. Chavey 00:12, 13 October 2017 (EDT)

Development update: source control changes

(If you are not a developer, you may want to skip this section.)

As some of you know, SourceForge, the Web site which we use for source/revision control purposes as well as to make our software publicly available, will stop supporting CVS at the end of November. We have to decide whether to move to another Web site which still supports CVS or to migrate to another source control package like Subversion (SVN), Git or Mercurial.

Considering how old CVS is (last update in 2008) and how few sites still support it, I think it's safe to say that we will need to migrate to another solution sooner rather than later. We might as well do it now since we have to do something anyway.

As far as choosing the new source control package goes, my thinking is:

  • SVN is more similar to CSV than other solutions, which is a double-edged sword. Although it's newer than CVS, it's almost legacy software at this point.
  • Git is by far the most popular package, but it's also the most complex one.
  • Mercurial is similar to Git since they are both decentralized source control systems, but it's supposed to be easier to use. I don't have first hand experience with either, so I have no opinion. Mercurial's two disadvantages are that it's less popular than Git and that it requires a specific version of Python, which will conflict with the version that we use. There are ways around Python conflicts, but it's something to consider.

I am currently in the process of setting up an SVN repository and a Git repository on SourceForge. I plan to experiment with both, including loading the history data from our current CVS repository, over the next few days. If you have any thoughts/ideas re: the process, please post them here. Ahasuerus 15:47, 11 October 2017 (EDT)

Git would probably provide the best flexibility. Also, should coding ever spread out again, candidate contributors are more likely to be familiar with it. BTW, Amazon's CodeCommit is Git. It's "free" for small numbers of users, although I don't know how much you have to buy in AWS to have access to it. --MartyD 16:56, 11 October 2017 (EDT)
Thanks for the heads-up about Amazon CodeCommit! Their "free tier" supports up to 5 users who, as near as I can tell, need to have an AWS account to access the data. I am concerned that it could be an issue for third parties trying to get a copy of our software. Ahasuerus 18:46, 11 October 2017 (EDT)
I've worked with CVS, SVN and Git for years now and I'd advise against moving to a website that still supports CVS. CVS is dead, and that's for some very good reasons. Once you're used to SVN or any other modern versioning system you will never look back to CVS (except for in horror about how cumbersome it was to use). The time to move to another CVS site is much better spent by choosing a site and versioning system which is future-proof for a while.
As for SVN, I wouldn't say that it is almost legacy software now: I'm quite sure it's still widely used - at least in-house in companies (in open source maybe not so much considering the GitHub hype). SVN is still actively developed and will be around for a while. It definitely has its place in world because it's easier to use than Git and has everything you need for simple teamwork and for keeping a good and consistent commit history. Git has a mind-boggling amount of features and options with a pretty high learning curve, and you likely never need most of it, especially if you want to keep the current development process as it is (with you doing most of the development). I'm mostly setting up new projects with Git nowadays, but I keep older ones in SVN because I currently don't feel any need to migrate them to Git.
To sum it up: both SVN and Git should have all features you need for ISFDB development. Choosing Git might be a better choice if you want to attract new developers in the long run. But I think the question may be rather: how much do you want to learn? :) BTW: Switching from SVN to Git later technically is quite easy because the migration (including the commit history) is quite simple (there's even a bi-directional bridge so you can mirror from SVN to Git and vice versa). If you need links to some good tools and documentation regarding SVN or Git let me know and I can post a few. Jens Hitspacebar 17:04, 11 October 2017 (EDT)
Very informative, thanks! Unfortunately, "how much do you want to learn?" is becoming the key question for me. I find it much much harder to learn new concepts these days, which is likely to become an ever larger issue going forward. Under the circumstances, "a mind-boggling amount of features and options with a pretty high learning curve" may not be the best choice. I remember being peripherally exposed to SVN in the 2000s and it seemed manageable. Ahasuerus 18:59, 11 October 2017 (EDT)
Yes, it should be manageable. If I remember correctly I found the transition from CVS to SVN quite easy and somehow "natural" when it happened about ten years ago. There was really no big problem, the main reason may be that they both are centralized versioning systems. Learning Git however was different beast because as a distributed version control some of its core concepts are completely different from SVN (and therefore from CVS). However, most of these differences only become relevant if you're not the only developer but working in a team and have branches and have to merge code and merge conflicts etc etc. Jens Hitspacebar 03:54, 12 October 2017 (EDT)
SVN is definitely much easier! --MartyD 20:59, 12 October 2017 (EDT)
SVN is easy to learn. Git is structured for Open Source programming, where you may have large numbers of potential contributors, and you have to be very careful about letting most of them submit anything. As a result, the algorithms for doing merges, "pull requests" (request that people look at your proposed changes), and handling merge conflicts is very sophisticated, and very hard to learn. We are not doing the kinds of development that Git is designed for, e.g. you tend to work on one thing at a time -- Git supports dozens of teams working in dozens of areas simultaneously. XKCD has done a couple of cartoons about the complexity of Git: https://xkcd.com/1667/ and https://xkcd.com/1597/. SVN, on the other hand, works plenty well enough for projects with a modest number of developers, and is far easier to work with. Chavey 00:31, 13 October 2017 (EDT)

(unindent) Thanks for your thoughts! I guess it will be SVN for now. Ahasuerus 10:04, 13 October 2017 (EDT)

The conversion has been completed and our Wiki-based documentation has been updated. If you find that anything is missing and/or wrong, please let me know. The CVS data hasn't been migrated and remains in the original CVS repository in a frozen state. Ahasuerus 15:45, 25 October 2017 (EDT)

Leonid Andreyeff

Excuse me for asking, but what is the difference between this Lazarus and this Lazarus by Leonid Andreyeff. MLB 19:55, 15 October 2017 (EDT)

The only difference between the two translations that I can see is that one spells the author's last name "Andreyeff" while the other one uses "Andreyev". Ahasuerus 21:22, 15 October 2017 (EDT)

Regular search changes - ISBNs/Catalog IDs

Please note that the way the regular search logic handles ISBNs/Catalog IDs has been changed to look for complete matches rather than for partial matches. For example, if you enter "#PP10" in the search box, it will now find #PP10, but not #PP1004-N, #PP107-N or #PP1008-N. If you need to look for partial matches, please use Advanced Publication Search instead. Ahasuerus 19:00, 26 October 2017 (EDT)

Last User Activity Date

The other day I was working on converting our source code repository from CVS to SVN and needed a guinea pig. I ended up implementing FR 1099 since it was a straightforward change. The following pages were modified:

One thing that I noticed after adding a "Last User Activity" column to these pages was that the Last User Activity date of some users was "None". For example, consider our third most active voter, User:Tmeekins. S/he has never created a submission, never edited a Wiki page and never verified anything. Since we don't store the date when a vote or a tag is added to the database, voting and tagging are not considered "user activity" for the purposes of calculating the "Last User Activity" date.

The question then is whether we want to change the name of these columns from "Last User Activity" to something like "Last Edit/Verification"? Ahasuerus 19:10, 26 October 2017 (EDT)

That seems like a good change. ···日本穣 · 投稿 · Talk to Nihonjoe 20:01, 26 October 2017 (EDT)

Adding Fanzines, question 1: Filk Songbooks

I haven't been around much doing "normal" things for the last several months, because I've been working towards entering a variety of fanzines into the system. Collating several collections of fanzines, cleaning up the data, and combining the data from these various collections into a format that I can transfer into our system has been quite time-consuming. I currently have about 15,000 fanzine titles, and well over 50,000 issues in my database, and my goal is to be able to get them to the stage that I can enter them into the isfdb database next summer. As I prepare these, though, I have three questions I want to get advice on in terms of inclusion/exclusion. I'll put these three questions into separate threads to simplify discussion. Your thoughts on any of these questions would be appreciated.

Question 1: Should I include fanzines that are Filking Songbooks? It seems that these are very similar to collections of spec fic relation poetry, so they should probably be included, but I was unable to find any policy on this. I was only able to find one filking book (Music of Darkover) and one issue of one filking fanzine (Xenofilkia), so it seems like we may have had some unwritten policy against such things. I've got listings for a little over 100 filking songbook fanzines listed, so it's not a large number, but I wanted to ask if I'd missed something. Chavey 01:39, 27 October 2017 (EDT)

The only reference to a filk-related discussion that I have been able to find is the following 2011 comment at the bottom of Help:Screen:Moderator:
  • There is currently a little question over whether Filk music reviews are "in".
The larger problem that I see with fanzines is that they form a spectrum from SF-centric to so far removed from SF as to be arguably ineligible for inclusion. I suspect that filking has a similar problem, but I don't know much about it. Ahasuerus 19:52, 27 October 2017 (EDT)
I had run across that "Filk music reviews" comment as well, but since that's the only comment either of us could find, it seems that the policy has not been discussed. There is a related question as to when we should include *any* listed fanzine. Some of the perzines I've seen really have no SF related info. Some of them come from important figures in fandom, and hence by analogy with "above the threshold" for regular writers, could be included on those grounds, but others that are listed in, say, the Ned Brooks index, really seem like they're irrelevant to our efforts. The challenge is that an awful lot of the zines on the regular "SF fanzine lists" have too little information to know whether they should be included or not. My inclination is to default to including them, along with a note of the form "subject material unknown to us"; and only discard titles if we *know* they are non-relevant. Chavey 20:21, 27 October 2017 (EDT)
I was thinking about perzines as well. Unfortunately, I don't know what percent of perzines are effectively "SF-free". Using our "threshold" as the inclusion criteria seems reasonable, but I am not sure what to do about the rest. If 80%+ (??) are only marginally related to SF, then it may be safer to exclude perzines by "under the threshold" authors unless there is some indication that they contain SF content. Ahasuerus 15:27, 28 October 2017 (EDT)
Fancyclopedia and ZineWiki do a pretty good job of identifying perzines within their listings. Fanlore organizes things by fandoms, and hence seems to exclude perzines. But the massive listings in libraries, with collections like the Ned Brooks, Eaton, Horvat, Lenny Kaye, Paskow, Coslet-Sapienza, Ransom, etc. don't seem to distinguish perzines from other types of fanzines. Those are the ones that are going to be harder to decide what to do with. My general sense though, from the 2,000 or so fanzines I have, is that even the titles that are listed as perzines are usually (but not always) SF related in some form, even if it's just the inclusion of a convention report. Chavey 00:42, 29 October 2017 (EDT)
I have no objection to including perzines – I think it's mostly a meaningless distinction to us if they have SF content, although I agree only the SF content should be indexed which is a rule of thumb I follow when indexing any fanzine. They often also have letters from authors waaay above the threshold, and perzines do win awards too. My two cents. PeteYoung 05:11, 29 October 2017 (EDT)

Adding Fanzines, question 2: Artists Portfolios

The fanzine world included a lot of fan art portfolios, including portfolios that had no artwork published in regular fanzines, such as Pern Portfolio. Our included SF "art books" tend to be limited to artists who have their art regularly published in traditional SF publications. Should I include fan art fanzines, even if I have no information about what art was actually used in other publications? Chavey 01:39, 27 October 2017 (EDT)

Adding Fanzines, question 3: Fanzine Novels

Once the Fan Fic movement got going strong (pretty much with Star Trek fandom), lots of fans published standalone novels such as Alkarin Warlord (or novellas, etc.). These were referred to as fanzines, had the "stapled, letter-size" format of a fanzine, and were distributed in much the same ways as periodical fanzines. But from a modern viewpoint, these seem like the same thing as current self-published books, which we enter as novels. Should these self-published novels be entered as novels, or as single-issue fanzines? Chavey 01:39, 27 October 2017 (EDT)

If they are novels (or novellas. novelettes, etc.), they should be entered as such. If something is already entered incorrectly, we should correct it. I agree that the publications you're discussing as simply early self-published works. ···日本穣 · 投稿 · Talk to Nihonjoe 01:57, 27 October 2017 (EDT)
As long as these self-published novels/novellas were not related to established fanzines, I think it would be best to enter them as NOVELs/CHAPBOOks. Those that were published as "special issues" of pre-existing fanzines (assuming that there were any) may be better handled as fanzines with serialized content. Ahasuerus 19:42, 27 October 2017 (EDT)
I agree with this. For example, R & R Special Editions were a couple of solo novels, but I would list them with the R & R fanzine. Similarly, Probe had a set of three specials that were a "title series" of novels, but it's still natural to list them with "Probe". Something Hidden is a standalone novel (not a fanzine special issue), but is one of a series of three novels ("The Explorer Series"). In analogy with current self-published books, we would list them as three NOVELS in a title series. Does that seem right to you? Chavey 20:51, 27 October 2017 (EDT)
Sounds reasonable to me. Ahasuerus 15:21, 28 October 2017 (EDT)

Advanced User Search changes

Advanced User Search has been modified to display the user's moderator flag. A similar change has been made to the "Top Contributors" pages, which will be regenerated in about an hour. Ahasuerus 00:05, 28 October 2017 (EDT)

Changes to post-submission and moderator review pages

The field names displayed on post-submission and moderator review pages have been tweaked to match what is displayed on bibliographic and edit pages. Ahasuerus 13:12, 29 October 2017 (EDT)

Binding -> Format

All known occurrences of the word "Binding" have been changed to "Format". Please let me know if I missed any. Ahasuerus 18:13, 29 October 2017 (EDT)

Mr(.) Skelmersdale in Fairyland

The H.G. Wells short story "Mr Skelmersdale in Fairyland" is one of many works for which we have Titles titles that differ in punctuation only (see also spacing and caps), namely as "Mr" T1029245 (as 1903-02) and as "Mr." T63081 (as 1901). At the moment neither is a variant of the other.

I don't doubt that 1903-02 is the correct date, as we have stated there with a source, and I have in the queue Submission 3607144 which gives complete information on the February 1903 publication with link to its first page at HathiTrust.

For what it's worth, the story is presented there as "MR Skelmersdale in FAIRYLAND", incorporated in the head illustration pp. 3; Skelmersdale is "Mr." in the text; the running header is "MR. SKELMERSDALE IN FAIRYLAND".

In the second Title record, as "Mr.", we say "First published: The Strand, 1901", no source. The FictionMags Index reports complete issue Contents for The Strand at FictionMags and it does not report Skelmersdale except as we say in the first Title record, as "Mr". Nor did I find him under 'M' in the index at the back of bound volumes 21 and 22 (1901) at HathiTrust: The Strand Magazine.

--Pwendt|talk 17:52, 31 October 2017 (EDT)

Technical problems - 2017-10-31

We are currently experiencing technical problems. I am currently looking into them. There is no ETA right now. Ahasuerus 19:43, 31 October 2017 (EDT)

OK, I think the problems have been resolved. Sorry about the hassle! If you find anything unusual, please let me know. Ahasuerus 19:53, 31 October 2017 (EDT)
Away for a couple of days, and can now log into my talk page/wiki but not the DB?? Keeps coming up with http://cgi-bin/dologin.cgi?index.cgi+0 which doesn't seem right. My browser just laughs .... --~ Bill, Bluesman 12:39, 1 November 2017 (EDT)
Okay, played with it a little and www.isfdb.org is being 'dropped' from the URL. By inserting it before cgi-bin I can log on. --~ Bill, Bluesman 12:49, 1 November 2017 (EDT)
It's doing this for every page I try to open on the DB side. --~ Bill, Bluesman 12:51, 1 November 2017 (EDT)
Thanks, let me take a look... Ahasuerus 13:02, 1 November 2017 (EDT)
I have tweaked some configuration settings. Does it look better now? Ahasuerus 13:13, 1 November 2017 (EDT)
Nope, still drops the www.isfdb.org/ in the page address. Let me refresh/restart and try it again. --~ Bill, Bluesman 13:26, 1 November 2017 (EDT)
Yes, now works. I did the refresh/restart before posting but no effect then, so your changes were the key. Thanks! --~ Bill, Bluesman 13:31, 1 November 2017 (EDT)
Great! Ahasuerus 13:36, 1 November 2017 (EDT)

Mouseover Help bubbles added to publication formats

As per this discussion, mouseover Help bubbles have been added to publication formats. If I missed anything or introduced a new bug, please let me know. Ahasuerus 17:49, 1 November 2017 (EDT)

At the moment, we have an option in "My Preferences" for "Do not display mouseover help on Edit pages", can something like that be added to cover the new ones as well? [The 'Do not display' we have doesn't work on the External ID section]. --~ Bill, Bluesman 18:09, 1 November 2017 (EDT)
I'll second that. Too much visual distraction. --Marc Kupper 01:37, 4 November 2017 (EDT)
In thinking about is more - I don't mind the mouseover help on the edit pages but would like a way to disable the "?" images that are now next to the binding on all of the display pages. Leave the span with the title in place. Someone hovering on a "pb" will see the hovertext. --Marc Kupper 02:16, 4 November 2017 (EDT)
My preference is with Marc. Remove the "? images and just have the mouseover. I'd prefer that on all pages. -- JLaTondre (talk) 08:30, 4 November 2017 (EDT)
Well, if the problem is that the current question mark icon is too big and too intrusive, how about we make it smaller and less intrusive? For example:
  • tp, or
  • tp
That way new users will still be able to tell that mouseover Help is available while experienced users won't be bothered as much. What do you think? Ahasuerus 13:14, 4 November 2017 (EDT)
I'd much rather be able to turn it off. Same with the '?' in the editing window for External IDs. At least for most things the 'help' that appears is small, but for that one it's the whole list and because the '?' is so close to the drop-down it jumps out all the time. If that one can't be turned off like the rest of the '?'s in the editing window could it at least be moved to the left side? Another thought: couldn't the "Format" in the header become an active link to the definitions page for same instead of each individual designation having the '?' ? --~ Bill, Bluesman 17:07, 9 November 2017 (EST)

External IDs

Is there a consensus on if we should be moving things out of the notes into External IDs? Many records have an OCLC number and often link as the last line. I assumed the reason these had not been automatically migrated was there's little consistency to the formatting. Some records have LCCN links. Should those be migrated too? Obviously, if there's a pub note that mentions an LCCN or OCLC as part of the note we'd leave that note in place but do we copy the numbers into External IDs? --Marc Kupper 02:56, 4 November 2017 (EDT)

Yes, supported external IDs should be moved to the external ID section. If there is a note that references the external ID, the link should be removed & the note edited (as necessary). There are some cleanup reports (see Notes/Synopses section) that capture items to be worked off. Mostly though, seems people just fix them when they run across them. -- JLaTondre (talk) 08:27, 4 November 2017 (EDT)
Ahasuerus was going to work on a macro to move most of these automatically, so I (at least) had stopped doing them manually -- except when I'm editing a Pub record. Ahasuerus: Do you expect to do that macro, or should others (like me) get more aggressive about moving them? Chavey 01:11, 5 November 2017 (EDT)
I updated that Community Portal section a few weeks ago, but it may have been lost in the shuffle. Sorry about that, I should have created a new section. Here is what I wrote at the time:
  • I have pretty much abandoned my attempts to convert OCLC links automatically, at least for now. There were just too many permutations and gotchas.
Ahasuerus 08:45, 5 November 2017 (EST)
I'm not surprised! I'll work on it my free time, if I can find any. Chavey 22:56, 5 November 2017 (EST)

Advanced Publication Search - Publication Formats are now a drop-down list

Advanced Publication Search has been changed. If you select "Format" in the leftmost column, the rightmost column changes from a "free text" input field to a drop-down list which includes all supported formats. The middle column, which has values like "is exactly", "contains", etc, hasn't been changed.

If everything looks OK, we can make similar changes to other fields, e.g.:

  • publication types
  • title types
  • title lengths
  • languages
  • the 4 title flags:
    • juvenile
    • non-genre
    • novelization
    • graphic format

Ahasuerus 16:44, 4 November 2017 (EDT)

Unsupported HTML

[strike]Today[/strike] is valid html (but with chevrons, now it is more like bbcode). [small][/small] does the job too, but is not in the list of official html. [size=-1] is, but would that get flagged down too?--Dirk P Broer 19:06, 4 November 2017 (EDT)

"strike" is no longer supported in HTML5, but "del" and "s" are. The last two are also fully supported by the ISFDB software, so I would recommend using them instead of "strike". Ahasuerus 11:06, 5 November 2017 (EST)
both "strike" and "s" are mentioned as 'obsolete' here. BTW "small" is mentioned there as valid HTML5, but gets flagged as unsupported in ISFDB -while it works.--Dirk P Broer 17:17, 5 November 2017 (EST)
"s" has a complicated history. As the site that you linked explains:
  • The "s" tag was deprecated in HTML 4, but it has been redefined in HTML5.
As far as "small" goes, I see no reason why we couldn't support it. Give me an hour or two and I will add it to the list of allowed tags. Ahasuerus 17:39, 5 November 2017 (EST)
Done. Ahasuerus 18:37, 5 November 2017 (EST)
Thanks!--Dirk P Broer 20:16, 5 November 2017 (EST)

Blocking malicious IPs

We have seen an unusually high number of traffic spikes over the last few weeks. A closer examination of our access logs determined that they come from the same group of IP addresses which send up to 100 simultaneous requests to the ISFDB server. I am not sure if they are trying to scrape our data or doing something else, but it's been affecting system performance since they do not respect the limits that we set for data mining robots.

I don't like blocking individual IP addresses because they change all the time, but in this case we have little choice. We'll see what happens next. Hopefully no innocent bystanders have been affected. Ahasuerus 11:15, 6 November 2017 (EST)

Can you set an expiry on the blocks? If so, you could set it for a few days or a week to block what's happening now. ···日本穣 · 投稿 · Talk to Nihonjoe 16:41, 6 November 2017 (EST)
There are tools that can modify lists of banned IP addresses dynamically. As far as I can tell, they take a certain amount of time to configure and monitor. If things deteriorate, we may have to give them a try. Ahasuerus 20:01, 6 November 2017 (EST)

series: Broken Empire vs. Broken Empire Universe

The short fiction collected in the first edition of Mark Lawrence's Road Brothers was all assigned to his Broken Empire series. The latest edition, containing four additional stories, has been placed more generally in the Broken Empire Universe. Are there any Lawrence readers here who would care to figure out where exactly the various short fiction belongs? --Vasha 01:44, 7 November 2017 (EST)

I seem to be the sole verifier of Lawrence's magnificent series, but I don't read e-books. So when I get the new hardcover collection, I'll take a shot at organization. My first thought is that "Broken Empire Universe" should be renamed "Broken Empire" and the sub-series should be changed from "Broken Empire" to "Game of Thorns" or some such. If the stories involve Jorg Ancrath or activities related to him, they belong in the sub-series, otherwise in the overall series. More when I read the new collection. Bob 12:11, 8 November 2017 (EST)

Advanced Search changes - 2017-11-10

Advanced Search has been further modified to display drop-down lists of valid values when specifying title types, publication types and title lengths. The 4 title flags (juvenile, non-genre, etc) are still outstanding. Ahasuerus 18:20, 10 November 2017 (EST)

Dav%Stevenson

Would anyone happen to know if Dave Stevenson, David Stevenson and David G. Stevenson are the same artist? Ahasuerus 16:24, 11 November 2017 (EST)

SFE3 seems to think the first and the second are the same, but I wouldn't be surprised when more than one artist lives under these three names, given the difference in technique between this David Stevenson (SFE3) and this David Stevenson (www.david-stevenson.com).--Dirk P Broer 18:36, 13 November 2017 (EST)
1st/2nd most likely [Pocket books connection] and the styles are the same. --~ Bill, Bluesman 18:40, 13 November 2017 (EST)

2017-11-12 downtime

The Web server was down between 5:54am and 9:05am due to technical problem. Everything should be back up now. Ahasuerus 09:09, 12 November 2017 (EST)

Nominating user Hitspacebar for moderator

See Moderator Qualifications#Becoming a moderator for the nomination process.

I would like to nominate user Hitspacebar (talkcontribs) for moderator. Right from the beginning he had a steep learning curve that has enabled him to manage most types of submissions. Also he has offered help for and done many betterments for all kinds of structural or informational changes, including software. So I think he meets the Moderator Qualifications and he has accepted the nomination. Stonecreek 10:59, 12 November 2017 (EST)


Support

  1. Support, as nominator. Stonecreek 10:59, 12 November 2017 (EST)
  2. Support. I was thinking the same thing myself earlier this month. Ahasuerus 11:02, 12 November 2017 (EST)
  3. Support. -- JLaTondre (talk) 11:39, 12 November 2017 (EST)
  4. Support. --Ron ~ RtraceTalk 13:09, 12 November 2017 (EST)
  5. Support. Deserved. Rudolf Rudam 13:11, 14 November 2017 (EST)
  6. Support. I don't think I handled many of his submissions, but looking at his talk page/contributions I see no issues/problems. --Willem 15:19, 14 November 2017 (EST)
  7. Support. Albinoflea 12:56, 16 November 2017 (EST)
  8. Support. Always good submissions, covering a variety of areas and complexity. --MartyD 20:06, 16 November 2017 (EST)
  9. Support. ···日本穣 · 投稿 · Talk to Nihonjoe 20:10, 16 November 2017 (EST)

Oppose


Comments/ Neutral

  1. I can't recall ever handling a single edit, so should remain neutral. Good luck! --~ Bill, Bluesman 18:36, 13 November 2017 (EST)

Outcome

The nomination passes. The moderator flag has been set. Congratulations! Ahasuerus 11:04, 17 November 2017 (EST)

Fantastic. Thanks a lot for the support and the trust in my capabilities. Jens Hitspacebar 12:25, 17 November 2017 (EST)

OMNI Magazine back in print

OMNI Magazine is back in print.[1] Should the new issue be added to the old Omni Magazine series ? --Auric 12:43, 12 November 2017 (EST)

There is certainly precedent, e.g. we list the two runs of Super Science Stories / Super Science Novels (1940-1943 and 1949-1951) and the many incarnations of Weird Tales as single series. Ahasuerus 00:21, 13 November 2017 (EST)
Okay, thanks.--Auric 10:38, 17 November 2017 (EST)

Advanced Search changes - 2017-11-13

The 4 title flags (juvenile etc) and title languages have been converted to drop-down lists. In addition, Advanced Search should no longer error out with a Python error when processing complex OR searches. Ahasuerus 00:16, 13 November 2017 (EST)

Authors' working languages have been converted as well, which I believe wraps up this FR. Ahasuerus 00:36, 13 November 2017 (EST)

Mouseover help changes

Based on a recent discussion, all mouseover bubbles displayed on bibliographic pages have been changed to somewhat fancy question marks. This includes all bubbles with transliterated values as well as publication format codes -- e.g., see this page. Mouseover bubbles displayed on edit pages haven't been changed.

Let's see if this change makes the bubbles less intrusive and addresses the initial concern.

P.S. The newly added "fancy question marks" became an internet standard in 2010, so hopefully everyone's browsers are able to display them in 2017. If not, we may have to use plain vanilla question marks instead. Ahasuerus 18:08, 13 November 2017 (EST)

Are they supposed to look like very miniscule rectangles? If so then they're displaying fine. Certainly less intrusive. But if we have to go 'plain', I prefer strawberry ..... --~ Bill, Bluesman 18:34, 13 November 2017 (EST)
Do the minuscule rectangles look like miniaturized versions of the huge black-and-white rectangle on this Wikipedia page? If so, then your browser doesn't recognize the "fancy question marks" added in the 2010 standard and reverts to a generic "replacement character". Perhaps it's not such a bad thing after all since these "replacement" rectangles have a question mark embedded. Ahasuerus 18:51, 13 November 2017 (EST)
No, just a blank rectangle maybe 1mmx3mm , sort of [] this but a little smaller and <sup>. It's 'quiet'. --~ Bill, Bluesman 23:20, 13 November 2017 (EST)
That's ... odd. Oh well, so much for fancy question marks and quirky browsers! I have changed it to a plain vanilla question mark (surely vanilla > strawberry?), so hopefully things should look better now. Ahasuerus 16:09, 14 November 2017 (EST)
I didn't mind the wee rectangles [or the vanilla ?], much easier on the eyes. --~ Bill, Bluesman 23:27, 14 November 2017 (EST)

Nominating user Dirk P Broer for moderator

See Moderator Qualifications#Becoming a moderator for the nomination process.

I would like to nominate user Dirk_P_Broer (talkcontribs) for moderator. For several years now he has been a considerable source of bibliographical updates, finding mistakes and omissions. In his huge amount of submissions he bettered many entries, and also found useful information about authors and etablished overlooked pseudonyms. He obviously also knows about the intricacies ISFDB has in store for the user. I think he meets the Moderator Qualifications and he has accepted the nomination. Stonecreek 09:12, 14 November 2017 (EST)


Support

  1. Support, as nominator. Stonecreek 09:12, 14 November 2017 (EST)
  2. Support. Long overdue. Rudolf Rudam 13:12, 14 November 2017 (EST)
  3. Support. I think he would do a good job. ···日本穣 · 投稿 · Talk to Nihonjoe 14:25, 14 November 2017 (EST)
  4. Support. The first nomination came too soon and the second stranded in miscommunication. The third time may be a little late, but is well deserved. --Willem 15:26, 14 November 2017 (EST)
  5. Support. He's been ready for quite some time. --Ron ~ RtraceTalk 20:47, 14 November 2017 (EST)
  6. Support. One suggestion: when adding notes to records, "As of 2017" is a better choice than "Now" or "Currently". Try to put yourself in the shoes of a "naive" 14-year-old user examining our records in 2031 to see if they will still make sense to her :-) Ahasuerus 21:25, 14 November 2017 (EST)
  7. Support. I think he would do a good job, and I'd like to see more moderators. Bob 11:54, 15 November 2017 (EST)
  8. Support. Albinoflea 12:56, 16 November 2017 (EST)
  9. Support. My thoughts haven't changed. This is long overdue. --MartyD 20:11, 16 November 2017 (EST)

Oppose

  1. I don't see any fundamental change from the first nomination. --~ Bill, Bluesman 23:29, 14 November 2017 (EST)

Comments/ Neutral

Outcome

The nomination was successful. The moderator flag has been set on the account. Congratulations! Ahasuerus 23:52, 20 November 2017 (EST)

Joe Hill's Graphic Novels

I'm holding several submissions of graphic novels by Joe Hill. In addition, this author already has about 8 graphic novel works entered in the database. Per our standards, ISFDB does not index graphic novels, comics, etc. We have made exceptions (though not stated in policy) for high profile authors (ex. Neil Gaiman), but we've generally kept the threshold for this well above what we use for non-genre works. Hill doesn't seem to fit this level in my opinion. However, we're now up to about 400 graphic works[2] in the database and many are not by significant genre authors (ex. single story). So:

  1. Is it time to revisit the inclusion criteria?
  2. Should we just reduce the exception level? Treat it more like non-genre? Include anything in an otherwise genre publication?
  3. Accept these submissions w/o addressing the above two?
  4. Deny these edits and delete Hill's other graphic works?

What are people's thoughts? -- JLaTondre (talk) 17:22, 15 November 2017 (EST)

Excellent questions. I don't have many answers, but I'll make a couple of observations:
  • Graphic novels have been growing by leaps and bounds as of late. Many popular SF series have graphic offshoots. Some of them are adapted by the original author while others are apparently done by a third party with some guidance by the creator of the universe.
  • Although I still tend to think of Joe Hill as a new kid on the block, he has won a dozen (!) major awards and has another couple of dozen nominations under his belt.
Ahasuerus 18:28, 15 November 2017 (EST)
A few thoughts only. I read a lot of graphic novels (well.. I buy them and shelf them anyway...) and I still think that if we open the door for them here, we are going to double the scope of the project. Which is not a problem on its own - except that we have major gaps in the text-based fiction. And if we start with the graphic ones, we need to start asking of radio-plays and movies based on stories and novels also belong (how is a graphic novel different from a radio-play adaptation?) and we will end up somewhere down the rabbit hole. So as much as I like them, I do not think that we should have any graphic novels - major authors or not. But that ship had sailed. So at this point, I would say to treat them as non-fiction - if we would add a non-fiction book by the author, we add their graphic novels as well; if we won't, we don't. That would make the Hill ones eligible. Annie 18:40, 15 November 2017 (EST)
My opinions are similar to Annie's: generally I'm not for including them, but, alas, they have crept in. I don't see much choice as to allow them for above-the-threhold authors, as we try to capture all of their works. There is a difference to radio plays and movies, though, as they usually are authored by genre authors, whereas the former mostly are 'adapted from' their work (at least that's the way for the modern comics / graphic novels as Joe Hill's). To treat them as non-fiction would be the best way to handle them, but maybe a new category 'non-fiction graphic' would be appropriate? It also seems that Joe Hill is becoming above-the-threhold. Stonecreek 23:24, 15 November 2017 (EST)

I find it a bit ironic that the only feedback has been from those opposed to graphic novels, but recommending their inclusion. ;-) I have accepted the edits. -- JLaTondre (talk) 17:19, 19 November 2017 (EST)

My 2¢: To me, most graphic novels seem to be long comic books with better binding. Many are reprinted comic books originally done in parts, or arcs. Unless we want to be another Grand Comics Database, I would disallow them. 500 existing GN's doesn't seem that many to delete. Some cases might be arguable, but that's what we do sometimes. Anyway, my vote is get rid of them all (and I'm guilty of entering a couple myself!) Doug / vornoff Vornoff 14:25, 25 November 2017 (EST)
Many novels have a critical graphic component. Do we start ignoring works because they have too many graphics? IMHO, you can exclude manga, comic books, and reprints of comic book in better bindings without ignoring graphic novels. It would be more reasonable to specifically exclude periodically printed comics and their reprints, than to create an arbitrary graphics to words ratio. Limiting to author's above the threshold seems a reasonable constraint as well. TAWeiss 10:30, 26 November 2017 (EST)
This is not a compelling argument. There is a significant difference between an illustrated novel and a graphic novel. And for the relatively very, very few works that straddle the border, it's no different than the genre / non-genre border (make a case by case decision with an leaning towards inclusion) which is a far, far more common issue that we handle fine. -- JLaTondre (talk) 07:37, 27 November 2017 (EST)
Careful with the "published as individual issues first" as well - more and more graphic novels get split and published in chapters before they get collected in a single book - without them being real comics issues. It is a very slippery slope and finding where the border betwen a GN and comics series is may not be as trivial. Annie 11:21, 27 November 2017 (EST)
I could live with something like that, with the exclusions you mention. I did mention some cases would be arguable. Just what is a graphic novel, then, anyway? Just a work that looks like a comic book but hasn't been printed before as a comic book? Or would there be some other criteria for inclusion? The Dark Knight Returns, for example, looks like it would be out for me. It was compiled from four comics issues and Miller doesn't look like he's over the threshold. If we left it in, shouldn't we enter the four comics and the other reprints of this work as well? On the other hand Coraline by Neil Gaiman, I would leave in. A cursory look seems like it was from his novella of the same name, plus he would definitely be over the threshold. No doubt it would be work to delete the comic book-y titles but as long as the editors were aware of this new direction, if agreed upon, they would get deleted over time by various editors/moderators as they come across them. No doubt this could open a can of worms; I can see where some verifiers might get bent out of shape if their entries were chosen for deletion so obviously, thought would have to be put into the decision to move ahead or not. Doug / Vornoff 00:24, 27 November 2017 (EST)
As well as Dark Knight, would Watchmen, V for Vendetta, From Hell, League of Extraordinary Gentlemen go as Alan Moore isn't that well known outside of comics? (Though the League GNs also have text stories. Would we add those stories separately?) Would books on art of comics and GN go as well? Mythology: The DC Comics Art of Alex Ross.

Title merge enhancements

As per FR 1095, Title Merge has been changed to ignore author order. For example, when merging two titles co-written by Niven, Barnes and Pournelle, the display logic will show Author values as identical regardless of the order in which they appear in the database. Ahasuerus 13:22, 16 November 2017 (EST)

Thank you! -- JLaTondre (talk) 17:34, 16 November 2017 (EST)
Sure thing! Unfortunately, FR 1096 is trickier to implement, in part due to the fact that a Title Merge submission can merge titles with different title types. Ahasuerus 18:21, 16 November 2017 (EST)

Award year validation

As per FR 1084, award year validation has been enhanced to disallow award years with months and/or days. Ahasuerus 16:30, 16 November 2017 (EST)

Family Name -> Directory Entry

As per Roadmap 2017, the field formerly known as "Family Name" has been changed to "Directory Entry". Please let me know if I missed anything. Ahasuerus 19:38, 16 November 2017 (EST)

Po-up validation enhancements

The error message which used to read:

  • For new/added publications the reference title should not be entered in the Content section. It will be added automatically at submission creation time.

has been fine-tuned. Here is what it looks like for magazines and fanzines:

  • When creating a new MAGAZINE publication, an EDITOR title should not be entered in the Regular Titles subsection of the Content section. It will be added automatically at submission creation time.

and for omnibuses:

  • When creating a new OMNIBUS publication, an OMNIBUS title should not be entered in the Regular Titles subsection of the Content section. It will be added automatically at submission creation time.

Ahasuerus 10:14, 17 November 2017 (EST)

Those are good. --Vasha 15:50, 17 November 2017 (EST)

Catalog IDs: support for multiple IDs per publication?

One of the enhancements that Roadmap 2017 calls for is adding support for multiple ISBNs per publication. The primary prerequisite is separating ISBNs and Catalog IDs which currently share the same field.

Here is what the road map currently says about Catalog IDs:

  • Move catalog IDs to a separate field, which will help with pubs that have both an ISBN and a catalog ID, e.g. book club publications.

As stated, it would be a reasonably straightforward change. I have already finalized the design document.

However, it occurs to me that it's possible for a publication to have multiple catalog IDs. An Advanced Publication Search on "ISBN/Catalog # starts with #%/" finds approximately 150 publication records, many of them published in Germany and Poland, although some of them appear to contain legitimate slashes. I also recall certain older (1960s) American paperbacks which had what looked like 2 different catalog IDs, but I never looked into it.

It's certainly possible to make "Catalog ID" a multiply occurring field similar to Authors, Transliterated Titles, etc. It will take more time to code, but, once finished, it should have no impact on performance or anything else. The question then is whether, based on everyone's experience with catalog IDs, it is worth doing. Thoughts? Ahasuerus 15:55, 17 November 2017 (EST)

I guess we'll go with a single-value field. Ahasuerus 00:01, 24 November 2017 (EST)

Poe's detective stories

Why are "The Mystery of Marie Roget" and "The Purloined Letter" not marked as non-genre (though "The Gold-Bug"is)? --Vasha 02:06, 19 November 2017 (EST)

Because noone had got around to it yet? :) Annie 17:03, 20 November 2017 (EST)
If no one objects, I will make the change a few days from now. --Vasha 19:18, 20 November 2017 (EST)
Sometimes the quality of our data for some of the best known stories is not very high. Many of them were entered early on, in the 1990s, which is to say before we had moderation, the "nongenre" flag, etc. Ahasuerus 19:22, 20 November 2017 (EST)
Would we consider Poe "above the threshold"? He's certainly one of the most well known authors in the world. ···日本穣 · 投稿 · Talk to Nihonjoe 21:14, 20 November 2017 (EST)
I would say that the creators of the genre are in this group automatically. When you think of Poe, do you think of him as a genre writer or as a crime writer? That's what this whole threshold is all about (when we stop counting the stories and whatnot) - is it a genre writer that sometimes works outside of it or a non-genre one with a few works that fit here. Annie 21:23, 20 November 2017 (EST)

(after edit conflict)

The Rules of Acquisition say:
  • The goal here is to avoid cataloging everything ever published by James Fenimore Cooper, Robert Louis Stevenson, Honoré de Balzac and other popular authors.
In other words, the "threshold" was designed to exclude some of the most well known authors in the world :-)
However, that's because most of their works are not SF. Poe is arguably different since many of his works are speculative. I am not sure what the percentage is, though. Ahasuerus 21:28, 20 November 2017 (EST)
As a matter of fact, a few of Poe's very well known stories aren't speculative per se, for example "The Pit and the Pendulum" (and "The Murders in the Rue Morgue," unrealistic orangutan notwithstanding). Nonetheless, they are among the founding texts of the horror and fantastic genres, and so they do, in my opinion, count as "genre" -- something that is as much a matter of context as content. "The Murders in the Rue Morgue" is reprinted in horror anthologies whereas the other Dupin stories aren't, so it's the only "genre" story in the series.
It isn't even necessary to take context into account to realize that the great majority of Poe's fiction is speculative. --Vasha 22:04, 20 November 2017 (EST)
Which is why I said "the creators of the genre" (should have been genres :) ). It is not about popularity, it is about the genre works :) Poe is in a different category because he is not just a popular author, he is one of our popular authors. I do not know - for an author that I know and had read, it is very easy to see if they should be above or bellow... I have a feeling we all are on the same page here :) Annie 22:24, 20 November 2017 (EST)

NOMMO awards?

Earlier today I saw a new SF award, the Nommo Awards for African Speculative Fiction, given by the African Speculative Fiction Society. I am not familiar with this recently formed organization, but Locus and Orbit seem to think that they are legitimate. Should we add them? Ahasuerus 15:29, 20 November 2017 (EST)

I'd say "yes". Annie 15:46, 20 November 2017 (EST)
I entered most of the shortlist as title notes a few days ago based on this tor.com article (which lists the winners whereas the organization's website hasn't been updated yet). There were a few short stories that I wasn't able to find much information about if anyone wants to do some digging. If this is added as a supported award, the title notes should be removed when converted to awards. -- JLaTondre (talk) 16:44, 20 November 2017 (EST)
OK, a new award type has been created. Moderators should be able to create new categories. Ahasuerus 20:36, 23 November 2017 (EST)
I've entered the short list. They also have a long list from which we're missing a lot of content if anyone has time to dig into it. -- JLaTondre (talk) 09:46, 24 November 2017 (EST)

Arthur Conan Doyle "above the threshold"?

A number of people in this community believe that the Sherlock Holmes stories belong in this database because, although they're not speculative, they've been thoroughly incorporated into a speculative metaverse that numerous writers draw on. Personally, I have no objection to that idea. But I'm wondering, should Doyle's other non-speculative fiction be catalogued wherever it occurs--is he "above the threshold"? On the one hand, his speculative fiction is percentagewise a small portion of his output; on the other hand, he did give us Professor Challenger and a few extremely popular supernatural stories such as "The Captain of the Polestar," "The Great Keinplatz Experiment," and "The Horror of the Heights." H. G. Wells is proof that an author can be above the threshold even if a minority of their writing is speculative. So what do you folks think? Here are some numbers about Doyle to help you decide:

  • NOVELS. Speculative (5) Holmes (4) Other (10)
  • SHORT FICTION Speculative (39) Holmes (56) Other (120)

--Vasha 22:22, 20 November 2017 (EST)

I am going to play the rebel today - under the threshold with an exception for anything Doyle's Holmes related (aka adopt Holmes into our cannon as non-genre because of the influence)). That opens the door to ask if modern Holmes stories belong here (and I would say - "no" unless they qualify on their own)... Annie 22:28, 20 November 2017 (EST)

Maurice Level -- how do we decide what's genre?

Maurice Level is a (fairly) renowned horror writer. As such, he has lots of stories in this database. However, his stories are strictly non-speculative, preferring what one commentator calls a "brutal naturalism". He writes of desperate people, sick & twisted people, gruesome scenes, horrible murders, and cruel ironies. Even his famous novel L'ombre/Those Who Return is a tale of madness that's only ambiguously supernatural, and it's his only work I know of that includes such hints.

So, I'm confused about how to decide which of Level's writings are "genre". Should we say that none of them are except Those Who Return? But some of the stories have been reprinted places like Weird Tales; does everything in Weird Tales automatically become genre? Or how about genre anthologies that contain both supernatural and non-supernatural horror, does appearing in one, several, or a dozen of those mean that a story should not be marked non-genre?

Yours uncertainly, --Vasha 15:41, 23 November 2017 (EST)

I think it's important to distinguish between two separate issues when evaluating a story/novel:
  • Is the work speculative fiction based on our definition of speculative fiction?
  • Does the work appear in a publication all of whose contents we would normally catalog, e.g. a genre magazine like Weird Tales?
If the answer to the first question is "yes" or "ambiguous", then we consider the story "genre". If the answer is "no", then we consider it non-genre.
If the answer to the second question is "yes", then we will include the story even if it is non-genre. However, the fact that we include it in the Content section of a Weird Tales issue doesn't automatically mean that we will list all of its appearances in other publications.
In the case of Maurice Level, L'ombre is presumably a genre work since it contains ambiguous speculative elements. On the other hand, his works of psychological horror are non-genre because our current definition of SF explicitly excludes "Purely psychological horror works that feature no supernatural or fantastical elements ". Ahasuerus 16:04, 23 November 2017 (EST)
OK, I've marked everything NG except L'ombre and the underground-city novel La cité des voleurs. (I forgot about that one because it hasn't been translated into English. I bet Black Coat Press is going to get to it soon.) --Vasha 17:31, 23 November 2017 (EST)

Step-by-step guide for entering translation added to help

Kurst had suggested a step-by-step guide a week ago about how to enter translations (see the howto's talk page). I have now added this guide with several corrections and additions by me to the Help:How_to_enter_translations page. Jens Hitspacebar 13:12, 24 November 2017 (EST)

Washington Irving's fiction

It seems to me that Washinton Irving is above the threshold--true, his books largely consist of travelogues and sketches, with a few tales thrown in, making speculative fiction very much a minority. But if you look at just his fiction then it's a majority. I've identified about 45 stories, of which 25 (including his most famous ones) are speculative. So I'm wondering if all of his fiction should be entered into the database. It wouldn't be hard; collections they would be in are already entered with partial contents. (Putting in all the nonfiction would be a lot more work and I don't personally feel like doing it.) Thoughts?--Vasha 00:24, 27 November 2017 (EST)

2017-11-28: Duplicate authors bug fix

The other day Darrah discovered a way to create title/pub records with the same author listed 2+ times. The functionality is not supported by the software, which is why the bug was causing display issues and even Python errors. The software has been fixed and I am currently in the process of correcting the affected records. If you see anything unusual, please let me know. Ahasuerus 17:07, 28 November 2017 (EST)

The data has been fixed. Ahasuerus 18:54, 28 November 2017 (EST)

Robert E. Howard Books in Foreign Languages

I'm going to start entering Howard books in a number of foreign languages, using HowardWorks.com as the source. I've asked for and received permission from Bill Thom, who runs the site, to use his scans for the covers. Is there something else I need to do to use these scans? Bob 00:32, 30 November 2017 (EST)

I have added Howard Works to Template:Image Host Sites. Give me an hour or so to update the software and then we should be good to go. Ahasuerus 10:44, 30 November 2017 (EST)
Done. Submit away! Ahasuerus 10:58, 30 November 2017 (EST)

Conan Universe Organization

There are three types of Conan stories: (1) those by Robert E. Howard, (2) those by REH and coauthors and (3) pastiches. Currently, the “Conan Universe” has subseries “Conan” and “Conan Pastiches”. The “Conan” subseries has one further subseries “Conan of Cimmeria”. These are “pure” Howard, no coauthors. Another important series was first published by Lancer, all of which are anthologies, with stories coauthored by L. Sprague de Camp and Lin Carter. There are numerous copies of these pubs in foreign languages and omnibus editions of multiple volumes in that series. The other prominent series was published by Gnome Press. Some volumes in that series are “pure” Howard, most are anthologies with stories coauthored with L. Sprague de Camp, the series editor. There are also some very large collections/anthologies (over 600 pages), most quite recent print on demand pubs.

I’d like to bring some order to the “Conan Universe”. Maybe something like

Conan Universe: "Robert E. Howard’s Conan", sub "The Complete Conan of Cimmeria"; "Coauthored Conan", sub "The Lancer Conans" and sub "Massive Conan Anthologies" (over 600 pages); and of course, "Conan Pastiches".

I’d like suggestions on how the “Conan Universe” might be organized. Bob 20:52, 4 December 2017 (EST)

We've currently got the Lancer/Ace, Gnome, Berkely, Sphere and Grant issues of the books as publication series and I think that is where they belong. If you try to make them into title series, you'll run into problems of how to treat the same title which appears in multiple title series (which can't be done with the existing software). For example Conan the Conqueror/The Hour of the Dragon is #5 in the Gnome series, #9 in Lancer/Ace, #1 in Berkley, #4 in Sphere and #11 in Grant. There really isn't a canonical ordering of the stories that I'm aware of. Even with the the organization that you describe, I believe you'll end up with anthologies that are a mixture of stories by Howard collected with those by other hands. You could certainly organize the stories as you suggest, but I think the books present a problem. --Ron ~ RtraceTalk 22:00, 4 December 2017 (EST)
Thanks for the input, Ron. I agree that "Conan the Conquerer"/"The Hour of the Dragon" is a problem. I think it's the only one because it's the only Conan novel Howard wrote (all Howard). The thing that kicked this off was the discovery of so many foreign language versions of the Lancer/Ace series. None of the other series seemed to generate this sort of interest. And there are too many pubs titled "Conan" and "Conan the Barbarian" that are completely unrelated. I'd just like to find some sort of organization. Bob 00:30, 5 December 2017 (EST)

Attention: Another Amazon Alarm!

Recently happened (or came to light) what maybe was to be expected: a link to a cover image supplied via an amazon link had suddenly a different image (one for a more actual printing).

To reconstruct the history: upon verifying, the image for this had been there, so I missed out on uploading a cover scan (but that I did just a few moments ago). The displayed / changed image can be seen when clicking on the DNB link supplied as External ID (the image is false for all three publications displayed at DNB). So it's quite obvious that amazon (as DNB) displays one image per ISBN / EAN: our amazon-displayed images WILL CHANGE as soon as amazon replaces the old version with a new one! Stonecreek 10:04, 5 December 2017 (EST)

That's kind of normal - Amazon changes their covers a lot for books that are in print - sometimes it is because of a new printing, sometimes it seems to be just on general principle (LibraryThing (my other playground:) ) has a similar system as ours for linking to local covers or Amazon covers and the reported issues with the Amazon ones are almost a daily occurrence). Annie 11:25, 5 December 2017 (EST)

Book Id Request

i'm looking for an entire book series which i cannot find. the titles of each of the books in the series are years that are crucial in world history, and the main character i believe time travels from a contemporary point in history, to one of these particular years. i found this series randomly on an application like two or so years ago. i didn't even download any of the books, but now i'm really curious to read them because they sounded really fun.

unfortunately too, i can't remember what the years were in history, or what years they covered.

they seem to be recent, and i believe the books were done as ebooks due to maybe lack of being big publications?

i don't necessarily know how they would be classified, they're time travel / historical fiction.

tomes. —The preceding unsigned comment was added by Kallistar (talkcontribs) .

Sorry, but I have no idea what this series may be. A quick search for five various important years came up with nothing. Stonecreek 11:59, 10 December 2017 (EST)
The best known time travel series whose titles contain years is probably Eric Flint's 163x: 1632, 1633, 1634: The Galileo Affair, etc. However, it's about a town in West Virginia getting sent back in time, so the plot is a very different.
On the other hand, Bob Mayer's recent (and apparently self-published) Area 51: Time Patrol is about time travelers going back to certain crucial points in the past, but the titles are different. Ahasuerus 14:03, 10 December 2017 (EST)

2017-12-15 server downtime

The ISFDB server will be unavailable due to patch installation starting at 8:30pm server time. I expect the downtime to take less than 10 minutes.

The patch will add a "Catalog ID" field to publication records. All ISBN/catalog ID values starting with a "#" sign will be automatically moved to the new field. Various and sundry cleanup reports, search pages, Help bubbles, etc will be updated to reflect the new functionality at the same time. I will update Help once the dust settles. Ahasuerus 20:05, 15 December 2017 (EST)

Everything should be back up. If you encounter any issues, please let me know. Ahasuerus 20:38, 15 December 2017 (EST)
Help has been updated and clarified. Ahasuerus 21:16, 15 December 2017 (EST)
Saw a warning that the ISBN/Catalogue was missing when it wasn't. See this pub. Doug H 21:41, 11 January 2018 (EST)
The link takes me to Lin Carter's "Foreword (The Land That Time Forgot)". Could you please list the editing steps that resulted in the warning message? Are you able to recreate them? Ahasuerus 22:51, 11 January 2018 (EST)

What to enter in the Catalog ID field

Checked the 'Help" and didn't notice anything specifically applying to the new field. Then again there are so many pages I probably missed the relevant one ...  ? --~ Bill, Bluesman 23:46, 15 December 2017 (EST)

Here is the rewritten template for ISBNs: Template:PublicationFields:ISBN. The template for the "Catalog ID" field (Template:PublicationFields:CatalogID) is brand new and doesn't have much information yet. Once we sort out various issues like the ones that you raise below, we can update it with the results of the discussion. Ahasuerus 12:54, 16 December 2017 (EST)

Since there are now two fields would it be prudent to split the ISBN/Catalog# column in two? I ask because there are possibly a multitude of cases where there can be both on a given edition [think about Berkley printings from the early 70s where there was almost universally a letter-4 digit number on the cover but an SBN/ISBN on the copyright page; Bantam is another prime example as well as DAW]. I'm thinking we probably assumed it would be one or the other, though I see the software is set up for the 'pairing' potential and puts ISBN/Catalog# just like that with the '/' between. --~ Bill, Bluesman 23:46, 15 December 2017 (EST)

That's right. If a publication has both an ISBN and a Catalog ID, they are both displayed in the "ISBN/Catalog ID" field with a slash separating them. Granted, we could create a separate column for Catalog IDs, but it would be empty most of the time while still taking up valuable "real estate" on the Web page. From the software development standpoint, it would be a minor change, so we can do it either way depending on what the consensus is. Ahasuerus 13:00, 16 December 2017 (EST)

In line with that probable assumption, I can see a situation/use of the new field that I hope we can forestall/eliminate before it happens. Incorporated in every ISBN is a catalog # [just remove the language/publisher prefixes and the checksum] but I think we should plainly state in the Help that that is NOT what the field is for, otherwise it will be an endless partial repetition for no real purpose. Also, there are publishers that had numbers on the covers that had NO relation to the ISBNs at all [hundreds of Avon printings for example]. We really have no way of proving that they are indeed catalog #s but the new field seems a perfect place for them [some Avon printings actually had two ISBNS on the spine, one 'real' and one incorporating the number on the cover ... so go figure ...]. A perfect example is the edition I just added this evening, [here], with the publication page now looking like [this]. Tried as an experiment to see how the display works. Just throwing these notions out there, no real preferences yet [other than completely stopping any over-eager editor/mod from dissecting every ISBN just to put part of it into the new field]. I do think that any letter-number combinations, even where there are ISBNs, are a perfect fit for the new field. Might eliminate/simplify quite a few notes as well. --~ Bill, Bluesman 23:46, 15 December 2017 (EST)

Bill's Avon example is exactly what I had in mind when I was implementing this enhancement.
Similarly, here is an SFBC pub which I tweaked a few minutes ago by copying the SFBC ID from the Note field to the Catalog ID field. We may want to create a new cleanup report to find all SFBC publications with an ISBN and no catalog ID.
As far as the issue of ISBN/Catalog ID duplication goes, there may be 2 different scenarios here. The first one is more common: a book has an ISBN and no stated catalog ID. I agree that we shouldn't be trying to extract the publisher's ID from the ISBN.
However, let's consider the second scenario. Back in the early-mid 1970s paperback publishers took their time transitioning from raw catalog IDs to ISBNs. Some used "SBNs", i.e. ISBN without a leading 0. Some printed both their catalog ID and the ISBN: sometimes on the cover, sometimes on the spine, sometimes on the copyright page. Some ISFDB editors capture this information in notes, but sometimes what we have is inconclusive. For example, consider this van Vogt pub. Our publication record includes the book's ISBN (0-441-25980-4), but if you examine the cover, only the publisher's catalog ID is visible. Perhaps the full ISBN is stated on the spine or on the copyright page, but it's also possible that the ISBN was derived from the catalog ID. This is uncommon -- see this list of Ace books published in 1973 -- but we'll need to decide how to handle this type of cases. Ahasuerus 19:52, 16 December 2017 (EST)

New cleanup repor: invalid BNF IDs

As previously discussed, BNF changed its URL format a few months ago. 427 ISFDB records were affected by this change. Most of them have been fixed manually, but 36 remain outstanding. I have coded and deployed a new cleanup report, Publications with invalid BNF identifiers, to help find them. The data will become available around 1:20am ISFDB time. Ahasuerus 17:26, 17 December 2017 (EST)

Fixed them. --Zapp 14:33, 18 December 2017 (EST)

New cleanup report: SFBC publications with an ISBN and no Catalog ID

As per this discussion, a new cleanup report, "SFBC publications with an ISBN and no Catalog ID", has been coded and deployed. The data, approximately 1,070 records, will become available tomorrow morning. Moderators will be able to "ignore" publications which legitimately lack a catalog ID. Ahasuerus 18:25, 17 December 2017 (EST)

Robert Louis Stevenson--delete Non-genre?

Someone went to a lot of trouble adding Robert Louis Stevenson's poems and nonfiction to the database, and I rather wish they hadn't. Despite the great fame of Stevenson's speculative fiction, there sure isn't much of it--two novels, nine short stories, and a collection of fables over the course of a 40-year career (eleven other novels and much else). You can see the whole of his works here. Does it really make sense to have all this in the database? --Vasha 04:26, 18 December 2017 (EST)

There is a fair amount of history here. Robert Louis Stevenson was an early victim of our first robot. The robot went out, interrogated a number of online catalogs and came back with hundreds of ISBNs, which were merrily added to our database. It happened back in the early 2000s when we were yet to grasp the implications of entering robotically generated data into the database.
Once the current moderation system was implemented in 2006, we went back and deleted RLS's book-length non-genre titles. Some of his non-genre short fiction is still in the database, in part because many collections of RLS's stories contain both his SF and his non-SF. Also, some stories are not really speculative, but contain macabre elements, e.g. the Prince Florizel series.
We may want to start by setting the non-genre flag as appropriate. Once that has been done, we will be in a better position to judge what (if anything) we want to delete. Ahasuerus 12:08, 18 December 2017 (EST)
That is done. Here's a summary of everything I think is genre-relevant:
  • Novels: Strange Case of Dr. Jekyll and Mr. Hyde and The Master of Ballantrae
  • New Arabian Nights/Prince Florizel: "The Suicide Club," "The Rajah's Diamond," and The Dynamiter
  • Short fiction: "Will o' the Mill," "Thrawn Janet," "The Body Snatcher," "Markheim," "Olalla," "The Bottle Imp," "The Isle of Voices," "The Tale of Tod Lapraik," "The Waif-Woman," "The Plague-Cellar," and fourteen fables
  • Poems: the ballads "Ticonderoga" and "Heather Ale"
  • Essays: "A Gossip on Romance" (defense of fanciful fiction), "A Chapter on Dreams" (discussion of the psychological themes in Dr. Jekyll and Mr. Hyde), and correspondence about Dr.Jekyll
--Vasha 13:15, 19 December 2017 (EST)
Nicely done, thanks! My thinking is that the next step would be to examine each collection/omnibus to determine whether it's primarily non-genre. For example, The Strange Case of Dr. Jekyll and Mr. Hyde and Related Poems contains The Strange Case of Dr. Jekyll and Mr. Hyde, an SF novel, and 21 non-genre poems. The title of the publication suggests that the poems are related to The Strange Case of Dr. Jekyll and Mr. Hyde, but I am not sure what the nature of the relationship is. We should probably check with User:Rtrace, who OCLC-verified this pub in 2010, to see what he knows about the pub. If he agrees that the poems are not directly related to the novel, we could remove them from the book and delete any orphans. Rinse and repeat :) Ahasuerus 14:34, 19 December 2017 (EST)
I have no special knowledge about that collection. I suspect that I added the contents because they were detailed in the Worldcat listing. Absent evidence to the contrary, I would give credence to the title as to the inclusion of the poems. As an inclusionist, I'd rather we err on the side of inclusion even if that lets a few non-genre works in as opposed to mistakenly delete a work that should be here. --Ron ~ RtraceTalk 21:10, 19 December 2017 (EST)
You can read all of those poems online and convince yourself that they have no apparent speculative content and no obvious connection to the novel. Unless someone can come up with a copy of the limited-edition book, what the editor had in mind will remain unclear. My feeling would be, since there is doubt, Mark them non-genre but don't delete them. --Vasha 23:08, 19 December 2017 (EST)

(Unindent; quoted from another discussion) I notice that the various editions of A Child's Garden of Verses have the following note "Contents entered are only those that may liberally be interpreted as speculative fiction (12 poems, all 7 publication records as of August 2017)." No indication of who said that.... I would have thought these poems weren't relevant to the database because we don't include writings for young children. And I don't see an indication that these poems have been reprinted in a clearly speculative context. --Vasha 17:14, 19 December 2017 (EST)

Where did you get the idea that works for young children aren't eligible for inclusion? We do have a specific exclusion for "Animal books for very young children", but that wouldn't apply here. Neither is there a requirement that eligible works be reprinted in a speculative context. I have no memory of writing that note and it doesn't sound like I would phrase it. Regardless, I'd vote for letting the poems stay. Some of the titles certainly sound speculative, e.g. "Fairy Bread", "The Land of Nod" and "The Unseen Playmate". --Ron ~ RtraceTalk 21:20, 19 December 2017 (EST)
It might be because we don't include picture books. Other than that, I'm not aware of any works for young children we don't include. ···日本穣 · 投稿 · Talk to Nihonjoe 13:24, 20 December 2017 (EST)
Okay then, those twelve poems are speculative. --Vasha 14:52, 20 December 2017 (EST)

Incidentally, Dr. Jekyll and Mr. Hyde is 25,000 words long. I'm not seriously suggesting breaking the long tradition of calling it a novel... Just noting the fact.

Anyway, the Stevenson page has been tidied up a lot and I feel like it'll do for now. --Vasha 07:31, 21 December 2017 (EST)

Goodreads Awards

Goodreads is a Web site which has been increasingly prominent in the social cataloging space. They have been administering "Goodreads Choice Awards", which are voted on by their users, since 2009. In 2017 3,887,950 votes were cast, including over half a million votes for SF titles. Here are the relevant categories as of 2017:

The last two categories also cover non-SF, but SF accounts for roughly 50% of their nominees/winners. The category structure has changed over the years and the number of nominees per category has fluctuated between 10 and 20, which is not uncommon.

Given Goodreads prominence -- global Alexa rank 324, US rank 170 -- I think we may want to add this award to our menagerie. Opinions? Ahasuerus 17:05, 20 December 2017 (EST)

Works for me. ···日本穣 · 投稿 · Talk to Nihonjoe 17:28, 20 December 2017 (EST)
Agreed. -- JLaTondre (talk) 17:29, 20 December 2017 (EST)
Without objection, so ordered. Goodreads Choice Awards is now live. I have added the 2017 winners and at least one award/nomination per superseded category so that they wouldn't appear on our cleanup reports. Ahasuerus 15:50, 22 December 2017 (EST)
Goodreads awards have all been entered. -- JLaTondre (talk) 18:49, 26 December 2017 (EST)
Great, thanks! Ahasuerus 18:57, 26 December 2017 (EST)
Updated the wiki page, too. ···日本穣 · 投稿 · Talk to Nihonjoe 20:11, 27 December 2017 (EST)

John Martin

While working on Fixer's latest catch, I noticed that our John Martin titles may be the work of multiple people. I suspect (but don't know for sure) that the 1970s editor, the 1950s/1970s fiction author and 1970s/2010s illustrator are all different people. Would anyone happen to be familiar with the many faces of John Martin? Ahasuerus 15:06, 22 December 2017 (EST)

https://csul.iii.com/record=b41651439~S0 identifies the author of Sci-Fi Junior High as Martin, John, 1963-. I've e-mailed him to for thoughts on this thread. --Marc Kupper 02:59, 29 December 2017 (EST)

Question on Collection vs. Anthology; author vs. editor

The 1721 publication of "A Collection of Novels and Tales of the Fairies" by Comtesse d'Aulnoy is a historically important work, and not just because it's the first appearance of the term "Fairy Tale" in English, and the first appearance of Cinderella in English. So I'd like to get this right. Essentially all of the references, including almost all OCLC records and L.W. Currey list the author of this book as Countess d'Aulnoy (or some variation of that name). However, the 1721 edition includes 4 other stories by Henriette-Julie de Castelnau, Comtesse de Murat, which are uncredited to her, and are listed as if they were by Countess d'Aulnoy. I've credited those stories directly to de Murat, instead of "uncredited" and then aliased to her, just out of simplicity (with a note to that effect). I've listed this as a collection, partly because everybody else does, but also because two of the stories by Comtesse de Murat were incorporated into framing stories written by Comtesse d'Aulnoy, so it seems that it fits under the "collaboration" exception for Collections. (It seems very likely that de Murat was a regular member of d'Aulnoy salon, in which they focused on such fairy tales.) However, the 1728 edition and the 1737 edition (along with other later editions) include a 3rd volume that is also *implied* to be by Comtesse d'Aulnoy, is uncredited, and of which all but one of the stories are by Louise de Bossigny, Comtesse d'Auneuil (one other story remains anonymous). And none of *these* stories are in collaboration with the Comtesse d'Aulnoy, at least in the sense of the framing stories being used. So it seems that, regardless of Currey and most WorldCat records, we should credit this to all 3 authors and call it an Anthology. (Princeton's OCLC record does this.) Of course that means that I can't variant the 2nd and later editions to the first edition (which everyone else does). And if I list Comtesse d'Aulnoy as the "author" of these later printings (as everybody else does), then that implies that she was actually the editor. And it is almost certainly the case that she had *nothing* to do with the selection of stories for an English edition based on her works, so calling her an editor would seem to be wrong. So I'm stuck! Any recommendations? (Aside: This seems to me like it's more of a Help Desk question, but our Help Screen on this issue is very explicit that I should ask it here.) Chavey 14:59, 26 December 2017 (EST)

I ran into a problem like this with Charles Lindley/Lord Halifax and his collections of ghost stories. I had a great deal of discussion with the moderators at the time -- you may be able to search and find some of them. In Lord Halifax's Ghost Book, nothing had been published previously, and despite some semi-anecdotal credits to individual writers/contributors (see the pub notes), I credited everything directly to Charles Lindley and made the book a COLLECTION. But in Further Stories from Lord Halifax's Ghost Book, some had been published previously (by different authors), so I ended up keeping those author credits and making this book an ANTHOLOGY, and I made Charles Lindley the editor, even though it's apparent his son had some hand in the assembly of this book. See the pub notes there as well.
Your case is a little different. Technically, since those later editions have different content, you should not have them as variants of the first edition anyway. But in the absence of a mechanism to record derivative relationships, I'd be tempted to treat her as the (sole) editor and do the variants. --MartyD 07:39, 31 December 2017 (EST)
I think it would be fair to abandon doing what everyone else does and instead do what makes the real situation clearest. It does sound as if the later editions should be an anthology with uncredited editor. As for the 1721 edition -- is it a translation of a French edition with a framing story by Aulnoy? --Vasha 13:04, 31 December 2017 (EST)
Yes, the 1721 edition is a translation of a French edition, presumably the 1698 "Les contes des fées", which we do not yet have included. The framing stories are by d'Aulnoy, and were in the original edition. The earlier French edition had ONLY stories by d'Aulnoy, but the (unknown) English editor replaced 4 of those stories with stories by Murat. I am unclear as to whether the framing story had to be altered by the translator to support those replacements. Chavey 03:24, 1 January 2018 (EST)
Here is a description of d'Aulnoy's 1697 collection, and here is a scan of a 1708 edition which seems,to be the same. There is no framing narrative. ----Vasha 04:41, 1 January 2018 (EST)

Cover image wiki page templates and help update

After this question in the help I got curious what was wrong and just learned that we have multiple templates for the cover image, depending on the number of cover artists, in the wiki. As always, you always learn :) However, I couldn't find anything in the help about this and therefore created Help:How_to_update_cover_artist_data and integrated it into the Help:How_to. This sounds like we probably need a cleanup report "Number of cover artists <–> cover image wiki template mismatches" or so if we want to capture cases where the wrong wiki template is used. Jens Hitspacebar 09:50, 27 December 2017 (EST)

Prometheus Award -- preliminary nominees

The Prometheus Award for Best Novel list has been recently (?) updated with information about each year's preliminary nominees. Looking for volunteers who would be interested in entering this information into the database. Ahasuerus 23:32, 29 December 2017 (EST)

Done. Most years already had them. Added the few that didn't have them. -- JLaTondre (talk) 10:51, 30 December 2017 (EST)
Thanks! Ahasuerus 13:09, 30 December 2017 (EST)

Happy New Year, all!

And all best wishes for 2018... --Vasha 21:36, 31 December 2017 (EST)

Happy New Year! Ahasuerus 22:43, 31 December 2017 (EST)
Bonne année ! Linguist 04:28, 1 January 2018 (EST).
Thanks! And a happy new year also to you all! Stonecreek 10:10, 1 January 2018 (EST)
Happy New Year, everyone! Annie 16:09, 1 January 2018 (EST)
A few days late, but 明けましておめでとうございます。 ···日本穣 · 投稿 · Talk to Nihonjoe 20:21, 4 January 2018 (EST)

"The Ghost Club" pseudonyms -- how to handle?

A confusing book, would like to consult. William Meikle published The Ghost Club: Newly Found Tales of Victorian Terror, in which he writes "new" stories by famous authors. Looking inside the book, each story has at the top of it Title over Fake-Author (e.g. To the Moon and Beyond [over] Jules Verne). I thought that this would be entered into the database with Jules Verne as a "pseudonym" of William Meikle ("To the Moon and Beyond" by William Meikle [as by Jules Verne]). But Christian changed my submission to instead have the content records like so: "Jules Verne: To the Moon and Beyond" by William Meikle. This form of the title doesn't appear anywhere in the book. What do all you folks think? --Vasha 16:46, 1 January 2018 (EST)

While we credit per the pub, based on what you describe, these aren't real credits. Instead they are part of the framing of the story. I would treat them as subtitles (so 'Title: Fake Author Name') and add a note explaining the format to the pub notes. That seems to match the pub & our standards the best. -- JLaTondre (talk) 17:18, 1 January 2018 (EST)
(after resolving an edit conflict) If Jules Verne had nothing to do with writing the story, then making the author a pseudonym of him does not make much sense to me. Otherwise anyone that ever wrote a pastiche will need to be pseudonymed as well.
So if I understand what the book contains, I am with Christian on it - the ":" is similar to how we record subtitles - the name we use then is nowhere in the book either in a lot of cases. Add a note in the collection can explain how the names are actually written. Annie 17:21, 1 January 2018 (EST)
OK, so noted. --Vasha 17:33, 1 January 2018 (EST)
Yeah, the other possibility would have been to establish one-time pseudonyms like (for example) Jules Verne (I), Jules Verne (fraud), Jules Verne (William Meikle), or others, and that would have been not really been nicer, or would it? Christian Stonecreek 10:36, 3 January 2018 (EST)

Fabian

I just opened a can of worms after discovering that http://www.stephenfabian.com/ is signed with Stephen E. Fabian Sr.--Dirk P Broer 08:30, 3 January 2018 (EST)

Wikipedia also has the 1930 illustrator as being Sr.
We did have Jr. as the main Fabian and Sr. as an occasional contributor, but it is the other way around. In collaborations the main Fabian names himself Sr. and his son then takes the main name his father uses...--Dirk P Broer 06:27, 4 January 2018 (EST)
Oh, great! It didn't occur to me that there are two artists (father and son) involved. Christian Stonecreek 06:42, 4 January 2018 (EST)
But how do we correct this? Stephen Fabian as presented is Stephen E. Fabian Jr., and Stephen E. Fabian Sr. is 'our' Stephen Fabian.--Dirk P Broer 19:30, 4 January 2018 (EST)
I don't know if it would work, but perhaps we could simply make Stephen Fabian a pseudonym for each of the Sr. and Jr., having those be canonical. Whenever we get an instance of "Stephen Fabian", we'd have to determine which it is and make the appropriate variant. Similar to how we handle house names. I think that treatment would allow everything to be separated appropriately and appear in a meaningful way everywhere. --MartyD 10:29, 7 January 2018 (EST)

Fireside Magazine

Back last January, I stopped cataloguing Fireside because they said they were changing from a monthly magazine to a more blog-like format. But it looks like they haven't changed that much: they still publish one featured story and several shorter pieces of fiction every month, with cover date of "February 2017" or whatever; the difference is just that the stories appear one at a time interspersed with other material. I am thinking I should catalog it after all. But fiction content only, the same as with Diabolical Plots, which is also a hybrid of a blog and a monthly fiction magazine. Agreed? --Vasha 13:03, 4 January 2018 (EST)

I would say yes, continue as before until they actually do change format. ···日本穣 · 投稿 · Talk to Nihonjoe 20:20, 4 January 2018 (EST)

isfdb in the news

Rocket Stack Rank’s 2018 Campbell Award-Eligible Writers page is up now. It lists 166 short fiction writers that are eligible for the 2018 Campbell Award. "Eligibility was determined by looking up the author on ISFDB.org and doing web searches for those without an ISFDB entry." Chavey 09:57, 5 January 2018 (EST)

How could anyone possibly be considered if they didn't have an entry? Doug H 10:00, 5 January 2018 (EST)
We are not as complete, nor as current, as we wish we were. And, as an example on the "completeness" side, I believe we are more stringent on which electronic magazines we accept than the Campbell Awards are. Chavey 12:40, 5 January 2018 (EST)
Their eligibility rules are rather complex. Given that they have to examine hundreds upon hundreds of stories, it seems likely that there will be some "out of scope" works which make their authors eligible for the Campbell award.
That said, their rules mention certain "within the scope" categories which we are likely to overlook, e.g. "a short story that appeared in our large-circulation daily newspaper". Ahasuerus 14:23, 5 January 2018 (EST)
I think (hope) the contents of all genre pro magazines in 2017 have been added now. 2016 wasn't as complete. As for anthologies, I have a list of about twenty in 2017 I haven't yet added, plus who-knows-how-many I don't know about, plus lots and lots that are in the DB don't have contents...
This is a good moment to point to the Anthology and Collection Tracker and the Magazine Issue Tracker. Light-brown and white squares indicate missing information. Any of you have copies of those pubs sitting around? (2018 trackers to be posted soon!)
I am very, very pleased that our efforts are useful in this way. --Vasha 14:18, 5 January 2018 (EST)
I have been thinking about these trackers lately. They are very useful, but take time and effort to maintain. I wonder if we could generate them directly from the database the way we generate various directories, cleanup reports, etc. It would be easy to create a magazine/anthology/collection issue grid for a given year, but certain types of notes like "Has contents but they need work" would require software changes.
Perhaps we could create new Notes templates, e.g. {{CompleteContents}} and {{IncompleteContents}}, which would be used by the software to color-code cells? Ahasuerus 14:31, 5 January 2018 (EST)
I must mildly disagree with you. The purpose of the magazine tracker (especially) is to keep an ahead-of-time schedule of issues that need to be added. An automatically-generated list of issues that are in the DB wouldn't add much of anything since the table already contains links to the issue grids, and the work of marking contents complete/incomplete takes no time at all compared to the work of predicting schedules and adding issues to the DB.
As for the Anthology and Collection Tracker.... I keep a private list of all the 2017 A/Cs, including ones (lots) that aren't in the tracker because they don't contain new stories, and periodically check for additions to the list. Having that automated would indeed be helpful. But there'd have to be a way of marking an item as "not of interest" that would permanently remove it from the list. And each new item on the list would have to be examined by a human before being added to the tracker (this is something I'd really like help with--I'm currently well behind).
That's only part of the purpose of the A/C tracker, though. It also records state of completeness of the contents, and furthermore contains A/Cs that have been announced as forthcoming, to be added to the DB when/if they appear. So, complete automation impossible. --Vasha 15:05, 5 January 2018 (EST)
The idea of having complete/incomplete contents templates isn't a bad one, but there's no way that they'd be used consistently by all the people who add to the DB. Once again, someone would find themself going through everything in the grid checking for accuracy and trying to figure out what isn't in the grid. --Vasha 15:13, 5 January 2018 (EST)
I see. Well, if the magazine tracker is mostly used to keep track of forthcoming issues that are yet to be added to the database, then there is not much we can do about it on the software side. Even the most advanced software can't report on what's not there :-)
Re: the issue of automating the A/C tracker, I am thinking of a new report accessible via the Cleanup Reports menu. Unlike regular cleanup reports, it would allow its users to specify selection criteria similar to the way Advanced Search works. Here are the selection criteria that I am currently considering:
  • Year. Accepts any YYYY year.
  • Publication type. A drop-down list with the following values:
    • Collections
    • Anthologies
    • Both (Magazines can be added later)
  • Fiction contents (i.e. not COVERART/INTERIORART/ESSAYs.) A drop-down list with the following values:
    • Any
    • Present
    • None
  • Template in Notes:
    • CompleteContents
    • IncompleteContents
    • Any
    • None
This will allow editors to search for different permutations, e.g. collections/anthologies which have no contents and no relevant templates in Notes. It should make it easy to find recently added collections/anthologies which haven't had a template assigned yet. You could also select a combination of selection criteria which will replicate the functionality of the currently existing tracker etc. Ahasuerus 18:28, 5 January 2018 (EST)

(Unindent) Yes, if your proposed templates+search allowed a/cs to be assigned to the categories "complete contents" "incomplete contents," "needs contents," and "no new contents," and if it allowed juvenile works to be excluded, it would replicate most of the function of the tracker. The tracker could then be for recording forthcoming items. I would redesign it somewhat if that was its primary purpose. Must think more about this. --Vasha 18:56, 5 January 2018 (EST)

Sounds good. I have created FR 1120 to document the requirements. Ahasuerus 15:24, 8 January 2018 (EST)

Heads-up about the "Add" buttons

I am working on the following RoadMap item:

  • Move “Add Authors/Transliterated Titles/etc” buttons to the right to free up screen real estate. This will become more important with the addition of new multiply occurring fields

It's a surprisingly complex task due to the way the software is currently written. There will be multiple successive patches to get us where we want to be. I am about to deploy the first patch, which should result in no user-experienced changes. If you notice any of the "Add ..." buttons behaving in unexpected ways, please post your findings here. Ahasuerus 19:53, 5 January 2018 (EST)

The second round of changes has been deployed. The changes affected all of the "Add ..." buttons except the "Add [person]" buttons. For example, "Add Transliterated Title" was affected but "Add Author" wasn't. Once again, the changes were strictly "behind the scenes" and there should be no user-experienced changes. If any "Add" buttons do not behave as expected, please do a full Web page reload (Control-F5 in most browsers.) If that doesn't help, please post your findings here. Ahasuerus 18:21, 7 January 2018 (EST)
Another day, another patch. More behind the scenes fixes plus one new bug: invalid Web page URLs no longer generate pop-up errors, although they still get caught by the submission validation logic. The bug should be fixed in the next 24 hours. Ahasuerus 18:54, 9 January 2018 (EST)
The bug has been fixed. You may need to do a full page reload (Control-F5 in most browsers) in order for it to take effect. Ahasuerus 14:14, 10 January 2018 (EST)
Yet another patch has been installed and once again you may need to do a full page reload using Control-F5. I apologize for the hassle, but I am in the process of undoing a certain decision which was made in 2004 and which has become deeply embedded in the software over the last 14 years. Ahasuerus 21:11, 13 January 2018 (EST)

Publication merge proposal

I'm thinking that a publication merge feature would be useful. For example, for Dolphin Island by Arthur C. Clarke we have

As a feature what I'm thinking is that a merge would only be allowed if all of the fields are identical including the link to the cover image. An editor would need to consolidate the notes for example and to have the same notes on both records.

The intent of the merge would be to move primary/transient verifications from the record being deleted to the one being retained. Keep the older of the two for secondary verifications that conflict. --Marc Kupper 17:22, 10 January 2018 (EST)

I can see the value in the ability to move primary verifications of an about-to-be-deleted publication to a different pub if some of the verifiers are no longer active. However, a general purpose "Publication Merge" option can be dangerous. Perhaps we could implement the desired functionality in a different, less risky way. The first thing that comes to mind is a moderator-only option to move primary verifications to a different pub, although that too could be risky. Ahasuerus 20:21, 10 January 2018 (EST)
Agreed - it should be moderator-only or maybe even something only you can do so that we get many eyes on the records before tossing a black hole at verified publication records. The main intent is to move primary verifications in those cases where someone missed that publication record was already available and created a new one. I could see some value to being able to move secondary verifications too. --Marc Kupper 01:02, 12 January 2018 (EST)

1973 Time magazine

I know it's a long shot, but does anyone here collect old Time magazines? Specifically, the April 9, 1973 issue? You can see the cover here. ···日本穣 · 投稿 · Talk to Nihonjoe 02:23, 11 January 2018 (EST)

You can access that issue as a subscriber or pay $2.99/mo to read it. I am a subscriber, so I can look up and clip articles for you. Not too many please.--Rkihara 13:28, 11 January 2018 (EST)
Just one article: "Message from a Star...". Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 17:04, 11 January 2018 (EST)
Sent.--Rkihara 18:56, 11 January 2018 (EST)
Thanks for your help. ···日本穣 · 投稿 · Talk to Nihonjoe 01:45, 12 January 2018 (EST)

Geffen Award

In the interest of internationalization (or should that be internationalisation?), I think we should add the Geffen Award to the list. The official site is mostly in Hebrew, but there is an English version listing all the winners. The Wikipedia page also has a list of winners. ···日本穣 · 投稿 · Talk to Nihonjoe 15:42, 26 January 2018 (EST)

An award presented by a national convention is certainly eligible. Do we have editors who would be able to enter Hebrew publications? Ahasuerus 17:53, 26 January 2018 (EST)
I can copy and paste if needed, but that's about it. ···日本穣 · 投稿 · Talk to Nihonjoe 14:59, 29 January 2018 (EST)
Well, it would be easy enough to create a new award type. However, if we don't have anyone who could enter/massage the data, it would either become an "orphan" or our data would be inaccurate. Perhaps we could request outside help? Ahasuerus 09:28, 31 January 2018 (EST)
I just contacted someone in Israel who says yes, they can find someone to do it --Vasha 11:28, 31 January 2018 (EST)
Excellent, thanks! Once that person creates an account and posts here, I will create a new award type. Ahasuerus 12:09, 31 January 2018 (EST)

Can anyone here read Chinese?

There's a magazine published in China called Future Affairs Administration which is a SFWA-qualified market. Their website is mostly in Chinese (although they plan to publish in English as well as Chinese) and I can't tell if they've produced an issue yet. Can someone check this out? --Vasha 21:19, 26 January 2018 (EST)

IIRC, User:Uzume knows a fair amount of Chinese, but he is currently unavailable. He expects to be back in a few months; perhaps leave a message on his Talk page? Ahasuerus 16:07, 29 January 2018 (EST)
I looked at it and couldn't figure it out. My Chinese is very limited. ···日本穣 · 投稿 · Talk to Nihonjoe 19:12, 30 January 2018 (EST)

My Pending Edits page updated

The "My Pending Edits" page has been update to display the number of pending submissions by all users. Ahasuerus 17:39, 30 January 2018 (EST)

Does it show it if a person has no pending edits? ···日本穣 · 投稿 · Talk to Nihonjoe 18:32, 30 January 2018 (EST)
At this time it just says "No submissions present". However, I think it would be better to display the count even if there are no pending edits. Let me see... Ahasuerus 18:42, 30 January 2018 (EST)
Done. Thanks for pointing out the issue! Ahasuerus 19:04, 30 January 2018 (EST)
Looks good. This is a very helpful idea. :) ···日本穣 · 投稿 · Talk to Nihonjoe 19:11, 30 January 2018 (EST)
Yes, looks good. Thanks a lot. Jens Hitspacebar 04:09, 31 January 2018 (EST)

Duplicate Catalog ID warnings

The software has been updated as per FR 1124, "Display a warning when a catalog ID is about to be duplicated". The new warnings are supposed to work the way duplicate ISBN warnings currently work. Ahasuerus 19:48, 30 January 2018 (EST)

Thank you! -- JLaTondre (talk) 21:27, 30 January 2018 (EST)

Potential award to add: Premio Ignotus

The Premio Ignotus is the most prestigious award for published science fiction, fantasy, and horror in Spain (there are also two prestigious prizes for unpublished work, the Premio UPC and the Premio Alberto Magno). It has been awarded since 1991 by the AEFCFT (the Spanish Fantasy, Science Fiction, and Horror Association), and is voted by the members of the Association and members of the HispaCon. Works published in Spain in the previous year are eligible in the following categories (in some but not all of the categories, the work must be originally written in one of the languages of Spain and first published in Spain): best novel, best novella (17,500-40,000 words), best story, best collected work (anth/collection), best nonfiction book, best article, best illustration, best audiovisual production, best graphic novel, best work of poetry, best magazine, best foreign novel, best foreign story (includes novellas), best web site. The page: AEFCFT --Vasha 23:57, 30 January 2018 (EST)

Seems to be a natural. Would you like to go ahead with it? Stonecreek 02:18, 31 January 2018 (EST)
Sure, I volunteer to do the data entry when the award is created. 26 years x all those categories + lots of works to add = a long time. Don't expect it to be finished soon. --Vasha 11:36, 31 January 2018 (EST)
Great! I have created the award type and 10 categories. We can add/adjust categories and verbiage as needed. Ahasuerus 13:12, 31 January 2018 (EST)
How come you left out the poetry category? --Vasha 13:17, 31 January 2018 (EST)
Oops! I thought I created it, but apparently not. Sorry about that, it has been added now. Ahasuerus 13:23, 31 January 2018 (EST)
Some notes:
  • There is an award for lifetime contribution to the field (referred to as "Premio a la labor de una vida" 1991-1993, "Premio especial Gabriel" 1994-present) which is decided by the board of the AEFCFT's publishing arm, Pórtico, rather than by vote.
  • In 1993, no awards were given except for the Premio a la labor de una vida.
  • In 1992, an award was given for "Mejor obra de no ficción;" starting in 1994, the two current categories "mejor libro de ensayo" and "mejor artículo" were established.
  • In 1991 and 1992, fiction was divided into "mejor novela" and "mejor relato." I can't find what the length definitions of these categories were, although I strongly suspect that the "relato" was everything less than 40,000 words. In 1994, the "relato" category was renamed "mejor cuento" and the award went to a novella-length work. As of 1995, this category was subdivided into "novela corta" (17,500-40,000 words) and "cuento" (under 17,500).
  • Dates when the other categories were first established: illustración - 1991; novela extranjera, cuento extranjero, revista, obra poética, producción audiovisual - 1994; antología, sitio web - 2001; tebeo - 2003
--Vasha 13:52, 31 January 2018 (EST)

(unindent) Descriptions of each category:

  • Mejor novela (best novel): Fiction at least 40,000 words long, written in one of the official languages of Spain and first published in Spain.
  • Mejor novela extranjera (best foreign novel): Fiction at least 40,000 words long which was first published (in any language) outside of Spain.
  • Mejor novela corta: Fiction between 17,500 and 40,000 words long, written in one of the official languages of Spain and first published in Spain.
  • Mejor cuento: Fiction less than 17,500 words long, written in one of the official languages of Spain and first published in Spain.
  • Mejor cuento extranjera: Fiction less than 40,000 words long which was first published (in any language) outside of Spain.
  • Mejor libro de ensayo: A nonfiction book, which may have been first published either within or outside Spain.
  • Mejor artículo: A nonfiction article which has not appeared in book form, and which may have been first published either within or outside Spain.
  • Mejor antología (I think a better English name for this would be "Best Collected Work"): A book containing at least three stories, original or translated, by one or several authors.
  • Mejor obra poetica: A work of poetry of any length, written in one of the languages of Spain and first published in Spain.
  • Mejor illustración: A single image, which may have appeared in any type of publication.
  • Mejor revista: A professional, semi-professional, or amateur magazine published in Spain.
  • Mejor sitio web: A website, at least part of whose contents are in one of the official languages of Spain.
  • Mejor producción audiovisual: A work of film, television, theater, or radio produced in Spain.

--Vasha 14:23, 31 January 2018 (EST)

Thanks for looking into this! Back when the current award editing system was implemented, we limited the ability to create new award types to bureaucrats in order to minimize the likelihood of creating duplicate types. We also limited the ability to create/delete/edit award categories to moderators. However, based on what we seen over the last few years, I don't think it makes sense to prevent editors from creating "Edit Award category" submissions. I have changed the software, so next time you land on an award category page, you should be able to edit it. Let's see how well it works. Ahasuerus 16:04, 31 January 2018 (EST)
I have made a change to "Mejor libro de ensayo". It is not a continuation of "Mejor obra de no ficción" (that award went to an article the one time it was given). You should create a separate category for "Mejor obra de no ficción." Also, I think it would be better to make "Mejor relato" a separate category; this meant something different than what "Mejor cuento" means in all but one of the years that "Mejor cuento" has existed. (If that is done, I think the 1994 award should be listed in the category Mejor relato, with a note. This is always going to be somewhat confusing, no matter what we do, but I think what would be clearest would be the category names "Mejor relato/Mejor cuento (1991-1994)" and "Mejor cuento (1995-)" --Vasha 16:34, 31 January 2018 (EST)
I have created new categories for "Premio especial Gabriel - Special Gabriel Award" and "Mejor obra de no ficción - Best Nonfiction".
Re: "Mejor relato/Mejor cuento", let me make sure that I understand the eligibility rules and the terminology correctly. It looks like there were three periods:
  • 1991-1993: Mejor relato -- all short fiction works under 40,000 words were eligible
  • 1994: Mejor cuento -- ditto, i.e. all short fiction works under 40,000 words were eligible
  • 1995-: Mejor cuento -- only works under 17,500 were eligible; works between 17,500 and 40,000 words were eligible for "Mejor novela corta"
Is this correct? If it is, then perhaps we could create three categories: "Mejor relato", "Mejor cuento (1994)" and "Mejor cuento (1995-)". Ahasuerus 12:07, 1 February 2018 (EST)
Sure, makes sense. --Vasha 13:36, 1 February 2018 (EST)
OK, I think I got it, although the wording may need to be tweaked. Ahasuerus 14:59, 1 February 2018 (EST)
There is a footnote to the AEFCFT page about Gabriel Bermúdez Castillo winning the Premio al labor de un vida in 1992: "Domingo Santos fue galardonado con un premio especial de la organización de la convención denominado “Ignotus”, aunque en realidad no se trataba de un auténtico premio Ignotus" (The organizers of the [HispaCon] convention gave Dominigo Santos a special award called "Ignotus," athough in reality it wasn't an actual Ignotus Award) -- How to record this? --Vasha 13:47, 1 February 2018 (EST)
Reading the explanation:
  • Premio a la labor de una vida: Gabriel Bermúdez Castillo (simultáneamente, los organizadores de la HispaCon de Gadir’92 otorgaron un premio con el mismo nombre a Domingo Santos, pero no deben confundirse ambos galardones).
it would appear that the award given to Domingo Santos was not really a Premio Ignotus award. I guess we could either document it in notes or create a separate category just for it with an explanation of what happened. A rather peculiar case. Ahasuerus 14:59, 1 February 2018 (EST)
I suppose add a note to the '92 Premio especial as they do, and record it in the notes of Domingo Santos's author page (which I just submitted a bio to). --Vasha 15:07, 1 February 2018 (EST)
Approved and massaged. Thanks! Ahasuerus 16:35, 1 February 2018 (EST)

(unindent) A couple errata on the award page: you have two categories in display order 30, and the capitalization of HispaCon in the description needs fixing. --Vasha 17:26, 1 February 2018 (EST)

Changed, thanks. Having two categories share the same display order value is harmless as long as their eligibility years do not overlap, but we might as well make it look pretty :) Ahasuerus 18:14, 1 February 2018 (EST)

Fuzzy searching

I have been experimenting with "fuzzy searches" based on what was mentioned during this Rules and Standards discussion last week.

The idea is that the regular search logic should ignore punctuation so that a search on "Steven H. Silver" would also find "Steven H Silver". (Note that for now I am defining "punctuation" as apostrophes, single and double quotes, commas, periods and parentheses.)

I have made the requisite software changes on the development server and it looks like they do not have a significant negative impact on response times. However, I have discovered that using "fuzzy" search logic can lead to somewhat unexpected results. A name search on "h. silver", which currently finds a single author record on the main server, finds 7 records on the development server: "Judith Silverthorne", "Kenneth Silver", etc. It seems like it may be an acceptable trade-off, but I wanted to run it by everyone before deploying the changes. Ahasuerus 18:34, 1 February 2018 (EST)

I don't think those few extra search results will be a problem. What about the other issue, the question of whether authors write their first initials "X. X." or "XX"? How many extra results would be generated by ignoring spaces also? The other way to get the same result would be to change the standard for the punctuated form to "X.X." --Vasha 19:33, 1 February 2018 (EST)
At one point I tried stripping spaces as well. However, it resulted in a lot of false positives. For example, a name search on "ajm" finds 4 names on the live server. When stripping spaces, it finds 37 names. I'll have to think of a way to normalize initials when searching. Ahasuerus 20:04, 1 February 2018 (EST)
If you are going forward with this, can it be made an option? I would find the false positives annoying. -- JLaTondre (talk) 20:38, 1 February 2018 (EST)
I'd also like the ability to decide when to use fuzzy search. Annie 20:57, 1 February 2018 (EST)
There is no hurry, we can take our time figuring out the desired behavior. Perhaps a "Fuzzy" checkbox within the search box? Ahasuerus 21:22, 1 February 2018 (EST)

Ballantine's Classic Library of Science Fiction - call for volunteers

A couple of Usenet posters have pointed out that our Ballantine's Classic Library of Science Fiction listing is incomplete. For example, it's missing The Best of Leigh Brackett and The Best of Lester del Rey. Calling for volunteers to identify and add the missing publications. Ahasuerus 08:46, 2 February 2018 (EST)

Is there a full list of them somewhere? ···日本穣 · 投稿 · Talk to Nihonjoe 14:21, 2 February 2018 (EST)
It looks like the following Advanced Publication Search: Title contains "Best of" AND Publication Type is exactly COLLECTION AND Publisher contains "Del Rey" finds all of them. Ahasuerus 14:36, 2 February 2018 (EST)
Looks much better, thanks! Also, it would appear that some pubs in this series are currently entered as by "Ballantine Books" as opposed to "Del Rey / Ballantine" -- e.g., see http://www.isfdb.org/cgi-bin/pl.cgi?407039. Ahasuerus 10:05, 3 February 2018 (EST)

Nominating user Boskar for moderator

See Moderator Qualifications#Becoming a moderator for the nomination process.

I would like to nominate user Boskar (talk) for moderator, with the understanding that he would limit his approvals in the beginning to his own edits, as brought up as a general proposal in the third discussional paragraph of this item. Boskar has proofed to manage all of the intricacies that are mandatory for his submissions of German and international publications & titles of all kinds of types. He has been very careful in handling them, and has proven good communication skills whenever it was necessary. So I think he meets the Moderator Qualifications and he has accepted the nomination. Stonecreek 06:58, 3 February 2018 (EST)


Support

  1. Support, as nominator. Stonecreek 06:58, 3 February 2018 (EST)
  2. Support. No objections on the self-moderation front. I have not run into any issues with his submissions in over a year. --MartyD 11:51, 3 February 2018 (EST)
  3. Support. No objections as well. The number of submissions I've handled since I've become a moderator myself just a few months ago isn't huge, but the ones I did handle were of good quality. Jens Hitspacebar 11:19, 4 February 2018 (EST)
  4. Support. No objections - I cannot remember even one issue with his submissions in the last months. Although I will miss the German practice I was getting while following his submissions :) Annie 18:52, 5 February 2018 (EST)

Oppose


Comments/ Neutral

  1. Neutral. I have only handled a few submissions by Boskar. If I remember correctly, they were OK, but I don't have enough experience with them to vote one way or the other. I suspect that it may become a more common scenario as the number of editors and moderators grows and contributions become more specialized. Ahasuerus 09:47, 3 February 2018 (EST)
  2. Neutral. Neutral, no experience with his submissions.--Rkihara 12:14, 4 February 2018 (EST)
  3. Neutral. As far as I know, I haven't dealt with any of his submissions. From what I've seen of his posts, he seems solid. ···日本穣 · 投稿 · Talk to Nihonjoe 14:23, 5 February 2018 (EST)
  4. Neutral. I have no experience with his submissions. I see no issues on his talkpage, so positively neutral. --Willem 16:33, 7 February 2018 (EST)
  5. Neutral. I've not had experience of his submissions. PeteYoung 16:37, 7 February 2018 (EST)

Outcome The nomination passes. The moderator flag has been set on Boskar's account and he should be able to approve his own submissions going forward. Ahasuerus 11:50, 9 February 2018 (EST)

Edit form changes - Part 2

I am in the process of working on another round of edit form changes. Like the last round, it shouldn't affect the behavior of the edit forms. If some of the "Add" buttons appear to stop working, please do a complete page reload (usually Control-F5). If that doesn't help, please post your findings here. Ahasuerus 16:27, 3 February 2018 (EST)

Another day -- another patch -- same patch notes as above. Ahasuerus 17:10, 4 February 2018 (EST)
Yet another day, yet another patch. Same drill. Ahasuerus 13:36, 5 February 2018 (EST)
Another patch, which changed the way the "Cover Art" section of New Publication submissions is handled behind the scenes. If you see anything unusual and Control-F5 doesn't fix, please let me know. Ahasuerus 17:47, 5 February 2018 (EST)
The changes to the Cover Art section which were started yesterday have been completed. There should be no editor-experienced changes. Ahasuerus 15:45, 6 February 2018 (EST)
Yet another patch has been installed. It affected the rest of the Content section. As before, there should be no user-experienced changes. If you see anything unusual and Control-F5 doesn't fix, please let me know. (With luck, I hope to be able to wrap up these annoying tweaks in the next few patches.) Ahasuerus 12:04, 7 February 2018 (EST)

(unindent) It looks like the last patch introduced a new bug which makes editing most pubs impossible. Investigating... Ahasuerus 12:41, 7 February 2018 (EST)

The immediate bug has been fixed. Some types of pop-up validation are currently unavailable, but it shouldn't prevent editors from creating submissions. Working on it... Ahasuerus 12:55, 7 February 2018 (EST)
OK, the pop-up validation should be fixed now. Sorry about that! Ahasuerus 13:54, 7 February 2018 (EST)
Yet another patch has been installed. We are getting closer to the finish line. Ahasuerus 17:49, 7 February 2018 (EST)
A big patch was installed a few minutes ago. The same rules -- if you see anything unusual and Control-F5 doesn't fix, please let me know -- apply. There is some additional cleanup that will need to be done over the next couple of days, but the bulk of the changes is now live. Ahasuerus 14:34, 9 February 2018 (EST)

New Mistakes

I just submitted the ebook version of Year's Best Hardcore Horror: Volume 2 to this site. I had to copy the contents off of Amazon by hand, and, of course, I made a few spelling mistakes. If my submission is accepted, I will correct the misspellings of the authors Michael A. Arnzen, Eric LaRocca, Jasper Bark, and Adam Cesare. I am so ashamed. MLB 20:34, 8 February 2018 (EST)

Say three "Hail Fixer's" and you will be absolved of this sin. It's been approved. ···日本穣 · 投稿 · Talk to Nihonjoe 20:40, 8 February 2018 (EST)

Udon Entertainment Japanese art portfolios

I picked up a couple of these some time back. Specifically this and this and these seems to be a plethora of these game related art portfolios. I sounded Hauck out about the validity (or not) of entering these kinds of publications. His reply to my query about computer game art portfolios was (quoted verabtim):

Hello, my understanding is that we're a "fiction" database, "fiction" seemingly taken in the implicit sense of "written". Note also that by our ROA, comics are explicitely out, so are graphic novels. For art books, my answer would be the same, they shouldn't included. Striking a blow to clarity, in the case of SFF artists, huge numbers of them (complete art books) have been entered (like here), that may have had some bibliographical justification at the time (to identify some covers) but then there was another slip and such portfolios were also entered (even if their relation to spec fic was nil). AFAIC, I'm strictly against inclusion of non-mainly-written-spec-fic items, including the portfolios that you evoke. To be a database of comics, "bandes dessinées", portfolios, artworks, even if SF-themed is, IMHO, not our present project. We can't (and don't want to) compete with such sites. But, as there is no pilot in the cockpit and we are in a time where any submission has a chance to be accepted regardless of its pertinence, you can try your luck, I won't reject your submissions. Remember that all this is just a personal rant, so perhaps can you bring the matter to one of our common spaces (even if this will be without me).

Some of these computer game art portfolios are already included in ISFDB - eg this which has a limited edition version (and probably numerous others.

I know there will be the "purists" (not a bad thing and not meant as an insult) who will have the view that none of these publications should be included - even art portfolios for well known SF artists like Chris Foss etc.

Would there be a definitive answer as to what material of this nature would/should be allowed or will it come down to the individual moderator.

I stated to Hauck that "I'll take my chances and if they're rejected then I'll accept that and move on."

Your thought/rants/comments.--Mavmaramis 14:18, 9 February 2018 (EST)

I have entered I think hundreds of SF/F art books, although none that are game-based. One thing I noticed though, is that Spectrum 24 contains significantly more game-related artwork in the "Books" category than earlier volumes of Sectrum, and it seems to outnumber the cover and interior art examples. SF/F artists seem to be making a larger fraction of their income from games than they used to. Obviously, I'm an advocate of art books and portfolios by noted fantasy artists, but I would not include game-based or movie-based art books unless the artist is also noted for illustrating genre books. But of course, I'm not a moderator. Bob 17:11, 9 February 2018 (EST)
I think that sf/f/h art books ought to be included, regardless of whether the person is known for illustrating genre books or magazines. Keep in mind that many of the artists who do game design and art in Japan (speaking to this example in particular) also regularly do sf/f/h magazine and book covers and interior illustrations. A very large percentage of them (far higher than in the States). ···日本穣 · 投稿 · Talk to Nihonjoe 18:25, 9 February 2018 (EST)
Also, there are a number of Hatsune Miku novels out, too, and since she's basically science fiction, I think she's in no matter what for novels and art books. ···日本穣 · 投稿 · Talk to Nihonjoe 18:27, 9 February 2018 (EST)

proposed canonical name change: Standish James O'Grady

Standish James O'Grady, currently with the canonical name Luke Netterville, should probably be changed to O'Grady, as this was both his usual name and the one under which he published some fairly-well-known children's fantasies. Netterville was his pseudonym for a single obscure science fiction novel. He is in SFE3 as O'Grady.

I have already changed "Dr. Douglas Hyde" to the more commonly used form Douglas Hyde. --Vasha 07:32, 11 February 2018 (EST)

Yes, as I processed some of your submissions, I had the same thought about O'Grady vs. Netterville. --MartyD 06:07, 12 February 2018 (EST)
All done. --Vasha 17:58, 12 February 2018 (EST)

Robert Anton Wilson

We have an editor wishing to add Robert Anton Wilson's non-fiction work which is non-genre. What is the community thought on Wilson's threshold status for non-genre works? -- JLaTondre (talk) 18:00, 11 February 2018 (EST)

Hi, I added a few books because I saw there were a number of his other non-fiction works already listed - including a few (Prometheus Rising, Ishtar Rising, the three Cosmic Trigger books) erroneously listed as novels. Bob is a strange case - his truest Science Fiction series was the Schrodinger's Cat trilogy, which dealt with multiple dimensions. His Illuminatus Trilogy and related works are sometimes considered science fiction for want of a better category. His non-fiction work is highly speculative and touches on many issues related to science fiction - space travel and life extension, in particular. --Donnachadelong 18:09, 11 February 2018 (EST)
Note that The Illuminati Papers is already in the database.--Dirk P Broer 10:54, 13 February 2018 (EST)

Since there has been no input, I'll remove the hold on these and leave it to another moderator. I'm not convinced Wilson is above the threshold, but would prefer to not to take unilateral action given the varying thoughts on this manner. -- JLaTondre (talk) 21:09, 14 February 2018 (EST)

Verification for scans, PDFs, etc.

I want to second Marty's proposal here: "I've been thinking that we should have some sort of "electronic copy" verification. It could encompass PDFs/OCR scans, Look Inside, Google Books, and perhaps others. In fact, maybe each ought to be its own separate item (since, for example, Look Inside is partial, while a PDF might be the full publication). Actually confirming facts out of one of these things often will be as good as, or better than, secondhand information from some of our other secondary sources. Yes, it's a little tricky to make sure the electronic copy represents the same publication and edition, but it still seems worthwhile, and the cost of someone goofing up such a verification is low."

Currently when I have entered data from an online scanned copy I mark this as a transient verification. I know some other people mark it as a permanent variation, and still others as not a verification. It would be nice to have a solid way of recording this. It should go along with a way of linking to the scan on Google Books, Archive.org, Hathi Trust, etc. --Vasha 13:51, 13 February 2018 (EST)

I think it's a worthwhile idea. Let's consider how we can implement it within the existing framework of primary and secondary verifications. Let's review the current functionality:
Primary verifications are subdivided into "Permanent" and "Transient". Any number of users can primary-verify a publication, but a user can only choose one primary verification sub-type. If you choose "Permanent", you can't choose "Transient" and vice versa.
Secondary verifications are "reference"-specific: Locus1, OCLC, etc. A publication can be "secondary-verified" against a reference by one and only one user. For example, once a pub has been verified against Locus1 by a user, no one else can do it.
The important distinctions are:
  • Primary verifications allow multiple users to verify the same publication while secondary verifications allow only one user verification per pub per reference.
  • Secondary verifications support multiple references per pub while primary verifications are generic.
Given these limitations, let's consider whether electronic verifications are more like primary verifications or more like secondary verifications. I guess it depends on the specific functionality that e-verifications will need to support. If the priority is to allow multiple users to e-verify the same publication, then we could add an "Electronic" sub-type to Primary verifications. If, on the other hand, the priority is to support multiple references like Google Books, Archive.org, and Hathi Trust, then we could add them to the list of supported references.
As far as linking to Google Books, Archive.org, Hathi Trust, etc goes, the current verification system doesn't support linking to other Web sites. We do have a couple of options, though:
  • Third party sites which allow linking by ID can be easily added as "external identifiers".
  • Third party sites which do not support linking by ID and require an explicit URL can't be supported at this time. However, we can support them once we implement FR 957, "Add a repeatable "Web Pages" field to Publication records".
Ahasuerus 16:03, 15 February 2018 (EST)
If treated as a secondary verification, there still only needs to be single "Electronic" (perhaps "Scan" would be better as eBooks are electronic also) type. Hathi Trust hosts the Google scans from university libraries so many (most?) times they are the same copies as Google Books. I'm not sure, but I'm under the impression that Archive.org copies the Google scans also. So in the end, there is only one source for many of these. While those are the main sites, there are plenty of other sources out there (I've seen authors host them on their website). A single type and stating the source in the pub notes should be sufficient. For sites without stable IDs and links, URLs can continue to be embedded in the notes via HTML before FR 957 is implemented. -- JLaTondre (talk) 19:44, 15 February 2018 (EST)

Harry Farjeon's "Exit"

This Usenet question revealed that we are missing Harry Farjeon's SF story "Exit" and the 2000 collection "That untravelled world: a collection of science fiction short stories". Looking for volunteers to do more digging and enter the missing data. Ahasuerus 14:37, 13 February 2018 (EST)

Edit form changes - Part 3

Yet another round of edit form changes has been deployed. Like the previous rounds, it shouldn't affect the behavior of the edit forms. If some of the "Add" buttons appear to stop working, please do a complete page reload (usually Control-F5). If that doesn't help, please post your findings here. Ahasuerus 16:07, 13 February 2018 (EST)

A new patch has been deployed. It shouldn't require Control-F5. Ahasuerus 18:34, 13 February 2018 (EST)
Another patch has been deployed. Ahasuerus 22:53, 15 February 2018 (EST)
Yet another patch has been deployed. It revamped the way "Edit Title" works behind the scenes and made the following quality of life changes:
  • The submission review page displays all authors, including reviewed and interviewed authors, at the top of the page instead of at the bottom
  • Blank fields for multiply occurring values (authors, transliterated titles, Web pages, etc) only appear when requested, which should save "screen real estate"
Ahasuerus 15:58, 16 February 2018 (EST)
An additional, hopefully completely transparent, patch has been installed. Ahasuerus 19:16, 16 February 2018 (EST)

(unindent) A new patch has been installed. Unlike the last few patches, it may require a Control-F5 reload (depending on your browser settings.) Ahasuerus 14:46, 17 February 2018 (EST)

Canonical name change: Alfred Tennyson

I don't know why Tennyson stil has the (incorrect!) name "Lord Alfred Tennyson" at the top of his page but a look at the page shows that "Alfred Tennyson" by far the commonest name-variant encountered. I will change it tomorrow if no one objects.

Also, it turns out that although "William Butler Yeats" is the canonical name, there's only 12 titles using that pseudonym versus 72 for "W. B. Yeats"! I knew the latter was commoner but didn't realize the difference was that much. Should we change it? --Vasha 20:26, 18 February 2018 (EST)

Tennyson is done. I guess if no one has anything to say about Yeats, I'll change the canonical form to "W. B." in a few days. --Vasha 18:40, 19 February 2018 (EST)
Sounds good to me. ···日本穣 · 投稿 · Talk to Nihonjoe 20:37, 19 February 2018 (EST)

"Black author" tag

(Moved from [here])

Hello, I'm voicing here my concern about this tag. Here (in France) we're quite touchy about questions of privacy and of data gathering. In a nutshell, to enter and/or to store and/or to give public access to data about the racial characteristics of an individual (or his/her sexual orientation or political opinion) is simply illegal and may be severely punished. I know we're not a french-law-abiding outfit but, to be frank, I'm quite ashamed to be associated with such an endeavour. Why not add such tags as "Muslim", "Leftist", "Gay", "Have cancer", "Of short heigth", "Redhead" etc... (I spare you the Godwin point "To be eliminated") It is the same for Afro-american and even the fact of passing the tag to "private" is not enough (for me and the french law, perhaps the european one), it's the simple constitution of a database with names that is illegal without it being declared. I was already quite worried about our forays into gender typification, but this is worse. Hauck 04:20, 23 February 2018 (EST)

The same is valid for 'African author' and likely some others. In addition to the things Hervé wrote, these labels totally ignore that tags are for titles, and not for authors: it may be okay to have 'Black protagonist', but no thing in this vein to stigmatize authors. Stonecreek 05:08, 23 February 2018 (EST)
I hear your concerns. I started that project last year because of the #BlackSpecFic report. People there were tracking where black authors got published and I wanted to help them. (I was thinking of contributing to Wole Talabi's list of African spec fix also but I never did much with that.) So you think it is not appropriate to do that tracking in a public manner? --Vasha 08:18, 23 February 2018 (EST)
I think it's totally contra-productive. It really smears all relevant information on thematical issues for an author like Samuel R. Delany to have 153 times characterized him as 'Black Author' (who would have thought that?). It has next to no thematic relevance, and I think this is the case for most authors characterized so or in a similar way. Who wants his works issued by color of his skin? Stonecreek 08:39, 23 February 2018 (EST)
The idea is not to tag the author but rather to get an idea of where and when works by black authors have been published--would it sound better to you to change that tag to "work by Black author"? Also, I am in touch with a couple of the people involved in #BlackSpecFic and if you like I could ask them whether they think it is useful or appropriate to use the database in this way.
I would also like to ask Darrah because of her experience trying to use the database to study women Writers. And I know she's given a lot of thought to privacy concerns. Unfortunately she's not around now. --Vasha 08:46, 23 February 2018 (EST)
No, this is a meaningless tag for our purposes. I really think it has to be deleted, or, if that's not possible, to mark it as 'Private'. You really seem to do some things without thinking them through: that's called actionism. A title tag has not to be misused as an author tag: and these examples seem to be a kind of putting authors into a drawing-box, and nobody has asked them if they want to be put in there. Stonecreek 09:03, 23 February 2018 (EST)

2016 #BlackSpecFic report. See surrounding discussion for why it is unfortunately necessary to talk about writers' race. It is not at all irrelevant to their chances of being published; it is the publishing industry that puts people into boxes. --Vasha 09:17, 23 February 2018 (EST)

We're not here to record the plight of black writers. Such tags attached to persons are simply unacceptable and may even be libelous. As for Darrah (who's a he), this page is IMHO already borderline. Both of you can make all the lists of black or feminine or transgenre or martian writers you want but you simply can NOT use the ISFDB which is a public space. Hauck 10:35, 23 February 2018 (EST)
This conclusion has been proven false, if not dangerous. Opening trenches between black / white or Jewish / Arian or any other thought-of dividing line between people has been never a tremendously good idea. You need to talk about the issues & sources of racism, and that what's literature for (foremost I think). But this can only be adressed in works, not by hammering differences into heads by showing this is the most important thing for a given author.
Also, you seem to flea the basic fault of the misuse of tags. Stonecreek 10:20, 23 February 2018 (EST)


(unindent) It looks like there are a number of issues here, including:

  1. The legality of certain tags
  2. The ISFDB software design which currently lets users add tags without moderator approval/oversight (potential accuracy issues)
  3. The ISFDB policy which determines which tags are made "Private" by moderators
  4. The feasibility of using title-level tags to enter author-specific data ("female author", "indie author", "Finnish author", etc)

As previously discussed, the laws that we have to abide by are the laws of the jurisdiction where the ISFDB server is currently located. (It's the same for projects like Project Gutenberg whose servers host different works depending on the country where each server is located.) There may be libel laws that we have to abide by, e.g. "written by a thief and a murderer" could be an issue, but at this time our data entry policies are not legally constrained otherwise. --Ahasuerus 10:41, 23 February 2018 (EST)

To reply to point four, is it really author-level data when you say "this particulat work was written by an Irish author"? Because that's how I intended it & maybe the name of the tag should be changed to make it more clear.
To the rest, I really want to hear from people! --Vasha 12:07, 23 February 2018 (EST)
I hope to provide some background info re: tags when I get back to my evil lair of evil secure undisclosed location later today. Ahasuerus 13:09, 23 February 2018 (EST)
Black for an American audience is not the same as black for a Brazilian one for example. Nationality is one thing. Colors and genders are a different thing. And I do not think we should be in the business of tagging people with colors (and genders - there is a reason why the last proposal to add gender to the DB was shot down again) - what percentage of your blood should come from black ancestors so you can be considered black? We are a FICTION cataloging site, let's not go around and put labels to people. Tagging works is ok based on the content of the text; tagging works based on author's characteristics has no place here in my book.
Even if something is legal, it may not be the correct thing for an international community as ours. The only question we should care about is if an author writes in our genres. I understand that some people may want to do research based on race and what's not but where do we stop? Just my 2 cents. Annie 14:46, 23 February 2018 (EST)
I know it's anathema to some people, but I couldn't care less about an author's race, gender, politics, sexual preference, or whatever else. I read fiction because the story interests me. Therefore, I see no valid reason to include tags that catalog those things here. ···日本穣 · 投稿 · Talk to Nihonjoe 15:09, 23 February 2018 (EST)

Using author information in title tags

Let's consider the larger issue of entering author-specific information via title tags. As previously indicated, we have tags like "indie author", "African author", "Finnish author", "Black author" and "female author". It seems clear that the reason they have been created is that we have no support for author tags and no structured way to enter this data in author records, which is why our editors who want to see this information recorded use title tags as a crutch.

I think this raises a number of potential issues:

  1. At least one of these tags ("female author") is effectively used to get around the fact that we have not been able to reach consensus re: adding a "Gender" field to author records.
  2. Some of these tags are inherently complex and ambiguous, e.g. "African author" -- consider the multi-faceted definition used by the Nommo Awards for African Speculative Fiction to determine who is and who isn't an "African author" for their purposes.
  3. If and when the tagged title's author(s) are disambiguated, author-specific tags may become invalid. For example, if we determine that the name "J. Smith" has been used by two people, "Joan Smith" and "John Smith", and move some of the titles to "J. Smith (I)", some author-specific tags may be moved to the wrong author.
  4. If our information about an author changes due to more in-depth research or because the author's circumstances have changed (e.g. an author may have published her first book as an "indie author" 10 years ago, but has been traditionally published since then), we'll need to go back and change a bunch of tags. Given that tags are editor-specific, it may not be easy to do if the tagger is unavailable.

Given these concerns, I think it may be prudent to make it our policy that all author-specific tags should be made "Private" by moderators. Editors will still be able to use them to compile lists of works by certain categories of authors for their own purposes. Ahasuerus 16:23, 23 February 2018 (EST)

Personal tools