ISFDB:Community Portal

From ISFDB

(Difference between revisions)
Jump to: navigation, search
(Displaying title tags on the Advanced Search Results page: ])
(Signing notes: FR 1238)
Line 2,076: Line 2,076:
:::: Wouldn't it be a good thing we'd '''always''' sign & date the notes? I'm working at a pharma company, and there it's mandatory to always sign & date every single change you make to a record ... [[User:MagicUnk|MagicUnk]] 09:49, 11 October 2019 (EDT)
:::: Wouldn't it be a good thing we'd '''always''' sign & date the notes? I'm working at a pharma company, and there it's mandatory to always sign & date every single change you make to a record ... [[User:MagicUnk|MagicUnk]] 09:49, 11 October 2019 (EDT)
::::: Only if we have a system that suppresses that from the everyday screens. You really do not want 20 lines of history for every line of a note. :) [[User:Anniemod|Annie]] 13:00, 11 October 2019 (EDT)
::::: Only if we have a system that suppresses that from the everyday screens. You really do not want 20 lines of history for every line of a note. :) [[User:Anniemod|Annie]] 13:00, 11 October 2019 (EDT)
 +
 +
:::::: Well, there is {{FR|1238}}, "Create an Edit History page for publications", which would create snapshots of verified publication records before they are changed. It would help with a lot of different issues. [[User:Ahasuerus|Ahasuerus]] 14:56, 12 October 2019 (EDT)

Revision as of 18:57, 12 October 2019


ISFDB Discussion Pages and Noticeboards
Before posting to this page, consider whether one of the other discussion pages or noticeboards might suit your needs better.
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
Roadmap: For the original discussion of Roadmap 2017 see this archived section. For the current implementation status, see What's New#Roadmap 2017.



Archive Quick Links
Archives of old 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 · 44 · 45 · 46




Contents

Windy City Pulpcon

If anyone here is attending the Windy City Pulpcon on April 12-14. Let me know if you'd like to meet up.--Rkihara 13:18, 1 April 2019 (EDT)

As usual, I'll be there. It's practically in my back yard. I'd be happy to meet up if you're interested. Bob 17:43, 1 April 2019 (EDT)
Sure, I'd like that. I'll be registered at the Westin as Ron Kihara and will be arriving Thurs. evening. I usually drive down I-80 from the Bay Area, so it's possible the weather might prevent me from making it.--Rkihara 19:20, 1 April 2019 (EDT)
Great, Ron. I'll get there about 11 a.m. Friday when the dealer room opens. I'll hang around the entrance for a while, then wander through the ballroom. I'm Bob or Robert Lumpkin. If we somehow miss on Friday, I'll be there about 10 am on Saturday with my adult son. I'll hang around with dealers like Greg Ketter (Dream Haven Books), Lloyd Currey, Andy Richards (Cold Tonnage Books), Graham Holroyd, and Bill Cavalier. There are usually a bunch of friends from the Robert E. Howard community as well. Bob 20:06, 1 April 2019 (EDT)
I'll be going into the dealer room about the same time. I'll be easy to spot, I'm Japanese, wear glasses, bald, and I'm told I look like a Kung Fu villain. The dealers I usually spend time with are; Steve Hafner, Larry Hallock, Craig Poole, Dave Kurzman and Steve Spilger. I'm also in and out of the hospitality room.--Rkihara 22:39, 1 April 2019 (EDT)

Template that can be used in the wiki

I've created Template:Notice float right for anyone that wishes to use it. If you visit the template page, it gives an example of how to use it. You can also see it in use on my user page. Enjoy! ···日本穣 · 投稿 · Talk to Nihonjoe 16:15, 5 April 2019 (EDT)

Changes to ISFDB links

The way ISFDB links -- including links to author, title, series, etc records -- are displayed has been changed behind the scenes. The new linking format is supposed to be supported by all browsers going back to Internet Explorer 6, so the change should be completely transparent to all users.

Still, given the variety of browsers out there, you can never be 100% sure. If a link doesn't work as expected, please post the offending URL and your browser version here. Ahasuerus 16:11, 6 April 2019 (EDT)

Does this address the issue of Facebook adding the "&fbclid=" bit onto URLs? ···日本穣 · 投稿 · Talk to Nihonjoe 12:08, 8 April 2019 (EDT)
I am afraid not. It's a different change addressing a different issue. Ahasuerus 14:02, 8 April 2019 (EDT)
Some users report issues under certain versions of Safari. Investigating. Ahasuerus 11:31, 14 April 2019 (EDT)
Unfortunately, there wasn't anything that I could do to address the Safari bug(s). For now, I have reverted the software changes to enable Safari users to access our data. Ahasuerus 11:57, 16 April 2019 (EDT)
Out of curiosity, what is the technical change? Uzume 19:29, 18 September 2019 (EDT)

Canonical name Thomas Blot or William Simpson (I)

In relation to the 1891 pseudonymous NOVEL The Man from Mars, as by Thomas Blot T1013407:
This week I added the 1900 Third Edition, Revised and Enlarged, by William Simpson (I), and the 1893 stated second edition, identical to the first but also under the author's real name. --both of those as publications one NOVEL by William Simpson (I) T2536737, although 80 of 280 pages are new in the Third Edition, a 60-page Preface and a new chapter. Most information on how the 1900 differs from the earlier eds. is now in the 1900 publication record P710623.

That is: 1893 change in credited author name; 1900 change in contents --but not so great that we should treat it as a new NOVEL, in my opinion. Right? If so, then one NOVEL title should be variant of the author, depending on which should be the canonical author name.

Neither Blot nor Simpson (I) is likely to enjoy a growing bibliography here. SFE3 knows the writer as Blot, from the 1st edition. Library of Congress doesn't know him at all.

Thomas Blot or William Simpson (I)? That governs the (re)location of all Author data and much NOVEL data to the appropriate parent record, and whether the Novel by Simpson should be dated 1891 or 1893? --Pwendt|talk 16:53, 6 April 2019 (EDT)

(fix template links, rewrite and expand) --Pwendt|talk 17:52, 6 April 2019 (EDT)
I think "Thomas Blot" is the way to go. As you said, there is little chance of either list growing significantly. Uzume 19:22, 18 September 2019 (EDT)

Arthur C. Clarke Award - Categories

The Arthur C. Clarke Award is currently set up to have the following categories:

  • Winner
  • Runner-Up
  • Shortlist
  • Honorary Award

It would appear that the first three categories as they are currently defined in the database are actually different "award levels" of the Best Science Fiction Novel "award category". My guess is that the editor(s) who originally created these categories were trying to get around the fact that -- the way the software currently works -- an award category has to be either a "Poll" or "Not a Poll". As luck would have it, the Best Science Fiction Novel category doesn't fit this mold. In 1987-1996 there was one "winner", one "runner-up" and a number of nominations every year, but "runner-ups" were eliminated in 1997. Whoever originally entered the date presumably tried to get around this problem by creating additional categories.

It occurs to me that there may be another way to organize the data. We could merge the three categories into one, "Best Science Fiction Novel", and set it up as a "Poll". Winners would then be entered as "Award Level 1", runner-ups as "Award Level 2" and regular nominees as "Finalists" (a supported "special award level".)

What do you think? Ahasuerus 19:54, 9 April 2019 (EDT)

Sounds good. It also looks like there were 3rd place winners announced in 1989 and 1990, per the SFADB. The Award's site lists only winners and shortlist for all years. --Ron ~ RtraceTalk 06:58, 10 April 2019 (EDT)
Done! Ahasuerus 15:46, 14 April 2019 (EDT)

Wiki page elimination

Moments ago I made three submissions related to the ISFDB wiki page Author:Jonathan Scott. That page gives a medium-length quotation from the British DNB.

  1. As the DNB quotation concerns Scott's edition of the Arabian Nights alone, [1] I copied all of it (typography unmodified) as a Note on Title 1340796.
  2. I copied most of the first sentence (typography modified) as a Note on Author Jonathan Scott.
  3. On the Wiki page I added a note, with signature and timestamp, explaining how the page is now redundant (entirely).

I have wondered about the procedure for wiki page elimination. I don't work on it systematically but I have added some Notes to the database that make some Author, Bio, and Publisher wiki pages redundant. Is it desirable to annotate the wiki pages as I have done here? Alternatively, is there another place to report that a page is now redundant?

The Author page is entirely redundant (pending approval of these submissions) in that all of its content is available in the database. Perhaps I should have included a link or some instruction in the Author Note --as I would have done for someone with a medium or long bibliography. For instance, minimally

  • "In 1811 Scott published the work by which he is chiefly known, his edition of the Arabian Nights" in six volumes (the 1811 work listed below).

--Pwendt|talk 14:39, 10 April 2019 (EDT)

When content from a wiki page has been moved to the database, add the following to the top of the page: {{Page transferred}}. That will tag the page for deletion and it will (eventually) get deleted. You can also use {{Deletion candidate|your reason here}} (filling in your reason) for pages that are not a simple transfer case. I have deleted the author's page. -- JLaTondre (talk) 16:06, 10 April 2019 (EDT)

New Help page organized

I created Help:How to (wiki) (which previously only existed as a list of pages in a category). I organized the pages by what they discussed. ···日本穣 · 投稿 · Talk to Nihonjoe 15:01, 12 April 2019 (EDT)

SFE3 Reconciliation - New Report

The way our author URLs are reconciled with SFE3 has been changed. A new cleanup report has been created and deployed. The new functionality includes:

  • All editors have access to the report (only moderators can "ignore" SFE3 URLs)
  • The cleanup report is now updated nightly with the latest SFE3 data
  • The software analyzes SFE3 article names and tries to guess what the corresponding ISFDB author should be

If everything goes smoothly, the new data will become available in a couple of hours when the nightly process runs. Ahasuerus 23:15, 13 April 2019 (EDT)

The nightly process ran successfully and added 4380 SFE3 URLs to the report. I have cleaned up the first 10 and the process seems to be working correctly. Things are looking up. Ahasuerus 11:33, 14 April 2019 (EDT)
Thanks!--Dirk P Broer 06:34, 2 May 2019 (EDT)
Glad it's useful :-) Ahasuerus 06:48, 2 May 2019 (EDT)

Inconsistent category names on Locus Poll Award in 2007, 2009 and 2010

I notice that there are a few instances of categories for the Locus Poll Award not being prefixed by "Best". This is visible at http://www.isfdb.org/cgi-bin/awardtype.cgi?28 , but a rough list of the offending categories/years is:

  • Collection (2010)
  • Fantasy Novel (2007, 2009, 2010)
  • First Novel (2007, 2009)
  • Nonfiction/Art Book (2010) - note that the "Best" category is "Best Nonfiction/Art", no "Book"
  • Novelette (2010)
  • Novella (2007, 2010)
  • Short Story (2010)

Anyone know if there's a good reason for this? It certainly seems less than ideal from a UX point-of-view. I'm a newbie here, so I dunno how easy it would be to merge these back into the more common "Best..." categories - it looks like there might be enough records for some sort of SQL bulk UPDATE to be better than multiple updates via the UI?

(FWIW, Wikipedia has them all as "Best Whatever", SFADB just has "Whatever". Locus' site uses "Best Whatever", at least for 2010: https://locusmag.com/2010/06/2010-locus-awards-winners/)

In a similar vein - but it is at least consistent within the award - the Nebulas are just categorized as "Novel", "Novella" etc, whereas on the SFWA site they are "Best Novel", "Best Novella" etc. e.g. https://nebulas.sfwa.org/award-year/2018/

--ErsatzCulture 19:19, 14 April 2019 (EDT)

I can think of two possible explanations:
  1. The person who created the additional categories overlooked the fact that related "Best" categories were already on file
  2. The person who created the additional categories was working with what s/he considered an authoritative source -- perhaps Locus issues? -- and s/he wanted to be as accurate as possible
Normally, a minor name change doesn't result in the creation of a new category, so my inclination would be to merge them and to document any discrepancies that we may find in the Note field. Let me check submission history to see who created these categories first... Ahasuerus 20:07, 14 April 2019 (EDT)
I have scanned the submission table looking for clues. Here is what I think is happening with these categories.
Originally, award types and categories were handled differently. Award types used to be hard-coded using two-letter abbreviations -- see "award_ttype" on this Wiki page. Award categories used to be manually entered every time an award was entered -- see "award_atype" on the same Wiki page.
It was very messy and I ended up rewriting the whole thing in 2014 -- see Development/Archive/2014 for details. I then migrated previously entered award categories to the new table structure. I think I tried to merge near-duplicates where feasible, but it looks like I missed quite a few.
I think it should be safe to merge these categories. Roughly 200 awards will be affected, which shouldn't be too hard to do manually. Ahasuerus 21:22, 14 April 2019 (EDT)
OK, I think I have merged everything that could be merged without losing specificity. Please let me know if I missed anything. Ahasuerus 21:30, 17 April 2019 (EDT)

Fantasy and Science Fiction mentioned ISFDB

I forgot to mention it back on April 4. See this post. ···日本穣 · 投稿 · Talk to Nihonjoe 20:08, 16 April 2019 (EDT)

Nominating Biomassbob as Moderator

I would like to nominate Biomassbob (talkcontribs) for moderator. He has been an editor for seven years and has 44530 edits as of this date, covering the full range of the database. I have always found his edits to be detailed and complete. I believe that he is highly qualified to be a moderator and he has accepted the nomination.

Support

  1. Support, as nominator.—Rkihara 10:41, 20 April 2019 (EDT)
  2. Support. No concerns; I can't remember the last time I had an issue with any of his plentiful and varied submissions. p.s. Thanks for making me feel old, Ron. --MartyD 17:05, 20 April 2019 (EDT)
  3. Support. About time :) Annie 17:55, 20 April 2019 (EDT)
  4. Support. I have no concerns. ···日本穣 · 投稿 · Talk to Nihonjoe 19:38, 22 April 2019 (EDT)

Neutral

  1. I haven't done a lot of moderating lately, so I'll abstain. Ahasuerus 19:16, 23 April 2019 (EDT)
  2. I can't remember having moderated enough of his submissions to make a judgement, therefore I stay neutral. Jens Hitspacebar 13:21, 25 April 2019 (EDT)

Outcome

The nomination was successful; the moderator flag has been set on the account. Congratulations! Ahasuerus 11:29, 27 April 2019 (EDT)

Cropping Amazon images

Dirk Stoeker has identified a way to crop Amazon images:

However, the way our software is currently configured, an Amazon image URL that contains "L._" generates a yellow warning ("Extra formatting in an Amazon URL") on post-submission pages. It also generates an exception on one of the nightly cleanup reports.

That said, it would be easy to modify the software to accept "L._CR" URLs, so the question is whether we want to allow them. On the one hand, "cropped" URLs can come in handy when dealing with unusual Amazon images. On the other hand, Amazon URLs are not always stable and this particular method has been apparently changed in the past. Here is what an overview of Amazon images says about it:

  • This crops the image, trimming away everything but a selected rectangular area. The first two numbers indicate the upper left-hand corner of the selection area (the first is how many pixels from the left, the second is how many pixels down from the top). The next two numbers are the width and height of the selected area, in pixels.
  • Note: this was a great way to trim away the excess white space left by other commands,,. which other commands used to create, although that was greatly reduced sometime around July 2007. You have to take the time to crop precisely for a given image. If you set the crop area to go beyond the edge of the image, the image will merely be resized with white space added.

Based on this history, I wonder if it may be better to create a cropped image on your computer and then upload it to ISFDB. What do you think? Ahasuerus 19:26, 20 April 2019 (EDT)

Amazon tends to change their formatting when it suits them and relying on it may end up with some weird side effects when they do in a few years. If the Amazon image is not good as is, then we either should live with it or find another one or fix it -- but relying on an undocumented and non-guaranteed external formatting is just asking for trouble. Annie 21:46, 20 April 2019 (EDT)
Downloading, cropping and uploading to ISFDB is not an valid approach, as there still can be copyright on the images (even if it is debatable if copyright can exist for a mere photo of a book). As far as history of Amazon shows they simply ignore old parameters when they change, so in this case the result would be an uncropped image again. --Stoecker 04:17, 21 April 2019 (EDT)
Copyright is irrelevant for this discussion. We use non-public domain cover images under fair use criteria. -- JLaTondre (talk) 08:32, 21 April 2019 (EDT)
Not for me. Even if the use of the motive is ok, modifying the photo taken by somebody else is a minefield I wouldn't want to step into, especially with the upcomming copyright changes in Europe. --Stoecker 11:58, 24 April 2019 (EDT)
Amazon covers are not always reliable, especially for older works. I'm not thrilled with the idea of locally uploading Amazon images, especially for works that were verified without a cover. If the cover is actually for the wrong edition, it creates a false record where it is harder to see that the image is not correct. Especially since uploaders are incredibly poor at properly changing the source field in the fair use template. The crop functionality has been around for sometime. If change were to occur, it would probably be that Amazon deprecates it, in which case, it would go back to showing the whole image so not much of a risk in my opinion. -- JLaTondre (talk) 08:32, 21 April 2019 (EDT)
While I don't care for relying on specifying the cropping in the URL, I don't see any harm in allowing it, and it does seem a little better to me in terms of preserving the provenance of the image than we might get if we recommend downloading and cropping and uploading. --MartyD 09:57, 21 April 2019 (EDT)
It looks like we have 3 editors in favor of allowing "L._CR" URLs, 1 opposed and 1 neutral (me). Would anyone else like to share his or her thoughts? Ahasuerus 11:06, 25 April 2019 (EDT)
If everyone else is in favor of it, I am not going to block the decision - so ignore the opposing editor. :) It is an easily reversible one if Amazon managed to make it a disaster with an update - because of the format, we can always strip those programmatically later. Annie 13:20, 25 April 2019 (EDT)

(unindent) OK, we have a consensus or as close to it as we are likely to get. I have changed the software to ignore "._CR" in Amazon URLs and updated Template:Image Host Sites with Dirk's instructions. In addition, the new cleanup logic is somewhat more robust than the old logic, so I expect it to find 30 new "problem" images tomorrow morning. Ahasuerus 19:59, 26 April 2019 (EDT)

Cleanup completed. Ahasuerus 14:14, 29 April 2019 (EDT)

Abridgements

There doesn't seem to be anything specific in the Help pages about abridgements, specifically what constitutes an abridgement, how to credit additional 'authors'/'editors' or note them if unspecified, varianting to the original title, effects of translation (e.g. abridgement of a translation vs. translated and abridged vs. abridged then translated) and distinguishing abridgements of an oft-abridged title. Are there rules or even consensus for these topics? If this turns into a discussion of actual rules, it should be moved to that forum, but for now I'm wondering about where we stand. ../Doug H 12:05, 29 April 2019 (EDT)

In my experience, it depends on the nature of the abridgement. If the differences are minor, they are documented in Notes. If they are major, a new title record is created.
If the person who performed the abridgement is known and a new title needs to be created, s/he is credited as a co-author -- see William Shakespeare's bibliography for examples. If the text was rewritten so extensively that it qualifies as a separate work (typically retellings for young children) then the original author is sometimes only credited in Notes. Ahasuerus 18:10, 29 April 2019 (EDT)
If we stick with crediting per the pub and it doesn't list the abridger on the title page, then it should be entered as by the author alone and varianted to as by the author and the abridger. We are inconsistent in doing that though. -- JLaTondre (talk) 18:45, 29 April 2019 (EDT)
If you variant to the title with the original author and abridger, then you can't variant that in turn to the original. Does that mean abridgements aren't supposed to be varianted to the original? Is there any way to keep track of the relationship? ../Doug H 19:37, 29 April 2019 (EDT)
Variants are the same work under a different title (or translations). They are not for variance in the work itself (other than translations). If an abridgment is different enough to be considered a new work, then it would not be varainted to the original work. If the changes are minor enough to not be considered a new work, then there would not be a separate title & it would be simply captured in the notes (as Ahasuerus referred to above). If it's different enough to have its own title record, then you would need to use the notes to link it back. The only exception would be if it's a minor abridgment and a title change (then a variant would be appropriate). -- JLaTondre (talk) 20:03, 29 April 2019 (EDT)

New External ID type for Science Fiction-Leihbuch-Datenbank?

The German SF database Science Fiction-Leihbuch-Datenbank has a fairly limited scope, but their data is solid and well organized. Their publication-specific pages can be accessed using "permalinks" like http://www.sf-leihbuch.de/index.cfm?bid=1691, which should make it easy to add the site as a new "External ID" type. Can anyone think of a reason not to add it? Ahasuerus 15:34, 3 May 2019 (EDT)

No reasons I can think of. --Ron ~ RtraceTalk 11:02, 5 May 2019 (EDT)
Me neither. Stonecreek 13:33, 5 May 2019 (EDT)
OK, FR 1274 has been created. Ahasuerus 14:12, 5 May 2019 (EDT)
A new External ID and a matching linking Notes template have been added -- see this record for an example. Ahasuerus 14:17, 6 May 2019 (EDT)

The Lord of the Rings / The Lord of the Rings (Boxed Set)

Hi, everybody! I just wonder why we have different titles for these: it seems to be somewhat illogical by our internal logic. After all, as for the titles involved in the respective OMNIBUSes, they should be identical; the form of publication shouldn't be reflected in the title, only in the publications. Am I missing something? Stonecreek 09:57, 5 May 2019 (EDT)

Agreed. It might relate to this discussion (at one time the single volume Lord of the Rings was apparently considered a novel). However, since they are both omnibuses, they should be merged. -- JLaTondre (talk) 10:10, 5 May 2019 (EDT)
Yes, I dimly remember that discussion. If no one objects, I'll do the merging: in any case, I'll wait until Tuesday before beginnin with it. Stonecreek 13:38, 5 May 2019 (EDT)
Can't find the original now that they're merged (without an long manual search), so a question - was the entry for the box itself, which had a separate ISBN and cover image than the books it contains? Should the other boxed sets be dealt with the same way? Two contrasting examples - James Blish Cities in Flight has a single publication for the set like Lord of the Rings, but A Game of Thrones: 5-Book Boxed Set will (hopefully) never see a single omnibus publication. ../Doug H 08:53, 8 May 2019 (EDT)
Re: GRRM. You never know, I think, though a printed one-volume edition will take a long time (and some muscles to carry home), there'll likely be an ebook someday.
Re: LotR. There were several boxes, some with one ISBN, some with separate ISBNs. I remember that there were several around the movie(s) release(s), apart from the one mentioned in the title note. Ideally, the page count should be split into three portions (XXX+YYY+ZZZ), though there was one (unverified) set that had only one number.
And yes, this was one example of the underlying principle, I should think. I have merged the Blish OMNIBUSes: thanks for pointing them out. Stonecreek 09:58, 8 May 2019 (EDT)
Lots more where those came from. Just pulled them from the list of 200+ titles with "boxed set" or "boxset". Didn't even try other variations. ../Doug H 19:29, 8 May 2019 (EDT)
To be merged, they would have to have the same contents and same title (minus the artificial box set disambiguation we add). If they had the same contents and different title, they would be varianted. There may be a few more cases out there if someone has the patience to go through them all, but I randomly poked through a bunch of the results from doing such a query and only found one. -- JLaTondre (talk) 21:58, 8 May 2019 (EDT)
So if there is no existing omnibus with a matching title when you enter the box set, the generated TITLE title would also include "box set", until such time as someone entered the matching omnibus without a "box set", at which point the box set should be merged with the second omnibus title? The inclusion of "box set" is acceptable as long as there isn't an existing non-boxed omnibus? In the case of Asimov's Foundation Trilogy, the box set title entry explicitly states it is not to be merged with the single volume omnibus. So it sounds like someone had a different opinion that a moderator agreed with. No verifications, so no names or dates. And a question for Stonecreek - what do you mean "by our internal logic"? And as a final observation - I thought box sets were different and were put in as OMNIBUS because that was the closest thing without coding a new type - which is the only reason I keep picking at this, I don't yet understand. ../Doug H 22:53, 8 May 2019 (EDT)
No, it seems that (box set) wouldn't belong to the title in general: it crept into our database just because vendors such as Amazon have that title, which only belongs to a specific publication, and is not to be found anywhere in or on the publication itself. Stonecreek 00:26, 9 May 2019 (EDT)

Crediting cover scans - proposed design changes

Here is how cover scan credits work at this time. When we link to a cover scan, we credit the organization operating the server that hosts the scan. The credit line reads "Cover art supplied by X", where [X] is the name of the organization (Amazon, Galactic Central, etc) linked to the organization's home page. Clicking the image takes you to the raw image hosted by the third party. Here is an example of how it currently works.

There are two partial exceptions to this rule. First, images hosted by the SFE3 Picture Gallery are handled differently. Due to SFE3 requirements, clicking an SFE3-hosted image takes you to the Gallery Web page that hosts it as opposed to the raw image. For example, if you click the cover of this publication, you will see its Gallery page, not the raw image. This was done in order to support SFE3 requirements -- see Template:Image Host Sites, which reads, in part, "all SFE3-hosted images must have a link to the associated "Gallery" page added after a "pipe" ("|") character", for an explanation of the technical details. (Ignore the fact that the linked publication page currently credits "Encyclopedia of Fantasy" instead of SFE3. It's a recently discovered bug which needs to be fixed. The bug has been fixed.)

The second partial exception is ISFDB-hosted images. As expected, clicking an ISFDB-hosted image takes you to the raw image. However, clicking the word "ISFDB" on the "Cover art supplied by ISFDB" line takes you to the ISFDB Wiki page for the image as opposed to the ISFDB home page. Here is an example of how it works.

This is inconsistent design because seemingly identical links behave differently depending on whether the cover scan is hosted by ISFDB, SFE3 or another party. To make it worse, recent discussions with the administrator of Science Fiction-Leihbuch-Datenbank have identified additional requirements for linking to their cover scans. They would like "Cover art supplied by" lines to link to their publication-specific pages the way they currently link to ISFDB Wiki pages for ISFDB-hosted scans. We could modify our software to accommodate their requirements, but it would further muddy the waters. Besides, who knows what kinds of additional linking requirements other sites may have in the future?

An e-mail discussion of this issue with the administrators of Science Fiction-Leihbuch-Datenbank and SFE3 ultimately resulted in the following proposal:


1. Change the software so that clicking a cover scan (including SFE3-hosted cover scans) should always take the user to the raw image file.

2. Modify the credit line to have two links instead of one. The first link will continue to be a link to the site's home page. The second, new, link will be to a publication-specific page (if defined.)

3. The new wording of the credit line will be "Cover art supplied by X on this Web page" where:

  • "X" will be the name of the organization hyperlinked to its home page, and
  • "this Web page" will be a link to a publication-specific page, specifically:
    • for SFE3-hosted images it will be the cover scan's Gallery page
    • for ISFDB-hosted images it will be the ISFDB Wiki page of the cover scan
    • for Science Fiction-Leihbuch-Datenbank-hosted images it will be the publication-specific page associated with the cover scan
    • all other cover scans won't have the words "on this Web page" displayed because there is no page to link to

For now we will keep the current way of entering links to publication-specific pages, which uses "|" to separate the cover scan URL from the URL of the publication-specific page. If the current proposal is accepted and implemented, I expect that we will revisit this issue at some point in the future and probably create a separate field for the data that is currently entered to the right of "|", but that's another discussion.

So, what do you think? Ahasuerus 13:56, 5 May 2019 (EDT)

Sounds overall very good to me! Stonecreek 13:45, 7 May 2019 (EDT)
To me as well. Jens Hitspacebar 14:50, 7 May 2019 (EDT)
Me three Annie 19:53, 7 May 2019 (EDT)
Sounds good to me. ···日本穣 · 投稿 · Talk to Nihonjoe 20:22, 7 May 2019 (EDT)
Thanks, folks. FR 1277 has been created. Ahasuerus 20:47, 7 May 2019 (EDT)

(unindent) The deed is done. (Hopefully without messing up anything else.)

BTW, there is another site that requires that we link to the underlying page -- Smashwords. We have a cleanup report that hunts invalid Smashwords links, but we need to add a yellow post-submission warning. I'll take care of it when I add Science Fiction-Leihbuch-Datenbank. Ahasuerus 17:53, 8 May 2019 (EDT)

Couldn't we get something similar for Amazon-linked cover art? I regularly get into a situation where I want to check out the actual page, so it'd be nice if I could click on a link and it'd bring me directly to the actual pub instead of to the frontpage where I then have to search for the pub I'm looking for...
Additionally, could it be done to have a link to any one of the Amazon sites, depending on which site is used to link to? For example, linking cover art from Amazon UK (or DE, or...) would link to Amazon.co.uk instead of to Amazon.com MagicUnk 14:00, 9 May 2019 (EDT)
Publications with ISBNs display links to the supported Amazon stores under "Other Sites" in the navigation bar on the left. Publications with ASINs link to the supported Amazon stores in the External ID section.
Having said that, we could add an explicit Amazon link to the credit line for Amazon-hosted images. Unfortunately, there is no way of telling which Amazon store the image is associated with because all Amazon images use the same URL structure (at least to the best of my knowledge.) I guess it means that we will have to link to Amazon.com. Ahasuerus 14:55, 9 May 2019 (EDT)
Ah yes, I could have found out myself if I'd paid attention. My apologies. Regards MagicUnk 17:56, 9 May 2019 (EDT)

Linking to SF-Leihbuch

SF-Leihbuch has been added to the list of third party Web sites that we are allowed to link to.

The software module responsible for post-submission warnings and cleanup reports has been rewritten to handle the three Web sites that require additional links for images -- SFE3, Smashwords and now SF-Leihbuch -- consistently. If you encounter any issues, please post your findings here. Ahasuerus 21:33, 9 May 2019 (EDT)

Crediting third party sites - Part 2

Now that the dust has settled and various queued up software tweaks have been implemented, I'd like to revisit the issue that I raised a week ago.

At this time there are three third party Web sites that require us to link to a particular Web page when displaying their images: SFE3, Smashwords and SF-Leihbuch. The way we do it is by entering the URL of the image, then the "pipe" character ("|"), then the URL of the third party Web page associated with the image -- see Template:Image_Host_Sites for site-specific details. In the future, we may be given linking permissions by other sites that have similar requirements.

I don't think the current design is a good long term solution. It's not consistent with how other data entry fields work, it can mess up Advanced Publication Searches and it makes the software more convoluted than it needs to be.

I propose that we create a new field for these links. Instead of entering something like "http://sf-encyclopedia.uk/gal/clute/FrazerS-Blow.jpg|http://sf-encyclopedia.uk/gallery.php?id=FrazerS-Blow.jpg" in the "Image URL" field, we would be entering "http://sf-encyclopedia.uk/gal/clute/FrazerS-Blow.jpg" in the "Image URL" field and "http://sf-encyclopedia.uk/gallery.php?id=FrazerS-Blow.jpg" in the new field. The new field would be optional, but we may be able to add pop-up validation to make it mandatory for the sites that have this requirement. Post-submission warnings and cleanup reports would be adjusted accordingly. Existing data would be migrated automatically.

I am not sure what a good name for the new field would be. Something like "Web page of the image"? Any reasons not to add it? Ahasuerus 18:39, 11 May 2019 (EDT)

Breaking it out into a new field seems reasonable, but it needs to have a name that is not confusing. Most users (especially new ones) will not be familiar with the unique requirements for this field. The name should be something that makes it clear it's only needed in certain cases (or another option, though more complicated, would be to use JavaScript to only make the field displayed when an URL for one of those sites it entered). I'm struggling to come with an good option, though. -- JLaTondre (talk) 10:50, 12 May 2019 (EDT)
Unfortunately, I can't think of a good name either. For now, I have created FR 1282, Create a new publication field for crediting 3rd party sites. Based on the limited feedback, I assume that it's relatively low priority. Ahasuerus 13:35, 24 May 2019 (EDT)

"SF Calendar" added

A new Web page, SF Calendar, has been added as per FR 998. It is accessible from the "Other Pages" section of the navigation bar on the left. After selecting a day, you are taken to the list of authors who were born and who died on the selected day.

Let's wait a few days to let everyone get used to the new functionality and address any page layout issues that may arise.

Once the layout is stable, we can revisit the discussion which prompted the creation of this FR in 2017: what do we want to appear at the top of the home page? Now that the birth/death data is available on a separate page, can we remove it from the front page? Pare it down? Replace it with a link to "SF Calendar"? Should we display other links at the top of the home page, e.g.:

  • Author Directory
  • Award Directory
  • Publisher Directory
  • Magazine Directory
  • Statistics/Top Lists

? Ahasuerus 17:01, 13 May 2019 (EDT)

Enhancing the "Chronological Bibliography" page?

A new Feature Request, FR 1280, "Create a noncategorized-chronology page for authors?" has been created on SourceForge. It reads:

I frequently want to view all of a given author's works in a single chronological list, rather than a set of categorized chronological lists. Is there already a way to do that? (When I click "Chronological", I get categorized lists.)

Currently, "Chronological" author pages ignore series groupings. However, they display novels, collections, anthologies, magazines, short fiction, essays, etc in separate sections. "Alphabetical" pages use the same approach.

The proposal, as I understand it, would add a new Web page for author bibliographies. The new page would display all titles chronologically without regard for the title type. I can see how it could be useful in certain cases, e.g. if it's not immediately clear whether a short novel appears in the NOVEL section or the SHORT FICTION section. At the same time I suspect that combining reviews, chapbooks, EDITOR records, etc in one long list may be confusing. Perhaps displaying the title type next to each title would help? Ahasuerus 14:58, 20 May 2019 (EDT)

That's a nice feature. The title type should definitely be displayed. Moreover, information about language and variants are quite helpful. The display format could be a mixture of the way titles are displayed on the author page and in the "Contents" section on a publication page, which would also include title series and alternate name information. For example, the line for the German translation of Iain M. Banks' Consider Phlebas could look like this:
  • "Bedenke Phlebas [German] (1989) • [Culture • 1] • novel (trans. of Consider Phlebas 1987) [as by Iain Banks]"
Also, the list could be organized into a separate section per year. Then the year could be omitted per line:
  • "Bedenke Phlebas [German] • [Culture • 1] • novel (trans. of Consider Phlebas 1987) [as by Iain Banks]"
Jens Hitspacebar 15:43, 20 May 2019 (EDT)
I would love to have a "All fiction in order" section - but if you add the collections and anthologies and magazines and reviews and interviews, it will become way too long to be useful. Maybe if you can chose which types you see? I assume that variants will remain filterable in that new view as well? Annie 16:49, 20 May 2019 (EDT)
We could have multiple different Chronological" pages, e.g.:
  • "Chronological by Title Type", which would be the same as the current "Chronological" page
  • "Strict Chronological" -- or whatever we decide to call it -- which list everything chronologically
  • "Fiction (Chronological)", which would be limited to novels, short fiction and poems
  • Possibly more, at which point we will probably need an intermediate menu page for different options
That being said, it looks like Jens may be proposing an additional layer of layout changes that would apply to all "Chronological" pages. Is my understanding correct? Ahasuerus 09:14, 21 May 2019 (EDT)
No, my layout proposal was for the "noncategorized-chronology" feature request only, not for the already existing chronological summary author bibliography. But now that you mention it, making the layout per title more consistent across all chronological pages could be a good idea. Or maybe not, I'm not sure yet. The existing chronological author summary bibliography puts the kind of variant at the beginning:
  • Translation: Bedenke Phlebas [German] (1989) [as by Iain Banks]
which, if used the same way on the new "Strict Chronological" page as well, could probably make the list hard to read if many lines would begin with Translation: or Variant:. On the other hand, changing the existing chronological summary author bibliography to also use "(trans. of ...)" instead of Translation: doesn't make sense there because it's obvious that it's a "translation of" due to the way parent and variants are formatted using the bullet lists.
Jens Hitspacebar 16:15, 21 May 2019 (EDT)

Notifying Primary Verifiers

Where do we stand on notifying primary verifiers of changes to their pubs? The "My Changed Primary Verifications" will show what field changed, but not what the change was (both columns show the new value). When that feature was implemented, the community was still saying that primary verfiers should be informed of what the change was. Since the "Note to Moderator" is displayed on the change page, that became the more frequently used method over posting to the verifiers' talk pages (as the moderator's note is easier and remains with the edit history, that makes sense). There is the added complication that not all changes (title level, import/remove) are shown in the changed verifications report. We used to be pretty stringent on the notification requirement. However, a recent conversation has indicated this is no longer the case. What are people's opinions on this? It doesn't help editors for moderators to be inconsistent so we should update practice and/or our documentation based on the community's current position. -- JLaTondre (talk) 17:37, 20 May 2019 (EDT)

Let me try to address the technical side. FR 1237 says:
  • Enhance "My Recently Changed Primary Verifications" to capture Import/Export and Remove Title submissions associated with the user's primary-verified publications.
Once it's been implemented, the report will become more comprehensive and more useful. Also, Roadmap 2017 says:
  • 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.
It will take more time to do, but it will make the report much more useful.
Both features are close to the top of my list, but I need to code and deploy the new Amazon interface first. Fixer takes so much of my time these days that making at least some of what I do publicly available (and thus freeing up more development time) is my top priority at the moment. It's kind of a Catch-22: working on other features means that I can't work on the Amazon interface, which means that I have to spend most of my time working on Fixer, which means that I don't have time to work on other features... Ahasuerus 09:06, 21 May 2019 (EDT)
Yes, I wasn't looking for a technical solution (though that will be nice when it arrives), but rather how we deal with it before the full technical solution comes. I appreciate the enhancements you have been giving us. -- JLaTondre (talk) 19:43, 21 May 2019 (EDT)
I am sorry, I didn't mean to come across as defensive, but I guess I did. Too many balls in the air and not enough productive hours, especially these days :=\ Ahasuerus 21:31, 21 May 2019 (EDT)
I do not notify when I am doing what I consider house-keeping actions (moving identifiers, adding transliterations, editing any of those elements, fixing typos in Notes, fixing a broken HTML or a formatting issue, updating a title or an author name to comply with the capitalization policy and so on) - but I always add a moderator note (and if I am replacing, I tend to add the old and new value in that note). Outside of that, I do notify - one by one if needed; as a group when I am working on a bigger project (swapping a canonical name for example). I also try to notify when making a big change on a title level that impacts a PV'd publication - with a note of what changed to what so we can backtrack if needed. Import/export also causes a notification in my book. For any big changes (changing an author name, fixing page numbers and so on), I try to discuss before I do the change. Annie 19:55, 21 May 2019 (EDT)
Annie's approach may be a viable compromise solution, at least for now: minor changes are listed in the Moderator Note field while significant changes require a note on the Talk page. Approving moderators get to decide what qualifies as a "significant" change and make sure that the Moderator Note includes the "before" version of the data for "minor" changes (since it would be lost otherwise.) Ahasuerus 21:05, 21 May 2019 (EDT)
Edit: On second thought, we probably need to create at least some standards re: what is considered a "minor" edit. Annie's list -- typos, transliterations, HTML fixes, etc -- would be a good start. Ahasuerus 05:44, 22 May 2019 (EDT)
One of the problems when approving someone else's update is that if they do not add the note, you cannot add it. In a lot of cases, the record has at least one more thing that needs fixing but now with most of the identifiers out of the way, we start running into no-issues publications. If I see something else needing fixing, I would fix it and add both edits to the moderator note - so the PV does not wonder what happened. Any chance of making the moderator note editable during the approval? That will also help with the other two issues we have - the fact that the moderator note is visible (a moderator will be able to clean up the private data if any) and the inability to add anywhere any notes when you approve something that at first glance looks weird. Add the ability to add the change if the original submitter had forgotten and we will have a much better system. :) Annie 22:40, 21 May 2019 (EDT)
Letting approving moderators change Moderator Notes would be possible although non-trivial. However, I am not sure it would be optimal. There would be potential for confusion re: who wrote what.
It may be better to create a new field and call it something like "Approval Note" or "Note by the approving moderator". We could then display the Moderator Note and the Approval Note side by side on submission history pages. We may also want to change the name of the "Moderator Note" field to "Submission Note", "Comment about the Submission" or something similar to avoid confusion with the new field. Ahasuerus 05:57, 22 May 2019 (EDT)
Sounds great! "Submission note" is fine, I'd think: it's short and right on the head of the nail. Stonecreek 08:35, 22 May 2019 (EDT)
That works as well :) Annie 17:19, 22 May 2019 (EDT)
I also like "Submission Note" and "Approval Note". ···日本穣 · 投稿 · Talk to Nihonjoe 17:31, 23 May 2019 (EDT)
OK, FR 1281, Add an "Approval Note" field; change "Moderator Note" to "Submission Note", has been created. Ahasuerus 11:12, 24 May 2019 (EDT)

I've gone ahead and approved the submissions I had on hold over this. I'll encourage the submitter to add moderator notes in the future, but given the low level of participation in this discussion, I don't feel it is appropriate to reject their edits when other moderators have been approving them. Thanks. -- JLaTondre (talk) 19:07, 22 May 2019 (EDT)

Sorry to be late to the party. I have not been paying attention to what others do. I don't worry about moving of identifiers or correcting of minor deviations from capitalization standards, but otherwise I would have done as you did: put on hold and ask that the PV(s) be notified. But I also prefer to avoid rejecting and losing work, so sometimes I will notify and/or check with the PVs myself and accept and remind the submittor to follow notification protocol the next time. It is a bit tough to be strict about notifications with so many inactive PVs and PVs with special notification instructions. --MartyD 22:30, 23 May 2019 (EDT)

Proposed Software Changes

Ahasuerus: As these "describe again what is obvious" type of requests is constantly causing me useless trouble: Would you accept a solution to this issue as patch? My solution would be to add the existing values at the time of accepting a request to the stored request, so that a historic view could show the accepted difference situation (That still can differ from the actual value at submission time in case of inbetween changes). --Stoecker 13:39, 23 May 2019 (EDT)

[P.S. later] I'd really like to have such a feature, because even moderators create request without any comment like this http://www.isfdb.org/cgi-bin/view_submission.cgi?4270746 and even moderators can be totally wrong. In this example the price was correct before with €15.00. It seems it is now €18.00, but I payed €15.00 as stated on the website at that time (plus €11.00 delivery) for that one Foster story (and the remaining useless contents :-) --Stoecker 13:55, 23 May 2019 (EDT)

We had a fairly extensive discussion of this issue in early 2017. Let me copy-paste the relevant parts of my comments:
Re: "a snapshot of all OLD values". Unfortunately, it would require a significant effort. Granted, it would be easy to do for fields like "ISBN" and "Price". However, consider publishers. The way the database works, we store publisher numbers (1, 2, 3, etc) in publication records. Then, when we display a publication, we retrieve the name of the publisher, including its transliterated name(s), from other parts of the database and display them.
This works well when displaying current data. However, suppose we were to save a verified publication record as it existed prior to submission approval. We would store publisher ID 12345 in the saved record. Then, a few months later, publishers 12345 and 12346 are merged, so publisher 12345 no longer exists in the database. When the original verifier goes back to check this pub's history, there is no publisher 12345 to display. The same thing can happen to publication series, authors and titles. Actually, it can get even more complicated with authors and titles if we want to preserve the pseudonymous/VT/series relationships as they existed at the time.
The ultimate way to address this issue would be to build a snapshot of the then-current version of each about-to-be-changed Publication Web page and store it in a separate database location prior to submission approval. We could then have a list of snapshots for every publication and display them on demand.
We'll need to add a warning about potentially broken links, but the textual part should be very close to what the data looked like as of the time of the edit.
[comments about the impact of this change on disk space omitted since testing has shown it to be manageable, especially if we compress the data]
[related comment:] We started work on a "history" system -- basically a log of changed data -- in 2007. However, we quickly ran into the problems outlined above and more. I spent many man-hours trying to get it to work in the early 2010s, but eventually had to give up because the underlying approach was flawed. Ahasuerus 18:42, 10 January 2017 (UTC)
As far as I can tell, the summary above is still our best bet.
I guess the first step would be to rewrite the software that displays Publication pages. Currently it displays the data as it retrieves it from the database, mostly one line at a time. We would need to change the code so that it would retrieve all of the data first, then build the Web page and then display it. Once it has been done, the next step will be to modify the submission approval process. For primary-verified pages, it will invoke the part of the Publication display code that retrieves and formats the data. It will then compress it and file it in a "history" table within the database. [Need to run, may post more later]. Ahasuerus 15:05, 23 May 2019 (EDT)
Having a complete version history like Wikipedia including the chance to revert edits and show history would be fine, but from my current perspective this would be impossible to do. My proposal only includes a small subset. At the time of a change request acceptance I'd simply add for changed fields the old and new information, so that the changes can be displayed with proper information. Maybe there could even be a cleanup process to drop that additional space wasting information after some time. --Stoecker 15:29, 23 May 2019 (EDT)
I agree that implementing Wiki-like "version history", including the ability to revert changes, would be time-consuming and not feasible at this time.
However, what I outlined above is more modest in scope. The filing software would simply capture the HTML version of each about-to-be-changed primary-verified publication record. It would then compress the HTML "blob" and store it in some table. Each "blob" would be linked to its publication ID and its submission ID. That way the HTML would be accessible from View Submission pages as well as from Publication pages. That shouldn't be too difficult to implement, I hope. Ahasuerus 17:00, 23 May 2019 (EDT)
P.S. I have added the above to FR 1238 "Create an Edit History page for publications". Ahasuerus 11:04, 24 May 2019 (EDT)
This proposed HTML method is nothing where there is a chance of successful cooperation between you and me. It is much to complex to get it to work. --Stoecker 04:34, 30 May 2019 (EDT)
I am not entirely sure I am parsing the 2 sentences above correctly. Are you saying that capturing the body (HTML) of primary-verified Publication pages is a complex process? It would appear to be fairly straightforward since the software already builds all the requisite data when a Publication page is displayed. We just need to capture and store it. Could you please clarify the nature of the complexity? Ahasuerus 10:09, 30 May 2019 (EDT)
Still my proposal is an option - a submission is stored as XML in the database. My proposal is to extend this XML at the moment of accepting the submission and store the old values for changed elements (and only these) in the XML. That way a diff can not only show new, but also old values of a change. That will help a lot to verify a change even if it does not cover all possibilities. This adaption needs only small modifications at submission acceptance and display of a change request and not a major new functionality. And it does also not conflict with any future plans. --Stoecker 04:34, 30 May 2019 (EDT)

Magazine issue navigation

How easy would it be to implement the "Previous ← Current → Next" issues navigation as a database feature, as shown in the Notes field here? Could it use the existing database table information used to create the issue grid, so any missing issues would be skipped? And maybe indicate that one or more issues were skipped (if it can easily see that)? Just curious, as it would benefit browsers of the site so they wouldn't have to visit the issue grid between each issue. ···日本穣 · 投稿 · Talk to Nihonjoe 21:05, 20 May 2019 (EDT)

How would you determine "next/previous" in the cases where we do not have months in the date field? Finding missing issues will be even more problematic - some magazines have 2 issues in some months, some are irregular... and not all magazines have nice numbers on all their issues. Not a bad idea but just wanted to make sure we cover all the logistics... :) Annie 22:19, 20 May 2019 (EDT)
That's why I asked. There's some sort of sorting happening in the issue grid, so that's why I suggested that as a possible method, if it can be used for that. ···日本穣 · 投稿 · Talk to Nihonjoe 02:28, 21 May 2019 (EDT)
Let's consider Science Fiction Quarterly (UK), which has 6 issues whose months of publication are known and 4 issues whose months of publication are unknown. January 1952 through August 1953 should be doable since all of the issues have months. However, what should we do about December 1953 and later issues? A human can figure out that the sequence should be:
  • December 1953
  • #6
  • #7
  • February 1955
  • #9
  • #10
but it would be difficult to replicate the logic programmatically. Ahasuerus 07:14, 21 May 2019 (EDT)
To add to that: The grid does not really do well if it needs to stick more than one issue in the same cell (the no month one for example for a specific year) - it does its best but... it can get a bit wonky and it does look weird for some magazines. Which is not a problem on a grid (you see them all, the slightly off order is not that problematic) but becomes a problem on a straight list. Annie 16:06, 21 May 2019 (EDT)
Perhaps a way to turn it on for magazines that have all the issues entered? Or some way to enable it on a per-issue basis? It could then be left turned off for any that don't line up correctly. Or, program it to display the out of order issues according to which was entered into the system first? Just tossing out ideas. :) ···日本穣 · 投稿 · Talk to Nihonjoe 17:27, 23 May 2019 (EDT)
We could use publication record numbers to determine the order in which issues were entered. However, there is no guarantee that magazine issues are necessarily entered in the order of their publication.
Having said that, it may be possible to piggy back on FR 1202, "Add fields for magazine issue numbers to pub records":
  • Add the following new fields for magazine issue numbers to publication records: Volume, Issue, Whole Issue. On the data entry side, we could have a single line for capturing this information, e.g.:
  • Volume: [field] Issue: [field] Whole Issue: [field]
For example, the proposed "Whole Issue" field could support the "|" format, which we currently use for page numbers. The first part of the value would be displayed to our users. The part after the "|" character would be a "sorting" value used to create "Previous" and "Next" links. Ahasuerus 10:45, 24 May 2019 (EDT)

Change ID

Good evening. I am sure this is covered somewhere but cannot for the life of me find the article. How can one go about changing their username on this site? Thanks in advance. Zybahn 23:22, 28 May 2019 (EDT)

I am afraid the ability to change user names is not supported by the ISFDB software. Ahasuerus 00:03, 29 May 2019 (EDT)

Guerra eterna (The Forever War) - Joe Haldeman - broken link

Hi, there's a broken cover picture link to the Mondourania website here, where they've probably reorganised their site. As I don't read Italian, can I ask that someone else rectify this. I think the link should be as below, but I may be wrong.

Covers page here: http://www.mondourania.com/collezione/collezione%201-20/uraniacollezione1-20.htm

Relevant cover here: http://www.mondourania.com/collezione/collezione%201-20/uraniacollezione10.htm

Thanks. BanjoKev 16:38, 30 May 2019 (EDT)

Fixed. The problem was an extra space in the string we had - so chances are that was never a correct image to start with. :) Annie 18:04, 30 May 2019 (EDT)
Thanks Annie! BanjoKev 19:06, 30 May 2019 (EDT)

Project Gutenberg publication records

Concerning publication records for Project Gutenberg ebooks:

1. Do we discourage formal verification? Only a small share of our records are Primary Verified, now 4 of about 50 for year 1998 and none at all 2017 to 2019.

2. Do we discourage use of the Cover field, in reference to the first image that is displayed at Gutenberg.org in HTML format (and probably some others)? Commonly the first image displayed, if any, is an image of the front book cover or front dust jacket. For instance, Ebook #25609 at Gutenberg (HTML format) and at ISFDB (COVERART credited). We now credit no cover artists for year 1998 and only four 2017 to 2019. (We link no cover images for those years.)

3. We have some old-fashioned records whose Notes contain an HTML link to the reported Ebook (not a different one) at Gutenberg.org. This one also provides an explicit list of available formats P291428. (That one is Primary Verified by User:DESiegel60, retired.) Do we recommend either retaining or deleting such notes?

--Pwendt|talk 16:54, 5 June 2019 (EDT)

As I have been the one keeping up with entering new PG publications, I will give you my rationale on how I've been entering them.
  1. We don't discourage formal verification. I have not been verifying them as PG's are a special beast in my opinion. They are regularly updated with corrections, etc. and I don't see the point of creating different records for minor edits (we have had this debate about ebooks before & haven't really come to a decision on how to handle). There is also the issue that PG ebooks are available in a variety of formats. So what is actually meant by verification? One format checked? All formats checked? Anyone can use the PG link to view the ebooks so verification hasn't been important in my mind. I'm basing my entries on the HTML versions.
  2. We don't discourage covers. However, I personally don't enter the image or the cover credit. My rationale is that PG ebooks are available in a variety of formats; some of which have "cover" images and some of which don't. But then again, I have been adding the interior art even though not all formats have that either. So I'm not necessarily consistent in my approach. ;-)
  3. As we have the built in link, there is little point to the manual link. Likewise, PG formats are pretty standard (though possibly not all older ones have every format) so little point in the format note. There is no harm to them and as long as they are valid (ex. formats haven't changed since the note was written), there is no point in removing. In the case of the one you link to, I don't believe the Plucker format is available anymore so it should probably be changed to reflect that.
Those are just my opinions. You are free to verify, add covers, etc. as you see fit. -- JLaTondre (talk) 21:38, 5 June 2019 (EDT)

Star Wars (and other franchises)

I noticed someone recently took the series Star wars: Nonfiction out of the parent series Star Wars universe with comment "As this lists titles about the Star Wars franchise, it is not a part of the Star Wars universe". This may be true but imo all "Star Wars" should be consolidated under one parent series. Such a separation would also affect any franchise like Star Trek, Doctor Who, etc. And where to set boundaries: Is a character encyclopedia not fictionous enough to be part of a "Universe"?

As a user i'd like to find all of a "Universe" on one page, not divided to a few.

What do you think? WernerWelo 13:46, 6 June 2019 (EDT)

Hi, Werner! It was me who made that differentiation. The reasoning behind this is that if a title is truly in-universe it is fiction (not nonfiction). It's misleading to have a title like this in the same category as this one, where in fact fictional stories about spaceships are told. Stonecreek 14:03, 6 June 2019 (EDT)
What about Star Wars chronologies and 'nonfiction' ? ../Doug H 14:23, 6 June 2019 (EDT)
These are in-universe and as such part of the franchise. Stonecreek 14:43, 6 June 2019 (EDT)
I‘m all for differentation, however all of Star Wars should imo be combined under one parent series wether its fiction, fictionous nonfiction or real nonfiction. Has not to be the „Star Wars Universe“, maybe a parent like „Star Wars Franchise“ or similar could do it. Under such a series the differentation could be made in universe and others. Welo 14:53, 6 June 2019 (EDT)
I see no issues with having Star Wars: Nonfiction having the parent Star Wars universe because the nonfiction is about the Star Wars universe. Since it's all contained within a series that specifically states it's nonfiction, I doubt anyone will ever get confused. ···日本穣 · 投稿 · Talk to Nihonjoe 18:13, 6 June 2019 (EDT)
I have been struggling with this issue for years. Some non-fiction books are "in universe" and some are not. Some are written or co-written by the author(s) (like Roger Zelazny's Visual Guide to Castle Amber), some are "authorized" and some are completely unauthorized. I am yet to come up with a hard and fast rule to determine when a non-fiction series should be entered as a subseries of the main universe. Ahasuerus 00:12, 7 June 2019 (EDT)
I'd say distanced pieces like "Star Wars on Trial" or movie compendiums that rely more on the filmical aspects shouldn't be part of a universe. 'Encyclopedias' of characters, planets and space-ships should be imo. Stonecreek 02:59, 7 June 2019 (EDT)
The problem is, that makes it very subjective as to what is included. One person may think it clearly should be, and another doesn't. I think it's best to simply lump all non-fiction together in a non-fiction series, and then put it under the main umbrella series for the entire series. That makes it easy for everyone to remember, and we'll not have to debate it every time something new comes up. Straightforward solutions work best, IMHO. ···日本穣 · 投稿 · Talk to Nihonjoe 13:19, 7 June 2019 (EDT)

(unindent) Ironically I came here looking for a link about a book entitled The Land of Unreason by Fletcher Pratt and L Sprague de Camp. —The preceding unsigned comment was added by Chris humel (talkcontribs) . 18:31, 6 June 2019 (EDT)

How is it ironic? Also, the book you're looking for is here: 18960. ···日本穣 · 投稿 · Talk to Nihonjoe 22:24, 6 June 2019 (EDT)

Publication pages - tweaks behind the scenes

I have installed a minor patch to change the way Publication pages are built by the ISFDB software. I expect a number of additional patches to be installed in the near future. Once everything has been completed, these changes will facilitate the process of creating "history" Web pages for primary-verified publications.

These changes should not affect the "look and feel" of Publication pages. If you come across any issues, please post your findings here. Ahasuerus 16:09, 14 June 2019 (EDT)

There's a tiny glitch with "(View All Issues)(View Issue Grid)" in the "Publication:" line on a magazine's publication page. The pub's title and these two links are not separated by space any more. Example: Interzone, #263: "Interzone, #263 March-April 2016(View All Issues)(View Issue Grid)". If I'm not completely mistaken there used to be some space between these three elements, which made it easier to read, and it looks like this change has been introduced with one of the recent commits, probably rev. 435. Jens Hitspacebar 04:10, 15 June 2019 (EDT)
Good catch; fixed. The latest patch also fixed links for magazines issues whose EDITOR records are not in a series (not that we are supposed to have any.) Ahasuerus 10:30, 15 June 2019 (EDT)

"Transliterated values" fixes

There was a bug with the way "transliterated values" were being displayed on some pages. I have installed a patch to address the issue. If you see anything unusual, please let me know. Ahasuerus 15:45, 16 June 2019 (EDT)

Pseudonym display

Would it be too difficult to change the display of titles with multiple authors, and where one or more has a pseudonym, to reduce the space and make it easier to read. Right now, it's something like this:

Title by Author One and Author Two and Author Three and Author Four and Author Five and Author Six and Author Seven and Author Eight and Author Nine and Author Ten and Author Eleven and Author Twelve and Author Thirteen and Author Fourteen and Author Fifteen and Author Sixteen and Author Seventeen and Author Eighteen and Author Nineteen and Author Twenty [only as by Author One and Author Two and Author Three and Author Four and Author Five and Author Six and Author Seven and Alt Eight and Author Nine and Author Ten and Author Eleven and Author Twelve and Author Thirteen and Author Fourteen and Alt Fifteen and Author Sixteen and Author Seventeen and Author Eighteen and Author Nineteen and Author Twenty]

If it could be something like this, it would make viewing the information much simpler and much easier to read:

Title by Author One and Author Two and Author Three and Author Four and Author Five and Author Six and Author Seven and Author Eight (as by Alt Eight) and Author Nine and Author Ten and Author Eleven and Author Twelve and Author Thirteen and Author Fourteen and Author Fifteen and Author Sixteen and Author Seventeen (as by Alt Seventeen) and Author Eighteen and Author Nineteen and Author Twenty

Would this be reasonably easy to do? ···日本穣 · 投稿 · Talk to Nihonjoe 23:45, 18 June 2019 (EDT)

I agree that it would be desirable and I looked into it at one point.
What I found was that it would be possible to do if "variant-specific authors" and "alternative author names" -- which are entered and stored in the database separately -- matched in predictable ways. Granted, in many cases they do, e.g. Robert Heinlein is always an alternative name for Robert A. Heinlein, Maxine Reynolds is always an alternate name for Mack Reynolds, etc. However, we also have many collective pseudonyms and house names that overlap unpredictably. For example, consider The Whispering Gorilla series. The Whispering Gorilla omnibus of 3 novels is currently listed as follows:
"Alexander Blade" was a house name used by both "unknown" and by Don Wilcox, so one might think that this line could be rearranged as follows:
However, if you examine the publication record, you will note that David V. Reed is credited explicitly, so the line above would be in error.
The only way to get it to work correctly for all records (that I could think of) would be to change the way we capture variant authors. Instead of simply entering the canonical names on the "Make Variant" Web page, we would have to specify which canonical name each alternate name corresponded to. It would require extensive changes to the way the software works and a subsequent review/revamp of thousands upon thousands of variants. Ahasuerus 11:19, 19 June 2019 (EDT)

Displaying the parent title's authors on the variant title's page

The recent discussion of Dates on COVERART variants has resulted in a software change. Variant title pages whose parent title's date is different from the variant's now display the parent's date on the "Variant Title of" line. For example, "The Skull" by Vincent van Gogh displays the following information:

  • Author: Vincent van Gogh
  • Date: 2011-00-00
  • Variant Title of: Kop van een skelet met brandende sigaret (1886) (by Vincent van Gogh )

Note the bolded date.

While working on this issue, I noticed something that we may want to address. As you can see above, the "Variant Title of" line includes the authors associated with the parent title. The information is useful if they are different from the variant title's authors, but we display the parent's authors even when they are the same.

The "Contents" section of Publication pages only displays the parent title's authors if they are different from the authors of the variant title. It seems to make sense to change the Title page to do the same.

Are there scenarios that would be adversely affected by the proposed change? Ahasuerus 20:53, 25 June 2019 (EDT)

The change has been made as per FR 1287. In addition, the "Variant Title of" line has been changed to display the language of the parent title if it's different from the language of the VT. Ahasuerus 19:05, 29 June 2019 (EDT)

Advanced Author Search - Author Image URL

FR 1286, "Add Author Image URL to the Advanced Author Search", has been implemented. Ahasuerus 19:31, 29 June 2019 (EDT)

Python error

Was trying to add an image to an entry and got this python error: Image:Screen Shot 2019-07-13 at 7.28.13 PM.png. Oddly, everything seems to have gone through okay as the publiscation in question (723121 seems to be fine (it has the cover I uploaded). ···日本穣 · 投稿 · Talk to Nihonjoe 21:32, 13 July 2019 (EDT)

Got the same error again when updating the cover artist for this pub: 264000. ···日本穣 · 投稿 · Talk to Nihonjoe 21:35, 13 July 2019 (EDT)
That was an unintended side effect of the post-approval change discussed on the Moderator Noticeboard, now fixed. The bug only affected the way things were displayed, not the data added to the database. Sorry about that! Ahasuerus 22:50, 13 July 2019 (EDT)
No problem. You want the image deleted? I don't know if it has any potential security stuff in it. ···日本穣 · 投稿 · Talk to Nihonjoe 23:08, 13 July 2019 (EDT)
Nope, no security stuff. Just a graphic reminder to re-test all changes before installing them :-) Ahasuerus 23:18, 13 July 2019 (EDT)

Yiddish Speakers?

I added story by Der Nister which is a translation from the original Yiddish. While I can find a scan of the original Yiddish edition at Hathi Trust (Vol. I and Vol II, I was unable to determine the original title in the proper alphabet and thus added this title reflecting the Yiddish language parent but noting that the original title hasn't been found yet. Are there any editors that are fluent enough with Yiddish to find the proper title? The English title is "At the Border". I'll also note that the collection Gedakht is described by Wikipedia as a "collection of fantastic stories", and may itself be eligible for inclusion here if someone wants to take that on as a mini project. Thanks. --Ron ~ RtraceTalk 12:06, 14 July 2019 (EDT)

It looks like "Gedakht" is געדאכט, but that's all I can figure out. ···日本穣 · 投稿 · Talk to Nihonjoe 15:22, 14 July 2019 (EDT)
Google Translate says "At the Border" is אין די גרענעץ, but I have no idea how accurate that is. ···日本穣 · 投稿 · Talk to Nihonjoe 15:23, 14 July 2019 (EDT)
I searched the text for גרענעץ and found four entries: אויף גרענעץ, which apparently means "On (the) border", זיך א גרענעץ, which means "set a boundary", זיין גרענעץ, which means "its boundary", and קיין גרענעץ, which means "no border". One of those might be the title. ···日本穣 · 投稿 · Talk to Nihonjoe 15:30, 14 July 2019 (EDT)
According to Language help, Linguist knows a little Hebrew. It's not the same thing, but uses the same alphabet, so he might be able to offer some help. ···日本穣 · 投稿 · Talk to Nihonjoe 15:33, 14 July 2019 (EDT)
I think I found it. The table of contents in Volume 1 is on page 243. I carefully reconstructed the title of the story that appears on page 137 using the Hebrew alphabet Wikipedia page and ended up with אויפן גרעניץ which Google translates as "On the borders". Assuming that I identified the correct letters (exporting the scan to a pdf didn't work with cut and paste), I think I've identified the title correctly. Thanks again. --Ron ~ RtraceTalk 17:44, 14 July 2019 (EDT)
Saw the post a bit late. I have added a transliteration to the title : oifn grenits, the equivalent of German auf der Grenze « on the border ». Cheers, Linguist 05:19, 16 July 2019 (EDT).
Maybe you can add Gedakht to the database? You might be able to it more quickly than anyone else here. ···日本穣 · 投稿 · Talk to Nihonjoe 19:54, 17 July 2019 (EDT)
Will do (in due time, as there seems to be quite a lot to do…). Or maybe Ron can enter the romanized titles, and I'll Hebraize what I can ? Lekhaim ! (! לחיים) Linguist 04:43, 18 July 2019 (EDT).
I went ahead and took a stab at it. It would be an excellent idea for someone to check the entries against the scans to ensure I didn't get them wrong. Also to add missing transliterations. I also went ahead and made the Yiddish name canonical. I was able to link up all but one of the English titles. I suspect that "The Fool and the Forest Demon" may be "דער נאַר און דער ןןאַלד־רוח" as Google translates that as "The fool and the old age" which is a close title. Thanks. --Ron ~ RtraceTalk 18:35, 19 July 2019 (EDT)
I have linked "The Fool and the Forest Demon" and "נאַר און דער ןןאַלד־רוח, that was OK, and corrected the transliteration (der nar un der vald-rukh; you had confused ו (v) and ן (final n), which do look almost alike; and in Yiddish, א at the beginning of a word usually indicates it starts with a vowel, so און is just un “and” (= German und), not aun. I have started looking at געדאכט: ערשטער באנד : your first entry, the title of the story on p. 7, דער נזיר און דאָס ציגעלע, was written backwards (each word correct, but in the wrong place, if you see what I mean). I have corrected it, but there is one word I don't undestand : "The hermit and the stigele". Does that sound familiar ? Linguist 05:56, 20 July 2019 (EDT).
Same thing for the next one (words were inversed) : צוס בּאַרג, tsum barg = German zum Berg “at the mountain”. Linguist 07:41, 20 July 2019 (EDT).
Ditto for אין וואַלד, in vald, "in the forest" (German im Wald). Linguist 07:58, 20 July 2019 (EDT).
Went through both volumes, updated a few more transcriptions, gave translations when possible. There are two titles I don't understand fully, and three not at all. Sorry ! :o( Linguist 08:38, 20 July 2019 (EDT).

Votes and counting

I've been adding a lot of votes recently and have been amused at how, sometimes, my feeling about a story/book fails to match up with others' votes. Amused, but not surprised--we all read, and enjoy, different things. However, I do think it would be interesting in the Statistics/Top Lists to expose those stories that have the farthest deviation from the mean (i.e., stories that some people absolutely hated while others loved). The one I ran into today was Philip Jose Farmer's "Riders of the Purple Wage," originally published in Harlan Ellison's Dangerous Visions anthology. I thought the story was quite good, especially when re-reading it recently because the concept of a universal basic income has become more prevalent. It already had two votes when I added mine: a 1 star and a 4 star. Somebody really hated it. Gengelcox 16:24, 14 July 2019 (EDT)

It's certainly doable. A simple database query (select std(rating),title_id from votes group by title_id having count(user_id)>3 order by std(rating) desc) creates a list of controversial titles starting with:
Ahasuerus 18:29, 14 July 2019 (EDT)
That's exactly the kind of thing I was hoping might pop up -- I recall reading The Time Traveler's Wife with a book club and how it really creeped some readers out while others loved it. Gengelcox 10:20, 15 July 2019 (EDT)

On the subject of the Stats page and votes, I'm wondering if we could expose the top lists for novella, novelette, and short story along with the list already being run for novel? Gengelcox 16:24, 14 July 2019 (EDT)

We could create a "top" list for SHORTFICTION titles, which would include short fiction without a "length" designation. We could also create separate sub-lists for novellas, novelettes and short stories. Ahasuerus 18:29, 14 July 2019 (EDT)
I'd be interested, but obviously not a priority. Gengelcox 10:20, 15 July 2019 (EDT)
OK, FR 1289 has been created to document these suggestions. Ahasuerus 11:04, 16 July 2019 (EDT)

Jerry Ahern

Someone had started shifting the canonical name to a name that was never used in print (Frédéric Charpier for them (at least in our DB) and unless I am missing something, is not even connected to these books). Can someone share light on what they had been doing here (and add notes to the authors names). If noone can explain what this is all about, I will break the pseudonym in about a week and clean these pages up a bit. Annie 15:27, 17 July 2019 (EDT)

Apparently Charpier translated a bunch of Ahern books into French, then continued the series himself while continuing to use the Ahern name. At least that's what the French Wikipedia article states:
Frédéric Charpier a par ailleurs traduit en français un grand nombre de romans d'inspiration survivaliste de la série Le Survivant de Jerry Ahern (en), puis continué lui-même la série en français, sous le couvert d' « adapations » qui sont des créations originales (numéros 29 à 53, aux éditions Vaugirard).
Not sure if that's what this is about, but it's what I could find. ···日本穣 · 投稿 · Talk to Nihonjoe 16:07, 17 July 2019 (EDT)
This site seems to support what I wrote above. ···日本穣 · 投稿 · Talk to Nihonjoe 16:11, 17 July 2019 (EDT)
Aha. That explains it. Thanks - somehow I missed that when I was looking through the wiki articles earlier and trying to figure out what is going on. Why can't people just use their own names (I know, I know...). I will add some notes on the author's page... :) Annie 16:12, 17 July 2019 (EDT)
I'm the culprit (only saw the post today). As Nihonjoe explained, Charpier went on writing the series himself after translating the first ones, but indeed his name never appears except as a “translator / adapter”  : a common practice among the Plon / Presses de la Cité / Vaugirard / Vauvenargues publications (see some of the Blade / Jeffrey Lord pubs, with a host of French shadow-writers). I thought I had made a general note somewhere about it, but had apparently only meant to… But such notes appear on each individual title, see here. And by the way, the Wikipedia passage quoted by Nihonjoe was written by me ! Linguist 04:15, 18 July 2019 (EDT).
Come to think of it, I had written the note for Original Richard Blade Adventures in French, but somehow the one about the Survivalist got waylaid. Linguist 05:23, 18 July 2019 (EDT).
Yeah, once I got on the right track, it started clearing up. Thanks for the explanation! I wonder if we should not use Jerry Ahern (Frédéric Charpier) for these books (or something along these lines) so it is clear it is not really Ahern - we had done that for other wrong attributions in the past. Annie 13:33, 18 July 2019 (EDT)
I would use "Jerry Ahern (II)" and include a note on the author page explaining things. ···日本穣 · 投稿 · Talk to Nihonjoe 14:45, 18 July 2019 (EDT)
That also works :) Even "Jerry Ahern (I)" would work. I find descriptive differentiator easier but the nummeric ones work as well. Annie 15:21, 18 July 2019 (EDT)
I like the numeric ones better because they are shorter. ···日本穣 · 投稿 · Talk to Nihonjoe 15:29, 18 July 2019 (EDT)


(unindent) Our software supports uncredited co-authors (e.g. see this series), but it doesn't have native support for ghostwriters. We have FR 346, "Add support for ghostwriters", but it wouldn't be easy to implement. For now, the best we can do is add a note at the author level like we have done with V. C. Andrews and her ghostwriter Andrew Neiderman.

Speaking of the "Richard Blade" series by "Jeffrey Lord", Wikipedia says:

  • In the early 1990s the Russian publishers could secure the rights to only the first six books in the series, and approached the translator - Mikhail Akhmanov - to write the further adventures of Richard Blade.[3] Together with then young sci-fi author Nick Perumov and others, Akhmanov wrote over sixteen sequels[4] to the adventures of Richard Blade, and then, after writing Russian sequels to the saga of Conan, went on to create numerous original characters and plots.

The Russian Conan project was apparently similar. Ahasuerus 15:30, 18 July 2019 (EDT)

Publication Series Name

Several hours ago the new publication series 8383 Queen's Treasure Series was created in the usual way, by adding its first publication record, which remains the only one. I added a pub series Note. Publication Series: Queen's Treasure Series. The keyword Treasure should be plural.

First Question: If the pub series name is modified in its one publication record, will the existing pub series record be deleted, and its information lost --either immediately on approval, or by some automated cleanup, perhaps overnight? (I understand that is the case when a publisher name loses its only publication record.) --Pwendt|talk 21:57, 17 July 2019 (EDT)

Publication series get deleted as soon as they loose their last book. This is one of the cases when you need to find a moderator and ask them to change the name. Which I just did. :) Annie 22:16, 17 July 2019 (EDT)
Thanks. So this one is now named "Queen's Treasure Series", which happens to be the name displayed on half-title page, rather than "The Queen's Treasure Series" as atop the description and list on the next page and in newspaper advertisements by the publisher. In some newspaper articles "Queen's Treasures" in quotation marks may be found.
Second Question: Do we have good reasons to prefer shorter or longer versions of names such as "[The] Queen's Treasures [Series]"? --Pwendt|talk 17:57, 18 July 2019 (EDT)
It's actually "Queen's Treasures Series" now - as requested :)
In a lot of cases, it is down to the editor's preference. The word Series is rarely in the name of a series I create for example - except for older series where I may add Series if I expect that someone can consider it a magazine. Or something. So... no real good reason - unless one specific spelling is used a lot more often in both books and reference materials. If I was adding it, I probably would have added it as "Queen's Treasures" but that is just me - and I would not change it when someone adds it the way you added it. Adding a note in all spelling being used (and where) is never a bad idea. Annie 18:18, 18 July 2019 (EDT)

Slow Server

Is it just me or is the server extremely slow at the moment? Annie 18:26, 19 July 2019 (EDT)

Facebook links

I'm guessing a fix was put in to allow links from Facebook to link correctly by stripping out the "&fbclid=" part. However, it looks like there's a tiny error still. The links come through with an equal sign (=) added to the end, so they still don't work. This link should go to 724848, but the equal sign prevents that. ···日本穣 · 投稿 · Talk to Nihonjoe 17:56, 24 July 2019 (EDT)

I don't recall any recent software changes that should have affected the way URLs are processed. Could you please post a sample "native" Facebook URL? Ahasuerus 22:10, 24 July 2019 (EDT)
They look like this:
https://l.facebook.com/l.php?u=http%3A%2F%2Fwww.isfdb.org%2Fcgi-bin%2Fea.cgi%3F134396%26fbclid%3DIwAR3OXAEyuHT1_DnUzX-Tcv034vhq_SK67RFSzaq9YPKJ-Rn0XK4sDu_e1no&h=AT3ko-M8HhxEAD8-AyA0tPcBCjgKDxL9prtGNFq-RVARTQoeLpsqJ3qmQuAuSSKwOYYGPkJl63JIBO_WyePJ7VehGAC_0QxggDWv-uOzULagbxUoyQq2eCiyf3jw_X3a2y6nXdZnMYgaXj_HFNiofqqf3xV_1nFkysIT-w-Q-NdK9PnjoR2-6Usdi8UkDKGAcfHdFc01i8wjGkUc22KidBMRxAm6hRrPHcxYvXJQ7z_oXXCZQLK5pgW1m_346YrJ0LQFPzZOzICFCMsngFmZoq_G3sjl4zH6jefD5FbY1B88uOv8hl4gz_B_tazNKu5uG9c5L4uG6pSFgHQ-4peRtlpt_Sow8v9nyKLLZGPapKOrOWRK9wEoQSJu1A_BuPFIjxX-7KZPi871ZFCbl4qecl4Ndw0dM3zUvPq9EXhZLE02Vs1lk1LcYOgZpLL-VHbYOCsyqLqjU5GLHl_i1geqN882tKZQG2N49Fcl7KWypQnmcn2PSlcDHlUkaIYV6GhQ5B0Up2FopsRdBBKAbEnY6K4OltweDFaOGwt66V4-Pmt_RTWAzT4h10b88kcNYuYxaIdBbtg8ak8qDcdxCqMLlrK3rxparnymcMPObbFeClppvLX3M3HgjTyhOUzIfzCZJbyeic8
That translates to http://www.isfdb.org/cgi-bin/pl.cgi?724848= somehow (note the equal sign at the end). ···日本穣 · 投稿 · Talk to Nihonjoe 13:50, 25 July 2019 (EDT)
It translates to
http://www.isfdb.org/cgi-bin/ea.cgi?134396&fbclid=IwAR3OXAEyuHT1_DnUzX-Tcv034vhq_SK67RFSzaq9YPKJ-Rn0XK4sDu_e1no
for me btw - which is not even the same author ID. And it is consistent with any link like that from Facebook since we started having this problem awhile back (after they changed how their links work). Are you sure you copied the correct link? 14:35, 25 July 2019 (EDT) —The preceding unsigned comment was added by Anniemod (talkcontribs) .
Depends if you are signed into Facebook or not. If you are signed into Facebook, it resolves to Nihonjoe's link. If you are not, it resolves to Anniemod's. -- JLaTondre (talk) 17:11, 25 July 2019 (EDT)
I see the same regardless if I am signed in (Firefox, I can see my name and so on) or not (Clean Chrome browser where I had never logged in). It may have something to do with having access to the link or not... Annie 17:14, 25 July 2019 (EDT)
I don't know about access. I just clicked the link above. Regardless, it is clear Facebook adds different parameters under different conditions when you click the "Follow Link" button. And ISFDB resolves those different parameters differently (no surprise). In an ideal ideal world, Facebook wouldn't add parameters to the URL, but that won't change. In an ideal world, ISFDB would handle extra parameters correctly (i.e. ignore them). I remember there was discussion about software issues with that. Guess the question is do we think the benefit (possible more exposure, new users) is worth the effort. I don't have an opinion on that - especially since it is Ahasuerus' time, not mine ;-). -- JLaTondre (talk) 17:48, 25 July 2019 (EDT)
That's right, we would have to change the way ALL ISFDB Web pages process and validate parameters. It should be doable and would also result in certain additional benefits. Unfortunately, it would be time-consuming :-( Ahasuerus 20:14, 25 July 2019 (EDT)
is the site behind an apache proxy or equivalent? Seems like something that could be done at that level, by not passing that parameter to the processor at all. --Unapersson 13:05, 22 August 2019 (EDT)
That's a good point. It may be possible to configure the Web server to pre-parse and modify incoming requests, stripping Facebook-added parameters. The downside is that it would also mean making configuring an ISFDB server more difficult and restrictive. Food for thought... Ahasuerus 12:57, 23 August 2019 (EDT)
This is exactly the type of reason why it would be very good to reduce the number of entry points in our code. By this I mean ideally it would be nice to have a very few CGI files (one would be great) and have that parse the path and query string after that (loading Python libraries as needed, etc.). That sort of transition would allow for greater flexibility (it might be possible to deploy without CGI, e.g., WSGI, etc.) and security (moving all the Python libraries out of the web browsers executable/CGI path and instead into Python packages and modules, etc.). Uzume 18:59, 18 September 2019 (EDT)
Sorry, I gave two different URLS. The example "native Facebook" link is for a different one than the other. I mixed up two different ones, somehow. :) ···日本穣 · 投稿 · Talk to Nihonjoe 18:55, 25 July 2019 (EDT)
Lol I didn't even notice the author ids were different (despite Annie saying that). I focused on the parameter differences. -- JLaTondre (talk) 19:25, 25 July 2019 (EDT)

"Translations without Notes" - Reorganizing the report

Here are the numbers behind Translations without Notes:

+--------------------+--------------------+
| English            |               6476 |
| Italian            |               6386 |
| French             |               4907 |
| German             |               4227 |
| Dutch              |               2554 |
| Portuguese         |               1108 |
| Spanish            |                774 |
| Polish             |                255 |
| Finnish            |                233 |
| Romanian           |                207 |
| Serbian            |                178 |
| Swedish            |                115 |
| Croatian           |                114 |
| Hungarian          |                112 |
| Danish             |                 68 |
| Turkish            |                 45 |
| Czech              |                 44 |
| Lithuanian         |                 44 |
| Japanese           |                 41 |
| Slovenian          |                 20 |
| Slovak             |                  8 |
| Esperanto          |                  8 |
| Chinese            |                  6 |
| Norwegian (Bokmal) |                  5 |
| Galician           |                  5 |
| Korean             |                  4 |
| Norwegian          |                  4 |
| Latin              |                  4 |
| Scots              |                  3 |
| Middle English     |                  2 |
| Catalan            |                  2 |
| Estonian           |                  2 |
| Russian            |                  2 |
| Hebrew             |                  1 |
| Albanian           |                  1 |
| Mirandese          |                  1 |
| Scottish Gaelic    |                  1 |
| Thai               |                  1 |
| Malay              |                  1 |
| Old English        |                  1 |
| Icelandic          |                  1 |
+--------------------+--------------------+

Based on editor feedback, my tentative plan is to create separate cleanup reports for English, Italian, French, German, Dutch, Portuguese, Spanish and Japanese translations. The first 7 have more than 500 affected translations while Japanese is the 7th most common language in the database and sees a lot of turnover.

The remaining languages will continue to be covered by the existing cleanup report whose name will be tweaked. We can also split it into language-specific tables and have a "table of contents" at the top of the page to make it easy to jump to the language of your choice.

Does this sound like a good idea? Anything else that we need to change or tweak while we are at it? Ahasuerus 15:52, 25 July 2019 (EDT)

How did Portuguese sneak up that high on the list... :) I like the plan - it will make that project a bit more manageable. I am not sure that we need the Japanese on its own - they always show up close to the top of the generic report anyway - but if it is not that hard to split it, I guess we might as well. Annie 16:30, 25 July 2019 (EDT)
I like this idea. Looks like I need to work hard to bring the number of Japanese titles even higher. Maybe I can get it to the top five. :) ···日本穣 · 投稿 · Talk to Nihonjoe 16:34, 25 July 2019 (EDT)
How hard it will be to add the parent's title or language (or both) in the report alongside the title (similarly to how advanced search shows titles for example)? It is not critical but sometimes it is helpful to work per language pair (mainly because the names of the translators become familiar and I usually have their pages open on the side...) Annie 21:04, 25 July 2019 (EDT)
It should be eminently doable. Ahasuerus 21:53, 25 July 2019 (EDT)

(unindent) OK, it's decided then. I have another 550 Fixer-harvested ISBNs to process and then I will work on this FR. Ahasuerus 15:42, 31 July 2019 (EDT)

The Shaver Mystery Compendiums - Call for Volunteers

6 Shaver Mystery Compendiums have been added by Fixer. Looking for a volunteer willing to add Contents-level items. (July and August are Fixer's busiest months because September is typically the busiest month in the publishing world.) Ahasuerus 17:45, 28 July 2019 (EDT)

I will add them later today - will see if I can figure out if all of the art items from the magazines were reprinted or just some of them. Annie 13:35, 29 July 2019 (EDT)
Fiction imported; will work on covers, ebooks and so on and a second check if we need any variants instead of the main titles later today. Annie 14:14, 29 July 2019 (EDT)
Thanks! Ahasuerus 15:10, 29 July 2019 (EDT)

Collected Short Stories (H. P. Lovecraft)

I am planning to convert this to a publication series - it looks weird with all the variants and so on - and it will allow us to actually mark the correct editions as part of the series. Anyone can see a reason not to? Annie 14:45, 29 July 2019 (EDT)

Never mind, they already have a pub series. Still something does not feel right... Annie 21:00, 31 July 2019 (EDT)
It appears that they should be in two publication series: Tales of Mystery & the Supernatural and "Collected Short Stories", which we don't support. I do think that the use of a title series which really only applies to the Wordsworth printings here is not appropriate. I would suggest removing the title series and putting these in a new publication series with a notes on both pub series records explaining a parent child relationship. This would be workable unless Wordsworth reprints these as part of Collected Works but not as part of Tales. However, I suspect that it unlikely. --Ron ~ RtraceTalk 06:52, 1 August 2019 (EDT)

What to do with pubs that are not spec-fic?

Hi. I wanted to check with the community what the feeling is about submitting deletions of pub records for which no evidence can be found they are spec-fic, and for which the author is not above the 'threshold' (yeah, I know, quite vague... :)?
I am well aware that it's not fun for the editor(s) that have put all the work in to get their work erased (I wouldn't like it myself). But if we let these records stay because someone happened to have put a lot of work in it, where will it end? It becomes trickier if a verified pub record turns out to be not spec-fic. I tend to leave these alone but these too should - strictly speaking - be deleted from the database (after contacting/notifying the PV first). Thought? Suggestions? MagicUnk 08:05, 31 July 2019 (EDT)

Well, we do not want chemistry and language textbooks (yep, we had some :) ). Can you share a few links of books you consider deleting? Annie 12:40, 31 July 2019 (EDT)
Yup, much easier to make a determination if we know which book(s) you are considering. ···日本穣 · 投稿 · Talk to Nihonjoe 18:02, 31 July 2019 (EDT)
Here are a couple of examples:
Out of the Dark (even though it's primary verified - this one's borderline, perhaps?)
Born of the Sun (even though it's recorded in Locus1 - unjustified imo)
The Bailey Game (did ask ClarkMCI, no feedback to date)
The Raging Quiet (here ClarkMCI made a note that he has suspicions this is not spec-fic)
I checked Goodreads reviews, and not really one that I could find that convinced me these contained spec-fic elements warranting inclusion. MagicUnk 07:51, 2 August 2019 (EDT)
All of these novels are by established SF authors, so I would flip the "non-genre" flag and document what we know about them in notes. It's been my experience that it's more beneficial to keep borderline cases in the database and explain why they are "borderline SF" or "not really SF even though they may look like SF" in Notes. Not only does it help our users, but it also prevents robots and new editors from creating "genre" records in the future. Ahasuerus 10:50, 2 August 2019 (EDT)
Hmm, that's something we could do of course. But then we are ignoring the 'authors above a threshold' rule (assuming they are non-genre), don't we? Afaik the authors of the examples above aren't really above the threshold - even though they are established SF authors. We can update/remove that rule of course... What are the other editors' practices related to non-genre publications? MagicUnk 10:59, 2 August 2019 (EDT)
Are we? The idea of the threshold rule is not to count someone’s work as if they are tomatoes and go by percentages but to allow us not to burden the DB with the collected works of a prolific author just because they happened to write 1 genre novel. If the author is considered a genre one, they are above the threshold IMO. There is a reason why that rule does not say anything about exact numbers. Annie 11:25, 2 August 2019 (EDT)
PS: And borderline spec fic belongs here for two reasons: the boundary will always be in the eye of the reader and so that we do not need to make all the explanations of why it is not. And especially for those borderline ones, I really do not feel like we should be removing them based on GR reviews alone (if you look at The Amazing Mycroft Mysteries and their reviews, you would decide that they are not genre either. And yet, as I read the first two lately, I can tell you that they belong (the first one has killer-bees and enough science around them to qualify it at least as a border case, the second one has a working medium). I would opt on the inclusion side unless either someone reads the book and can make a decision or at we find another way to make sure they are not speculative at all. Or that the author is not a predominantly genre one - if they are, then the works are in anyway. Annie 12:50, 2 August 2019 (EDT)
As to the threshold rule. The relevant parts of the policy say Works (both fiction and non-fiction) which are not related to speculative fiction, but were produced by authors who have otherwise published works either of or about speculative fiction over a certain threshold [are included], and, Works that are not related to speculative fiction by authors who have not published works either of or about speculative fiction over a certain threshold. [are excluded]. Mirriam-Webster has Threshold: a level, point, or value above which something is true or will take place and below which it is not or will not. For me, this means that the policy says that even for established SF authors who have not produced 'much' (ie are below the threshold), the non-spec-fic should not be entered. Cursory reading of the threshold-related discussions over the years seem to support this interpretation. Now, we may not want the policy to mean 'above a certain value', but then I would like to see the rules of acquisition rephrased to somehow include 'non-spec-fic work of established SF authors' (whatever established really means :-))
Now, having said that, there's - as you rightly point out - the other side of the medal; when is a work borderline SF, and thus eligible for inclusion? I concur that it is difficult to figure out without having read the book whether they have to stay or not from reviews and synopses alone. So, concluding, and erring on the side of caution, I surmise the current consensus is to leave them in when there's not unambiguous proof of the contrary, but add a note as to the debatable SF contents of the work. If it can be unambiguously established the contents is non-spec-fic, and the author is below the threshold (even when an established SF author), it has to go out. In practice, the latter case may not happen at all - unless someone actually reads the work, right? MagicUnk 16:31, 2 August 2019 (EDT)
How do you define the threshold? Not what the dictionary says, how do you understand it in ISFDB? We read that rule somewhat differently I think - mainly on how threshold is defined. It sounds like you are trying to find a numeric representation of that value. What I am reading when I see that is exactly what you want to rewrite it as - an author over the threshold is not just about the numbers, it is about "is that one of our authors?". I just cannot see a difference between what we have and what you are proposing to change in the wording. Some authors are ours under one pseudonym (Nora Roberts and her Robb pseudonym for example), some are ours under all names. Annie 16:59, 2 August 2019 (EDT)

[unindent] Wellll... I think that's the issue - we're not interpreting the current rules exactly the same way, which is a Bad Thing™. If I may paraphrase; you and Ahasuerus seem to imply that whenever the author is an SF author -ALL- publications should be recorded. This implies - amongst other things - that however small their output may be, all publications have to be recorded irrespective, as long as they are a recognized SF author. Whereas I'm interpreting the rules as "even if the author is an established/recognized SF writer, if his/her SF output is 'low', everything that's non-SF should -NOT- be recorded". Do note that new editors do not have all the ISFDB history, so are likely to go and check the Mirriam-Webster definition of threshold, and will end up with interpreting it as a numerical threshold - as I did. Going back to the examples given above, authors such as Gillian Cross, Welwyn Wilton Katz, and Sherryl Jordan, notwithstanding having produced SF works, can hardly be considered established SF authors imo. Now, I'm not having a preference one way or the other 'per se', but the rules should be as unambigous as we can make them, especially so because we do not want new editors to interpret the rules - nor the moderators for that matter. So, either we have

  1. If it's a 'recognized' SF author, record ALL (ie SF -and- non-SF) his/her works, or
  2. If it's an author with SF works above a certain threshold (numerical), record ALL his/her works (and consequently, if the output is 'low', do NOT record all of his/her non-SF works)

I've been interpreting the rule as the latter (and I suspect many other editors & moderators with me - chime in and tell me if I'm wrong here ;). So, do we want to organize a poll as to clarify if it's 1, or 2 we want? If we want 1. we need to rephrase the rules, as it's confusing right now for new (non-English-speaking) editors (which I am), as they are likely to interpret it as 2. Also, if we would go with 1., the danger exists we end up with loads and loads of non-spec-fic works of 'established' SF authors, however low their output is/has been.

---And yes, I am well aware of the difficulty of coming up with acceptable definitions of 'recognized SF author', or, 'author with SF works above a threshold' - but that's food for another discussion...

Also, we should be careful not to muddy the waters by introducing the notion of 'borderline' SF, as these need their own set of rules, of which the most important one is (I think): if in doubt, you are allowed to add it to the DB, but make sure to add a clear note about its dubious SF contents. Regards, MagicUnk 11:49, 4 August 2019 (EDT)

Early on, we had multiple discussions about the "threshold". Lots of different definitions were proposed, including numerical ones like "at least 50% [75% etc] of the author's output is SF". We always ended up with exceptions and exceptions to exceptions, so we were never able to come up with an explicit rule. The best we could do was to state that:
  • ... "certain threshold" is hard to define, but we need to draw the line in a way that would exclude Winston Churchill, who published at least one work of borderline speculative fiction. The goal here is to avoid cataloging everything ever published by James Fenimore Cooper, Robert Louis Stevenson, Honoré de Balzac and other popular authors. Instead, we want to catalog their speculative fiction works only.
Note that it doesn't say anything about the author being "recognized". For example, take George Orwell. By any measure -- sales, critical recognition, public awareness, etc -- he is one of the most recognized authors of speculative fiction. Yet we do not list his non-speculative novels.
For what it's worth, here is how I usually approach the issue when deciding whether to include non-genre works for an author. I ask myself: "If an average reader comes across an unfamiliar work by this writer, will she suspect that it is SF?" In the cases listed above, the answer is "no", so their non-SF doesn't get included. If the answer is "yes" or "probably", I include the non-SF works.
A few other things to keep in mind. There was a time when the only non-genre works supported by the software were NOVELs. It made life difficult and sometimes editors used "creative" solutions. They had to be revisited once the software was able to handle all types of non-genre titles. Some of these "creative solutions" may still be lingering in dark corners.
Also, at one point the policy was to enter non-genre works if they had been reviewed in genre-publications. The policy was changed years ago, but, yet again, some ineligible non-genre works still exist in the database. Ahasuerus 11:02, 5 August 2019 (EDT)

Japan Fantasy Novel Award

Can we add this one? It was given from 1989-2103, and then started up again in 2017. It has the following categories:

  • Japan Fantasy Novel Award
    • Grand Prize
    • Award of Excellence
    • Nomination

I will add them once the award is created. Thanks! ···日本穣 · 投稿 · Talk to Nihonjoe 15:50, 6 August 2019 (EDT)

Done. As per this discussion of the Clarke Award, "Grand Prize" and "Award of Excellence" are not separate categories in the ISFDB world but rather different award levels within the same category. The category has been created and I have added a note about the first and the second places being called "Grand Prize" and "Award of Excellence" respectively. Ahasuerus 17:08, 6 August 2019 (EDT)
Based on your comparison to the Clarke Award, there should be two categories under win:
Win:
  • Grand Prize
  • Award of Excellence
Neither of these should have a "poll place" because you either win the Grand Prize or Award of Excellence or you don't. There is no ranking beyond winning one of those two or not. All of the nominees are eligible for both. The prize committee can choose to award in any given year either of those to one or more nominee (two is the most in a given year for either, at least historically), or choose to award only one of them to one or more nominees. The remaining nominees are just nominees for the Japan Fantasy Novel Award in general. I hope that makes sense. ···日本穣 · 投稿 · Talk to Nihonjoe 18:50, 6 August 2019 (EDT)
Originally, the Clarke Award was set up to have separate categories for winners, runner-ups, and shortlisted works. However, based on the outcome of the linked discussion, we changed it to have a single "Best Science Fiction Novel" category. Clarke winners are now entered as "Award Level 1", runner-ups as "Award Level 2" and other nominees as "Finalists" (a supported "special award level".)
If my understanding is correct, "Japan Fantasy Novel Award" has a similar hierarchy: "Grand Prize" is effectively "Award Level 1", "Award of Excellence" is "Award Level 2" and the rest of the nominees are "Finalists". Is this a reasonable approximation of the way the award works? Ahasuerus 19:49, 6 August 2019 (EDT)
Kind of. The problem is that both "award levels" are not awarded every year, as opposed to the Clarke Award where they are. If you win a Japan Fantasy Novel Award, you are awarded either the Grand Prize or the Award of Excellence. ···日本穣 · 投稿 · Talk to Nihonjoe 20:53, 6 August 2019 (EDT)
Let me double check that I understand correctly. "The Grand Prize" is a higher award level than "The Award of Excellence", right? And their recipients are selected from the same pool of books? Ahasuerus 10:22, 7 August 2019 (EDT)
It's always listed higher on websites that list awards, so probably? The recipients of both are from the same pool of nominees. ···日本穣 · 投稿 · Talk to Nihonjoe 12:53, 7 August 2019 (EDT)
There are some years where a Grand Prize is not awarded, so considering it "Award Level 1" seems weird in years where only the Award of Excellence was given. ···日本穣 · 投稿 · Talk to Nihonjoe 20:53, 6 August 2019 (EDT)
Well, similar things have been known to happen. For example, consider the 1976 and the 1994 John W. Campbell Memorial Awards, which had no winners and two (ranked) finalists. Or the 2015 Hugo awards, when "No Award" won in the "Best Novella" and the "Best Short Story" categories. Ahasuerus 10:22, 7 August 2019 (EDT)
I guess I'm most concerned about how it will show up on the title and author pages. As long as it shows "Grand Prize" and "Award of Excellence", then it should be fine. ···日本穣 · 投稿 · Talk to Nihonjoe 12:53, 7 August 2019 (EDT)
Maybe this is a feature request: to have the option of assigning names to the "places" for an award, or to just use numbers. That way, it displays "Grand Prize" instead of "1". I know this wouldn't be applicable in all cases, but it seems this (and perhaps the Clarke Award and other) are a bit of a hybrid between awards like the Hugo (which have multiple categories) and single awards with "poll places" (or rankings) since they give a name to the places, but share a common nomination pool. ···日本穣 · 投稿 · Talk to Nihonjoe 13:13, 7 August 2019 (EDT)
Well... The good news is that we already have a related FR, FR 656, Add a "Term used to described runner-ups" field to Award Category. The bad news is that it would be a fairly time-consuming change. Nothing too drastic, just a bunch of man-hours. Ahasuerus 15:49, 7 August 2019 (EDT)
Another wrench for the works: in 2005, a person who would have won the Award of Excellence declined it. There's not currently a way to indicate that since you can't select both "Poll place" (to indicate they would have won place 2, the Award of Excellence) and "Special" (to indicate they declined the award). ···日本穣 · 投稿 · Talk to Nihonjoe 13:44, 7 August 2019 (EDT)
That's right. The only way to capture both the fact that s/he was supposed be given the Award of Excellence and that s/he subsequently decline it would be to have a separate category for the Award of Excellence and, presumably, another one for the "Grand Prize". I guess we could do that, but where would the nominees/runner-ups go then? Ahasuerus 15:52, 7 August 2019 (EDT)
That's the big question, isn't it? This one really is a hybrid. ···日本穣 · 投稿 · Talk to Nihonjoe 17:23, 7 August 2019 (EDT)
A third category "Award Nominees"? When someone wins, they get removed and moved to the proper award (and we add notes on the three categories explaining the situation). Why won't that work? Annie 17:30, 7 August 2019 (EDT)
Well, it would work, but then we would have the same problem that we had with the Clarke Award prior to the changes. Having nominees in one category, Grand Prize winners in another category and Award of Excellence winners in yet another category would mean that the data for one award year would be scattered across multiple categories. It wouldn't be an accurate model of the way the award data is structured, so third party developers who use our award data would need to add an exception and so on.
Given our software limitations, I guess we need to decide whether it's more important to display the words "Grand Prize" and "Award of Excellence" on award pages or to have all nominated works listed within the same category (and use to document "Grand Prize" and "Award of Excellence".) Ahasuerus 08:15, 8 August 2019 (EDT)
I am less concerned about the award pages (we can add notes there) and more concerned on how things look on title pages -- we cannot add notes explaining what is what there. I know that we have challenges with the model (because it as built based on how US awards look like :) ) - but then we do have two types of consumers of the data - the people that look at the site and the people that use the DB itself. Just thinking aloud. Annie 12:16, 8 August 2019 (EDT)

(unindent) I guess we might as well do it right and change the software to support "Displayed Award Level". I assume that:

  • The new optional "Displayed Award Level" field will be at the award level, not at the award category level. This will support scenarios where the displayed award level changes from year to year.
  • The new "Displayed Award Level" field will be in addition to the currently existing "Award Level" field. The latter will continue to be used for sorting purposes.
  • The new "Displayed Award Level" field will allow arbitrary text like "Grand Prize" and "Award of Excellence" to be entered.
  • If no "Displayed Award Level" value is entered for an award, the value of the "Award Level" field will be displayed.

Does this look about right? Ahasuerus 09:55, 11 August 2019 (EDT)

I think so. Annie? ···日本穣 · 投稿 · Talk to Nihonjoe 16:20, 11 August 2019 (EDT)
Looks good. Do we want to have the award name always in English or do we want to allow a “Native / English” format as well? If the latter, we will also need the transliterations fields. As these names will show up on title pages, I’m inclined towards allowing the double name - especially because some of this names may not have an agreed upon English name and whatever is chosen will be the editors’ interpretation. Annie 02:52, 12 August 2019 (EDT)
At this time none of the award-related records, i.e. award types and award categories, support transliterated values. If we decide to add support for them, we should probably do it across the board. Ahasuerus 11:35, 12 August 2019 (EDT)
True. Which is why I brought it up. Because this one needs transliteration in its main form as well. Annie 12:34, 12 August 2019 (EDT)
I like the idea of adding transliteration fields for the awards. ···日本穣 · 投稿 · Talk to Nihonjoe 14:21, 12 August 2019 (EDT)

(unindent) OK, FR 1295 "Allow entering transliterated values for awards-related values" has been created. Also, FR 656 has been changed to reflect the "Displayed Award Level" functionality discussed above. Ahasuerus 13:59, 15 August 2019 (EDT)

Chen Jiatong

Would anyone happen to know the Chinese spelling of Chen Jiatong's name? I have found a bunch of English-language articles about him, including a recently added SFE3 article, but I don't know how his name is spelled in Chinese. Ahasuerus 17:01, 9 August 2019 (EDT)

Looks like it's 陈佳同 per this book. ···日本穣 · 投稿 · Talk to Nihonjoe 18:12, 9 August 2019 (EDT)
The National Library of Singapore agrees. Thanks! Ahasuerus 10:12, 11 August 2019 (EDT)
No problem. ···日本穣 · 投稿 · Talk to Nihonjoe 16:19, 11 August 2019 (EDT)

Dublin Worldcon

If someone is attending and wants to meet, let me know :) That also means I will be (mostly) out of pocket on the site after today and until late next week. Annie 14:26, 12 August 2019 (EDT)

According to Template:Moderator-availability, Rtrace headed to Dublin a few days ago. Watch out for leprechauns! Ahasuerus 17:51, 12 August 2019 (EDT)
I need to remember to update that thing when I am off... Well, I was considering catching a few and bringing them back to help with Fixer's queues... :) But then they are not that reliable I guess. Annie 19:12, 12 August 2019 (EDT)
Indeed I am. I’ve sent an email with my contact info. I recall that we had problems with the wiki email function in the past, so let me know if you don’t get it. Hope to see you there. --Ron ~ RtraceTalk 04:58, 13 August 2019 (EDT)
Got it - see you in Dublin. :) Annie 10:34, 13 August 2019 (EDT)

Support for 3 new language codes added

The software has been updated to support the following new languages:

  • Guarani
  • Interlingua
  • "South American Indian language", which, as per the ISO 639-2 standard, covers all South American Indian languages

Ahasuerus 17:45, 12 August 2019 (EDT)

"Translations without Notes" split into language-specific cleanup reports

As per FR 1291, the cleanup report "Translations without Notes" has been split into multiple language-specific cleanup reports. The original report is now called "Translations without Notes - Less Common Languages". Additional columns ("Original Title", "Language") have been added throughout.

Updated data will become available when the nightly process runs in a few hours. If you run into any issues, please post your findings here. Ahasuerus 22:48, 13 August 2019 (EDT)

Awesome! ···日本穣 · 投稿 · Talk to Nihonjoe 13:38, 15 August 2019 (EDT)
Thanks! And I love the new format of the reports with the extra columns. Annie 17:19, 27 August 2019 (EDT)
Glad to hear the new reports are useful! Ahasuerus 20:42, 27 August 2019 (EDT)

Professor Bernice Summerfield

For the series: Professor Bernice Summerfield (http://www.isfdb.org/cgi-bin/pe.cgi?12512) the first 6 books have usually different names for the original and the identical reprint. This series pops up in my cover art verification as different titles have the same cover link. For none I found a verified pub, so I'd like to unify them by dropping the "Professor Bernice Summerfield and" and cleanup the duplicate records. Objections? --Stoecker 11:00, 15 August 2019 (EDT)

The titles should be as they appear in each book. If that makes them not "unified", then so be it. We shouldn't be changing them just to fit a sense of order. ···日本穣 · 投稿 · Talk to Nihonjoe 13:38, 15 August 2019 (EDT)
Please read what I wrote before answering! None of the entries have a verified status, so the data is only based on the externally visible data and thus there should be no difference with identical covers (and identical ISBN BTW). --Stoecker 15:52, 15 August 2019 (EDT)
I did read what you wrote. The couple that I checked have the full title as we list it on the cover. If we don't have a PV, and there is no Look Inside available, that's what we have to go with. When the titles have "Professor Bernice Summerfield and the..." as part of the title, it's obviously part of the actual title, unless we can get a PV to show it's otherwise listed on the title page. It's exactly the same as all the "Harry Potter and the..." titles. ···日本穣 · 投稿 · Talk to Nihonjoe 16:41, 15 August 2019 (EDT)
OK. In this case I can change all the titles and covers by adding "Professor Bernice Summerfield and" to the titles without? --Stoecker 17:42, 15 August 2019 (EDT)
If that's what's on the covers, then yes. Lacking any PV to check with, we have to go by what's on the cover or Look Inside. If the cover doesn't have that as part of the title, then we shouldn't be adding it. ···日本穣 · 投稿 · Talk to Nihonjoe 18:29, 15 August 2019 (EDT)
So, for example, this one doesn't have the "and the" as part of the title, so the "Professor Bernice Summerfield" is just the name of the series. On the other hand, this one has "Professor Bernice Summerfield and the Dead Men Diaries " on the cover, so that's what we use for the title unless a PV or Look Inside shows differently. ···日本穣 · 投稿 · Talk to Nihonjoe 18:32, 15 August 2019 (EDT)
I only refer to 1-6. It seems since 7 the "and" is missing. That's why my first proposal was to also drop that for 1-6. But as you rejected that I'll unify 1-6 with the "and". If later a PV notes that this is wrong it at least will be much easier to correct it by simply editing the title records. --Stoecker 08:02, 16 August 2019 (EDT)

Imadjinn Awards

Here's a new award to consider adding. It has the following categories (those marked with * may or may not be genre for ISFDB):

  • Best Anthology *
  • Best Children's Book *
  • Best Historical Fiction (not genre)
  • Best Horror Novel
  • Best Fantasy Novel
  • Best Literary Fiction Novel *
  • Best Non-Fiction Book *
  • Best Paranormal Romance Novel
  • Best Romance Novel *
  • Best Science Fiction Novel
  • Best Short Story *
  • Best Short Story Collection (single author) *
  • Best Thirrler Novel *
  • Best Urban Fantasy Novel
  • Best Young Adult Novel *

Should we add them? ···日本穣 · 投稿 · Talk to Nihonjoe 14:28, 15 August 2019 (EDT)

The award is apparently given by a regional (Kentucky) convention which has been active for the last 6 years. It seems similar to Endeavour Award, which we support, so it appears to be eligible. Ahasuerus 14:35, 15 August 2019 (EDT)
Done. Ahasuerus 08:20, 25 August 2019 (EDT)
Thanks! I'll work on them when I get done with the Japan Fantasy Novel Award (only about 10 more years to enter). ···日本穣 · 投稿 · Talk to Nihonjoe 20:00, 27 August 2019 (EDT)

Spaces in Japanese names

Can we have a report that lists people with spaces in their names if they are Japanese? For example, "のの原 兎太" as opposed to "のの原兎太". The Canonical Name shouldn't have any spaces in it, but I haven't figured out how to do an advanced search that finds them. ···日本穣 · 投稿 · Talk to Nihonjoe 22:08, 22 August 2019 (EDT)

You can do "Canonical Name contains % %" plus "Working Language is exactly Japanese" to try to get at this. You might be able to reduce the list with a few more restrictions, but there is a bit of noise since you pick up the names that are in other languages. --MartyD 08:01, 23 August 2019 (EDT)
Right, that's exactly what I use to find errant spaces. Ahasuerus 12:51, 23 August 2019 (EDT)
Awesome. I've cleaned those up. ···日本穣 · 投稿 · Talk to Nihonjoe 19:24, 23 August 2019 (EDT)

Currency symbols and abbreviations

Do we have somewhere a reference list of currency symbols and abbreviations? Help:Screen:NewPub#Price contains several illustrations but no hyperlink to a more complete list at ISFDB.org, nor elsewhere.

E.g. France: For the currencies of France, we report prices "FR 6" (1865), "F50.00" (1947), and "4 F" (1966) --for publications of Voyages Extraordinaires #1 T7386 and Baltimore Gun Club #1 T7389. Wikipedia-EN implies a single franc (F or Fr or FF) from 1795 to 1960. --Pwendt|talk 13:01, 23 August 2019 (EDT)

This is the only list we have that I know of. The different French Francs are afaik grounded in the different restarts of the currency. It may be that Linguist knows more about the matter, and I'll post a link at his talk page. Christian Stonecreek 13:35, 23 August 2019 (EDT)
I just started Help:List of currency symbols, if anyone wants to add to it or correct it. ···日本穣 · 投稿 · Talk to Nihonjoe 20:46, 23 August 2019 (EDT)
We can also use this page for reference. ···日本穣 · 投稿 · Talk to Nihonjoe 20:49, 23 August 2019 (EDT)
As far as French francs are concerned, it seems to me that this database has mainly used "F" directly followed by the number, e.g. "F5.50", just like the dollar sign. At least, this was the standard rule Hauck and I followed. The difference in currencies does not really correspond to a difference in symbols, which vary greatly from publisher to publisher, as well as with publishing dates. Whenever I meet something like "FR 6" or "4 F", I change it to the standard F6.00 or F4.00. The only tricky cases are :
  • the period between 1945 and 1959, when the one franc piece corresponded to the lowest possible value (due to the dramatic post-war devaluation of the franc). I usually write prices indicated during that interval without any decimals, as these were just impossible. The only exception to this (unfortunately) comes form the lingering but rather rare use of the old 50 centimes piece until the end of the ’40s (mainly for newpapers and such).
  • the short period following 1960, when prices appeared in "new francs" usually noted as "NF" (one "new franc" equalling 100 "old francs"). But as this practice of noting NF disappeared with time, and there was nothing official in the NF notation, prices in new francs are just noted the standard way, with a commentary in the notes if necessary.
If no one objects, I'll add a line about French francs in the list, summing up our usual practice in the matter. Linguist 04:59, 24 August 2019 (EDT).
Which list? Help:List of currency symbols or Help:Screen:NewPub#Price? ···日本穣 · 投稿 · Talk to Nihonjoe 02:04, 25 August 2019 (EDT)
The latter (since the issue has been dealt with in the former). Linguist 04:08, 25 August 2019 (EDT).
Works for me. Feel free to add or correct anything on Help:List of currency symbols. ···日本穣 · 投稿 · Talk to Nihonjoe 13:28, 25 August 2019 (EDT)

So all non Latin characters are followed by numbers without a space character, but currencies in Latin characters use a space before the number. What I don't understand is: why doesn't the French F follow this rule as the only Latin currency.
And I miss the Swedish krona SEK in Help:Screen:NewPub#Price --Zapp 06:33, 26 August 2019 (EDT)

Summary of ISFDB Publications search > Price contains: fr --two characters
'Bfr ' -- the usual hit, these four characters uniformly
Sfr Sfr. sfr sfr. -- four different strings, always followed by a space (narrow SEARCH)
Fr Fr. FR -- three different strings, followed by a space
Fr.-- no space
The use of 'FR ' that I mentioned above happens to be unique, and also the only hit for this search prior to year 1900. --Pwendt|talk 11:22, 26 August 2019 (EDT)
Our database Advanced search evidently strips leading as well as trailing spaces. Search '9 f' will find some uses of closing currency symbol F.
Summary of ISFDB Publications search > Price contains: 9 f --three characters (SEARCH)
F FF Ft --three different strings, all at end of price.
Wikipedia evidently uses trailing currency symbols for denominations of notes and coins; thus, 20F price may be paid by two 10 F coins. --Pwendt|talk 12:02, 26 August 2019 (EDT)

Tauchnitz Edition format

Do we know that the Tauchnitz Collection of British Authors was published sometimes in hc or tp format, as our publication records report?

I infer pb format from the digital copy of #3440 that I viewed at HathiTrust yesterday. (Augmentation of the publication record P567383 is in progress and I will add a publication series Note, too.)

First, I look for confirmation or correction of my interpretation that HDL images 6-7 and images 288-89 of 299 show the original front outside/inside cover and back inside/outside cover, or page 1 to "page 4 of cover" in Tauchnitz terms. Meanwhile images 1-5 and 290-99 show --perhaps with buffer images-- the materials, including hard covers, in which the original was bound by Cornell University Library or donor Theodore Stanton (image 3, with terms of holding and circulation). Right?

Thus I infer paper covers. HathiTrust reports size 16cm, which implies pb format --if not digest or another paper format, but all our publication records for the series reports hc, tp, or pb.

The August 1900 list of Latest Volumes ends (page 4 of cover) with this statement now quoted in the publication record, and destined for the series Note:

"The Tauchnitz Edition is to be had of all Booksellers and Railway Libraries on the Continent, price M 1,60. or 2 francs per volume. A complete Catalogue of the Tauchnitz Edition is attached to this work."

From this I infer a single format, rather than multiple formats, as of August 1900.
(Because Tauchnitz is based in German, I entered price "ℳ 1.60".) --Pwendt|talk 13:44, 23 August 2019 (EDT)

Since paperbound and hardcover formats are verified it seems reasonable to assume that the titles were published - likely simultaneously - in both variants, at least for some time. This was the standard procedure for some German publishers at the beginning of the 20th Century. Stonecreek 01:45, 24 August 2019 (EDT)

Neffy Awards

Should we add the Neffy Awards (or "National Fantasy Fan Federation Speculative Fiction Awards")? According to this page, they have been awarded annually since 2005, and periodically since 1949. I haven't found a list of the older awards yet, but I'll keep looking. The 2019 list is here, and the 2018 list is here. Locus seems to report on them. ···日本穣 · 投稿 · Talk to Nihonjoe 02:13, 25 August 2019 (EDT)

I could have sworn that we discussed adding this award a few years ago, but I can't find anything in the archives. Perhaps it was a similarly named award?
In any event, checking the award history, it looks like it's mostly a media award. However, it also includes enough fiction categories to make it arguably worth adding.
As an aside, their naming conventions can be confusing. For example, in 2010 their "Best SF/F Author" award was given to "Suzanne Collins – Mockingjay". I assume it means that it was actually a "Best SF/F Fiction" category rather than a "Best SF/F Author" category. Oh well, nothing that we can't work our way around. Ahasuerus 14:03, 27 August 2019 (EDT)
OK, a new award type has been created. I'll leave category creation to the folks who will be working on this award. They change from year to year and can get rather convoluted. Ahasuerus 19:42, 9 September 2019 (EDT)

Maithili added

The software has been updated to support the Maithili language. Ahasuerus 16:17, 27 August 2019 (EDT)

Authors with Author Data and One Non-Latin Title

Can we make entries ignorable on this report? There are a number where they just need to be ignored. ···日本穣 · 投稿 · Talk to Nihonjoe 19:32, 29 August 2019 (EDT)

I support this request :) Annie 19:36, 29 August 2019 (EDT)
Especially since a large number of the names in the list are alternate names that are already correctly varianted to the canonical name. ···日本穣 · 投稿 · Talk to Nihonjoe 19:48, 29 August 2019 (EDT)
It would be easy to do, but the fact that it's explicitly designed to find authors with one non-Latin title makes me wonder. Anyone recall the original reasoning behind this cleanup report? Was it, by chance, one of those reports which were created to go after the "low hanging fruit"? Is it still relevant now that we have cleaned up a lot of language issues? Ahasuerus 21:51, 29 August 2019 (EDT)
I do not think that it is relevant or needed anymore - it was part of the reports that were built around the time we were assigning languages to titles and dealing with the non-Latin author names. There is nothing there that we do not get from other reports for the ones where we need to do some work on. If you would rather kill it instead of allowing the ignore, I am all for it. Annie 22:20, 29 August 2019 (EDT)
If there are no objections, I will zap this report in a couple of days. Ahasuerus 12:52, 31 August 2019 (EDT)

Verification of electronic copies

Does a scanned edition of a book count towards verification? Would Transient be used if viewed anpreed Permanent if a copy is possessed? To what extent would the associated catalogue information (assuming a library-like source) be acceptable for completing fields? There are templates for a few sources (e.g. British Library, Gallica), but should there be a more standard approach (template, wording, format) to documenting the reference / link? Bear in mind that some sources require membership or presence at a physical site for viewing. Doug H 08:28, 30 August 2019 (EDT)

National libraries are not only acceptable, but wishable, I'd say (more than a vendor like amazon). They have the same goal in cataloguing the correct data (their main aim isn't to sell), and I use them whereever I can lay my hands on them.
Using a scanned copy (instead of a physical one) should imho always be used for transient verifications. Christian Stonecreek 10:49, 30 August 2019 (EDT)
A number of sites (e.g. Gallica, Hathi) support 3 levels - searchable (tells you number of hits, but shows no text), viewable and downloadable. Is there a distinction between viewed and downloaded (as in "I've seen" vs. "I have") for an electronic copy? Or would they both be lumped in as Transient? A scan isn't really the book, so is it more of a secondary source? Would it make sense to include "Scanned Copy" along with Blieler and Locus? Templates allow me to reference some of the sites / copies directly (e.g. Gallica), but I have made use of some downloads from my university library that gives me access to collections not publicly available. Would a template to incorporate the name/link be a better approach than simply using notes? Doug H 15:17, 30 August 2019 (EDT)
I would still argue that "I have/own a scan of the book in e-format" is different from "I own the book". You can "permanent verify" an e-book you have that is published as an e-book; I book that is marked as HC can be "permanent verified" only from a HC copy. Even if you see the whole book in a scan. So in your case I would Transient and add a note that the Transient is from a downloaded copy that I have. Annie 15:25, 30 August 2019 (EDT)
I would prefer a verification category intermediate to Primary and Secondary --regardless whether images of original covers and endpapers (and more?) are contained or absent; the latter because a re-bound or damaged book has been scanned, or because the scan is incomplete.
And I do prefer, in each publication record with digital copy as source, we specify whether it contains original cover or not (if appropriate: "apparently", "probably", with description, etc).
As yet I have written or updated hundreds of publication records noting some data from online copies, usually at HathiTrust, and I have not formally Verified any of them. My boilerplate, modified to fit the occasion:
* HathiTrust Digital Library provides full view of four copies including one with original cover: [that one is the one linked somewhere below, a hopeful unstated convention]
HDL holdings rarely include spine scans, a lack that I rarely note.
--Pwendt|talk 16:28, 30 August 2019 (EDT)
Annie, I agree they are different. But permanent vs. transient is temporal - do you still have or no longer have access to the "book". Saying a scan you have possession of is transient seems wrong. I'd prefer pure notes to such an approach. The permanent vs. transient seems to apply just as well to the scanned copy. Which is why I wondered if "Scanned Copy" is a valid secondary source - not the real thing but a good substitute? That approach offers only a Yes/No verification though. So a category between Primary and Secondary sounds nice, but probably mucks around too deep in the code to be easily implemented. If such an approach is viewed as 'good' and 'likely', it there a way to 'tag'/'template' entries to minimize the effort of conversion in the future? Pwendt's boilerplate works for him, but a more standardized approach based on some discussion would be better. So I'm hoping this generates some of that discussion. Doug H 16:58, 30 August 2019 (EDT)
That is why I said "I would argue" :) I do not verify based on scans (not because I would not but because I have other stuff to do before that) so I just offered an opinion on what I would do. :) If we agree on a different approach, I am ok with it. I also know that the fact that I have a scan today does not mean I won't misplace it tomorrow (with books at least it won't end up in a disk I throw away by mistake). Neither fits but my gut feeling is that if I had not seen the book in the format it is published, it is not a Primary Permanent for me.
These days more and more books get digitized so maybe we need a "Primary/scanned" type of verification (as Pwendt kinda proposed as well now reading up) ... And “Monitor/Notify” while we are at that (I enter book on behalf of a user who does not speak English, being able to mark it as such will mean that I will get the "changed" AND people know that they can ask me questions about the book because I show as active while the actual PV is around only when I call him to check and verify.
Back to the scans - if enough people think it should be Permanent, I will be fine. You may want to ping Ron (Rtrace) - he has quite a lot of books verified based on scans. Annie 20:01, 30 August 2019 (EDT)

See this archived discussion for a prior discussion that had some implementation ideas on this topic. -- JLaTondre (talk) 08:16, 1 September 2019 (EDT)

Based on that discussion and this, I'd be good with an additional "Scanned copy" form of Secondary Verification. It's pretty generic and non-linking (there may be a FR to deal with that). How would we document the additional information in the Notes? I like using templates, it ensures a standard approach, simplifies searches and stats gathering and enables future enhancement. We only catalogue one publication of a book regardless of the number of copies, so theoretically would need only one verification / scan reference for a given publication. However, scans vary in quality, completeness and other factors, so I wonder if there needs to be allowance for more than one. And different publication sites (Hathi, Google Books, ...) may or may not be showing the same copy, but generally indicate the source copy owner or it is evident from the scan itself, which could help eliminate duplication. But do we document the publisher (and possible link) or owner (source)? Doug H 16:18, 1 September 2019 (EDT)

Margery Allingham

Our friends at SFE3 have updated their "Margery Allingham" entry. Would anyone be interested in leveraging it to flesh out her ISFDB bibliography? Ahasuerus 13:08, 30 August 2019 (EDT)

I can get that this weekend - I like Allingham anyway. I never think of her as one of ours. :) Annie 13:28, 30 August 2019 (EDT)
Thanks! And if I had a penny for every time I was asked "Wait, what? X wrote SF?", I would have a very impressive penny collection :-) Ahasuerus 14:30, 30 August 2019 (EDT)
I've actually read everything they had added over in SFE3. They just did not register as something I would add here or as something that I would even consider speculative (in my brain, they are in my mystery/crime part of my genres). Then I think about them and I realize that they DO fit. On the other hand, I had the same reaction after I read Reply Paid earlier this year -- I had to think on why we would be here (and once I figured it out (and it was obvious once I thought about it), I even added tags. For the first time in my life I think) But them I read extensively in both genres so "that is not from this genre" does not even register sometimes - it fits in one of my genres. I really need to pay more attention to that while reading early mysteries... :) Annie 14:58, 30 August 2019 (EDT)

"They" They 'They'

For several Kipling stories we now have not two but three SHORTFICTION that represent the title fashioned with double, or single, or no quotation marks. Some double-q-mark are canonical 21309 ("They") 1000842 and some no-q-mark are canonical 78266 97925 1000846. (Some single q-mark are parents with only one child.)

Previously I have used the double-q character where a publisher (or library record) uses the single-q, as one convention among several in a class to which it may not officially belong. Without recommending [a] consistent use of any typographical convention thruout the database, or [b] for all English-language titles, we might recommend [1] that either double- or single- but not both be used for any one title.

Regarding the hierarchy, we might recommend [2] that one among double, single, and no-q be the parent for all related titles (eg, SHORTFICTION, CHAPBOOK, ESSAY, INTERIORART, COVERART).

Perhaps I should have specified that the 1904 Kipling short story "They" &c was that year published in a magazine as They; and collected as 'They' (UK) and as "They" (US trade ed.), US subscription ed. not viewed. And published in 1905 chapbooks as 'They' (UK and US); the same US publisher who re-set the collection did not re-set the chapbook. --Pwendt|talk 18:07, 30 August 2019 (EDT)

--Pwendt|talk 17:00, 30 August 2019 (EDT)

'Romance language' added

'Romance language' has been added to the list of supported languages. Please note that it is only supposed to be used to enter titles written using Romance dialects and less well-known languages not supported by the ISO 639-2 standard. Examples include Picard, Lorrain and various Norman dialects. Ahasuerus 18:13, 30 August 2019 (EDT)

Series Names That May Need Disambiguation

Does someone remember what is the point of this report?

It seems to contain all the series names that have at least one partner that had been disambiguated (and which had not been ignored yet). Our policy (formal or not but implemented in practice) is that in such cases one series remains as the main one named so (no disambiguator) and everyone else gets disambiguated. So these series that are on the list will just need ignoring (and will always just need ignoring). Before I go and ignore all 589 of them:

  • Am I missing something? When do we need to do something with those?
  • If I am not missing anything, do we really need this report?

Thanks! Annie 20:52, 30 August 2019 (EDT)

It's been a few years, so I am not 100% sure what I was thinking when I created the reports. I suspect that the idea was that we have a lot more identically named series than we have identically named authors because series names are, quite frequently, commonly used words. For example, we have 10 series whose name happens to be "The Guardians" and 5 series whose name is just plain "Guardian". Moreover, none is an obvious favorite. Compare and contrast with Stephen King vs. Stephen King (I) vs. Stepehen King (artist) where it's clear which author doesn't need to be disambiguated.
Also, I find that having a disambiguating suffix comes in handy when processing new series. When the time comes to enter Book 1 in Jane Doe's "Guardians" series, the approving moderator will see a yellow warning because there will be no plain "Guardians" series on file. Ahasuerus 13:16, 31 August 2019 (EDT)
This actually makes a lot of sense - that is why I am asking before doing something cardinal. I will see if I can get some of those resolved. Any chance to get a matching Pub Series report? These use even less words usually and get repeated across publishers a lot. Annie 01:52, 5 September 2019 (EDT)
Sounds good; FR 1300 "Create a cleanup report to find pub series needing disambiguation" has been created. Ahasuerus 21:23, 5 September 2019 (EDT)
Done -- see this post for details. Ahasuerus 21:59, 4 October 2019 (EDT)

The Warrior's Apprentice Baen edition cover art

Hi, I merged the existing cover art title credited to Alan Gutierrez with the other one credited to Gary Ruddell: the cover artist was identified as Ruddell & the entries for the two printings preceding the 7th thus must have been erroneous. There was no note for those as to where the cover art credit stemmed from. Christian Stonecreek 10:10, 2 September 2019 (EDT)

The assumption is that unless otherwise stated it is from the pub. It would have been better to have checked with the verifiers before doing this. Willem's response to TAWeiss (see this discussion) implies the cover art for his printing was credited as Gutierrez in the publication. It is not unknown for publishers to change the cover and forget to update the credit until a later printing. You can explicitly ask him if you wish to double check, but otherwise, this needs to be reverted and a variant established instead. -- JLaTondre (talk) 10:18, 2 September 2019 (EDT)
I unmerged the titles again, added the right credit and varianted it to the Ruddell entry. Also added pub notes. --Willem 10:45, 2 September 2019 (EDT)

Handling erroneous tags

The way the current version of the ISFDB software ("ISFDB 2.0") works, almost all types of database changes are entered as "submissions" and have to be approved by a moderator. The major exceptions are verifications, votes and title tags. They are all user-specific, which is why Al and I originally thought that we wouldn't need moderator oversight for them. In the case of title tags we were also leveraging other Web sites' (Amazon, Goodreads, etc) experience which suggested that "crowdsourcing" tags worked well even without moderator oversight.

Overall it has worked reasonably well with a few caveats. First, we are primarily a bibliographic database, so tags like "read in 2013" are not something that we normally want to display to the world at large. A few years ago we got around this problem by letting moderators change tag status to "private".

Second, we never reached the volume that is required to make "crowdsourced" tags reliable. The vast majority of the 7% of our title records that have tags have been tagged by just one or two people. An incorrect tag entered by a single user is not too bad when dozens of other users have entered valid tags, but it becomes a problem when there are no valid tags to offset it.

Based on Christian's recent experience with incorrectly tagged titles, I would like to propose that we create a new moderator-only Web page. It will display a list of currently entered user/tag pairs for the selected title record, e.g.:

This will let moderators contact taggers and ask them about questionable tags. The proposed Web page will also let the reviewing moderator remove invalid tags for users who are not moderators.

How does it sound? Ahasuerus 12:29, 3 September 2019 (EDT)

I'm not really clear on tags, having never used them here (either entry or following). If you put this on the moderator's forum, I'd guess you wanted to know if such a tool were useful, but on a public forum I'm thinking you're asking if editors want moderators to perform this function. Doug H 08:18, 4 September 2019 (EDT)
Significant software changes are usually discussed on the Community Portal even if they mostly affect moderators. You never know when a non-moderator may notice a potential pitfall. Ahasuerus 12:48, 4 September 2019 (EDT)
I'm not sure about adding policing of tags to the list of moderator tasks, but as a tool for determining something and correcting it, it beats direct database manipulation. Doug H 08:18, 4 September 2019 (EDT)
It's possible to delete erroneous tags directly, but I should have mentioned that I haven't done it in years. Direct database manipulations are risky, so I rarely attempt them. Ahasuerus 12:52, 4 September 2019 (EDT)
I'd suggest that the deletion of a tag should be 'logged' for the Changed Primary Verifications. Doug H 08:18, 4 September 2019 (EDT)
Keep in mind that title tags are associated with titles, not publications. Title changes (edits, variants, etc) are not captured by the Changed Primary Verifications report at this time. Changes to tags would be even more difficult to link to verified publications because tags do not go through the standard submission approval process. Ahasuerus 12:45, 4 September 2019 (EDT)

(unindent) Seems good to me. In addition to this, two things has always bothered me as far as tags were concerned : 1) the fact that many "private" tags do appear from time to time, drowning the others completely (see here for instance); Linguist 10:43, 4 September 2019 (EDT).

Is the problem with this author the fact that tags like "C1 Nanzan 52" can be viewed by all users? If so, then a moderator can change their status from "Public" (their current status) to "Private", at which point only the tagging user will be able to see them. Ahasuerus 12:40, 4 September 2019 (EDT)
Oh ! I hadn't been aware of that… I now sadly realize that being a moderator doesn't mean being omniscient… ;o( ! Linguist 04:14, 5 September 2019 (EDT).

and 2) the number of quasi-identical tags (e.g. ghost / ghosts, werewolf / werewolves, fish-men / fishmen, etc.) which should be merged (if only it were possible…). Linguist 10:43, 4 September 2019 (EDT).

FR 911, "Allow moderators to edit and merge tags", should help address the issue of merging tags. We'll get to it yet! :-) Ahasuerus 12:42, 4 September 2019 (EDT)

But I suppose we'll just have to grin and bear it… Linguist 10:43, 4 September 2019 (EDT).

Do we really want to enhance tag support? What if we would not have any tags at all? Would that be a big loss? - or would it actually be a win - evil grin -:)? As far as I'm concerned, there's not much benefit in tagging, let alone start moderating them. Another argument against ISFDB tags altogether is the nonexistent support for them from our user base (see the volume - tagging is available, but just not used by our users - exceptions notwithstanding). MagicUnk 11:39, 4 September 2019 (EDT)
I should note that the fact only 7% of our titles (116K out of 1.66M) are tagged is a bit misleading. Certain title types - covers, essays, interior art, etc -- are rarely tagged for various valid reasons and variant titles can't be tagged at all. The 206,000 tags that we have is a fairly significant number since we have 178,000 novels and 464,000 short fiction works. Ahasuerus 13:01, 4 September 2019 (EDT)
How many of those tags are coming from the robots (Fixer and the older ones)? Annie 13:38, 4 September 2019 (EDT)
Well, I am responsible for 70,591 of the 206,000 tags that we have on file. Since most of my tagging is related to Fixer's activities, I guess the answer is "Around one third"?
Also, as I mentioned during the last iteration of this discussion, I find tags to be particularly useful in borderline cases like "magical realism", "surrealism", etc. It's a quick way to indicate why the title is in the database. (Of course, there are times when things are even muddier and I end up adding a note to the record explaining what we know about the title and what our sources are.) Ahasuerus 15:38, 4 September 2019 (EDT)
I was just curious - expected it to be even higher :) I agree that they are useful for these cases - I just do not want to end up in a "this is surrealism, no it is "magical realism" kinda wars. As long as we agree on what is removable and what is better to stay on, I am all for it. And we do need a way to clean after both mistakes and after non-so-nicely done tags (for one reason or another). Annie 15:45, 4 September 2019 (EDT)
True, moderators will need to be careful. Even common words may have different denotations when used by different people. For example, when a romance reader says that a story is a "fantasy", it may be a reference to "erotic fantasy" or to a trip to a billionaire's private tropical island. Back when Fixer was just starting, it took me a while to figure out why he was grabbing so many non-SF romance/erotica books... Ahasuerus 09:21, 5 September 2019 (EDT)
Implementing moderation is not going to change that (au contraire). We should just kill tags outright as our active editors base is just too small to make it viable. I would argue that we need to grow our editor base first and encourage them to actively start tagging before thinking in implementing additional tag functionality. Just my 2cents though MagicUnk 11:39, 4 September 2019 (EDT)
Well, it would help a lot! And tags are a wonderful way for users to become aware of titles they may be interested in, that's the reason they were invented for. Stonecreek 12:03, 6 September 2019 (EDT)
The primary concern here is accuracy. There are certain areas of the database that are relatively rarely used yet we have spent a significant amount of time and effort ensuring that the data that we do have is accurate. For example, less than 1% of our titles is "non-genre", but at one point I spent multiple man-weeks cleaning up the code to make sure that the database model could adequately support non-genre titles. Ahasuerus 13:06, 4 September 2019 (EDT)
Killing something just because noone does it is a bit... premature. I am not a huge fan of tags (mainly because of how I use the DB) but they can be valuable. And if you look at 3 years ago, someone may have proposed to kill the language field as well - it was not really used either. Things change - we need one editor to make it their pet-project and our numbers explode.
On the other hand, what worries me about tags is that they can be subjective. Especially around the borders of the sub-genres. I do not want to see a moderator deleting a bunch of "fantasy" tags because they think the book is something else (and vice versa). It won't be malicious but... most of us do have somewhat strong personalities and things can get heated. :) Merging was already brought up. Annie 13:38, 4 September 2019 (EDT)
I would only delete tags that are obviously erroneous, like the two I found (see over at Ahasuerus' talk page). If the editors are still active, communication about it should be mandatory and maybe it'd be possible to install a mandatory note to be filled in for the reason of deletion. Or, let's have a mandatory note on the community portal for tags to be deleted in case of inactive editors. I don't expect to encounter masses of erroneous tags, but the ones that are in the db still need fixing, I'd say.
I think the tags are one of the advantages of ISFDB: personally I'm more interested in certain authors but am also eager to encounter new ones, and if I find somebody with an interesting theme / tag (that may possibly have gained a high voting), I'll may want to give the text a try. Stonecreek 13:38, 5 September 2019 (EDT)
I was just thinking aloud through implications -- best laid plans tend to end up in disaster some times. I am sure that noone that is here today will go into tag-editing war. A few years down the road when everyone forgets why we can moderate them? Different story. :) Mandatory notes before deletion will help. Keeping a list of deleted tags somewhere so we can restore if needed will also help. But I do agree with you in general that we need the ability to remove tags that are really really wrong. Annie 12:23, 6 September 2019 (EDT)
It would be possible to capture, store and display deleted tags along with moderator notes. However, it would take significantly longer to implement than a simple "display/remove tags" page.
One temporary compromise would be to make this functionality bureaucrat-only. We are probably talking a few dozen tags a year, which shouldn't be too onerous. Ahasuerus 13:41, 6 September 2019 (EDT)
How easy will be to change the access-levels later if we decide to - aka build for bureaucrat-only, switch to moderator-also if the numbers get too big? Annie 13:46, 6 September 2019 (EDT)
It would be a simple change. Ahasuerus 14:01, 6 September 2019 (EDT)
I wonder if changing these from Public to Private won't be a good enough alternative for now? Annie 12:23, 6 September 2019 (EDT)
Unfortunately, erroneous tags can be perfectly valid in other contexts. For example, one of the tags that prompted this discussion was "parallel universe" and we wouldn't want to make it a "private tag". Ahasuerus 13:37, 6 September 2019 (EDT)
Yeah, somehow my mind blanked that part. You are indeed right. Annie 13:46, 6 September 2019 (EDT)
Bureaucrat-only is a good compromise in my opinion. Tags are very useful for anyone interested in a special theme or a certain author, but if you'd be interested in stories that deal with parallel universes and find titles that are tagged erroneously with that label, you are bound to lose some trust. Stonecreek 14:00, 6 September 2019 (EDT)

(unindent) OK, FR 1303, "Allow the removal of erroneous tags", has been created. Ahasuerus 14:50, 19 September 2019 (EDT)

Two author reports

Do we really still need both these reports:

While I had seen entries in just the first one once or twice, it had not been for a while and it seems like a waste to go through the authors twice. Maybe time to see if we can combine them? Annie 01:56, 5 September 2019 (EDT)

Hm, that's a very good question! If I recall correctly, the original reason why they were "separated at birth" was that one of them allowed moderators to "ignore" author records while the other one didn't. However, their logic has been tweaked a few times and neither one supports the "ignore" functionality at this point.
Checking the code behind the reports, I see that one of them is a superset of the other. There are some minor display differences, but I don't see anything that should prevent us from merging them.
FR 1299 "Merge 2 'Invalid Directory Entries' reports" has been created. Thanks! Ahasuerus 13:18, 5 September 2019 (EDT)
Can we keep the format of 189 (with the working language)? :) They did make sense when we had hundreds of non-Latin entries (I think the latter was deployed when we were cleaning the accented characters from the authors' directory names(?) but I may be wrong - the other one is one of the very old ones). Annie 13:32, 5 September 2019 (EDT)
Yup, the current plan is to keep the "Working Language" column. Ahasuerus 13:55, 5 September 2019 (EDT)

(unindent) Done. The remaining report has been changed to display "Working Language" values. Ahasuerus 21:55, 4 October 2019 (EDT)

Wiki page for table of translations

I have set up a wiki page here to help differentiate translations of Cinq semaines en ballon and help editors locate the correct title to add their publication to. I'd like to use it to as a template for similar pages for most of the other (50+) titles by Jules Verne. Before replicating this, I'd like some feedback in general and on a couple of particular questions. To whit:

  1. Should the translated text be in quotes? Many of the other books include dialogue already in quotes. Doug H 10:20, 5 September 2019 (EDT)
I do not see why - it is in its own column, so does not really matter. I would not use quotes if I were you :) Annie 12:47, 5 September 2019 (EDT)
  1. Should the translation follow the paragraph structure of the published form? Some translations (e.g. Mysterious Island) begin with about five paragraphs consisting of three or four words. Doug H 10:20, 5 September 2019 (EDT)
When we know what it is, why not? Makes those translations immediately obvious.Annie 12:47, 5 September 2019 (EDT)
  1. These are currently ordered by the earliest date for the translation, with the exception of the unspecified text. Other options include alphabetically by translation (to make searching for the right variant easier), and by translator (making searching easier where the translator is known - an not 'unknown'). Another option would be to split entries by title and duplicate the translation information and sort by title alone. What order would prove most useful? Doug H 10:20, 5 September 2019 (EDT)
I like the dates order Annie 12:47, 5 September 2019 (EDT)
  1. What should the order of languages be? As an English speaker and this being an English language site, I put it first, then followed alphabetically, which happened to be purely alphabetical in this case. One option is in decreasing order of the number of translations. There's also by the number of translation/title combinations. Or even in order of the number of publications (not visible, but I have the data).Doug H 10:20, 5 September 2019 (EDT)
I like it as it is now - and the menu at the top will allow jumping down. Inside of a table - strictly by date I would say. Annie 12:47, 5 September 2019 (EDT)

Your feedback would be greatly appreciated, as I've gotten too enmeshed to have a wider perspective. Doug H 10:20, 5 September 2019 (EDT)

Some notes above. That kinda illustrates my point from a couple of days ago about editors and per project - once someone starts working on something, we end up collecting a LOT of information. :) One thing though - instead of separate pages, I will keep the different titles per language together. Think of a Russian speaking editor who is trying to sort out the Russian titles - it is a lot easier to work off one page than moving between 50 or so - especially when working on omnibuses. Which begs the question on how to split and my gut feeling is pure alphabetical per language: A-D, E-G and so on (in whatever increments we need). Annie 12:47, 5 September 2019 (EDT)
Curious about where you're going with this, or rather where you think someone is coming from. Is your hypothetical user trying to figure out which French title their Russian title is a translation of? And that the title has never appeared before - otherwise they could do a text search on the Jules Verne Summary Bibliography page - same as I do when I want to find the original French title for something like "The Clipper of the Clouds"? ../Doug H 13:07, 5 September 2019 (EDT)
Length of that page - if we get 50 titles, 100 translations per title, we will need a split somewhere. And the Russian are an easy example because they tend to have as many translations as English does. And I like seeing all titles in the same language on the same page - easy to spot what someone may had forgotten to add. :) Annie 13:11, 5 September 2019 (EDT)
Are you saying the Summary Bibliography page might get too long and need to be split? This wiki page will do nothing for that. Also, isn't your scenario covered by the Preferences - Translations - Selected? Pick Russian and view the page and see all Jules Verne's Summary Bibliography in the French with only Russian translations? While a page per language (or language group - [would it be German or Deutsche?]) makes some tasks easier, verifying its correctness would be really tough - you'd end up having to scan the 50 pages for each language. Or generate a query for periodic reviews. But the same query would resolve the issue of seeing them together. So who makes the query - the maintainer or our Russian editor? And what if I only get the Extraordinary Journey series done, what happens to the other titles, or short stories. I'm convincing myself that keeping it to one wiki page per title with all languages is preferable. How's by you? ../Doug H 14:01, 5 September 2019 (EDT)
The wiki page, not the summary page :) The point of the wiki page is to give you an easy reference, right - that does not require either DB knowledge or opening 20 pages to see the beginning translations? I may be overthinking it. If you want to go title by title, that also works. I just would prefer it to be language based and not title based (but it is not my project). And once we have it, we can always reorganize. So... just thinking aloud :) Annie 14:11, 5 September 2019 (EDT)
Ah, you were referring to your suggested single page for all Jules Verne translations, not my suggestion that the Russian editor could find a title by searching the Summary Bibliography page using a browser search to find which French title corresponded to his Russian one. Different perspectives - my approach is intended for people wanting to add a publication and helping them locate the correct title, based on the actual text, not just the title, or even the translator, given there are sometimes multiple unknowns, each with a different translation. And the only reason for not following paragraph structure was to keep the page from getting too long. :) ../Doug H 14:29, 5 September 2019 (EDT)

Internet Archive "hotlinking"

Perhaps this is a better discussion for rules or moderators but I thought everyone might like to chime in on this.

We currently detail which sites we allow images to be hotlinked from at: Template:Image Host Sites. I was wondering if we could get the Internet Archive added to this list. We already have the Open Library which is one of their projects but I was interesting adding magazine covers like:

I realize in this example Starlog it not considered fully an "in genre" magazine and thus we often do not concern ourselves with providing covers (much-less full contents) but this is just an example.

Notice there is some information about what not to link to here: https://archive.org/services/docs/api/items.html#archival-urls This is why I mentioned:

and not:

which is the target of the redirect (at least for now; presumably they do not want people linking to this as it might change)

It should also be noted this has come up on some of their online discussions before and they have approved it, e.g.:

Do we need to specifically ask them (and is someone volunteering) or can we have them added to the list of possible hotlink hosts we accept?

Thank you, Uzume 17:38, 5 September 2019 (EDT)

The linked discussion and https://archive.org/services/docs/api/items.html#archival-urls suggest that stable Internet Archives look like https://archive.org/download/<identifier>/<filename> . However, the Starlog scover scan URL linked above is https://archive.org/download/starlog_magazine-004/004_jp2.zip/004_jp2/004_0000.jp2?ext=jpg , a different format. Also, it requires a query string, "ext=jpg". Is there any kind of Internet Archive documentation about this format? Ahasuerus 16:41, 9 September 2019 (EDT)
Actually that is the filename/filepath. "004_jp2/004_0000.jp2" is the filepath inside the 004_jp2.zip file. I could not find any documentation on ext query string that seems to convert the file type. We could link to the .jp2 file but I found it did not render in my browser (so was not a good choice as an img src). You can poke around in here to get the idea how I came to the link I provided above: https://archive.org/download/starlog_magazine-004 As a test, I made this submission. Uzume 17:23, 18 September 2019 (EDT)
I am putting this on hold. We still do not have permissions to link but I will leave it so Ahasuerus can see how it looks like - although if you wanted to use it as a test, your moderator note should have added that to the link :) Annie 17:34, 18 September 2019 (EDT)
I thought there were more than adequate notes here so just linked to here in the mod note (and I did not expect it to be approved directly; thanks for holding it). As for a perhaps more realistic example (instead of Starlog which is only partially here under acquisition rules), I was thinking of something like https://archive.org/download/equality00belluoft/equality00belluoft_jp2.zip/equality00belluoft_jp2/equality00belluoft_0001.jp2&ext=jpg for 270798. Uzume 17:59, 18 September 2019 (EDT)

(unindent) I guess we have two overlapping questions here:

  • Are we allowed to deep-link to images hosted by the Internet Archive?
  • Are their images and image URLs stable?

Historically, the main problems with deep-linking have been:

  • bandwidth
  • revenue

In the not so distant some past popular sites like Wikipedia didn't want third party sites to link to "their" images because it increased their hosting costs. This issue appears to be less important now that bandwidth is less expensive.

Revenue is something that we have run into with SFE3: their sponsoring organization wanted us to link to a separate Web page which was set up to make it easier for customers to buy books.

In the case of the Internet Archive, their staff posts suggest that they do not have bandwidth or revenue problems with deep-linking. Uzume's research suggests that the URLs that he uses are at least moderately stable.

Based on the analysis above, I wouldn't be opposed to adding "archive.org" to the list of supported third party sites, although an explicit e-mail permission dated "2019" would be nice to have.

That said, do we expect a significant number of links to their site? If not, it may be easier to upload the images in question to the ISFDB server and link them. Ahasuerus 15:03, 19 September 2019 (EDT)

As a workaround, I have just been pulling Internet Archive images across to their Open Library project and then hotlinking here (since we already allow that). That works alright for books as I just have to ensure the right entry exists there but I am not sure about magazine covers which would likely not be in their Open Library project. Uzume 12:36, 22 September 2019 (EDT)

Not One of Us – magazine or fanzine?

I wonder do we need to make a decision about whether we list the all-fiction/poetry publication Not One of Us as a magazine or a fanzine? From 1986–2001 we have it listed as a fanzine, then from 2002–2018 it is alternately a magazine or a fanzine. Their website refers to it being "a hardcopy zine", with a pull-quote saying it's a "stalwart of the zine scene", yet their Guidelines page has "The editorial philosophy of the magazine...".

We also have a small problem with format, sometimes A5 but mostly octavo, although from the website's description "a digest-sized (5.5 x 8.5 inch, 52-page) publication" it is clearly octavo.

All this makes for inconsistency and I think we could present the information better. The majority of publications are unverified, so I'll spend a bit of time knocking the series into shape if we can come to a group decision on the Magazine/Fanzine question – my own preference would be for Magazine. I'm particularly looking for feedback from verifiers Hkauderer and Kevin Hardy, although I'm uncertain if either of those editors are still around. Thanks. PeteYoung 08:10, 6 September 2019 (EDT)

I would have never thought that it is not a magazine. Just as a data point, ~15 years ago (probably a bit more), Not One of Us was one of the not so many English language small magazines which were available from a distributor website that was willing to ship to Bulgaria - together with some other books from small presses. Everything else on their list of magazines was a clear magazine, this one was in the same section. Does not mean that it could not have been something else but... I probably should find the boxes with all magazines next time I am back at my mother's apartment. Annie 11:23, 6 September 2019 (EDT)

Deprecating {{A}}

In Help:Using Templates and HTML in Note Fields#Linking Templates we list {{A}} as a supported template stating:


[...]
At this time the following linking templates are supported:
Template Functionality Example
A Links to an Author record within ISFDB using the author's name {{A|Jules Verne}}

[...]

However, later in the same page at Help:Using Templates and HTML in Note Fields#Links to Other ISFDB Records we state:


[...]
If you choose to link to another ISFDB record, avoid using the following types of links:
[...]

[...]

Should we deprecate and/or update/replace {{A}}? Thank you, Uzume 19:14, 8 September 2019 (EDT)

"A" links to the search page, not to the author page itself - so if you pass "Test" as a parameter, it goes to http://www.isfdb.org/cgi-bin/se.cgi?arg=Test&type=Name&mode=exact. There is just a forward in the Advanced Search that IF there is only one result, it opens the author page instead of the results page (using the ID and not the name). Try the A template with an author that does not exist :) So A is quite safe. Annie 19:40, 8 September 2019 (EDT)
PS: If you do not believe that the template is tied that way, look at any note with it and hover over the link or view the source code :). It effectively links to the author but not by getting the name and creating a URL out of it. Annie 20:20, 8 September 2019 (EDT)
That's right. The "A" template works fine with non-Latin characters and so do all other templates. Ahasuerus 21:18, 8 September 2019 (EDT)
Oh, I never said it did not work just that it did not link by author id. The link section recommends we should link by the author id and not the author name itself (although we guarantee those are unique). Perhaps we should recommend linking by author id *or* the exact search? Also since we have redirects why not make the author lookups that use a name instead of an author id redirect to the exact search and/or at least redirect to the author pages referenced by author id (for cases that don't have Unicode issues and currently work). Uzume 15:08, 9 September 2019 (EDT)
I mean why not have:
redirect to:
to solve the problem or at least directly to (since "Ray Bradbury" stays within Latin-1) to:
So when people copy the URL it causes them to not want to link to name reference entry (because it gets redirected).
I just am concerned why the difference in the same page about how to reference an author. If it can be solved in code—so be it. Uzume 15:16, 9 September 2019 (EDT)
Let me try this again because I do not think that you read any of the messages (or clicked on any of the links. :)
We do NOT link to http://www.isfdb.org/cgi-bin/ea.cgi?Ray_Bradbury via the template. We link to http://www.isfdb.org/cgi-bin/se.cgi?arg=Ray Bradbury&type=Name&mode=exact. Which searches and then links based on the ID (to http://www.isfdb.org/cgi-bin/ea.cgi?194), not using the name.
Just go to ANY of the notes that have a name and look at where the address is pointing to. The format you are concerned with is not used at all.
Linking to an ID when the ID is not presented will mean that the lookup needs to be done during the save and replaced (as opposed to building a URL using what the user presented). Doable? Yes. But what do we do when there is no user to be linked? Now it sends you to the search page (so if one day there is one, it will connect). And how about cases where the author ID changes - who is going to update all the links? The warning for the "bad URL" is valid but the template does not use it so... it is not relevant. :) Annie 15:40, 9 September 2019 (EDT)
I think you did not read what I am trying to say (perhaps I worded it poorly) and that is there is a discontinuity in how these two situations are handled. I like the way the template links to the exact search (and forwards to the author id URL). I just think that case and the one where an author name (vs. its id) is in the "ea.cgi" link (which is discouraged) should be handled the same way. So why not make that redirect to the exact search URL (or directly to the author id URL since the exact search often ends up there anyway). This would further dissuade others from using the author name in the "ea.cgi" link as the redirect will cause the URL to not stay in the browser's address bar. That is what I meant by solve issues with code vs. my original idea to perhaps have the templates link to the author id (necessitating a change and possibly deprecating {{A}}).
I was never confused how they worked technically (I can directly read the Python and SQL code), I just feel they ought to be handled in a similar manner. Since we are on the topic if the deprecated author name "ea.cgi" (and other author page) linking. It should be possible to log such accesses (and any HTTP referer [sic]) and create something like a nightly report such that others can then investigate and attempt to rectify such inbound links from elsewhere. We might also be able to warn the user using such an inbound link and perhaps they would be able to change the external link source. Uzume 16:42, 10 September 2019 (EDT)
I did read what you said - having the template being mentioned all over the place confused the whole thing a lot though. :) Deprecating the template won't do anything for what you are trying to solve here - because the template is technically the correct way to do these links. :)
So to summarize without all the irrelevant parts and window dressing above: you propose ea.cgi to be modified so that when someone uses a name in the URL, it emulates the template and goes via search with the name (thus going for the ID link at the end). Is that the current state of the idea? Annie 17:24, 10 September 2019 (EDT)
I actually looked into this issue back when the Search software was rewritten and templates were implemented. I ran into some technical issues and ended up abandoning the effort. It may be worth trying again. Ahasuerus 17:46, 10 September 2019 (EDT)
@Annie: Yes, that is exactly what I am saying (and it applies to the other author page formats as well, like awards, alpha, chrono, etc.). It might also be useful to document the ability to link to the exact author search but I am not sure we want to support such inbound URLs. One can already see the issue of author name linkage in ea.cgi: wikipedia:Special:LinkSearch/www.isfdb.org/cgi-bin/ea.cgi (be sure to page down past the author numeric IDs; Mediawiki does not allow more complex external link searches that would let me filter those out). Uzume 17:16, 18 September 2019 (EDT)
Thanks, that's a useful URL to be familiar with. I suspect that officially supporting deep links using our Search URLs would invite problems. We have changed the URL format a few times and may change it again, so URLs are not guaranteed to remain stable. Ahasuerus 15:09, 19 September 2019 (EDT)

If we do decide to work on making redirects for these (e.g., the creation of an FR, etc.), lets not forget this needs to also be extended to series, e.g., currently this is still legal (though methinks it too should be discouraged): The Stainless Steel Rat should redirect to The Stainless Steel Rat I am not sure if there are other similar candidates. Uzume 00:35, 1 October 2019 (EDT)

Award Name Changes

The editor for Analog has announced that the John W. Campbell Award for Best New Writer has been renamed as "The Astounding Award for Best New Writer". Those in charge of two other awards have also been discussing name changes. The Gunn center has announced that the John W. Campbell Memorial Award will be renamed, but the new name has not yet been chosen. The administrators of the James Tiptree, Jr. Award have discussed changing the name and have decided not to do so at this time, though they describe their thinking as "ongoing and tentative" and the name may change in the future. I've been thinking about how we should handle these changes in the database. All three of these awards have several years of history and I would suggest that we keep the existing award pages as is for those years under the old name. Going forward I would suggest a new award page for the new name with links in the notes of each award page linking to the other. That way folks searching for either name would be able to find things. I don't know if we want to set up a new page for the Astounding award now, as it would be blank until next year when the nominations are announced. We will need to wait for the new name for the second Campbell award and for the Tiptree, only if they decide to change the name. Are there any other ideas on how to reflect this award data? --Ron ~ RtraceTalk 22:45, 8 September 2019 (EDT)

I can think of at least one award whose name has changed since it was created, the Ditmar Award. It was originally known as the "Australian Science Fiction Achievement Award". The full name that we currently use is "Ditmar Award / Australian Science Fiction Achievement Award" while the short name is "Ditmar". We could use a similar approach in similar cases, although it may become unwieldy when dealing with longer award names: "The Astounding Award for Best New Writer / John W. Campbell Award for Best New Writer" may be a bit too much.
Alternatively, we could create new award types and document the relationship in Notes. Ahasuerus 23:14, 8 September 2019 (EDT)
If that becomes highly common and complex we might eventually decide to support some sort of award relations not unlike our current pseudonym and title variant systems (I can already see where this would be very useful to support publisher relations of some sort). Uzume 17:04, 18 September 2019 (EDT)
Well... The "variant title" system lets us capture title/author information as it exists in publications as well as its "ideal" or "canonical" state. That's a good thing, but it also adds a great deal of complexity to the software.
It's been occasionally suggested that we use a similar system for other parts of the database, e.g. to capture publisher names both as they appear in publication and in their "canonical" form. It's doable, but it would require a significant amount of work, both on the software side and on the data entry side. We would also need to make sure that our design is very solid because we really wouldn't want to do re-do again. For publishers, it would mean figuring out how to handle imprints and books co-published by multiple publishers, which is liable to add extra levels of complexity. Ahasuerus 17:40, 19 September 2019 (EDT)
I understand such a change is nontrivial but we have made nontrivial changes in the past and I am sure they will happen in the future. Are we ready for this type of change now? I do not know but I thought I should bring it up as it is a possibility that impacts the issues being talked about. As a side note, I got a good laugh out of thinking about "canonical publishers" (isn't that some sort of oxymoron?). Uzume 12:30, 22 September 2019 (EDT)

Final Fantasy

I think I got them all. "The Spirits Within" is a weird one. I've seen a number of places that Dean Wesley Smith wrote the novelization (with cover images and everything), and then other places that say John Vornholt wrote it (also with cover images). Both have the same ISBN, and it hasn't been verified here, so I have no idea which might be the actual author. Any ideas which is actually correct? ···日本穣 · 投稿 · Talk to Nihonjoe 00:35, 10 September 2019 (EDT)

After digging a little bit more, it looks like Vornholt did a juvenile novelization, and Smith did an adult novelization. It would be helpful if the covers indicated this, and if Amazon didn't have the entries merged. ···日本穣 · 投稿 · Talk to Nihonjoe 02:09, 10 September 2019 (EDT)
Well, where will be the fun in that? :) Annie 02:37, 10 September 2019 (EDT)
Novelizations have always been iffy, e.g. see this UK novelization of Capricorn One vs. its US counterpart. However, things have been slowly getting worse over the last 10-20 years. For major intellectual properties, the rights owners routinely commission 2+ different novelizations for different age groups. Factor in the fact that novelizations aimed at younger children tend to obscure their authors and you end up with a big mess. Ahasuerus 10:28, 10 September 2019 (EDT)
That is highly confusing, especially since the Dean Wesley Smith adult novelization was translated to Japanese (instead of creating Japanese novelizations directly from the Japanese movie) and the NDL indicates all three publications are children's literature. Uzume 16:46, 18 September 2019 (EDT)
Wouldn't be the first time people varied on whether something was or wasn't for kids. Until we get PVs for them, we can't really do more than what we have. ···日本穣 · 投稿 · Talk to Nihonjoe 02:10, 19 September 2019 (EDT)

pb, tp and the dropdown

After explaining/clarifying for a third time this week that "pb" does not really mean paperback, it kinda reminded me that we really need to do something about this - or new editors will continue to be confused. So why don't we change the drop down values:

  • pb -> "pb (mass market paperback)" or even "pb (7'x4.25' / 18x11 cm or smaller)"
  • tp -> tp (7'x4.25' / 18x11 cm or larger)"

We do have the space (the audio formats are long) so... why not help the editors visually? Other idea for naming are welcome. Annie 16:13, 15 September 2019 (EDT)

I've always thought of pb as pocket book. Not that it is really any better defined but narrows down the argument by eliminating the trade-sized publications. ../Doug H 22:13, 15 September 2019 (EDT)
While closer, that will include a lot of Eastern European books which are too big to be called pb in the DB - even with generous exceptions (apparently we used to have bigger pockets than the Western world - or something). But still - it will be better than now (and this is only for the front end - the codes behind the scenes are fine) :) Annie 22:57, 15 September 2019 (EDT)
That's not only the case for Eastern Europe but for Western Europe as well (maybe with the exception of some islands in the West): we had this discussion a while back. So this specific dropdown would also be misleading. The best I can think of is an explaining comment that these measures are valid for (most of) the English publishing world, but may vary for other countries. Stonecreek 02:22, 16 September 2019 (EDT)
And put the comment where? :) If people will read the help pages, that is easy enough - we have the info there. They don't. So I am thinking of changing the drop down so it is easier to see that this type is not what they think it is. Annie 02:38, 16 September 2019 (EDT)
I just saw that we have that infobox right near the 'Format' field (when you come on the question mark). This one should be specified, I'd say. Christian Stonecreek 15:10, 16 September 2019 (EDT)
Sure, it is there - but people will still not look at it - or hover on it - hovering from a touch screen is a bit... problematic :) The list looks logical - and if they press "p" to get to "paperback" and see "pb", most people won't look further. Not what we wish people were doing but no reason to just hide our heads in the sand and pretend that it is not happening. The field looks self-explanatory - and most people do not read the help pages for this kind of fields. Annie 15:31, 16 September 2019 (EDT)
Which one would you select for wikipedia:Bunkobon? I usually go with "pb" since it is paperback A6. It would be nice if we had "A6" since we have "A4" and "A5". Uzume 16:52, 18 September 2019 (EDT)
You are mixing up things here - you cannot use A4/A5 for books under the current rules. Look at the help page. We have clear separation between formats for magazines and formats for books - the A4/A5 formats are ONLY for magazines so even if we had A6 in the magazines section it would not have been the correct format for a book. So yes, a paperback book as small as that one is "pb". Does not mean that we do not have a handful books of each of the A4/A5 formats (uncleaned data, magazines converted to anthologies, stuff that was just missed, special cases or 3 and so on) but technically, the magazine formats should not be used for anything besides magazines and fanzines. Until we decide to change the rules anyway.
However the thread above is not about changing the rules - it is just for providing a visual help to editors so let's not go into "let's change the formats" discussions. If you want to reopen that can of worms, start another thread. :) Annie 23:11, 18 September 2019 (EDT)
I was not confused nor soliciting for rule changes as much as just underscoring the state of things as it applies to a certain common publishing format by asking a rhetorical question (hoping to ensure such were considered). My point was to perhaps include "bunko" along with "mass market" in the description; something like:
pb -> "pb (mass market/bunko paperback; 7'x4.25' / 18x11 cm or smaller)". Uzume 12:11, 22 September 2019 (EDT)

(unindent) Getting back to the "Format" drop-down list, it would be possible to make each choice more descriptive along the lines of "tp (7'x4.25' / 18x11 cm or larger)".

However, keep in mind that we already have a few different ways of conveying this information to our users. On the publication display page we have mouse-over help for each format code. For example, if you hover your mouse over the question mark next to "pb" on this page, you will see "Paperback. Typically 7" by 4.25" (18 cm by 11 cm) or smaller, though trimming errors can cause them to sometimes be slightly (less than 1/4 extra inch) taller or wider/deeper."

When editing a publication record, the blue bubble next to the word "Format" says, among other things, "pb for 7" by 4.25" (18 cm by 11 cm) paperbacks".

In other words, we already display this information in 2 different places, although the wording is somewhat different. Displaying it yet again doesn't seem like the best way to handle the problem. Perhaps we could use the empty space to the right of the drop-down list to display a message along the lines of "Hover your mouse over 'Format' on the left for more details"? Ahasuerus 15:59, 20 September 2019 (EDT)

And my point is that people do not hover over things or read help pages for things they believe to be intuitive and clear - we can have it in 10 places, put a "hover for details" in 10 more and people still won't do it because they do not think they need details because "pb" sounds like the logical choice. People see "pb" and are done with reading - we may have it in 2 places but people just do not look there. Annie 17:00, 20 September 2019 (EDT)
I am pretty sure there is a really good British joke about how much people hover over things in here. Uzume 12:14, 22 September 2019 (EDT)
The problem is actually only the fact that in the help pb and tp is defined with the length and width and that is apparently wrong (Google translator).--Wolfram.winkler 04:37, 23 September 2019 (EDT)

Custom database query request

Earlier today I received the following e-mail request from the SF reviewer James Nicoll:

Is there a way to construct a search of isfdb so that it returns a list of the Best Novel Hugo for each year, with a list of the cover artists for the American editions of the stories in the first year of publication? (basically, looking for patterns in cover art and nominations)

The short answer is that yes, it should be possible, but it would require a custom database query. I haven't been feeling well lately, so I am not in a good position to create one. Would anyone be interested in giving it a try? Ahasuerus 14:20, 16 September 2019 (EDT)

See the following table. There are some caveats:
  • It is using the title id the award is linked to. It could be expanded to look for variants / parents in case also appeared under different form in that time frame.
  • It does not include serialization. Some of the early ones were in magazines. It could be expanded to include those, but the cover would not necessarily be for the story and the artist would likely change over the serialization.
  • It bases it being an American edition of the price being US $. It therefore ignores pubs w/o a price.
  • It does not include the Retro Hugo, but that could be included.
If any tweaks are desired to the above caveats, I will see what I can do. -- JLaTondre (talk) 21:16, 16 September 2019 (EDT)
Award Year Author(s) Title Pub Year Publisher Price Artist(s) Image
1959 James Blish A Case of Conscience 1958 Ballantine Books $0.35 Richard Powers Image
1960 Robert A. Heinlein Starship Troopers 1959 G. P. Putnam's Sons $3.95 Jerry Robinson Image
1961 Walter M. Miller, Jr. A Canticle for Leibowitz 1959 J. B. Lippincott $4.95 Milton Glaser Image
1961 Walter M. Miller, Jr. A Canticle for Leibowitz 1960 J. B. Lippincott $4.95 George Sottung Image
1962 Robert A. Heinlein Stranger in a Strange Land 1961 G. P. Putnam's Sons $4.50 Ben Feder Image
1962 Robert A. Heinlein Stranger in a Strange Land 1961 G. P. Putnam's Sons / SFBC $1.70 Ben Feder Image
1963 Philip K. Dick The Man in the High Castle 1962 G. P. Putnam's Sons $3.95 Robert Galster Image
1963 Philip K. Dick The Man in the High Castle 1962 G. P. Putnam's Sons / SFBC $1.20 Robert Galster Image
1964 Clifford D. Simak Way Station^Here Gather the Stars 1963 Doubleday $3.50 Ronald Fratell Image
1964 Clifford D. Simak Way Station^Here Gather the Stars 1963 Doubleday / SFBC $1.20 Ronald Fratell Image
1965 Fritz Leiber The Wanderer 1964 Ballantine Books $0.75 Bob Abbett Image
1966 Frank Herbert Dune 1965 Chilton $5.95 John Schoenherr Image
1967 Robert A. Heinlein The Moon is a Harsh Mistress 1966 G. P. Putnam's Sons $5.95 Irv Docktor Image
1968 Roger Zelazny Lord of Light 1967 Doubleday $4.95 Howard Bernstein Image
1969 John Brunner Stand on Zanzibar 1968 Doubleday $6.95 S. A. Summit, Inc. Image
1970 Ursula K. Le Guin The Left Hand of Darkness 1969 Ace Books $0.95 Diane Dillon + Leo Dillon Image
1970 Ursula K. Le Guin The Left Hand of Darkness 1969 Walker & Co. $4.95 Jack Gaughan Image
1970 Ursula K. Le Guin The Left Hand of Darkness 1969 Walker & Co. / SFBC $1.98 Jack Gaughan Image
1971 Larry Niven Ringworld 1970 Ballantine Books $0.95 Dean Ellis Image
1972 Philip José Farmer To Your Scattered Bodies Go 1971 G. P. Putnam's Sons $4.95 Ira Cohen Image
1972 Philip José Farmer To Your Scattered Bodies Go 1971 Berkley Medallion $0.75 Richard Powers Image
1973 Isaac Asimov The Gods Themselves 1972 Doubleday $5.95 David November Image
1973 Isaac Asimov The Gods Themselves 1972 Doubleday & Company / SFBC $1.98 David November Image
1974 Arthur C. Clarke Rendezvous With Rama 1973 Harcourt Brace Jovanovich $6.95 Hal Siegel Image
1974 Arthur C. Clarke Rendezvous With Rama 1973 Harcourt Brace Jovanovich / SFBC $1.49 Hal Siegel Image
1975 Ursula K. Le Guin The Dispossessed 1974 Harper & Row $7.95 Fred Winkowski Image
1975 Ursula K. Le Guin The Dispossessed 1974 Harper & Row / SFBC $2.49 Fred Winkowski Image
1976 Joe Haldeman The Forever War 1975 St. Martin's Press $7.95 UNKNOWN Image
1977 Kate Wilhelm Where Late the Sweet Birds Sang 1976 Harper & Row $7.95 M. C. Escher Image
1977 Kate Wilhelm Where Late the Sweet Birds Sang 1976 Harper & Row / SFBC $1.98 M. C. Escher Image
1978 Frederik Pohl Gateway 1977 St. Martin's Press $8.95 Boris Vallejo Image
1978 Frederik Pohl Gateway 1977 St. Martin's Press / SFBC $2.49 Boris Vallejo Image
1979 Vonda N. McIntyre Dreamsnake 1978 Houghton Mifflin $8.95 Stephen Alexander Image
1979 Vonda N. McIntyre Dreamsnake 1978 Houghton Mifflin / SFBC $2.98 Stephen Alexander Image
1980 Arthur C. Clarke The Fountains of Paradise 1979 Harcourt Brace Jovanovich $10.00 Paul Bacon Image
1980 Arthur C. Clarke The Fountains of Paradise 1979 Harcourt Brace Jovanovich / SFBC $3.50 Paul Bacon Image
1981 Joan D. Vinge The Snow Queen 1980 The Dial Press $10.95 Leo Dillon + Diane Dillon Image
1981 Joan D. Vinge The Snow Queen 1980 The Dial Press $10.95 Leo Dillon + Diane Dillon Image
1981 Joan D. Vinge The Snow Queen 1980 The Dial Press / SFBC $4.50 Leo Dillon + Diane Dillon Image
1982 C. J. Cherryh Downbelow Station 1981 DAW Books $2.75 David B. Mattingly Image
1982 C. J. Cherryh Downbelow Station 1981 DAW Books / SFBC $4.50 Rego Image
1983 Isaac Asimov Foundation's Edge 1982 Doubleday & Company $14.95 Joe Caroff Image
1983 Isaac Asimov Foundation's Edge 1982 Whispers Press $50.00 UNKNOWN Image
1983 Isaac Asimov Foundation's Edge 1982 Doubleday $14.95 Joe Caroff Image
1983 Isaac Asimov Foundation's Edge 1982 Doubleday $14.95 Joe Caroff Image
1984 David Brin Startide Rising 1983 Bantam Books $3.50 Jim Burns Image
1985 William Gibson Neuromancer 1984 Ace Books $2.95 James Warhola Image
1985 William Gibson Neuromancer 1984 Ace Books $2.95 James Warhola None
1986 Orson Scott Card Ender's Game 1985 Tor $13.95 John Harris Image
1987 Orson Scott Card Speaker for the Dead 1986 Tor $15.95 John Harris Image
1987 Orson Scott Card Speaker for the Dead 1986 Nelson Doubleday / SFBC $7.98 Vincent Di Fate Image
1988 David Brin The Uplift War 1987 Phantasia Press $22.00 Wayne D. Barlowe Image
1988 David Brin The Uplift War 1987 Phantasia Press $60.00 Wayne D. Barlowe Image
1988 David Brin The Uplift War 1987 Bantam Spectra $4.50 Michael Whelan Image
1988 David Brin The Uplift War 1987 Nelson Doubleday / SFBC $9.98 Richard Powers Image
1989 C. J. Cherryh Cyteen 1988 Warner Books $18.95 Don Maitz Image
1989 C. J. Cherryh Cyteen 1988 Warner Books / SFBC $9.98 Don Maitz Image
1990 Dan Simmons Hyperion 1989 Doubleday Foundation $18.95 Gary Ruddell Image
1990 Dan Simmons Hyperion 1989 Doubleday Foundation $8.95 Gary Ruddell Image
1991 Lois McMaster Bujold The Vor Game 1990 Baen Books $4.50 Tom Kidd Image
1991 Lois McMaster Bujold The Vor Game 1990 The Easton Press $45.00 UNKNOWN Image
1991 Lois McMaster Bujold The Vor Game 1990 GuildAmerica Books / SFBC $9.98 Dean Morrissey Image
1992 Lois McMaster Bujold Barrayar 1991 Baen Books $4.99 Stephen Hickman Image
1993 Vernor Vinge A Fire Upon the Deep 1992 Tor $21.95 Boris Vallejo Image
1993 Vernor Vinge A Fire Upon the Deep 1992 Tor / SFBC $10.98 Boris Vallejo Image
1993 Connie Willis Doomsday Book 1992 Bantam Spectra $22.00 Tim Jacobus Image
1993 Connie Willis Doomsday Book 1992 Bantam Spectra $10.00 Tim Jacobus Image
1993 Connie Willis Doomsday Book 1992 Bantam Spectra / SFBC $12.98 Tim Jacobus Image
1995 Lois McMaster Bujold Mirror Dance 1994 Baen $21.00 Gary Ruddell Image
1995 Lois McMaster Bujold Mirror Dance 1994 Baen / SFBC $9.98 Gary Ruddell Image
1996 Neal Stephenson The Diamond Age 1995 Bantam Spectra $22.95 Bruce Jensen Image
1996 Neal Stephenson The Diamond Age 1995 Bantam Spectra / SFBC $8.98 Bruce Jensen Image
1997 Kim Stanley Robinson Blue Mars 1996 Bantam Spectra $22.95 Don Dixon Image
1997 Kim Stanley Robinson Blue Mars 1996 Bantam Spectra / SFBC $10.98 Don Dixon Image
1998 Joe Haldeman Forever Peace 1997 Ace Books $21.95 Bruce Jensen Image
1999 Connie Willis To Say Nothing of the Dog 1998 Bantam Spectra $23.95 Eric Dinyer Image
1999 Connie Willis To Say Nothing of the Dog 1998 Bantam Spectra / SFBC $11.98 Eric Dinyer Image
1999 Connie Willis To Say Nothing of the Dog 1998 Bantam Books $6.50 Eric Dinyer Image
2000 Vernor Vinge A Deepness in the Sky 1999 Tor $27.95 Bob Eggleton Image
2000 Vernor Vinge A Deepness in the Sky 1999 Tor / SFBC $13.98 Bob Eggleton Image
2001 J. K. Rowling Harry Potter and the Goblet of Fire 2000 Arthur A. Levine Books $25.95 Mary GrandPré Image
2001 J. K. Rowling Harry Potter and the Goblet of Fire 2000 Scholastic $25.95 Mary GrandPré Image
2001 J. K. Rowling Harry Potter and the Goblet of Fire 2000 Scholastic / SFBC $12.98 Mary GrandPré Image
2002 Neil Gaiman American Gods 2001 William Morrow / HarperCollins $26.00 Kamil Vojnar Image
2002 Neil Gaiman American Gods 2001 William Morrow / HarperCollins / SFBC $12.98 Kamil Vojnar None
2003 Robert J. Sawyer Hominids 2002 Tor $25.95 Donato Image
2004 Lois McMaster Bujold Paladin of Souls 2003 PerfectBound / HarperCollins $11.99 UNKNOWN None
2004 Lois McMaster Bujold Paladin of Souls 2003 Eos / HarperCollins $24.95 David Bowers Image
2004 Lois McMaster Bujold Paladin of Souls 2003 Eos / HarperCollins / SFBC $12.99 David Bowers Image
2005 Susanna Clarke Jonathan Strange & Mr. Norrell 2004 Bloomsbury USA $27.95 Portia Rosenberg + William Webb Image
2005 Susanna Clarke Jonathan Strange & Mr. Norrell 2004 Bloomsbury USA $27.95 Portia Rosenberg + William Webb Image
2005 Susanna Clarke Jonathan Strange & Mr. Norrell 2004 Bloomsbury USA / SFBC $14.99 UNKNOWN None
2005 Susanna Clarke Jonathan Strange & Mr. Norrell 2004 Bloomsbury USA $27.95 Portia Rosenberg + William Webb Image
2006 Robert Charles Wilson Spin 2005 Tor $25.95 Drive Communications Image
2006 Robert Charles Wilson Spin 2005 Tor / SFBC $12.99 Drive Communications Image
2007 Vernor Vinge Rainbows End 2006 Tor $25.95 Stephan Martiniere Image
2007 Vernor Vinge Rainbows End 2006 Tor / SFBC $13.99 Stephan Martiniere Image
2008 Michael Chabon The Yiddish Policemen's Union 2007 HarperCollins $26.95 Will Staehle Image
2008 Michael Chabon The Yiddish Policemen's Union 2007 HarperCollins $26.95 Will Staehle Image
2008 Michael Chabon The Yiddish Policemen's Union 2007 HarperAudio $39.95 Will Staehle Image
2008 Michael Chabon The Yiddish Policemen's Union 2007 HarperLuxe $26.95 Will Staehle Image
2008 Michael Chabon The Yiddish Policemen's Union 2007 HarperCollins / SFBC $14.99 Will Staehle Image
2009 Neil Gaiman The Graveyard Book 2008 HarperAudio $18.99 Dave McKean Image
2009 Neil Gaiman The Graveyard Book 2008 HarperCollins $17.99 Dave McKean Image
2009 Neil Gaiman The Graveyard Book 2008 Harper Children's Audio $22.95 Dave McKean Image
2009 Neil Gaiman The Graveyard Book 2008 HarperCollins $18.89 Dave McKean Image
2009 Neil Gaiman The Graveyard Book 2008 HarperCollins / SFBC $11.99 Dave McKean Image
2010 China Miéville The City & The City 2009 Del Rey / Ballantine $26.00 FWIS Image
2010 China Miéville The City & The City 2009 Del Rey / Ballantine $11.99 UNKNOWN Image
2010 China Miéville The City & The City 2009 Del Rey / Ballantine $26.00 FWIS Image
2010 China Miéville The City & The City 2009 Del Rey / Ballantine / SFBC $14.99 FWIS Image
2010 China Miéville The City & The City 2009 Subterranean Press $75.00 Vincent Chong Image
2010 Paolo Bacigalupi The Windup Girl 2009 Night Shade Books $24.95 Raphael Lacoste Image
2010 Paolo Bacigalupi The Windup Girl 2009 Night Shade Books $6.00 Raphael Lacoste None
2010 Paolo Bacigalupi The Windup Girl 2009 Night Shade Books $15.99 UNKNOWN Image
2011 Connie Willis Blackout 2010 Spectra / Ballantine Books $11.99 UNKNOWN Image
2011 Connie Willis Blackout 2010 Spectra / Ballantine Books $26.00 Charles Brock Image
2011 Connie Willis Blackout 2010 Spectra / Ballantine Books / SFBC $14.99 Charles Brock Image
2011 Connie Willis Blackout 2010 Subterranean Press $75.00 J. K. Potter Image
2011 Connie Willis Blackout 2010 Spectra / Ballantine Books $26.00 Charles Brock Image
2011 Connie Willis Blackout 2010 Spectra / Ballantine Books $16.00 UNKNOWN Image
2011 Connie Willis Blackout 2010 Brilliance Audio $39.99 UNKNOWN Image
2011 Connie Willis Blackout 2010 Brilliance Audio $29.99 UNKNOWN Image
2011 Connie Willis All Clear 2010 Spectra / Ballantine Books $26.00 Charles Brock Image
2011 Connie Willis All Clear 2010 Spectra / Ballantine Books / SFBC $15.99 Charles Brock Image
2011 Connie Willis All Clear 2010 Spectra / Ballantine Books $11.99 UNKNOWN Image
2011 Connie Willis All Clear 2010 Brilliance Audio $39.99 UNKNOWN Image
2012 Jo Walton Among Others 2011 Tor $24.99 Kamil Vojnar Image
2012 Jo Walton Among Others 2011 Tor $9.99 Kamil Vojnar Image
2013 John Scalzi Redshirts 2012 Tor $24.99 Peter Lutjen Image
2013 John Scalzi Redshirts 2012 Tor / SFBC $14.99 Peter Lutjen Image
2013 John Scalzi Redshirts 2012 Tor $9.99 Peter Lutjen Image
2014 Ann Leckie Ancillary Justice 2013 Orbit (US) $15.00 John Harris Image
2014 Ann Leckie Ancillary Justice 2013 Orbit (US) $9.99 UNKNOWN Image
2015 Cixin Liu The Three-Body Problem 2014 Tor $25.99 Stephan Martiniere Image
2015 Cixin Liu The Three-Body Problem 2014 Tor / SFBC $26.99 Stephen Martiniere Image
2016 N. K. Jemisin The Fifth Season 2015 Orbit (US) $15.99 Lauren Panepinto Image
2016 N. K. Jemisin The Fifth Season 2015 Orbit (US) $9.99 UNKNOWN Image
2017 N. K. Jemisin The Obelisk Gate 2016 Orbit (US) $15.99 Lauren Panepinto + Arcangel Image
2017 N. K. Jemisin The Obelisk Gate 2016 Orbit (US) $9.99 UNKNOWN Image
2018 N. K. Jemisin The Stone Sky 2017 Orbit (US) $16.99 Lauren Panepinto + Arcangel Images Image
2018 N. K. Jemisin The Stone Sky 2017 Orbit (US) $11.99 Lauren Panepinto + Arcangel Images Image
2018 N. K. Jemisin The Stone Sky 2017 Orbit (US) $16.99 Lauren Panepinto + Arcangel Images Image
2019 Mary Robinette Kowal The Calculating Stars 2018 Tor $18.99 Jamie Stafford-Hill Image
2019 Mary Robinette Kowal The Calculating Stars 2018 Tor $9.99 UNKNOWN Image
Thanks, I'll let James know! Ahasuerus 22:25, 16 September 2019 (EDT)

Sciencewood - non-genre non-fiction - a second set of eyes needed

Does anyone see a reason to keep this one in the DB? It looks like a popular science book, no speculative element and not about speculative content -- which is out of scope if you are not above threshold. Or is there something I am missing about the author or the book? Thanks to anyone that would lend an eye (or 2) to that book and help determine what to do with it. Annie 18:00, 17 September 2019 (EDT)

Not seeing anything that would indicate it belongs here. The author image is from Amazon, so we won't need to remove it if the book is deleted (since this is the only entry for the person). ···日本穣 · 投稿 · Talk to Nihonjoe 18:36, 17 September 2019 (EDT)
And it had been nuked from space - both pub and title. Even if we had to delete the image manually, it is not that hard either. Thanks for verifying for me. :) For anyone interested who lands here after the deletion, this is the book in question. Annie 18:51, 17 September 2019 (EDT)

Curious records in awards table for Tiptree Award

The records in the awards table for the Tiptree have (at least) 8 "meta" records that seem to be purely for display purposes, and aren't actually works that were nominated for the award. This SQL query will show them (at least the ones I'm aware of):

   select * from awards where award_title  like '%Honor List%' or award_title like '%Long List%';

Alternatively, you can see them at these URLs:

I've not noticed these for any other award, and they only seem to exist for 2009, 2011, 2013, 2014 and 2015 in the Tiptree. (I noticed them a while ago, but it's only now that I need to deal with them for a project I'm working on that uses a local copy of the database.)

Are these a legitimate/expected way of using ISFDB? If not, could/should they be deleted?

As far as I can tell they have use award_level values - as opposed to some special value that might indicate they are meta-records - which were chosen so that they appear in a particular place on the award pages (such as the first of the three links above). However, they seem somewhat redundant functionally, as the "proper" Finalists and Honorable Mentions headings also appear on that page. Perhaps someone felt those terms were inaccurate, and created these other records to try to address that? —The preceding unsigned comment was added by ErsatzCulture (talkcontribs) .

I am not sure I am getting the question - so let me try to clarify :)
The Honorable list is an official list and we do record Long lists and so on when an award announces them - so their availability depends on the award itself. So what exactly is bothering you? Annie 19:08, 18 September 2019 (EDT)
If you look at the 45437 & 45438 links provided above, you will see that the "--- Honor List ---" and the "--- Long List ---" headings on the year page have been implemented by creating fake award records. It's a rather unorthodox attempt to add the award's actual names for these subcategories instead of the ISFDB's standard award levels. -- JLaTondre (talk) 19:18, 18 September 2019 (EDT)
(JLaTondre has basically given my response - much more succinctly - but here's what I was about to submit before getting into an edit conflict):
My question isn't with the awarded works being in the database, but that there are additional records - the ones with these headings - that have also been added, which don't exist for any other awards, or indeed for most years for the Tiptree. This makes writing code that queries the database a bit of a hassle - not impossible to deal with, but more than I'd like, especially if the offending records shouldn't have been there to start with.
Let me give an example. If I wanted to know how many finalists there were in 2018 - a year which isn't affected by this issue - I would do this database query:
   select count(1) from awards where award_type_id = 43 and award_cat_id = 523 and year(award_year) = 2018 and award_level = 90;
and get back the answer 11, which is correct. (Just for clarify: 43 and 523 are the IDs for the Tiptree and its main category respectively; 90 denotes finalist.)
If I switch the year value to 2015, and run the query, I get back 12. That's one too many, because "--- Honor List ---" is counted as one of the finalists. To get the right answer, I would have to add some logic to filter these out, but that requires foreknowledge of what these "meta-values" are, and for all I know, next year someone could enter them with different text, or as italic, and then my code will be wrong again.
Don't get me wrong - I'm certainly not asking or expecting ISFDB to change functionality just to suit my personal project, but my suspicion is that perhaps these "meta-records" weren't something that was expected when the award functionality was designed or implemented.
Can I ask a question in turn - is there any particular reason you know of why 2009, 2011, 2013, 2014 and 2015 might have these records, but no other years of the Tiptree? —The preceding unsigned comment was added by ErsatzCulture (talkcontribs) . 16:30, 18 September 2019 (EDT)
It looks like someone was wanting it to say "Honor List" and "Long List" instead of the system-generated "Finalists" and "Honorable Mentions" (respectively), even though these last two are still displayed. So, pretty much what JLaTondre wrote. I think these should be removed. ···日本穣 · 投稿 · Talk to Nihonjoe 19:35, 18 September 2019 (EDT)
(after resolving conflict) Ah, now that you pointed it out, I see it - I was loking at the lists themselves and not getting it. :) Sorry - had a slow moment here :) Isn't that similar to a question we had earlier this month about special names and what's not (wasn't it one of the Japanese awards?) - except someone went and did them that way instead of asking how? I would say that they should be removed and we need to discuss that "custom naming" thing for awards and levels.
As for why only some years - because noone added them? :) Different people add different awards at different times so... who knows - we keep adding missing information all over the place. Annie 19:37, 18 September 2019 (EDT)
I've deleted the bogus records. ···日本穣 · 投稿 · Talk to Nihonjoe 19:41, 18 September 2019 (EDT)
Thanks. —The preceding unsigned comment was added by ErsatzCulture (talkcontribs) . 16:47, 18 September 2019 (EDT)

eBook types

Hello all. Recently I re-visited this ebook pub record, to which an ASIN was added well after I created the record. Since this ASIN is for another ebook format (MOBI/AZW) compared to the original (EPUB), it started me thinking if we need/want to discriminate between different ebook formats, much like we have hc, tp, pb,... format types for paper books? Searching the internet a bit reveals that this particular pub, when bought from bol.com, is in EPUB format, whereas if it is bought from Amazon it will likely be in AZW3 (MOBI for older books I believe - not 100% sure on this one) format. So, since these are different formats, and since they generally are DRM protected and therefore not readily convertible into one another (at least not officially, that is...), I'd think it makes sense to add ebook formats to the dropdown list. Could be something like

  • ebook - EPUB
  • ebook - MOBI
  • ebook - AZW (all versions)
  • ebook - IBA (this one's Apple's, I believe)
  • ebook - other

as these are the most popular ones.
Thoughts ?
BTW, nice summary here: https://www.makeuseof.com/tag/ebook-formats-explained/
MagicUnk 09:04, 19 September 2019 (EDT)

A lot of books are available in multiple formats at the same time -- do we really want to add multiple copies to the DB? And unlike the physical books, conversions are possible very easily. I would much prefer we keep one type and either use Notes or add a new text field for format details... Just thinking aloud. Annie 10:33, 19 September 2019 (EDT)
I guess that if it would be so that electronic versions could be easily and without any restrictions and at all times be converted into each other (or be able to be read by whatever reading app/device), we could argue that the issue is moot and that for all practical intents and purposes the ebook formats can be effectively considered the 'same' (for as far as we're talking about the exact same contents of course). However, since this is not the case (not all formats can (should legally not) be converted, nor all ereaders can read all formats), I am inclined to think we should be able to make a distinction - preferably without having to clutter the notes, but if not, then the notes will have to do :).
Now, as for the how to distinguish, I see a couple of options. Adding more choices to the dropdown as listed above is just one possibility, and would be in line with what we already have for paper and even for audio formats too, albeit that these are physical formats, as opposed to electronic ones for ebooks. Adding an additional field to select the ebook (sub-)type is another, and even adding multiple additional types to a single ebook pub record (much like the possiblity to add multiple external ID's with the '+' button, or what we intend for the eagerly awaited multiple price enhancement - hint-hint :), could be a third. MagicUnk 11:23, 19 September 2019 (EDT)
I think we should handle this at the publication level as we always have. A publication is a salable product (and can and often is assigned an ISBN, UPC or other product code). Does one have to buy the item again to get another format? Then it likely is a separate publication. If not, then it is more like another part of the same product/publication (and we do not explicitly track that though a note in the pub record stating which formats a publication supports might be nice). Uzume 11:51, 22 September 2019 (EDT)
Well, that's the thing, isn't it? You cannot normally convert/are not allowed to convert between EPUB and MOBI, so you'd have to buy a new version, hence end up with two pub records... MagicUnk 18:42, 23 September 2019 (EDT)
It is quite easy to convert between the two. There are sellers who also provide multiple formats for a single purchase. This suggestions adds a lot of complexity & work for no gain. -- JLaTondre (talk) 18:50, 23 September 2019 (EDT)
I am not aware that it is legal to do so in all cases, or is it?MagicUnk 01:57, 24 September 2019 (EDT)
As long as one does not sell the converted copy I believe this is legal under fair use. Uzume 00:25, 1 October 2019 (EDT)

Multiformat magazines and series grids

We have two separate ways to record magazines that publish their issues in more than one format (print + ebook for example or webzine + print + ebook).

  • Two separate series - 1 per format. This becomes a problem if one of the format is discontinued - someone needs to know to look for the other series. And if one of the two formats does not exist from the beginning, it looks incomplete. Plus the series are almost never kept in parallel (example and its digital brother.
  • One series, all formats merged into a single early record. This can make the grid look heavy (and repetitious) but gives the full picture of the magazine. example.

Which one is used depends on the editor who sets the series basically - so we have both all over the place (and sometimes the wrong formats end up in the wrong place when someone does not pay attention). It is probably a good idea to at least try to decide what the standard should be. :)

I prefer the second way (we can always add some filters on the grid page if they start looking heavy) - especially now that we show the non-common formats so it will be clear why we have multiples. What does everyone think? Annie 23:14, 19 September 2019 (EDT)

Magazine handling is a little strange. It would be easier from a data management/maintenance perspective if we put it all into one. I wonder if we could somehow use a combination of pub series and title series to help organize. --MartyD 07:39, 20 September 2019 (EDT)
I believe we still support super series, e.g. you can see Astounding / Analog (1937-1971) and Astounding/Analog (UK), however there is also Astounding/Analog which merges the two (albeit in this case with some other stuff but methinks you get the idea). Also in this case the UK reprints are not exactly exposing a different format so much but a super series could be made for both of Galaxy's Edge and Galaxy's Edge (digital edition) and then one would be able to view a merged issue grid if one wanted to. For something like Journey Planet, one would have to split the editorial records and make another series and then super series the both. Uzume 11:23, 22 September 2019 (EDT)
Since Marty brought it up again, I always thought magazines should be defined/handled as pub series instead of the hackish way we do it now using their editorial records (which I have never suggested should go away; for some reason whenever I mention magazines as pub series that is the first thing people seem to think I am trying to get rid of). I realize we have many such records and it would be a large effort to create pub series for all the magazines/fanzines, etc. but it does solve the complexity of our current issue grid (and having an issue grid view for other non-magazine pub series might be useful too; I wonder if we have an FR on that) as well as allowing us to merge editorial records for different formats/reprints/etc. but still allow one to record and present them each separately. I believe our software does not currently support super pub series (though that might also be a useful addition) Uzume 11:23, 22 September 2019 (EDT)

Biographical warnings for audio books

I think that awhile back we suspended showing bibliographic warnings for missing ISBN/Catalog ID when the format of the book is ebook. Can we extend that to the "digital audio download" type as well? Example where we have both types and just one warning: Yarn. Or is there a reason to always have it (because these will (almost) never have ISBNs and catalog numbers)? Thanks! Annie 04:02, 21 September 2019 (EDT)

I don't think it was a conscious decision. We (or at least I) didn't realize that "digital audio download" books typically didn't have ISBNs. Ahasuerus 09:12, 21 September 2019 (EDT)
I think it was a "let's try to clean this thing" kind of thing and as we did not have that many audio books, noone thought of it.:)
How about prices and webzines? The warning is there even though the vast majority of webzines do not have prices unless we really like typing 0.00. And since we opened the doors for them, we have more than we used to. Annie 11:51, 21 September 2019 (EDT)
FR 1306, "Suppress bibliographic warnings for digital audio downloads", has been created. Ahasuerus 10:55, 3 October 2019 (EDT)
FR 1306 has been implemented. Ahasuerus 15:36, 4 October 2019 (EDT)
Any objections to suppressing bibliographic warnings for webzine issues without a price? Ahasuerus 10:55, 3 October 2019 (EDT)
Hearing no objection, FR 1308 has been created. Ahasuerus 21:32, 10 October 2019 (EDT)

Question about the Twilight Zone Universe lists

Is there a reason for not including Richard Matheson's Twilight Zone Scripts volumes in the Twilight Zone Universe lists? —The preceding unsigned comment was added by Klepsis (talkcontribs) .

Good catch! I have put them in a new series, Twilight Zone Scripts (Richard Matheson), and turned it into a sub-series of Twilight Zone Universe. I've also disambiguated the name of Twilight Zone Scripts (Rod Serling) to avoid confusion. Thanks! Ahasuerus 18:46, 3 October 2019 (EDT)

Templates for incomplete contents

Can we create a template for incomplete contents to be used in publication Notes ({{Incomplete}} for example) that shows "The contents are incomplete" (or anything else we want for it to say)? We have a very good way to find magazines and collections and anthologies with no fiction content but once at least one story is added, the title drops from the report (and sometimes a non-fiction magazine that is ignored from the report still needs its no-fiction contents to be finished. And every editor seems to use their own string: "Content from Place; possibly incomplete.", "Incomplete Contents", "contents incomplete", "contents possibly incomplete" and so on. Trying to find all of them so someone can update them is... hard. Annie 16:30, 3 October 2019 (EDT)

Seems like a good idea. Can anyone think of a reason not to create {{Incomplete}}? Ahasuerus 11:30, 4 October 2019 (EDT)
FR 1309, "Create an 'Incomplete' template", has been created. However, before I do implement it, let's make sure that we are all on the same page. The use of this template is supposed to be limited to pubs missing eligible Contents items; it will not be used for pubs which exclude ineligible (typically non-genre) items on purpose, right? Ahasuerus 21:38, 10 October 2019 (EDT)
Yeah, that's the idea. Christian Stonecreek 02:49, 11 October 2019 (EDT)
Yep. Annie 03:12, 11 October 2019 (EDT)

Leading and Trailing spaces in External IDs

Can we modify the javascript check when submitting a publication to first trim() the data in the field and then check if it contains any of the characters we do not want (and then trim again on the back end before saving so we do not save them or pass back the trimmed value)? If I leave an extra space at the end of a title, ISBN or any other field, it just ignores it and cleans it up before saving. The External ID field makes me go and remove it before I can submit... :) Thanks! Annie 20:29, 3 October 2019 (EDT)

Makes sense. FR 1307, "Strip leading and trailing spaces in External IDs", has been created. Ahasuerus 12:12, 4 October 2019 (EDT)

New User

Greetings from Algonquin

I am a very new user to your App (isfdb.org).

What is the purpose of isfdb.org?

With Best Regards,

Jed Dowd (Algonquin) Mypoustinia@msn.com 3 October 2019 at 20:33 CDT —The preceding unsigned comment was added by Algonquin (talkcontribs) .

Welcome to the Database. We are a bibliography site - trying to cover the field of Speculative fiction and list all editions of all books published in the field (and it is a work in progress...). Look around - if you want you can just browse and use the data or you can decide to help - if so, here is the main wiki page with a lot of useful links on how to do things and what you can do and so on :) And you can always ask here.
From our main page: "The ISFDB is a community effort to catalog works of science fiction, fantasy, and horror. It links together various types of bibliographic data: author bibliographies, publication bibliographies, award listings, magazine content listings, anthology and collection content listings, and forthcoming books.".
How did you find us and what were you looking for? :) Annie 22:02, 3 October 2019 (EDT)

Add the rest of the Amazons where they are missing

Can we add the rest of the Amazons to the list of available sites for links? We have only 6 of them now (including missing one of the English language ones...)- and I keep needing some of the rest and having the direct link there will be very useful...

And just in case the list is stored separately, I also want them on the moderator screen so I can click on them for books with ISBNs while trying to research a submission.

And while we are at it, we are also missing two Amazons on the ASIN list - Turkey and UAE:) Annie 00:32, 4 October 2019 (EDT)

Sure, we can do that. Our list of supported Amazon stores hasn't been adjusted in years. A lot has changed in the last decade. Here are the stores that we currently support:
  • Amazon CA
  • Amazon DE
  • Amazon FR
  • Amazon JP
  • Amazon UK
  • Amazon US
Here are the Amazon stores that I think we could add:
Any others? Ahasuerus 11:59, 4 October 2019 (EDT)
You should have 16 - so one is missing: Netherlands. And it is one I need a lot - we have quite a lot of Dutch speaking editors these days:) Annie 12:07, 4 October 2019 (EDT)
Noted, thanks. Ahasuerus 12:15, 4 October 2019 (EDT)

(unindent) One problem that I think we are going to have is "screen real estate". Once we link all 16 Amazon stores, the "Other Sites" drop-down list on the Publication display page is going to become unwieldy. Perhaps we could move the Amazon links (all 16 of them) to another, adjacent, drop-down list? I guess we could call it something clever like "Amazon Links". Ahasuerus 22:39, 10 October 2019 (EDT)

Yes, this seems to be a better solution. Christian Stonecreek 02:48, 11 October 2019 (EDT)

Series content counts

How easy would it be to have the system do a count every day (or every week) of the number of unique titles in a series, then list those numbers at the top of the series page? Maybe even list the number of short fiction, collections, omnibuses, non-fiction, etc., too? This might be useful information on author/artist pages, too. ···日本穣 · 投稿 · Talk to Nihonjoe 13:27, 4 October 2019 (EDT)

I assume you mean mostly for series that are not numbered, right? I like the idea. :) Annie 13:32, 4 October 2019 (EDT)
It's more for series that are ginormous (like Star Wars or Star Trek), that have a bunch of different series within the series, as well as many collections, omnibuses, short fiction, and so on. Statistics are harder to manually do in those cases. ···日本穣 · 投稿 · Talk to Nihonjoe 13:58, 4 October 2019 (EDT)
I like the idea of displaying counts. Not just for series, but for the Contents sections of publication pages as well. However, I don't think we want to calculate them once a day/week because it would create discrepancies. Doing it on the fly shouldn't really affect performance.
The only problem that I can see is the way the software currently displays embedded series. It would likely require a medium amount of work to get everything working properly. Ahasuerus 15:03, 4 October 2019 (EDT)
I imagine there's a series table in the database, listing the title and parent for each series. If a new column (or columns, depending on how granular you want to do things) was added to each series for the count, then it would be a run through to do all the counts, then another that adds up counts for any subseries. The last one would repeat until all the sub-sub-sub-sub series are totaled. Something like that? ···日本穣 · 投稿 · Talk to Nihonjoe 15:15, 4 October 2019 (EDT)
True, there is a "series" table in the database. It contains each series':
  • title
  • parent series
  • relative position within its parent series
  • note
That said, adding a new column and updating it on a daily/weekly basis would result in series counts being potentially out of sync with what is actually displayed for up to 24/168 hours. I believe that would be Very Bad (tm).
Now, the problem with doing these calculations on the fly is that the software which displays series data is invoked recursively for embedded sub-series. Every time it is called, it doesn't know much about the parent series that invoked it; it just displays all titles in the series and returns to its parent, which then calls it again to display the next sub-series, etc. The software would need to be modified to keep track of the total number of titles that it has displayed for the whole series tree. In addition, it wouldn't have the totals until the last sub-series has been displayed, so it would have to be displayed at the bottom of the page -- unless we redesigned the software to accumulate all that display data first and display it second. To complicate matters, the same series display code is used by Series pages as well as by Author Bibliography pages. They display co-authors differently.
In other words, it's probably doable, but would require a certain amount of tinkering. Ahasuerus 15:51, 4 October 2019 (EDT)
How about a solution similar to how we solved the counts number in Advanced Search? Instead of showing it every time, add a button/link which calculates it and shows it on the fly when pressed? This way an editor can find the number easily but we do not calculate it for series that do not need it and we do not change all the underlying logic of the series pages. Annie 16:00, 4 October 2019 (EDT)
Another possibility: Have the count on each series update whenever a new item is added to the series? That would keep it current, and be a logical place for it to increment the count. Then, when it's incremented the count, have a subroutine that checks to see if the series has a parent. If so, then have it increment the parent's total count. Do that until there are no more parents. There would have to be something in place to prevent simultaneous incrementing (in case two or more different entries were updated/approved simultaneously), so it would do one, and then the other. When initially implemented, a single-run script could create the initial counts for each series, and then the updating could be done as I described above. ···日本穣 · 投稿 · Talk to Nihonjoe 16:03, 4 October 2019 (EDT)
This strikes me as a cool idea, but the developer part of my brain wonders about all the nasty edge cases. e.g. would it make sense to show The Night's Dawn *Trilogy* as comprising 9 novels - the 3 UK (?) volumes plus the 6 volumes from the "Night's Dawn Trilogy (paperback)" subseries, which are (AFAIK) the same content, just split into 2 smaller books? From my limited understanding of the database, I don't see anything that indicates the latter are a "variant series", other than text notes in the former stating that they were split into 2 volumes for US publication. ErsatzCulture 17:40, 4 October 2019 (EDT)
And you also have the moving of series under new parents (or removing them from parents) which will cascade calculations at update time as both the new and the old parent and the current series totals (plus all additional parents) will need to be recalculated... Annie 17:49, 4 October 2019 (EDT)
True. As a general observation, database designs that required keeping running totals of things were popular in the 1960s and 1970s, back when data was mostly stored on tape. They became less popular once cheap and reliable disk storage became available. And a good thing too because they were a pain to maintain. Ahasuerus 19:57, 4 October 2019 (EDT)

New cleanup report - Publication Series Names That May Need Disambiguation

A new cleanup report, "Publication Series Names That May Need Disambiguation", has been coded and deployed. The data will become available overnight. I expect approximately 75 records to be flagged. Ahasuerus 21:57, 4 October 2019 (EDT)

Displaying title tags on the Advanced Search Results page

Last year we had the following request posted on the Community Portal (FR 1166):

If you use Advanced Title Search and one or more of the results have too many tags, the tags column squishes the rest of the fields and takes over most of the screen. Here is an example.
Can we add a new user preference (similar to the "Do not display bibliographic warnings on Title pages") that controls if a user sees the tags in advanced search at all? This way if someone really does not care about the tags can stop them from cluttering the page.
Alternatively, we could set a hard width for the column using a percentage. ("Specific pixels width" wouldn't work for phones.)

Earlier today User:ErsatzCulture proposed the following solution:

  • If the user included a tag value in their search criteria, show all tags (on the logic that this is something they're definitely interested in, and we don't want to risk hiding the tag they searched on)
  • If they didn't include a tag in the search criteria, then we limit the number of tags - currently 5, which seems to limit to no more than 2 lines of tags on a 1280px wide screen.
Example screengrabs:

NoTagCriteria.png

WithTagCriteria.png


I rather like this approach. What do other editors think? Ahasuerus 10:35, 7 October 2019 (EDT)

I like the approach but I still think we need to do some restriction on the field and/or the ability not to see tags at all. I do most of my work from small screens - and having 5 tags is already making my screen too hard to see what I need. And the things get worse if one of the tags is extra long. Annie 12:11, 7 October 2019 (EDT)
Seems a bit counter-intuitive. If I search for titles with a particular tag, it's likely I know everything I need to about tags for my selection. And I'm assuming my tag is there for everything displayed. Unless you get into OR conditions ... ../Doug H 14:03, 7 October 2019 (EDT)
I think you're correct on a functional level, but I think human beings generally need reassurance that the results they've got back match what they expected to get. Compare this to when you do a search for something on Google or on a Kindle device, the displayed results highlight the searched keywords in bold or inverse video to show you the match, and to provide context etc.
My thinking for displaying all tags if the user searched for tags, is that searching on tags expressly signals an interest in tagging from the user, so it seems reasonable to show detail in that area.
Re. user controllable showing/hiding of columns, this would obviously be a preferable solution, but it would involve more work. Also, it feels like something which doesn't fit the current preference model, but rather something which should be configurable on the fly in the results page itself, probably with the prefs being stored on a per-device level. (If anyone's switching between using the site on a phone and an HD/4k screen on a regular basis, it's unlikely they'd want the same display settings in both environments.)
All of this is doable, it's just working out the tradeoffs between user needs and pain (and obv. the needs etc of casual users vs contributors vs mods aren't necessarily the same), developer time/interest, etc, etc.... ErsatzCulture 14:30, 7 October 2019 (EDT)
after thinking some more on that. A bit of a background. I proposed that FR awhile ago because the current results screen makes it very hard for me to do a lot of the things I do on a daily basis. This solution won’t solve the problem so if you want to implement it, go ahead and do that. Then I will ask for another FR that actually solves the problem I am having. If you have a big enough screen, the tags are not that big of a trouble - we do not have that many titles with more than 5 tags (can someone check how many)? And the problem is not just how many lines they take but also how much of the screen they take - they squeeze (squish in my original writing above) the columns I am looking for on a small screen taking more than half the screen. And let’s not ignore the fact that a lot of the casual browsers will be using touchscreens and other small screen devices - and making that page work only when you are at home and have your nice big screen is a little behind the times. :) Annie 15:09, 7 October 2019 (EDT)
Not directly related, but it might be good to start planning for a mobile version of the site that presents the options better for a smaller screen. Maybe we should make a page to start discussing that. It would basically present information based on the reported browser. ···日本穣 · 投稿 · Talk to Nihonjoe 16:04, 7 October 2019 (EDT)
It is not just mobile though - I have one of the small Chromebooks (11.6") which is perfect for what I need it for - but the screen is small :) And when a page is optimized for regularly sized screen, it does need to compensate somewhere :) But yeah - we probably should - although some of the pages will need too much of a redesign... Annie 21:30, 7 October 2019 (EDT)
If we spend the time hammering out a good design, it will make making changes more easy in the future. I've made this page to help facilitate discussions toward that end. ···日本穣 · 投稿 · Talk to Nihonjoe 15:17, 11 October 2019 (EDT)

Signing notes

Is there a simple way to sign and date notes (Title and Publication in particular) that is invisible on the display but easily seen and edited in the Editing window? ../Doug H 11:25, 10 October 2019 (EDT)

Nope. Which is why I add a very visible date when that is relevant.
Technically you can use {{BREAK}} which will send it to a separate screen so it will be there if someone wants to see it but not there on the main screen. Annie 11:51, 10 October 2019 (EDT)
So nothing like html comments <! > or null tags ../Doug H 12:35, 10 October 2019 (EDT)
Nope. If you use them, they will get flagged as not allowed HTML and cleaned up overnight if the accepting moderator does not catch them on submit and clean them there and then. If you think they will be useful, you can always open a discussion about allowing them but... I am not a fan of hidden text - way too easy to miss and way too easy to be exploited for someone's advertisement purposes or other mischief. And if we want to keep track of the notes, we probably should built it into the system as opposed to asking an editor to add the data. Annie 13:31, 10 October 2019 (EDT)
Wouldn't it be a good thing we'd always sign & date the notes? I'm working at a pharma company, and there it's mandatory to always sign & date every single change you make to a record ... MagicUnk 09:49, 11 October 2019 (EDT)
Only if we have a system that suppresses that from the everyday screens. You really do not want 20 lines of history for every line of a note.  :) Annie 13:00, 11 October 2019 (EDT)
Well, there is FR 1238, "Create an Edit History page for publications", which would create snapshots of verified publication records before they are changed. It would help with a lot of different issues. Ahasuerus 14:56, 12 October 2019 (EDT)
Personal tools