Difference between revisions of "ISFDB:Community Portal/Archive/Archive42"

From ISFDB
Jump to navigation Jump to search
(archive through January 31)
 
(archive through February)
Line 875: Line 875:
  
 
::I've added them to the series, but you could do that yourself-- the place for indicating the series is in the title record, right below the date. --[[User:Vasha77|Vasha]] 07:59, 30 January 2017 (UTC)
 
::I've added them to the series, but you could do that yourself-- the place for indicating the series is in the title record, right below the date. --[[User:Vasha77|Vasha]] 07:59, 30 January 2017 (UTC)
 +
 +
== Is Book Smugglers a blog? ==
 +
 +
It seems to me that the Book Smugglers' [http://thebooksmugglers.com/ online presence] is in the same situation as what we discussed for Tor.com -- they're a blog rather than a webzine. Would you folks agree? And they're only semipro, so no exception made to index their fiction. I've only been adding their online fiction if they also published it as a chapbook (they didn't always). --[[User:Vasha77|Vasha]] 22:03, 1 February 2017 (UTC)
 +
: As the current rules stand - yes, they are a blog so not eligible except if there is a special case for it. [[User:Anniemod|Annie]] 22:49, 1 February 2017 (UTC)
 +
 +
== 2017-02-01 server downtime at 9pm ==
 +
 +
The server will be unavailable between 9pm and 9:05pm server time. [[User:Ahasuerus|Ahasuerus]] 01:40, 2 February 2017 (UTC)
 +
 +
: The server is back up. [[User:Ahasuerus|Ahasuerus]] 02:02, 2 February 2017 (UTC)
 +
 +
== Language-less NOVEL cleanup ==
 +
 +
Here is where we stand as of this evening:
 +
 +
+------+----------+
 +
|letter|    count |
 +
+------+----------+
 +
| S    |    3001 |
 +
| T    |    1571 |
 +
+------|----------+
 +
 +
The letter 'A' will be added to the report shortly. [[User:Ahasuerus|Ahasuerus]] 02:21, 2 February 2017 (UTC)
 +
 +
: Done. [[User:Ahasuerus|Ahasuerus]] 04:12, 2 February 2017 (UTC)
 +
 +
:: "A" novels cleared. 2 French ones (waiting to be approved) and a few changes in cases when I spotted them which will get English this way when that gets approved... The rest can get automatic English. How about N-Q next? [[User:Anniemod|Annie]] 19:06, 2 February 2017 (UTC)
 +
 +
::: Approved and thanks. N-Q coming right up. [[User:Ahasuerus|Ahasuerus]] 21:22, 2 February 2017 (UTC)
 +
 +
:::: N-Q novels cleared. Nothing non-English in the whole list (some interesting titles though - the genre authors have weird imaginations...). So off with their heads... uhm - I mean, let's assign them all the language automatically. :) E-F next? [[User:Anniemod|Annie]] 18:11, 3 February 2017 (UTC)
 +
 +
::::: N-Q done. The cleanup report has been adjusted to include E and F . Also, I have removed the processed letters from the table above. [[User:Ahasuerus|Ahasuerus]] 02:38, 4 February 2017 (UTC)
 +
 +
:::::: E-F novels are cleared - all English. Even the French sounding one that gave me some hope that something is not English in this group... How about B next? [[User:Anniemod|Annie]] 16:28, 6 February 2017 (UTC)
 +
 +
::::::: E and F have been auto-assigned. Thanks for working on them! B should be available tomorrow morning. [[User:Ahasuerus|Ahasuerus]] 17:03, 6 February 2017 (UTC)
 +
 +
:::::::: All language-challenged "B" novels are English... :) "C" next? [[User:Anniemod|Annie]] 17:57, 8 February 2017 (UTC)
 +
 +
::::::::: Done. "C" is in progress and will be available tomorrow morning. [[User:Ahasuerus|Ahasuerus]] 18:47, 8 February 2017 (UTC)
 +
 +
:::::::::: "C" are all cleared. English all the way down the list (a few formatting issues fixed while I was looking at them anyway). M and R next? [[User:Anniemod|Annie]] 16:17, 9 February 2017 (UTC)
 +
 +
::::::::::: Done and thanks again. M and R will be available tomorrow morning. [[User:Ahasuerus|Ahasuerus]] 18:38, 9 February 2017 (UTC)
 +
 +
:::::::::::: M and R are cleared. S next? Once they are cleared as well, then it will be all remaining novels (T plus the non-letter ones).[[User:Anniemod|Annie]] 18:16, 10 February 2017 (UTC)
 +
 +
::::::::::::: Done and done! [[User:Ahasuerus|Ahasuerus]] 18:54, 10 February 2017 (UTC)
 +
 +
:::::::::::::: S novels are all clear for automatic English. Time for all novels :) [[User:Anniemod|Annie]] 17:05, 13 February 2017 (UTC)
 +
 +
::::::::::::::: Done and done. Once the rest of the novels are wrapped up, we will have the following title types left:
 +
 +
+--------------+----------+
 +
| COVERART    |    2961 |
 +
| INTERIORART  |    5546 |
 +
| ESSAY        |    4331 |
 +
| SHORTFICTION |    4154 |
 +
+--------------+----------+
 +
 +
::::::::::::::: The actual counts will be somewhat lower due to auto-assignment. Also, we could break them down further. For example, a search on ESSAYs which contain words like "Introduction" "Foreword" and "Note" results in approximately 600 hits . [[User:Ahasuerus|Ahasuerus]] 20:17, 13 February 2017 (UTC)
 +
 +
:::::::::::::::: Except that I had found some Introductions and Forewords that had the English word but were not English (mainly Italian). I'd fix the name on some when I know how but still... So I am not comfortable relying on that alone to call them English... If you mean to allow splitting, a 4K or so list is workable so no need to split those. I think I would take short fiction next - later this week I suspect :) Then Essay, then cover art and at the end we can start chipping at the interior art. [[User:Anniemod|Annie]] 20:47, 13 February 2017 (UTC)
 +
 +
::::::::::::::::: Right, I am aware of non-English "Introductions". I was thinking that making the lists shorter would facilitate the review process, but if lists with 4K titles are manageable, then so much the better! [[User:Ahasuerus|Ahasuerus]] 21:01, 13 February 2017 (UTC)
 +
 +
:::::::::::::::::: Sorry, I had been just playing with all those languages and my mind is going to automatic languages immediately :) I prefer smaller ranges (1-2K) but 4K is manageable for a few ranges :) [[User:Anniemod|Annie]] 21:32, 13 February 2017 (UTC)
 +
 +
:::: And any chance I can get the 234 Undefined ones at some point so I can vet which ones are not English and we can have the rest reset to English? :) [[User:Anniemod|Annie]] 20:45, 3 February 2017 (UTC)
 +
 +
::::: I plan to change all empty string ("Undefined") values to nulls, which will make them appear on the cleanup report. [[User:Ahasuerus|Ahasuerus]] 02:38, 4 February 2017 (UTC)
 +
 +
:::::: Ah, that makes sense (and will probably make it easier to fit with the current process. Thanks! [[User:Anniemod|Annie]] 02:57, 4 February 2017 (UTC)
 +
 +
::::::: Any idea when? :) [[User:Anniemod|Annie]] 19:10, 10 February 2017 (UTC)
 +
 +
:::::::: I plan to convert them right before I change the nightly cleanup report to include all NOVELs. [[User:Ahasuerus|Ahasuerus]] 19:49, 10 February 2017 (UTC)
 +
 +
::::::::: Should have waited one more day and it would have become clear. :) Always happens this way it looks like. Thanks! [[User:Anniemod|Annie]] 20:54, 10 February 2017 (UTC)
 +
 +
(unindent) Too many dots so time to unindent. All novels are cleared and require a bit of magic to turn them back into pumpkins... :) Next - short stories.
 +
 +
: Excellent! Here is where we stand with the NOVEL titles out of the way:
 +
 +
+--------------+----------+
 +
| COVERART    |    2186 |
 +
| INTERIORART  |    5411 |
 +
| ESSAY        |    4283 |
 +
| SHORTFICTION |    4143 |
 +
+--------------+----------+
 +
 +
: Given the much reduced number of COVERART titles, do you still want to attack short fiction first? [[User:Ahasuerus|Ahasuerus]] 22:26, 14 February 2017 (UTC)
 +
 +
:: Art gives me a headache with the number of splits and multi-language books - so I would rather deal with the text elements and clear all of them first and then brave the arts :) [[User:Anniemod|Annie]] 22:32, 14 February 2017 (UTC)
 +
 +
::: Sounds good -- short fiction coming right up! [[User:Ahasuerus|Ahasuerus]] 22:48, 14 February 2017 (UTC)
 +
 +
:::: Short fiction cleared - there are 4 language assignments pending I think and a few deletions in my pending queue - anything else on the language-challenged list is English. I think I want COVERART next or I will end up with 7500 art titles left at the end... and that won't be much fun :) [[User:Anniemod|Annie]] 00:15, 18 February 2017 (UTC)
 +
 +
::::: 2,100 COVERART titles coming right up! [[User:Ahasuerus|Ahasuerus]] 02:10, 18 February 2017 (UTC)
 +
 +
:::::: Can you get me some approximate numbers for interior art and essays groups if we split each of them in 3 groups (A-H, I-S, the rest)? The 4K group is manageable but a bit overwhelming (plus this way I will be able to alternate text and art)... [[User:Anniemod|Annie]] 02:55, 18 February 2017 (UTC)
 +
 +
::::::: Sure thing. I'll post the breakdown tomorrow. [[User:Ahasuerus|Ahasuerus]] 05:05, 18 February 2017 (UTC)
 +
 +
By the way - did you ever reset the Undefined ones to null? If you did - then all of them are one of the internal types we still have hanging - as today's list did not bring any surprise visitors. [[User:Anniemod|Annie]] 21:55, 14 February 2017 (UTC)
 +
 +
: That's right, "undefined" [http://www.isfdb.org/cgi-bin/titles_by_language.cgi is no more]! [[User:Ahasuerus|Ahasuerus]] 22:26, 14 February 2017 (UTC)
 +
 +
:: So internal ones only - I do wonder if most/all were not parents or something. Oh well - not that it matters if they never show up again. And I see that Italian finally pulled in front of the unassigned. :) [[User:Anniemod|Annie]] 22:32, 14 February 2017 (UTC)
 +
 +
== 2017-02-02 downtime - 5pm ==
 +
 +
The server will be unavailable between 5pm and 5:03pm server time. [[User:Ahasuerus|Ahasuerus]] 21:23, 2 February 2017 (UTC)
 +
 +
: The server is back up. [[User:Ahasuerus|Ahasuerus]] 22:01, 2 February 2017 (UTC)
 +
 +
== Verified pubs without story lengths ==
 +
 +
This is not addressed to any one person... As I have been going through double-checking 2016 short fiction, I have been finding numerous (several dozen) anthologies and magazine issues marked "primary verified" which do not have the lengths of the stories indicated. That seems like a problem. Personally I would never mark a publication verified unless I had determined the story lengths. --[[User:Vasha77|Vasha]] 01:28, 3 February 2017 (UTC)
 +
 +
: Well, ideally all verified publications would have "length" values assigned. However, determining accurate word counts can be a time-consuming process due to different fonts, different page layouts, borderline cases, etc. If a potential primary verifier has the time to verify everything else about the pub but not each story's length, it may be better to have a primary verification on file even without the length values. [[User:Ahasuerus|Ahasuerus]] 01:48, 3 February 2017 (UTC)
 +
 +
::I agree that print books are a real headache-- I have spent way more time than I wanted to painstakingly counting in those. For ebooks, you have to know how to use Calibre or some such program in order to count. But at least there is no excuse for not cutting-and-pasting online stories into a word counting app!
 +
 +
::Maybe someone could chime in with some tips for how to get word counts from ebooks in various formats? --[[User:Vasha77|Vasha]] 01:53, 3 February 2017 (UTC)
 +
 +
::: If I have the book and I am not sure at the lengths, I would still verify it and then try to figure out lengths if I can after that. Some people verify without even adding content so your mileage may vary. :) The verification says "I have the book/magazine" so if someone needs clarifications, I can pull it out and answer questions. As for the e-books and sizes - depends on the publications - a LOT of them add number of words in their stories (good people); in other cases once you know the number of words of one of the story, you can figure out the rest based on kindle pages (for example). And sometimes just sending a note to a publisher or an author will get you that one example to allow to untangle the rest. [[User:Anniemod|Annie]] 02:32, 3 February 2017 (UTC)
 +
Note that as [http://www.isfdb.org/wiki/index.php?title=User_talk:Bluesman&curid=14182&diff=464413&oldid=464389#Corrections_in_Drowned_Worlds per this discussion], some caution on the subject must be in order so any excess of zeal must be avoided. [[User:Hauck|Hauck]] 07:21, 3 February 2017 (UTC)
 +
 +
== Add Gender to Author Edit? ==
 +
 +
I think things have cooled off enough to return to the subject of adding gender identification to the author edit. Hopefully we can have a calm discussion and a straight up/down vote on it.--[[User:Rkihara|Rkihara]] 19:35, 3 February 2017 (UTC)
 +
 +
: I'd say that it will be useful to have it as long as in addition to the standard two, we also have N/A (for pseudonyms and the like) and a category for the cases where someone is shifting or not declaring or whatever other reason for not knowing it (LibraryThing uses "other/contested/unknown" for example for this case) which is different from the value when no gender had been selected yet. Plus a policy about what the genre of an author should be set to if they changed it in their lifetime so we do not end up in a constant editing mode when editors disagree :) [[User:Anniemod|Annie]] 19:42, 3 February 2017 (UTC)
 +
 +
:: I am not seeing why this would be a particularly useful addition, actually. Yeah, sometimes people want to look up only stories by women or nonbinary people or whatever, but there are other sources for that. Having to figure out what gender people prefer to be referred to as, or figure out if they've changed, can be a headache. More importantly, if you get it wrong, people feel it as a personal insult, more so than if you get other biographical data wrong! --[[User:Vasha77|Vasha]] 20:17, 3 February 2017 (UTC)
 +
 +
:::: Some of these issues were discussed [http://www.isfdb.org/wiki/index.php/ISFDB:Community_Portal/Archive/Archive31#Add_Gender_to_Author_Edit.3F back in 2013], but the discussion became too heated and then petered out for lack of consensus. [[User:Ahasuerus|Ahasuerus]] 20:34, 3 February 2017 (UTC)
 +
 +
::: So are you against adding gender to the DB at all or against allowing gender to be set to something different than male/female for a person? If the latter, how do you propose to assign gender in non-binary cases? [[User:Anniemod|Annie]] 20:22, 3 February 2017 (UTC)
 +
 +
:::: No, I actually like having it not specified now. Yes, we should either list the exact gender that the person gives for themself or just neither of the above if we implement it. But I think the risk of getting it wrong is great, and the usefulness less by comparison.
 +
 +
:::: If we don't change the gender listed on a trans person's page it will seem like a refusal to recognize them, even if it's only the result of slowness. We already had a discussion where people didn't want to change a trans person's name until they had been using the new name for "a while" however long that was. Will the same apply to gender? --[[User:Vasha77|Vasha]] 20:36, 3 February 2017 (UTC)
 +
 +
::::: That was a discussion of an author's canonical name, not a legal name. Canonical names are [http://www.isfdb.org/wiki/index.php/Help:Screen:AuthorData "the most recognized name for that author. The canonical name may be a pseudonym"]. For this reason our canonical name for Alice Sheldon is "James Tiptree, Jr." [[User:Ahasuerus|Ahasuerus]] 20:59, 3 February 2017 (UTC)
 +
 +
:::::: OK fair enough, but there's still the headache of figuring out what gender people prefer. As I say, I'm against adding it at all, but I think that if we do, there should be a standard saying that it should not be specified on a ''living'' author's page unless determined from the author's website, a recent self-contributed bio, or some such. --[[User:Vasha77|Vasha]] 21:08, 3 February 2017 (UTC)
 +
 +
::::::: Let me quote something that I wrote during the last iteration of this discussion:
 +
 +
:::::::* When choosing new fields for a record, I consider a few factors. Fist of all, the data, if known, must be unambiguous. One could argue that the place where a person was "raised" is more important than the place where he or she was born -- and many authors list the former rather than the latter in their autobiographies -- but it's too nebulous for us to use because in many cases there is more than one location.
 +
:::::::* The second factor is relevance, e.g. DOB and DOD are relevant because they show when the author worked, help identify posthumous books, inform the reader that no, there probably won't be a sequel, and so on.
 +
:::::::* The third factor is how easy it will be to find the information. For example, a few years ago I considered proposing that we add a "Place of Death" field to the Author record, but concluded that it would likely remain largely empty for lack of data.
 +
 +
::::::: If identifying gender becomes, as you write, "a headache" -- e.g. see [http://docbrite.livejournal.com/2003/08/22/ this 2003 post by Poppy Z. Brite (who later changed the name to Billy Martin)] -- then it affects the first and the third criteria and the proposed addition goes to the back of the line (in my mind.) [[User:Ahasuerus|Ahasuerus]] 21:24, 3 February 2017 (UTC)
 +
 +
:::::::: Having a standard that the source of gender information must be the author themself would help make it unambiguous. Not quite as good as DOB which never changes. Aren't you phasing out the wiki pages? They're the natural place to put notes about sources of info. --[[User:Vasha77|Vasha]] 21:43, 3 February 2017 (UTC)
 +
 +
::::::::: Not the ones for authors (yet). I am sure that when we decide to tackle those, there will be space in the DB for the data from them. [[User:Anniemod|Annie]] 22:06, 3 February 2017 (UTC)
 +
 +
:::::::::: That's right. Once we wrap up the migration of series-, publisher- and publication-related notes to the database proper, we will tackle "Biblio" and "Bio" pages. We still need to decide whether they will be migrated to a single "Note" field or whether we will have a separate field for "Biography" (where does "Son of Anne McCaffrey" go?). All "Note" fields support the <nowiki>{{BREAK}}</nowiki> [http://www.isfdb.org/wiki/index.php/Help:Screen:EditTitle#Note syntax,] so any supporting information will appear on a separate page. [[User:Ahasuerus|Ahasuerus]] 22:56, 3 February 2017 (UTC)
 +
 +
:::::::::::Im absolutely against this field's addition. My position is simple: the only person that can specify at a given time the gender of an individual is solely him/her/itself and no-one else. As we can't guarantee this, we should not provide this data regardless of the bibliographic benefits that may derived from it.[[User:Hauck|Hauck]] 10:01, 4 February 2017 (UTC)
 +
 +
: I think it would be a useful addition. For authors who were born gender A and then legally changed it to gender B, we'll have to decide whether we should enter the last known gender like we do with legal names. [[User:Ahasuerus|Ahasuerus]] 20:23, 3 February 2017 (UTC)
 +
 +
:: Leaning on ''legal'' changes is not good practice when it comes to trans people. They can transition and live as their preferred gender long, long before the paperwork is straightened out (if it ever is). You just have to list a person's gender as what they say it is. --[[User:Vasha77|Vasha]] 20:46, 3 February 2017 (UTC)
 +
 +
::: That's another interesting question. The [http://www.isfdb.org/wiki/index.php/ISFDB:Community_Portal/Archive/Archive31#Add_Gender_to_Author_Edit.3F last time it came up], Darrah mentioned some of the difficulties involved:
 +
 +
:::* There are some author databases that keep track of gender (e.g. LibraryThing), and at least some of them distinguish between "male" and "presumed male". The question becomes, how much do we need to know about them before we assign them a gender? Do we have to have seen them? (Is that good enough?) Do we have to have a reasonably authoritative reference that uses a gendered pronoun for them? Can we assume someone named "Susan" is female? As a gender researcher, I also would like to see this data in the database. (So far, when I've identified a non-obvious gender identity, I've made sure to add something to the author's wiki bio age that includes a gendered pronoun.) But to add this feature (which I support), we need to decide what rules apply when we specify a gender as "known".
 +
 +
::: [[User:Ahasuerus|Ahasuerus]] 21:06, 3 February 2017 (UTC)
 +
 +
:::: Thanks for that! --[[User:Vasha77|Vasha]] 21:10, 3 February 2017 (UTC)
 +
 +
: The issue of joint/house pseudonyms is an interesting one and I don't think it has come up before. Obviously, something like {{A|Editors of Chicago Tribune}} would have to be "N/A". However, we'd have to decide what to do about {{A|The Brothers Grimm}}, {{A|Ilona Andrews}} (a husband and wife team), and house names all of whose authors are/were of the same gender. My first reaction is that it's safer to use "N/A" for all pseudonyms due to their shifting nature. Either way, it shouldn't stop us from implementing gender support. [[User:Ahasuerus|Ahasuerus]] 20:23, 3 February 2017 (UTC)
 +
 +
:: I wouldn't think this to be a useful addition, too. Gender can be a theme in some works, and though there may be a dominancy in some author's work, I'd think it would serve more towards to put an author into a certain basket, obscuring the many other facets the author's works may show (and should, if the work is more than one-dimensional). I'd think that the tag function is enough to mark the works of gender interest. (I am aware that it is a biographical marking rather than a bibliographical one but people tend to confuse those things; also, we than also could - or should? - index if an author is married or lives in a stable relationship, has children, his major occupation could be of interest etc.) [[User:Stonecreek|Stonecreek]] 10:35, 4 February 2017 (UTC)
 +
::Also, I fear that there wiil be some rush to assign gender to every author's entry, culminating into many false or questionable edits (similar to some of the novelizations markings we had: there were some that are part of a movie or tv series universe, but weren't novelizations of any film). [[User:Stonecreek|Stonecreek]] 10:43, 4 February 2017 (UTC)
 +
 +
:::I imagine it would be moderately useful, if and only if there were a foolproof way of adding the data. As there obviously isn't, and as this would raise too many problems, I think it is not really worth the trouble. [[User:Linguist|Linguist]] 15:47, 4 February 2017 (UTC).
 +
 +
::::Most bibliographies such as Austlit and Gale do list gender, and we already list info that is irrelevant to a purely bibliographical listing: DOB, DOD, Birthplace, Art, Artist, Interior Illustrations, etc.... There is no foolproof way to enter "any" data into the database, errors of gender should be no more common than errors of language, or DOB/DOD and Birthplace. The ISFDB is being more widely used as a research tool and gender information is useful in a lot of research. I solicited comment from Eric Leif Davin, author of Partners in Wonder, on the value of gender identification when I first proposed this. Unfortunately, this seemed to throw more fuel on the fire.
 +
 +
::::I have a personal interest in this, since I feel female writers bring a sensibility to the genre that most male writers lack. I know that Chavey is interested from an academic standpoint.--[[User:Rkihara|Rkihara]] 18:32, 4 February 2017 (UTC)
 +
 +
::::: I appreciate Ahaseuras quoting my concerns about the challenges this would lead to from our previous discussion. I have recently been involved in a couple of Wikipedia edit wars over a person's gender, including one editor who insisted that [http://www.isfdb.org/cgi-bin/ea.cgi?146327 Charlie Jane Anders] should be listed as male, because they had found an (old) posting in which she referred to herself as male. Those two edit wars make me a little more skittish about assigning gender, at least for living authors. If we did include "gender", I think we would need to have a mandatory field for the source of that info. What would we accept as sufficient justification? "They use an obviously male name"? "I've seen them at a convention, and they were obviously female"? "I found an online post where they refer to themselves as male"? "I heard them talk about transition to female, so they must be female now."  In each of these cases, I know counter-examples. Rkihara suggests that it's ok to have some mistakes in the database, but I don't think current authors will get as annoyed at us over an error on their date of birth, or being confused about assigning their "Richard Smith" art to the wrong "Richard Smith". Getting someone's gender wrong will, something like 50% of the time, get that person angry. And I think we should try to avoid that. So, while I previously supported the idea of including gender info, I think now that I can only support Vasha's suggestion about limiting gender information to deceased authors. And that would have to be enforced by the software. [[User:Chavey|Chavey]] 19:50, 4 February 2017 (UTC)
 +
 +
:::::: Agreed, because while an individual or team can gather information according to whatever standards they think is right, a crowd-sourced project like this one is bound to run into trouble. That's how this DB differs from other bibliographies. --[[User:Vasha77|Vasha]] 15:18, 5 February 2017 (UTC)
 +
 +
:::::::I agree that it's good idea to list the source of a gender assignment, but I feel restricting it to dead authors is a little too restrictive. I think that Chavey's estimate that 50% of people would be upset if we misidentified their gender is way too high. I think that most would be upset only if we refused to correct their gender identification when notified. Authors who use initials for their given names are probably used to this. People that are trans or self-identify as another gender are aware that it takes a while for this to be widely known, and would probably be glad to see this information more widely disseminated.--[[User:Rkihara|Rkihara]] 21:10, 5 February 2017 (UTC)
 +
 +
(unindent) Given the variety of opinions on this subject, reaching consensus may be difficult. Still, let me suggest a slightly different approach, which may, with luck, alleviate some of the concerns. How about the following values for the proposed drop-down list:
 +
 +
* Blank (default)
 +
* Female
 +
* Male
 +
* N/A (joint pseudonyms, house names, etc)
 +
* "See Note"
 +
 +
Since our underlying standard for author fields is "publicly available information", we could put whatever information is available publicly in the Note field and make it as detailed as necessary. Of course, we'll have to add support for author-specific notes first. [[User:Ahasuerus|Ahasuerus]] 21:54, 5 February 2017 (UTC)
 +
 +
: Publicly available when and where - is Wikipedia enough at the time it is added to the DB here? And which DBs update their records properly so we can rely on them for verification? Will all the moderators take it upon themselves to verify every attempt to change the gender before approving? What happens in 10 years when someone changes a gender? Who is going to bring the data uptodate when someone changes their gender 5 years after theirs is assigned here? And I am worried about new editors, not reading through the rules for gender and deciding to help by assigning genders on all authors they have books from. The more I think about it, the more it feels like Pandora's box and I would rather not have it in the record than deal with the possible issues... [[User:Anniemod|Annie]] 03:02, 6 February 2017 (UTC)
 +
::Still strongly opposed to this idea. [[User:Hauck|Hauck]] 08:18, 6 February 2017 (UTC)
 +
:::If it's a straight up/down vote, I'd say no. If you want a single reason - we've got enough to do filling gaps rather than creating new ones. [[User:Holmesd|Doug H]] 13:50, 9 February 2017 (UTC)
 +
 +
(unident) '''Outcome''': Consensus not reached. An FR will not be created. [[User:Ahasuerus|Ahasuerus]] 18:40, 17 February 2017 (UTC)
 +
 +
== 2017-02-03 server downtime - 3pm ==
 +
 +
The server will be unavailable between 3pm and 3:02pm server time. [[User:Ahasuerus|Ahasuerus]] 19:41, 3 February 2017 (UTC)
 +
 +
: Everything is back up. [[User:Ahasuerus|Ahasuerus]] 20:02, 3 February 2017 (UTC)
 +
 +
== Bill Longley ==
 +
 +
Raising a glass to 'our' [http://www.isfdb.org/cgi-bin/ea.cgi?113161 Bill Longley], who died three years ago today. We're carrying on with the good work, Bill! [[User:PeteYoung|PeteYoung]] 05:19, 5 February 2017 (UTC)
 +
 +
: Cheers! (I miss him.) [[User:Chavey|Chavey]] 09:06, 5 February 2017 (UTC)
 +
 +
::Me too: I wouldn't be as much involved hadn't he shown his loads of encouragement and good humour. [[User:Stonecreek|Stonecreek]] 09:59, 5 February 2017 (UTC)
 +
 +
::: Ars longa, vita brevis... [[User:Ahasuerus|Ahasuerus]] 17:31, 5 February 2017 (UTC)
 +
 +
::::  I had been meeting him virtually a lot the last few weeks while I had been transferring wiki data to the DB. Haven't realized that he is not with us anymore :( [[User:Anniemod|Annie]] 03:05, 6 February 2017 (UTC)
 +
 +
::::: Cheers! Miss you! [[User:Rudam|Rudam]] 10:27, 6 February 2017 (UTC)
 +
 +
== 2017-02-06 downtime - 12 noon ==
 +
 +
The server will be unavailable between 12 noon and 12:02pm server time. [[User:Ahasuerus|Ahasuerus]] 16:41, 6 February 2017 (UTC)
 +
 +
: The server is back up. [[User:Ahasuerus|Ahasuerus]] 17:02, 6 February 2017 (UTC)
 +
 +
== 2017-02-08 server downtime - 1:40pm ==
 +
 +
The server will be unavailable between 1:40pm and 1:42pm server time. [[User:Ahasuerus|Ahasuerus]] 18:23, 8 February 2017 (UTC)
 +
 +
: Everything should be back up. [[User:Ahasuerus|Ahasuerus]] 18:42, 8 February 2017 (UTC)
 +
 +
== Two different John L. Campbell? ==
 +
 +
Can someone look at {{A|John L. Campbell}}? I am pretty sure that we are dealing with two different people with the same name but will appreciate a second set of eyes. The author data belongs to the novelist from the 21st century. [[User:Anniemod|Annie]] 19:31, 8 February 2017 (UTC)
 +
 +
:I'd say you're right. The photo on the website link belongs to a guy who is 40 years at most, likely younger. Presuming that it is from the 2010s he can't be the one who published stories in the Sixties. [[User:Stonecreek|Stonecreek]] 19:44, 8 February 2017 (UTC)
 +
 +
:: Yeah and he does not claim the stories anywhere on the site either. Or anywhere else. Is that enough to get a disambiguation going? [[User:Anniemod|Annie]] 19:51, 8 February 2017 (UTC)
 +
 +
:::I'd say so! Good catch! [[User:Stonecreek|Stonecreek]] 19:52, 8 February 2017 (UTC)
 +
 +
:::: I've submitted the change in the authors for the two old stories (I think I got it anyway...). There is always a first time for everything. :) [[User:Anniemod|Annie]] 21:29, 8 February 2017 (UTC)
 +
 +
== 2017-02-09 server downtime at 1:15pm ==
 +
 +
The server will be unavailable between 1:15pm and 1:17pm server time. [[User:Ahasuerus|Ahasuerus]] 18:02, 9 February 2017 (UTC)
 +
 +
: The server is back up. [[User:Ahasuerus|Ahasuerus]] 18:16, 9 February 2017 (UTC)
 +
 +
== Author formatting? ==
 +
 +
I believe I have removed all extra spaces between author names. However, there is another discrepancy which we may want to address. Some of our pages separate co-author/co-editor names using commas while other pages use "and"s. For example, [http://www.isfdb.org/cgi-bin/pl.cgi?600424 this publication page] says:
 +
 +
* '''Editors''': C. L. Werner, Steve Lyons, Rob Sanders, Ben Counter
 +
 +
but [http://www.isfdb.org/cgi-bin/title.cgi?2111542 the associated title page] says:
 +
 +
* '''Editors''': Ben Counter '''and''' Rob Sanders '''and''' Steve Lyons '''and''' C. L. Werner
 +
 +
I think it would be desirable to use the same separator throughout the site. Personally, I prefer commas to "and"s. [[User:Ahasuerus|Ahasuerus]] 21:11, 9 February 2017 (UTC)
 +
 +
::Good point. And yes, I think commas look better too.
 +
 +
::Another suggestion: should the category "Shortfiction" on author summary pages be "Short Fiction"? --[[User:Vasha77|Vasha]] 21:55, 9 February 2017 (UTC)
 +
 +
::: I like the '''and''' on the author pages (being in bold and with the long enough string to split the authors)- it gives me an easy split in the long chain of names on some of the titles and makes the page easier to read. At the same time, I prefer the comma on publications and title pages. [[User:Anniemod|Annie]] 22:46, 9 February 2017 (UTC)
 +
:::: Agree with Annie on that, and also in tables of contents the '''and''' is good. --[[User:Vasha77|Vasha]] 14:36, 11 February 2017 (UTC)
 +
 +
== 2017-02-10 server downtime at 1:45pm ==
 +
 +
The server will be unavailable between 1:45pm and 1:47pm server time. [[User:Ahasuerus|Ahasuerus]] 18:30, 10 February 2017 (UTC)
 +
 +
: Everything is back up. [[User:Ahasuerus|Ahasuerus]] 18:46, 10 February 2017 (UTC)
 +
 +
== Changing the main search box? ==
 +
 +
As per feedback from an MIT librarian, the default search box, which requires the user to select a value from a drop-down list, may be sub-optimal. He proposed creating separate boxes for authors, fiction titles and series since they are the most common types of searches. I have seen similar comments on Usenet, so I am posting it here to see if there is broader support for this proposal. [[User:Ahasuerus|Ahasuerus]] 22:11, 10 February 2017 (UTC)
 +
:As an editor, I'm always having to scroll down to get to the Editing Tools (more so on publication pages due to the Other Sites - I have most turned off, but like to keep a few). I'd hate to see those even further down because the search box is now bigger. But I'd also like the convenience of making those searches easier. Maybe it's time for a whole revamp of the left menu? Have different menu layouts for logged in and non logged in viewers? If logged in, move the Editing Tools further up? --&nbsp;[[User:JLaTondre|JLaTondre]] ([[User talk:JLaTondre#top|talk]]) 22:28, 10 February 2017 (UTC)
 +
 +
:: What JLaTondre said - more boxes at the top will make it even harder for editing. Plus - I seem to be doing a lot of searching for ISBNs - especially when I am adding so I would vote for that being somewhat more visible as well... I wonder if two different left bars (editing/browsing) controlled by a user preference and a switch at the top are not in order (so logged in users that are mainly searching do not see the editing either)... [[User:Anniemod|Annie]] 00:17, 11 February 2017 (UTC)
 +
 +
::: Well, we have [https://sourceforge.net/p/isfdb/feature-requests/345/ FR 345], which reads:
 +
 +
:::* Convert the Navigation Bar [on the left] to drop-down menus across the top of each page.
 +
 +
::: I was a bit shaky on the technical side when I created the FR in 2012, but I think I can do it now. My recovery is progressing slower than expected, so more demanding/larger projects are currently on the back burner anyway.
 +
 +
::: A less drastic change would be to move the "3 by 5 catalog card" image to the top of the page, which will free up additional real estate within the search box. [[User:Ahasuerus|Ahasuerus]] 01:45, 11 February 2017 (UTC)
 +
 +
:::: If we are going to change that menu anyway, any chance to have the Editing tools on a publication a bit higher ( see {{P|523646|Dark Star}} - the "Other sites" box is a great thing (for someone I guess anyway) but it is in the way a lot... [[User:Anniemod|Annie]] 01:58, 11 February 2017 (UTC)
 +
 +
::::: Three responses:
 +
:::::* I say yes, if it's possible to have different menu bars logged in or not, the editing tools should be above the other sites.
 +
:::::* I am pondering the idea of dropdown menus at the top, and I think, on the whole, I don't like it. I think it would greatly slow down access for things that I click on a zillion times while editing.
 +
:::::*If you must have preset boxes for multiple categories, maybe just name and fiction title? Too many choices is confusing. I don't quite see how the design would work, with the search at the side, since the bar is too narrow to have labels next to the boxes.
 +
:::::--[[User:Vasha77|Vasha]] 07:08, 11 February 2017 (UTC)
 +
 +
::::::I agree with the opinion that the LHS is already rather packed, and multiple search boxes would add to the clutter.  A few observations/thoughts I have about the box:
 +
::::::* '''Name''' is not obviously author name.  (I realize "author" doesn't do justice to the nature of the possible roles). --[[User:MartyD|MartyD]] 13:20, 11 February 2017 (UTC)
 +
 +
::::::: How about "Person"? [[User:Ahasuerus|Ahasuerus]] 19:16, 11 February 2017 (UTC)
 +
 +
:::::::: In a Fiction DB? If I see "Person", I would think "character in a book". Author may not be exact but will not be confusing at least. [[User:Anniemod|Annie]] 19:40, 11 February 2017 (UTC)
 +
 +
::::::* I frequently do multiple searches for the same category of thing.  If that's not '''Name''', I have to (remember to) repeat the picking each time.  Something that records my last choice (could be a cookie -- doesn't have to be in the DB; no preference management needed) and starts the drop-down out with that choice selected might help some searchers.
 +
::::::* The flat list is sort of haphazard.  If the items in there were grouped (e.g., title things, publication things, etc.), it might be possible to make a different layout that takes advantage of those groups, perhaps via tabs + checkboxes/radios, pull-rights, or some other sort of progressive disclosure.
 +
::::::Ideally, search would just look at everything and then tell the searcher about all categories of matches.  Tough to do that efficiently in the current set-up.  --[[User:MartyD|MartyD]] 13:20, 11 February 2017 (UTC)
 +
 +
:::::::: It wouldn't be too hard to add a new type of regular search to cover "All" record types, where "All" would be defined as authors, titles, series, publishers, publication series, ISBNs, tags and awards. That's what Amazon, IMDB, [https://myanimelist.net/ MAL] and many other popular databases do. We could then display the results either as page sections (a la [http://www.imdb.com/find?q=heinlein&s=all IMDB] and [https://myanimelist.net/search/all?q=heinlein MAL]) or as tabs.
 +
 +
:::::::: However, as Marty suggested above, performance would be a concern. An author search on "heinlein" takes less than 0.1 second. A title search on "heinlein" takes approximately a second. An "All" search of all record types would probably take under 2 seconds. (That's assuming that we make the search smart enough to identify search strings that look like ISBNs and handle them differently.) Is it an acceptable trade-off? I guess we could create a User Preference to control whether "All" is the default selection in the regular search box. [[User:Ahasuerus|Ahasuerus]] 19:36, 11 February 2017 (UTC)
 +
 +
::::::: Personally, I like the IMDB.com search bar. It gives more options, and takes up less space. It's essentially the same as the Amazon search bar. (Hence it's something users are familiar with.) I prefer Amazon's listing of the categories on the left better than IMDB's listing it on the right, but I prefer IMDB's organization of the results better. [[User:Chavey|Chavey]] 13:43, 11 February 2017 (UTC)
 +
 +
(unindent) I agree that scrolling is a pain when editing records. How about an alternative approach? The most popular action when a logged-in user is viewing an ISFDB record is "Edit [record]". We already display the ISFDB record number on the first line of each page, [http://www.isfdb.org/cgi-bin/pl.cgi?604827 e.g.]:
 +
 +
Publication: Grave Markers: Volume Two        ISFDB Publication Record # 604827
 +
 +
How about adding an "Edit" link to the right of the number if you are logged in? And perhaps dropping the word "ISFDB" since it's redundant? The same line would then look as follows:
 +
 +
Publication: Grave Markers: Volume Two        Publication Record # 604827 [Edit]
 +
 +
Would that make life easier? [[User:Ahasuerus|Ahasuerus]] 18:15, 11 February 2017 (UTC)
 +
 +
: YES! That would be awesome. [[User:Anniemod|Annie]] 19:37, 11 February 2017 (UTC)
 +
:: Very good idea. --[[User:Vasha77|Vasha]] 21:56, 11 February 2017 (UTC)
 +
::: Very much like the idea of having an edit link there.
 +
::: One of my personal issues with search is that I frequently accidentally perform the wrong type of search because of the trackpad on my laptop, and then have to type and select my search all over again. It would be great if when a search returned zero results, there was a way to automatically re-submit the search as a different type. Aside from the "A search for 'xxxxx' found 0 matches" notification, there's nothing else on the screen, so there should be plenty of screen real estate to work with. [[User:Albinoflea|Albinoflea]] 03:39, 13 February 2017 (UTC)
 +
 +
:::: How about re-displaying the search string in the search box when you land on the "search results" page? That way you could select the correct value from the drop-down list and resubmit the search without having to re-type the search string. [[User:Ahasuerus|Ahasuerus]] 04:35, 13 February 2017 (UTC)
 +
::::: I'd love that. [[User:Anniemod|Annie]] 04:49, 13 February 2017 (UTC)
 +
:::::: Likewise. [[User:Albinoflea|Albinoflea]] 21:59, 14 February 2017 (UTC)
 +
 +
== DC Metaverse/Sandman Cleanup ==
 +
 +
[http://www.isfdb.org/cgi-bin/pe.cgi?43582 This series], [http://www.isfdb.org/cgi-bin/title.cgi?2049730 this pub], [http://www.isfdb.org/cgi-bin/title.cgi?31268 this title], [http://www.isfdb.org/cgi-bin/pe.cgi?7958 this series], [http://www.isfdb.org/cgi-bin/title.cgi?1108910 this title] and [http://www.isfdb.org/cgi-bin/pl.cgi?593188 this title] are all interrelated and should be part of a common series.  I would recommend " The Dreaming"  to match [https://en.wikipedia.org/wiki/The_Dreaming_(comics) this link].  The Sandman and Books of Magic should remain as part of that series.  Death should have her own series with [http://www.isfdb.org/cgi-bin/title.cgi?1108910 this title] and [http://www.isfdb.org/cgi-bin/pl.cgi?593188 this title].  How would I go about getting this resolved? [[User:Taweiss|TAWeiss]] 19:55, 12 February 2017 (UTC)
 +
 +
: I have added the titles that you listed (plus the ''Sandman'' series) to [http://www.isfdb.org/cgi-bin/pe.cgi?43582 ''The Dreaming'' series]. I have also turned it into a sub-series of [http://www.isfdb.org/cgi-bin/pe.cgi?33519 ''DC Comics Metaverse'']. Could you please check [http://www.isfdb.org/cgi-bin/pe.cgi?43582 the results] to make sure that everything looks OK? [[User:Ahasuerus|Ahasuerus]] 19:22, 13 February 2017 (UTC)
 +
:: Death of the Endless series which includes [http://www.isfdb.org/cgi-bin/title.cgi?1108910 this title] and [http://www.isfdb.org/cgi-bin/pl.cgi?593188 this title] should also be part of the DC Metaverse. [[User:Taweiss|TAWeiss]] 22:51, 26 February 2017 (UTC)
 +
 +
::: Done! [[User:Ahasuerus|Ahasuerus]] 23:25, 26 February 2017 (UTC)
 +
 +
== 2017-02-13 server downtime at 2:15pm ==
 +
 +
The server will be unavailable between 2:15pm and 2:17pm server time. [[User:Ahasuerus|Ahasuerus]] 19:10, 13 February 2017 (UTC)
 +
 +
: Everything is back up. [[User:Ahasuerus|Ahasuerus]] 19:17, 13 February 2017 (UTC)
 +
 +
== Kim Stanley Robinson's Mars Trilogy ==
 +
 +
I'd like to propose a change of the name for the series, from ''Mars Trilogy'' to ''Mars (Kim Stanley Robinson)''. The reason for this is that this ain't a trilogy no more, since it was expanded by a [http://www.isfdb.org/cgi-bin/title.cgi?38877 collection] and diverse short pieces. If there are any objections or ideas for a better name, please let me know. Christian [[User:Stonecreek|Stonecreek]] 18:57, 14 February 2017 (UTC)
 +
 +
: It is kinda the accepted name, despite it not being a trilogy though. If I come to look for the series, I would type Mars Trilogy. So I would not be very happy to lose the name from our listings. [[User:Anniemod|Annie]] 19:26, 14 February 2017 (UTC)
 +
 +
::Yeah... I agree with Annie, it is still most frequently referred to as the Mars Trilogy when it is discussed and written about... in fact, of his three trilogies this is the one with the least amount of confusion over its name.
 +
::As for the collection ''The Martians'', which in one edition or another collects all of the short pieces that aren't just excerpts from the novels, most of them don't relate to the main Trilogy story; many are set in a totally different timeline (e.g. ''Exploring Fossil Canyon'', ''Green Mars'' (novella), ''A Martian Romance'') or an alternate timeline with the same characters (e.g. ''Michel in Antarctica'', etc.)... the way I usually like to think of it, if the Mars Trilogy was a DVD, ''The Martians'' would essentially be the DVD extras portion of the disc. [[User:Albinoflea|Albinoflea]] 20:03, 14 February 2017 (UTC)
 +
 +
:::You are right with the collection, but should it be part of the series in the first place? A collection consisting of six pieces from Ray Bradbury's Martian Chronicles and seven other pieces wouldn't be considered as part of the series either. [[User:Stonecreek|Stonecreek]] 20:26, 14 February 2017 (UTC)
 +
 +
::::That's an interesting question... the collection was assigned as part of the Trilogy before I started editing here, and I've always taken it for granted. Advertisements and interviews from around the time it was completed frequently list it as "a companion volume" to the Mars Trilogy... the NY Times called it "a series of variations on the grand theme of the trilogy". Whatever it is, it's an edge case. [[User:Albinoflea|Albinoflea]] 21:25, 14 February 2017 (UTC)
 +
 +
::::: Perhaps this collection -- as well stories like "Michel in Antarctica" -- could be put in a "Mars Trilogy Metaverse", which would also include the original ''Mars Trilogy'' as a sub-series? [[User:Ahasuerus|Ahasuerus]] 21:48, 14 February 2017 (UTC)
 +
 +
== 2017-02-14 server downtime - 5:40pm ==
 +
 +
The server will be unavailable between 5:40pm and 5:42pm server time. [[User:Ahasuerus|Ahasuerus]] 22:29, 14 February 2017 (UTC)
 +
 +
: Everything is back up. [[User:Ahasuerus|Ahasuerus]] 22:41, 14 February 2017 (UTC)
 +
 +
== People relationships ==
 +
 +
There are some authors and artists that are related to others (Anne and Todd McCaffrey, J.R.R. and Christopher Tolkien, for example). Would it be a useful or desired feature to have a bio field where that relationship could be indicated? Perhaps, when editing a person's entry, have a field to enter the name/ID of the person to whom they are related, and then a drop-down for what that relationship is (parent, sibling, child, spouse, etc.)
 +
 +
Thoughts? ···[[User:Nihonjoe|<font color="darkgreen">日本穣</font>]] · <small>[[Special:Contributions/Nihonjoe|<font color="blue">投稿</font>]] · [[User talk:Nihonjoe|Talk to Nihonjoe]]</small> 01:27, 15 February 2017 (UTC)
 +
 +
: At this time author-specific biographical and bibliographical information is recorded in the Wiki, e.g. see {{A|Nicolaus Bornhorn}}'s Summary page which links to his [http://www.isfdb.org/wiki/index.php/Bio:Nicolaus_Bornhorn Wiki-based Bio page].
 +
 +
: There is a [https://sourceforge.net/p/isfdb/feature-requests/307/ Feature Request] to "Move "Author" and "Bio" pages from the Wiki to the database". I plan to implement it once we finish migrating all publisher-, publication- and series/magazine-specific pages to the database. We will need to decide whether we want to have two author fields -- one for biographical data and the other one for bibliographical data -- or whether we want to use just one "Note" field. [[User:Ahasuerus|Ahasuerus]] 01:38, 15 February 2017 (UTC)
 +
 +
== Search tweaks ==
 +
 +
The search box has been changed to keep the entered search value when displaying the search results page. For example, if you do a search on "farewell," and forget to change the search type from "Name" to "Fiction Titles", the resulting [http://www.isfdb.org/cgi-bin/se.cgi?arg=farewell%2C&type=Name empty search results page] will re-display "farewell," in the search box.
 +
 +
We can also change the software to preserve the last search ''type'' if it is deemed desirable. Opinions? [[User:Ahasuerus|Ahasuerus]] 21:50, 15 February 2017 (UTC)
 +
 +
: I'd love that. This way if the search failure is because of a typo (or because of non-Latin capitalization issue), I can change a single letter and try again immediately [[User:Anniemod|Annie]] 21:52, 15 February 2017 (UTC)
 +
 +
:I love that, and I would love the "type" idea, too. ···[[User:Nihonjoe|<font color="darkgreen">日本穣</font>]] · <small>[[Special:Contributions/Nihonjoe|<font color="blue">投稿</font>]] · [[User talk:Nihonjoe|Talk to Nihonjoe]]</small> 22:56, 15 February 2017 (UTC)
 +
 +
:Definitely, repeat last search type. --[[User:Vasha77|Vasha]] 23:47, 15 February 2017 (UTC)
 +
 +
:Yes. --[[User:MartyD|MartyD]] 13:36, 16 February 2017 (UTC)
 +
 +
:Yes please. [[User:Albinoflea|Albinoflea]] 16:58, 16 February 2017 (UTC)
 +
 +
:: The ayes have it! The software has been modified to remember the last search type.
 +
 +
:: P.S. Please note that these changes affect the ''search results'' page only. If a search finds a single matching record, the software will automatically redirect you to that record's page. If that happens, the last entered search value/type will not be preserved. [[User:Ahasuerus|Ahasuerus]] 18:04, 16 February 2017 (UTC)
 +
 +
::: A little fix needed - when you search for a non-Latin name, when you go to the next page, instead of the name, you have the codes of the characters. Search for "вежинов" (with the small letter). I would like to be able to just change the first letter to capital to search again but you cannot really do that (short of knowing the codes of course). Same happens with Japanese characters (and anythinh non-Latin1 I suspect (you can use 之 to see it with a Japanese character)[[User:Anniemod|Annie]] 00:13, 28 February 2017 (UTC)
 +
 +
:::: Thanks, I'll take a look. [[User:Ahasuerus|Ahasuerus]] 00:21, 28 February 2017 (UTC)
 +
 +
::::: OK, I think I got it now -- too much escaping is a bad thing! [[User:Ahasuerus|Ahasuerus]] 00:39, 28 February 2017 (UTC)
 +
 +
:::::: Confirmed for the regular ones. But there is an issue with finding [http://www.isfdb.org/cgi-bin/title.cgi?2148942 this title]. Thanks! :) [[User:Anniemod|Annie]] 00:52, 28 February 2017 (UTC)
 +
 +
::::::: Well, the problem with this title is that it's actually "& & # x 2 3 ; x 2 5 C D ;" (spaces added to defeat HTML). If you search for this exact string, you will find the record, but, of course, that's not something a regular user would think of doing. I am not sure we can fully support these "fake HTML" cases until we convert everything to Unicode. [[User:Ahasuerus|Ahasuerus]] 01:39, 28 February 2017 (UTC)
 +
 +
:::::::: Suspected as much - just could not resist trying that one. Thanks for checking! [[User:Anniemod|Annie]] 01:45, 28 February 2017 (UTC)
 +
 +
== "Edit" links added to all biblio pages ==
 +
 +
Based on the outcome of [http://www.isfdb.org/wiki/index.php/ISFDB:Community_Portal#Changing_the_main_search_box.3F this discussion], "[Edit]" links have been added to all bibliographic (author, title, publisher, series, publication series, publication) pages. The only exception is the three award pages (award, award type, and award category) which do not display ISFDB record numbers at this time. [[User:Ahasuerus|Ahasuerus]] 21:29, 16 February 2017 (UTC)
 +
: Awesome! Thanks for the quick fix:) [[User:Anniemod|Annie]] 21:59, 16 February 2017 (UTC)
 +
 +
== Stoker preliminary nominees to be added ==
 +
 +
I've added Award records for most of the [http://horror.org/2016-bram-stoker-awards-preliminary-ballot-announced/ Stoker preliminary ballot], but have decided to leave it to someone else to add the last of the nominated works that aren't yet in the DB. Here they are:
 +
 +
*NOVEL (Superior Achievement in a First Novel): ''The Monster Underneath'' by Matthew Franks; ''Until Death'' by Kari Kilgore; ''The Sanguinarian ID'' by L. M. Labat; ''The Black Goat Motorcycle Club'' by Jason Murphy; ''Wasteland Gods'' by Jonathan Woodrow; ''The Eighth'' by Stephanie Wytovich
 +
*NOVEL (Superior Achievement in a Young Adult Novel): ''Keep Away from Psycho Joe'' by Michael Brent Kelley
 +
*NONFICTION (Superior Achievement in Non-Fiction): ''Haunted: On Ghosts, Witches, Vampires, Zombies and Other Monsters of the Natural and Supernatural'' by Leo Braudy; ''A Season with the Witch'' by J. W. Ocker; ''Guillermo del Toro’s “The Devil’s Backbone” and “Pan’s Labyrinth”: Studies in the Horror Film'' by Danel P. Olson; ''The Turn to Gruesomeness in American Horror Films, 1931-1936'' by Jon Towlson
 +
*COLLECTION (Superior Achievement in a Poetry Collection): ''Brothel'' by Stephanie M. Wytovich; ''Small Spirits'' by Marge Simon; ''Two Drinks Away from Chaos'' by Aurelio Rico Lopez, III; ''Voices from Empty Rooms'' by Lisa Lepovetsky; ''Field Guide to the End of the World'' by Jeannine Hall Gailey; ''Children of God: Poems, Dreams, and Nightmares from the Family of God Cult'' by Craig DiLouie and Jonathan Moon; ''The Seven Yards of Sorrow'' by David E. Cowen; ''Corona Obscura: Poems Dark and Elemental'' by Michael R. Collings
 +
 +
--[[User:Vasha77|Vasha]] 20:46, 17 February 2017 (UTC)
 +
 +
:All done. --[[User:Vasha77|Vasha]] 04:01, 24 February 2017 (UTC)
 +
 +
== Weird pseudionyming ==
 +
 +
In [http://www.isfdb.org/cgi-bin/ea.cgi?19432 this one], shouldn't Gregory Francis be pseudonymed under the canonical name and not under another pseudonym? Or am I missing a case where this is the way to do it? [[User:Anniemod|Annie]] 22:28, 17 February 2017 (UTC)
 +
 +
: You are right, it's a joint pseudonym of two writers and should be "pseudonymed" to their canonical names. I have made the change. Thanks for spotting it! [[User:Ahasuerus|Ahasuerus]] 22:33, 17 February 2017 (UTC)
 +
 +
:: I would have fixed it - I just wanted to make sure we do not have an obscure rule for that somewhere :) I think the canonical name of this guy got shifted from the full name to the shorter one at some point or the two names were connected after the joined pseudonym was already in place without completely cleaning the field - I am cleaning a lot of leftover story variants as well for him. [[User:Anniemod|Annie]] 22:53, 17 February 2017 (UTC)
 +
 +
== 2017-02-17 server downtime ==
 +
 +
The server will be unavailable between 8:30pm and 8:32pm server time. [[User:Ahasuerus|Ahasuerus]] 01:17, 18 February 2017 (UTC)
 +
 +
: Everything is back up. [[User:Ahasuerus|Ahasuerus]] 01:31, 18 February 2017 (UTC)
 +
 +
== Language-less titles - the final push ==
 +
 +
Language-less ESSAY titles by first letter:
 +
 +
None. None at all.
 +
 +
Language-less INTERIORART titles by first letter:
 +
 +
+------+----------+
 +
| I    |      133 |
 +
| J    |      36 |
 +
| K    |      15 |
 +
| L    |      138 |
 +
| M    |      207 |
 +
| N    |      93 |
 +
| O    |      139 |
 +
| P    |      195 |
 +
| Q    |      25 |
 +
| R    |      127 |
 +
| S    |      445 |
 +
| T    |    1865 |
 +
| U    |      60 |
 +
| V    |      32 |
 +
| W    |      217 |
 +
| X    |        2 |
 +
| Y    |      13 |
 +
| Z    |        6 |
 +
+------+----------+
 +
 +
[[User:Ahasuerus|Ahasuerus]] 21:05, 18 February 2017 (UTC)
 +
 +
: All cover art entries are cleared. My pending queue has a few language assignments and a few unmerges (which will end up with new languages again). Once they are approved (and the language of the unmerges fixed), all remaining will be English :)
 +
 +
: Essays A-H next? [[User:Anniemod|Annie]] 23:11, 19 February 2017 (UTC)
 +
 +
:: All relevant submissions have been approved. The first ESSAY batch will become available in the morning. [[User:Ahasuerus|Ahasuerus]] 00:29, 20 February 2017 (UTC)
 +
 +
::: Thanks! I just sent in the last batch of followups - after the unmerges. :) [[User:Anniemod|Annie]] 00:51, 20 February 2017 (UTC)
 +
 +
:::: Looks good. I have auto-assigned the remaining COVERART titles, so we are down to ESSAYs and INTERIOART. [[User:Ahasuerus|Ahasuerus]] 02:09, 20 February 2017 (UTC)
 +
 +
::::: A-H (plus special characters and numbers) essays are either English or in my pending list with a different language or for deletion ( a lot of them are for deletion). A-H (plus special characters and numbers) in Interior art next? [[User:Anniemod|Annie]] 04:06, 21 February 2017 (UTC)
 +
 +
:::::: Nice! INTERIORART/A-H coming up next. [[User:Ahasuerus|Ahasuerus]] 04:33, 21 February 2017 (UTC)
 +
 +
::::::: Thanks! With a bit of luck (and no work travel on my part), we should be done with assignments before the end of the month. :) [[User:Anniemod|Annie]] 05:34, 21 February 2017 (UTC)
 +
 +
(unindent) All Interior art records that start with A-H or special characters had been inspected and are neatly separated in 3 groups:
 +
* Group one - Legitimate non-English single titles - Title Edits submitted for all of those.
 +
* Group two - Mixed language versions - unmerge submitted for all
 +
* Group three - All of the rest - aka the majority - the English ones.
 +
 +
At this point it is safe to assign English to the whole group - and further update where needed will deal with it:
 +
* Group one - if the automatic language goes before my update, my update will fix the language; if my update goes first, the automation won't update this title (as it now has a language)
 +
* Group 2 - I have the log to work back through so setting these to English will not lose them (regardless if it happens before or after the unmerge) and will not cause any permanent issues (more titles added to the multi-language report is the side effect but I am clearing that one anyway)
 +
* Group 3  - all clear
 +
 +
All of the above explains why I was not that eager to deal with 5K of these on their own :)
 +
 +
So... next group please - Essays I-S? :) In the meantime I am dealing with the followups of the unmerges. If you want to wait for me to finish the followups, just add the new essays and leave the old interior arts in for tomorrow -- but making them English won't harm anything (as shown above). Thanks! [[User:Anniemod|Annie]] 23:09, 21 February 2017 (UTC)
 +
 +
: Understood and thanks. All INTERIORART titles through H have been auto-assigned. The remaining ESSAYs will be making their triumphant appearance on the cleanup report by 1:30am server time. I can almost smell victory! :-) [[User:Ahasuerus|Ahasuerus]] 23:31, 21 February 2017 (UTC)
 +
 +
:: I will be happy to be done with this project - it had become a bit tedious in the last few batches (not enough to get me to stop so close to the end so no worries). Then I can go and kick some wiki entries out of the way again for a bit - does not seem like anyone is making any progress there. ;) [[User:Anniemod|Annie]] 23:41, 21 February 2017 (UTC)
 +
 +
::: Seems like the essays did not make it to the report tonight. :) oh well - tomorrow night then. [[User:Anniemod|Annie]] 07:06, 22 February 2017 (UTC)
 +
 +
:::: So was it intentional or did just the report get a bit confused? [[User:Anniemod|Annie]] 17:51, 22 February 2017 (UTC)
 +
 +
::::: Sorry, that was a snafu on my side. I had two patches almost ready to go, but only the first one was actually installed yesterday. All fixed now; the data will become available in the morning. (And yes, I can see how reviewing and fixing approximately 100,000 titles can get just a tad tedious ;-) [[User:Ahasuerus|Ahasuerus]] 18:02, 22 February 2017 (UTC)
 +
 +
:::::: No worries - figured out that either you never pushed it or messed up the filtering (I had been a bit creative on ranges in some types after all). Just wanted to make sure that there is no other reasons I am missing. I had not been counting how many I had been clearing - would not be surprised if it is close to 100 000 indeed. At least I did not send 99K or thereabouts "set to English" updates into the moderators' queue during the process - that would have caused a revolt. [[User:Anniemod|Annie]] 18:27, 22 February 2017 (UTC)
 +
 +
::::::: We had approximately 100,000 "bad" titles after the first, completely automated, round of auto-assignments. We are down to 6,000 as of this afternoon. Some auto-assigned titles were processed without a manual review, but my guess is that approximately half were reviewed. That's a lot of titles to eyeball! [[User:Ahasuerus|Ahasuerus]] 18:44, 22 February 2017 (UTC)
 +
 +
:::::::: The automatic assignment of essays and stories starting with "The " did catch a few non-English by the way (stories and essays still unmerged from their English parents) - but they all popped up on the multi-language report so we are good there. And a lot of titles had weird languages - working through some of them had been fun. It could have been a lot worse - if we had a lot more international titles from the days when languages were not supported (or not mandatory or whatever the case was). [[User:Anniemod|Annie]] 19:07, 22 February 2017 (UTC)
 +
 +
(unindent) Two essays are waiting for their language in my pending list (one German, one French in the process of unmerging) and a few more German ones are trapped in two German non-fiction collections (I am leaving them there merged for now until we find the content so we do not end up doing triple work). Anything remaining is English. And we are officially done with the text titles. :) Next: the remaining Interior Art titles (all 3700 or so of them). [[User:Anniemod|Annie]] 20:40, 23 February 2017 (UTC)
 +
 +
: Excellent! All eligible ESSAY titles have been auto-assigned and the remaining INTERIORART titles have been primed. I have also moved all language-specific title reports to the main "Title" section, but I should make a separate post about it in case things get lost here. [[User:Ahasuerus|Ahasuerus]] 00:13, 24 February 2017 (UTC)
 +
 +
(unindent) And done. I expect a few new titles to hit the multi-language report (I did not split a few because I saw more work to be done on their publications so I would rather see it on the other report and finalize them; anything that would not hit the report or does not need to is cleaned) but at this point all language-challenged titles that are not in my pending list can be assigned to English. And that means that the "To be Assigned" group lost the game and takes its rightful place at the end of the queue. It will take a few days to untangle some of the multi-language ones but outside of that the languages for titles cleanup project is officially finished. :) [[User:Anniemod|Annie]] 20:37, 27 February 2017 (UTC)
 +
 +
: Fantastic (no pun intended)! Many thanks! Everything has been assigned, so the cleanup report should be, well, "clean" tomorrow. I am going to revert the change to the "Titles without Pubs" report next and post a note on the Moderator Noticeboard next. [[User:Ahasuerus|Ahasuerus]] 21:10, 27 February 2017 (UTC)
 +
 +
:: Yep. Thanks again for pulling the two types out of the report temporarily. [[User:Anniemod|Annie]] 21:15, 27 February 2017 (UTC)
 +
 +
== Import coverart titles from publications? ==
 +
 +
Currently edit/importcontent.cgi (and presumably the .py too but I did not look at the source) allows one to import titles from another publication (i.e., so called "Option 1" which is nice improvement from before; I recall when there were no options but importing via a list of title numbers). That said, it does not seem to allow me to import coverart titles from one pub to another? Do we have a feature request for this? If not, can we get one? I did not test if edit/exportcontent.cgi would move covertart titles but I expect it won't (someone let me know if I am wrong). Thank you. [[User:Uzume|Uzume]] 04:33, 19 February 2017 (UTC)
 +
:I think I have seen that already done (but you'd have to be careful when there's an already existing coverart title). If you just want the art transferred you only need to copy the address into the Image URL field. [[User:Stonecreek|Stonecreek]] 11:52, 19 February 2017 (UTC)
 +
::The bulk import (Option 1) does not import the coverart. However, the individual import (Option 2) does allow you to import a coverart title. Option 1 already has a check box for including page numbers. A cover art check box could be added similar to the clone pub screen. --&nbsp;[[User:JLaTondre|JLaTondre]] ([[User talk:JLaTondre#top|talk]]) 13:22, 19 February 2017 (UTC)
 +
:::Yes, that makes a lot of sense, adding a checkbox if the cover artist is the same when bulk importing titles. --[[User:Vasha77|Vasha]] 14:37, 19 February 2017 (UTC)
 +
 +
:::: Sounds like a good idea. If there are no objections, we can create an FR. [[User:Ahasuerus|Ahasuerus]] 15:55, 19 February 2017 (UTC)
 +
 +
:::::Does the bulk import already bring over interior art? --[[User:Vasha77|Vasha]] 19:21, 19 February 2017 (UTC)
 +
 +
:::::: Yes, it does. We could add another check box for INTERIORART titles. [[User:Ahasuerus|Ahasuerus]] 19:28, 19 February 2017 (UTC)
 +
 +
::::::: Yes there should be for the same reason that Clone has it. (And the order of the boxes the same as Clone, no confusion.) --[[User:Vasha77|Vasha]] 20:38, 19 February 2017 (UTC)
 +
 +
:::::::: If Clone has this I would consider this a bug (albeit minor) and not an FR. [[User:Uzume|Uzume]] 00:54, 20 February 2017 (UTC)
 +
 +
(unindent) OK, [https://sourceforge.net/p/isfdb/feature-requests/980/ FR 980] has been created. I don't think it's a bug since the current behavior is intended. [[User:Ahasuerus|Ahasuerus]] 02:14, 20 February 2017 (UTC)
 +
:The current behavior is intentionally different from clone behavior? Why? [[User:Uzume|Uzume]] 03:06, 20 February 2017 (UTC)
 +
 +
:: Originally, there were no check-boxes. Then we added a "page numbers" check-box to the import process. Then we added an "intermediate" page with a couple of check-boxes to the cloning process. Then we added more check-boxes to the "intermediate" cloning page. Now we are considering adding more check-boxes to the Import/Export page. The fact that similar functionality is already available on the cloning side of things doesn't mean that there is a bug with the import/export software as it currently exists. [[User:Ahasuerus|Ahasuerus]] 15:49, 24 February 2017 (UTC)
 +
 +
== Wiki sidebar tools ==
 +
 +
I just recently noticed (though it probably was true sometime back) that the left sidebar tools in the wiki are now at the bottom of the page past all the rest of the content. What I mean it is the sidebars labelled "navigation", "search" and "toolbox". This makes it very cumbersome to use these links on long pages (like this one for example but it is still true on short pages like the main page). I am using Chrome if that makes a difference. Has anyone else noticed this and have any feedback, comments, or solutions? Thanks. [[User:Uzume|Uzume]] 16:39, 19 February 2017 (UTC)
 +
 +
: I have tried it with Chrome, Firefox and IE. All three show the sidebar tools on the left. Have you tried clearing the cache and other browser detox tricks? [[User:Ahasuerus|Ahasuerus]] 16:51, 19 February 2017 (UTC)
 +
::I have this with Chrome too. It's also occurs in incognito mode with no extensions active so seems pretty clear it's a browser issue. It's been that way for awhile. Using the developer tools, I can see Chrome and Firefox parsing the &lt;div&gt; indents differently which causes the issue in Chrome. I suspect that the wiki software is probably doing something that isn't 100% standard and Chrome is no longer tolerating it. --&nbsp;[[User:JLaTondre|JLaTondre]] ([[User talk:JLaTondre#top|talk]]) 01:53, 20 February 2017 (UTC)
 +
:::I appreciate the response. I might have to look into that and see where things are going awry (we might be able to fix it in a template or some such). [[User:Uzume|Uzume]] 02:46, 20 February 2017 (UTC)
 +
 +
== Author birthdays and dates of death on the front page ==
 +
 +
This one was discussed (perhaps not for the first time) when I noted that the list of birthdays on the home page was growing too long to be interesting. Instead of limiting that list to only "genre" authors, whatever they are, maybe limit it to authors with a certain minimum number of titles? --[[User:Vasha77|Vasha]] 21:30, 19 February 2017 (UTC)
 +
 +
: That's a good point. Perhaps the algorithm could be fine-tuned to select:
 +
:* award-winning and award-nominated authors
 +
:* authors with more than N titles
 +
: The only difficulty is that identifying these authors in the database will be time-consuming, so we'll need to move the data retrieval process to the nightly job. It shouldn't be a big deal since dates of birth/death change once every 24 hours by definition. [[User:Ahasuerus|Ahasuerus]] 23:48, 20 February 2017 (UTC)
 +
 +
:: That will give a huge preference to artists - they tend to have more works and the good ones have more awards. Not that I mind having some artists in the list but if the list gets nominated by artists, it will kinda be against what the DB is all about. [[User:Anniemod|Annie]] 23:57, 20 February 2017 (UTC)
 +
 +
::: True. It would also be biased against newer authors who haven't published many books/stories yet.
 +
 +
::: I guess we could add different weights or thresholds for different title types: you need to have more than N art titles to be included, but you only need M other types of titles to be included. [[User:Ahasuerus|Ahasuerus]] 00:05, 21 February 2017 (UTC)
 +
 +
:::: We need to be careful for authors like Chiang for example - he has a limited number of stories and nothing else (no novels for example) but he does belong to the list. Although awards will probably get him into the list anyway. [[User:Anniemod|Annie]] 00:13, 21 February 2017 (UTC)
 +
 +
::::: Even one award will be enough to get included, so he will be fine. Besides, he has close to 40 canonical titles, which will be plenty. [[User:Ahasuerus|Ahasuerus]] 00:22, 21 February 2017 (UTC)
 +
 +
:::: And new and upcoming authors will almost never make it - or international ones - where we do not have a lot of their works... Maybe have a different rule for ISFDB minority language writers so they can make it in and for authors that had had works only in the last 2-3 years? [[User:Anniemod|Annie]] 00:13, 21 February 2017 (UTC)
 +
 +
::::: The problem here is that we have no way of telling whether a non-English author has only one title on file because he never wrote any other SF or because we haven't cataloged his work yet. [[User:Ahasuerus|Ahasuerus]] 00:22, 21 February 2017 (UTC)
 +
 +
:::::: True but excluding them altogether is not a good idea either. Not sure how to solve that. Maybe reserve some number of spots for non-major language authors? At least this way we will have some non-major language presence...[[User:Anniemod|Annie]] 00:26, 21 February 2017 (UTC)
 +
 +
::::::: I guess the first question to answer is whether we are trying to make the front page a fair representation of:
 +
:::::::* what's in our database, or
 +
:::::::* worldwide speculative fiction
 +
::::::: The former would be hard enough, but the latter would be much much harder.
 +
 +
::::::: Another thought. Do we really need to display these lists on the front page? They are nice to have, but they are hardly our "core competency". We could move them to a separate page and link them from the front page. If we do that, we don't need to limit the current lists. Or perhaps leave the top 10 authors on the front page and link to the full lists? [[User:Ahasuerus|Ahasuerus]] 00:50, 21 February 2017 (UTC)
 +
 +
:::::::: You have something here. If we have only top 10 (or even top 5) and then a complete list (language-filterable if you feel like coding), that will be awesome. What I do not like is having a list that seems like a representation of the site that contains only English authors. [[User:Anniemod|Annie]] 01:13, 21 February 2017 (UTC)
 +
 +
::::::::: I think the easiest way to do it would be to limit the front page lists to award winners. As long as there is at least one eligible award winner, that is -- February 29 can be problematic :-) [[User:Ahasuerus|Ahasuerus]] 01:39, 21 February 2017 (UTC)
 +
 +
:::::::::::: Award winning authors who were born or died on Feb. 29th: Janet Kagan, Patricia A. McKillip, Tim Powers, Howard Tayler, Sharon Webb. That's 5 :-) [[User:Chavey|Chavey]] 03:04, 22 February 2017 (UTC)
 +
 +
:::::::::: Winners and nominees for when the winners are not enough or none are there? There may be someone often nominated and never won still standing... As for Fen 29 - just make a date exception - if it is Feb 29, get top 5? :) [[User:Anniemod|Annie]] 01:43, 21 February 2017 (UTC)
 +
 +
::::::::::: I'll need to run a few queries to be sure, but I expect that we will need just two cases:
 +
:::::::::::* all winners and nominees whose birthday/anniversary of death is today
 +
:::::::::::* if none are found, the 5 people with the most titles
 +
::::::::::: [[User:Ahasuerus|Ahasuerus]] 03:01, 21 February 2017 (UTC)
 +
 +
:::::::::::: I cannot imagine a case (short of no one's birthday being on this day) that won't work with those two (if you do not want to give precedence to winners compared to nominations). [[User:Anniemod|Annie]] 03:14, 21 February 2017 (UTC)
 +
 +
:::::::::::::As this would likely mean that for most days only award winners will be shown: likely neither Jules Verne, H. G. Wells, Franz Kafka or other of their contemporaries would show up. [[User:Stonecreek|Stonecreek]] 04:56, 21 February 2017 (UTC)
 +
 +
:::::::::::::: Actually, H. G. Wells has a number of title-based nominations, so he would appear on the short list. Jules Verne has [http://www.isfdb.org/cgi-bin/eaw.cgi?159 5 nominations], but they are not title-based. We could check for matching author names, though.
 +
:::::::::::::: However, I agree with your larger point. Any kind of automated algorithm which reduces the number of displayed authors from 30-50 to 5-10 is bound to be biased in some way. We can try to predict/identify the bias and adjust for it, but I don't think we can eliminate it. Awards are an imperfect indicator because there are many more awards now than in the past, but at least they are somewhat objective. [[User:Ahasuerus|Ahasuerus]] 05:14, 21 February 2017 (UTC)
 +
 +
::::::::::::::: I think we do need a list of creators who are displayed. This list would be filled with the criterions already stated: award winners/nominees and number of published works. I'd say both criterions should fill the list, and I propose a number of 7 works of fiction (or more?). That should be enough to downsize the number of displayed creators considerably. But we also need a list of eminent authors that would be left out: Mary Shelley or other early practitioners of the fantastic who didn't publish much but impacted on the genre. This list ideally should be moderated or would be accessible by moderators. Any comments? [[User:Stonecreek|Stonecreek]] 12:04, 21 February 2017 (UTC)
 +
::::::::::::::::Sorry to be blunt, but I've always felt that this list was completely useless. What's its intended purpose? (apart to have the pleasure to see your name in it once a year). Probably the relatives of the authors listed don't need us to remind their birthday. [[User:Hauck|Hauck]] 19:08, 21 February 2017 (UTC)
 +
 +
::::::::::::::::: A very good question indeed! I don't think there was any significance to the choice of author birthdays and dates of deaths when Al added them to the front page. As I recall, we just needed to put something on the front page that would change daily. Forthcoming books and author birthdays seemed like a good choice. The rotating banner was added around the same time.
 +
::::::::::::::::: As I mentioned earlier, I don't think they are our "core competency", but they have apparently become associated with the ISFDB -- or at least I have seen a lot of online discussions which mentioned them. It's really the only reason why I would be hesitant to replace them with a link to a separate Web page. [[User:Ahasuerus|Ahasuerus]] 19:27, 21 February 2017 (UTC)
 +
 +
:::::::::::::::::: Well for me (and I'd think for many others) it has the charm to find some new authors and even more to remind me of nearly forgotten ones. I do think it is one of the things that makes ISFDB special (and regardless of any condensed list, I'll try to maintain the long list: if there are more than 500 names on it per day, I'll likely revise this). [[User:Stonecreek|Stonecreek]] 20:04, 21 February 2017 (UTC)
 +
 +
(unindent)Well, if the problem is what to put on this page, perhaps a kind of "last entered publications" may be a good candidate to 1) showcase our contributor's efforts and 2) give a casual browser the full breadth of our coverage. [[User:Hauck|Hauck]] 08:12, 22 February 2017 (UTC)
 +
 +
: Now, that's an interesting thought! We already have this publicly available [http://www.isfdb.org/cgi-bin/recent.cgi list of recent edits] which links to updated records. Perhaps we could:
 +
:* limit the birthday column on the left to 10 award-winning and "marque" authors
 +
:* display a link to the full list of people who were born and died on this date
 +
:* replace the list of authors who died on this date (some users find it too morbid) with a list of the last 10 edited records
 +
: ? [[User:Ahasuerus|Ahasuerus]] 15:20, 22 February 2017 (UTC)
 +
 +
(unindent) One thing to consider is that the front page already handles a similar issue. Every month, hundreds of new publications are released. However, the front page displays just 22 pubs, a small subset of what's out there. They are called "Selected Forthcoming Books". The list of all forthcoming books is [http://www.isfdb.org/cgi-bin/fc.cgi linked to] from the front page.
 +
 +
How does the software that generates the front page decide which 22 publications to display? It limits the list to "marque" authors. How does it choose the "marque" authors? Every time a user views an author's bibliography, the software increments that author's "view" counter. That counter is then used to generate the [http://www.isfdb.org/cgi-bin/most_viewed.cgi?authors Most Viewed Authors Since 2005] report. Every January a bureaucrat runs a special script which checks all of the "view" counters. The top 2% of all authors are marked "marque". The process is completely automated, objective and doesn't require additional human input.
 +
 +
Since the "marque" data is already available and used on the same Web page, we can leverage it to select the subset of authors to be displayed and then link to a separate Web page which will display the full lists. [[User:Ahasuerus|Ahasuerus]] 15:10, 21 February 2017 (UTC)
 +
 +
:But would this also take regard of important French, Bulgarian or other non-English language authors? Also, it could be of interest to display artists like Hieronymus Bosch who certainly inspired great amounts of fantasists. Nevertheless your idea is certainly better than the long list we have right now. [[User:Stonecreek|Stonecreek]] 16:26, 21 February 2017 (UTC)
 +
 +
:: If it counts each click regardless of who does it, it may propel the Bulgarian ones to the top of the list in some days - the way I am checking on them sometimes :) More seriously though - maybe a mix between a currated list and the marque authors will do. The more international works and updates we get, the more clicks we will get on the non-English ones. Plus the full list will contain them all. [[User:Anniemod|Annie]] 16:34, 21 February 2017 (UTC)
 +
 +
::: I am concerned that a manually maintained/curated list of "important authors" would cause even more problems than the (in)famous "threshold". We all have different ideas about who is important and who isn't, so there is liable to be controversy. In addition, it will require continuing tweaking as the field evolves. Consequently I believe that objective criteria would serve us much better.
 +
 +
::: I can think of three objective criteria which we could use singly or jointly to promote "selected authors" and "selected forthcoming publications" to the front page:
 +
:::* The "marque" flag as discussed above; it's already being used by the "forthcoming publications" display logic
 +
:::* Awards and nominations
 +
:::* Total title count or possibly the count of certain title types, e.g. fiction
 +
 +
::: The last two criteria are computationally intensive, so we will need to move the selection logic to a nighty job to be run at midnight. As I mentioned previously, it won't be a big deal programming-wise. I also think that we should use the same selection criteria to select both authors and publications in order to be consistent. [[User:Ahasuerus|Ahasuerus]] 18:08, 21 February 2017 (UTC)
 +
 +
::::You're right about the differences about who is important, so I'd be okay with the criteria mentioned. [[User:Stonecreek|Stonecreek]] 20:04, 21 February 2017 (UTC)
 +
 +
:::::This is kind of tangential, but if the DOB/DOD is going to be reworked, it seems to me that the DOD should be sort by DOD, rather than the DOB as it is now.--[[User:Rkihara|Rkihara]] 20:06, 21 February 2017 (UTC)
 +
 +
:::::: I am not sure I care too much about this but to throw a little something on this fire, how about quantifying authors by counts of translations of their works (as a measure of how widespread the body of their work is since we cannot count books sold, etc.). That is an interesting (if not the only) measure of how important an author is seen by the the world at large (it is not without its flaws though of course). [[User:Uzume|Uzume]] 01:11, 25 February 2017 (UTC)
 +
 +
::::::: I like the idea in principle but we have so little non-majority languages ([http://www.isfdb.org/cgi-bin/titles_by_language.cgi statistics] books added that it will be almost meaningless at this point :( [[User:Anniemod|Annie]] 01:16, 25 February 2017 (UTC)
 +
 +
:::::::: True, it is sad we have more Ancient Greek titles than Arabic ones. [[User:Uzume|Uzume]] 01:39, 25 February 2017 (UTC)
 +
 +
::::::::: The old requirement to enter transliterated/romanized author names was a big disincentive for anyone who might have been interested in entering works in non-Latin languages. Hopefully things will pick up as we improve our multilingual support. [[User:Ahasuerus|Ahasuerus]] 02:09, 25 February 2017 (UTC)
 +
 +
== Title date display ==
 +
 +
Based on the outcome of a recent discussion, the Title page and the "Diff Publications" page have been changed to display the exact date of the title in the "Year" field. (There is another [https://sourceforge.net/p/isfdb/feature-requests/765/ FR] to change the word "Year" to "Date".)
 +
 +
Next question: is the current abbreviated date format used in the "Date" column of the Title page OK or should we use the YYYY-MM-DD format instead? For example, [http://www.isfdb.org/cgi-bin/title.cgi?1961377 this page] displays "Mar 2016" instead of "2016-03-15", "Feb 2017" instead of "2017-02-21", etc. [[User:Ahasuerus|Ahasuerus]] 15:57, 21 February 2017 (UTC)
 +
 +
:For me, "Feb 2017" (the month of a certain year) is absolutely sufficient. [[User:Stonecreek|Stonecreek]] 16:28, 21 February 2017 (UTC)
 +
:: If we have the actual date recorded, I'd rather see it without the need to go down into the publication. [[User:Anniemod|Annie]] 16:54, 21 February 2017 (UTC)
 +
:::I do like it this way with the month in words. But making it "2 Mar 2010" would be just as good. --[[User:Vasha77|Vasha]] 18:19, 21 February 2017 (UTC)
 +
 +
:::: Why not have both using hover over popup concept taken from transliterations? One issue to consider about date formats is internationalization and localization. "2010-03-02" is the same as "2010年3月2日", "2 Mar 2010", and "March 2, 2010". This is why I prefer the YYYY-MM-DD (ISO 8601) format we use elsewhere. I would prefer a unified style (why do we use "2 Mar 2010" on author birth and death dates but "2010-03-02" on title and publication record dates?)  if we do not support a user selection option. [[User:Uzume|Uzume]] 01:25, 25 February 2017 (UTC)
 +
::::: User-selected date display is certainly a good idea to consider. --[[User:Vasha77|Vasha]] 14:23, 3 April 2017 (UTC)
 +
 +
:::::: There are some inconsistencies in the way dates are displayed. Originally, the idea was that we would always use the international standard (YYYY-MM-DD) format instead of country-specific formats. Later on there was some concern that "YYYY-MM-DD" wasn't as user-friendly as something like "2 Mar 2010". There were a few attempts to make the display side of things more user-friendly, but the changes were inconsistent. A few years later I standardized the way the data entry forms behave, but I didn't touch the display side because I wasn't sure there was a consensus. It's definitely something to revisit at some point.
 +
 +
:::::: Also, although it's possible to add support for user-defined date formats, we will have to consider the potential for confusion when editors discuss dates on Wiki pages. [[User:Ahasuerus|Ahasuerus]] 19:55, 3 April 2017 (UTC)
 +
 +
== My Errored Out Edits ==
 +
 +
This was a feature introduced in 2015 as described [http://www.isfdb.org/wiki/index.php/ISFDB:Community_Portal/Archive/Archive36#New_report:_My_Errored_Out_Edits here]. This also seems to be about the only documentation of the feature. A recent submission ended up there, and some of the edits were incorporated, but others were not. In this case, it seems that the notes and contents updated, but not the reviews. As a non-moderator - what is my role at this point? Do I resubmit the reviews? Do I wait? [[User:Holmesd|Doug H]] 19:05, 21 February 2017 (UTC)
 +
 +
: Please add the Contents titles which didn't make it. It looks like the MIA titles include all of the regular reviews and 2 reviews-entered-as-regular-titles, right?
 +
 +
: On the technical side, I will be able to copy the errored out submission to the development server tomorrow morning. I will then try to re-approve it and see if it generates an error. Once we know what it is, we should be able to fix it. [[User:Ahasuerus|Ahasuerus]] 19:14, 21 February 2017 (UTC)
 +
 +
::That seems correct. I'll resubmit later this evening. [[User:Holmesd|Doug H]] 21:59, 21 February 2017 (UTC)
 +
::Resubmission worked. Looks like the second last ESSAY review partially succeeded - title/author but not the page number. So I'm good on this. Priority is your call. [[User:Holmesd|Doug H]] 13:28, 23 February 2017 (UTC)
 +
 +
::: Sounds good. I am in the process of restoring the latest backup file on the development server. We'll see what happens when I try to re-approve the submission. [[User:Ahasuerus|Ahasuerus]] 15:26, 23 February 2017 (UTC)
 +
 +
:::: After installing the latest version of the database on the development server, I tried to restore the data to the state that it had been in prior to the submission approval. I then re-approved the submission. Unfortunately, it didn't result in an error, so we are still in the dark :-( [[User:Ahasuerus|Ahasuerus]] 21:59, 23 February 2017 (UTC)
 +
 +
::::: Ah, but at least I can pretend it's not my fault. [[User:Holmesd|Doug H]] 16:34, 24 February 2017 (UTC)
 +
 +
== Order of story length options ==
 +
 +
Undecided discussion from [http://www.isfdb.org/wiki/index.php/ISFDB:Community_Portal#Further_.22story_length.22_changes here]: should the dropdown list of lengths be in increasing order (I vote yes) or stay as it is? --[[User:Vasha77|Vasha]] 21:08, 21 February 2017 (UTC)
 +
: I like the current order - less likely to confuse the two similar sounding ones. [[User:Anniemod|Annie]] 21:14, 21 February 2017 (UTC)
 +
::Me too. Keep it as it is please! [[User:PeteYoung|PeteYoung]] 22:38, 21 February 2017 (UTC)
 +
:::If excerpt got added to the mix it would be hard to "sort" that with the others. [[User:Uzume|Uzume]] 01:28, 25 February 2017 (UTC)
 +
 +
:::: I think you may be the only one who likes the idea of "excerpt" being a length... everyone else was talking about it being a title type --[[User:Vasha77|Vasha]] 01:32, 25 February 2017 (UTC)
 +
::::: That was not how I read things (but perhaps I misunderstood). As title type seems that seems problematic. As a length, I could usefully apply it to essays and nonfiction (like an excerpt of a bibliographic index, etc.) as well. [[User:Uzume|Uzume]] 01:43, 25 February 2017 (UTC)
 +
:::::: I read it as a support for Length as well :) However to your latest point here: length may be showing when you are working on publications for all titles but if you read the documentation and look at editTitle for example, it is valid only for Short Fiction. Making it usable elsewhere will probably require a lot of other changes (even if it becomes a length). [[User:Anniemod|Annie]] 01:46, 25 February 2017 (UTC)
 +
::::::: Taking this discussion back up to the proper section now --[[User:Vasha77|Vasha]] 02:10, 25 February 2017 (UTC)
 +
 +
== Sheree R./Sheree Renée Thomas ==
 +
 +
Would there be any objection if I switch the canonical name from Sheree R. (four publications) to Sheree Renée (eighteen publications)? I'm just about to add a collection with dozens more items under the latter name. --[[User:Vasha77|Vasha]] 03:27, 23 February 2017 (UTC)
 +
 +
: Sounds like a plan. It looks like she changed her working name ca. 2005, but we never updated her canonical name. [[User:Ahasuerus|Ahasuerus]] 03:35, 23 February 2017 (UTC)
 +
 +
:: Looking at my email with her, that's the name she's used in all professional settings since at least 2011. [[User:Chavey|Chavey]] 03:58, 23 February 2017 (UTC)
 +
 +
:::OK, pseudonym and variant changes submitted. Ready to delete empty titles as soon as those are approved. --[[User:Vasha77|Vasha]] 04:00, 23 February 2017 (UTC)
 +
 +
== Cleanup reports: Language-specific title reports ==
 +
 +
As you know, we are nearing the completion of the language assignment project or at least its title part. Language-less titles are becoming extinct and will be gone in a few days. It will still be possible to create a title without a language during certain unusual Edit/Clone Publication operations, but it will be rare. For this reason, I have folded the "Titles: Languages" section of the cleanup reports menu into the general "Titles" section. [[User:Ahasuerus|Ahasuerus]] 00:17, 24 February 2017 (UTC)
 +
 +
: You may want to change the text in that report from "No eligible container titles without a language found." to "No titles without a language found." [[User:Anniemod|Annie]] 00:41, 24 February 2017 (UTC)
 +
 +
:: Will do, thanks! [[User:Ahasuerus|Ahasuerus]] 01:05, 24 February 2017 (UTC)
 +
 +
::: Fixed. [[User:Ahasuerus|Ahasuerus]] 01:44, 24 February 2017 (UTC)
 +
 +
== Authors without a language ==
 +
 +
Once all of the title records have had a language assigned, the next step will be assigning a "working language" to our author records. Here is the current breakdown:
 +
 +
+------+----------+
 +
| A    |    2292 |
 +
| B    |    5692 |
 +
| C    |    4931 |
 +
| D    |    3249 |
 +
| E    |    1366 |
 +
| F    |    2431 |
 +
| G    |    3402 |
 +
| H    |    4754 |
 +
| I    |      401 |
 +
| J    |    1624 |
 +
| K    |    2816 |
 +
| L    |    3382 |
 +
| M    |    6200 |
 +
| N    |    1197 |
 +
| O    |    1039 |
 +
| P    |    3253 |
 +
| Q    |      159 |
 +
| R    |    3386 |
 +
| S    |    6670 |
 +
| T    |    2360 |
 +
| U    |      194 |
 +
| V    |      816 |
 +
| W    |    3764 |
 +
| X    |      23 |
 +
| Y    |      388 |
 +
| Z    |      381 |
 +
+------+----------+
 +
 +
I think it would be best to create a cleanup report covering a few letters to see how many are not English. If the answer is "0", then we can safely auto-assign "English" to all of them. [[User:Ahasuerus|Ahasuerus]] 00:39, 24 February 2017 (UTC)
 +
 +
: The answer will be "at least 1 in whatever letter we select" just to make things hard. At least if the titles were any indication... Can we get the pseudonyms cleared before we start manual inspections at least? :) [[User:Anniemod|Annie]] 00:43, 24 February 2017 (UTC)
 +
 +
:: Oh yes, pseudonyms. Sorry, it slipped my mind. I will see what I can whip up. [[User:Ahasuerus|Ahasuerus]] 01:04, 24 February 2017 (UTC)
 +
 +
::: There are 3,326 pseudonyms without a language even though their canonical name has a language code on file. Of that number, 233 canonical names are not English. I'll create an auto-assignment script later today. [[User:Ahasuerus|Ahasuerus]] 16:28, 24 February 2017 (UTC)
 +
 +
:::: All eligible pseudonyms have been auto-assigned. The table above has been updated with the new counts. [[User:Ahasuerus|Ahasuerus]] 18:14, 24 February 2017 (UTC)
 +
 +
::::: I hoped we will win more than that. On the other hand, 3,326 that won't need manual check are better than 0. Shall we proceed with a cleanup report of a letter or two and see how it looks like? I might as well keep my "language-challenged elements rescue" hat a bit longer and help with this one... :) [[User:Anniemod|Annie]] 18:37, 24 February 2017 (UTC)
 +
 +
:::::: Sounds like a good plan. Should we start with the letters [https://en.wikipedia.org/wiki/XYZ_Affair X, Y and Z]? :) [[User:Ahasuerus|Ahasuerus]] 18:49, 24 February 2017 (UTC)
 +
 +
::::::: Why not even though we will do our best not to have even a quasi-war with our French, right? :) [[User:Anniemod|Annie]] 18:53, 24 February 2017 (UTC)
 +
 +
:::::::: The XYZ affair was mostly about (attempted) bribery. Strangely, no one has tried to bribe us yet! :) [[User:Ahasuerus|Ahasuerus]] 19:07, 24 February 2017 (UTC)
 +
 +
(unindent) OK, a new cleanup report has been created. The "XYZ" authors will become available tomorrow morning. A cursory review suggests that a significant number of artists' last names will be of Japanese or Chinese origin. Of course, as we all know, it can be hard to tell what the working language is when dealing with artists. [[User:Ahasuerus|Ahasuerus]] 20:29, 24 February 2017 (UTC)
 +
 +
: Quick question - is there a moderator report somewhere that identifies pseudonyms and main authors that are assigned different languages? And if not, can a check be done to see if we have any couples like that? [[User:Anniemod|Annie]] 21:00, 24 February 2017 (UTC)
 +
 +
:: Not at this time. Checking the database I see that we have <del>53</del> [edit: 47] mismatches. Some of them are legitimate, e.g. {{A|Gregory Kern}}, but most aren't. Let me see if I can create a new cleanup report and allow moderators to ignore mismatches. [[User:Ahasuerus|Ahasuerus]] 21:12, 24 February 2017 (UTC)
 +
 +
::: Yeah, there will be a few legitimate but sometimes when the pseudonyms are connected late, there are real mismatches. [[User:Anniemod|Annie]] 21:42, 24 February 2017 (UTC)
 +
 +
:::: OK, the report has been coded and deployed. It's moderator-only since (a) it lets you ignore records and (b) there are only 47 mismatches to process. [[User:Ahasuerus|Ahasuerus]] 22:21, 24 February 2017 (UTC)
 +
 +
::::: Expected that to be the case. Someone will hopefully get to it. :) [[User:Anniemod|Annie]] 22:29, 24 February 2017 (UTC)
 +
 +
(unindent) So... what are we doing with authors like [http://www.isfdb.org/cgi-bin/ea.cgi?107679 this one] and his co-author. He wrote in Kaqchikel which we do not support... [[User:Anniemod|Annie]] 21:18, 27 February 2017 (UTC)
 +
 +
: Well, it is easy to implement a new language; we have added quite a few over the last 5+ years. My only concern is that Kaqchikel is not listed by ISO 639-2 (available on [https://www.loc.gov/standards/iso639-2/php/code_list.php this Library of Congress Web page]) and we use a subset of their language codes. It's basically our way of outsourcing various issues associated with languages -- e.g. is X a language or a dialect? -- to an established third party. I guess we were liable to run into a language not listed by ISO 639-2 sooner or later... [[User:Ahasuerus|Ahasuerus]] 23:41, 27 February 2017 (UTC)
 +
 +
:: Actually it is there - just hiding. It is one of the Mayan languages - so maybe we should implement "myn" (ISO 639-2 collected them all together due to their obscurity) and we will need to add a note to which one it is into the author page? [[User:Anniemod|Annie]] 23:48, 27 February 2017 (UTC)
 +
 +
:: PS: ISO 639-3 has it on its own but I do not think we need to go that deep... [[User:Anniemod|Annie]] 23:49, 27 February 2017 (UTC)
 +
 +
::: Thanks for looking it up! This is a pretty good example of the challenges created by choosing  between ISO 639-2 and ISO 639-3. Kaqchikel (aka "Cakchiquel" and a bunch of other spellings) is not directly supported by ISO 639-2. On the other hand, ISO 639-3 has codes for "Cakchiquel, Central", "Cakchiquel-Quiché Mixed Language", "Cakchiquel, Santa María De Jesús", "Cakchiquel, Southern", "Cakchiquel, Acatenango Southwestern", etc. In other words, ISO 639-2 is too little and ISO 639-3 is too much for our purposes. Perhaps there is a happy medium that I am unaware of because I am only peripherally familiar with these standards.
 +
 +
::: In this particular case I am tempted to go with "cak" for "Cakchiquel" (listed as "Cakchiquel, Central" by ISO 639-3), but I am concerned about establishing a precedent. [[User:Ahasuerus|Ahasuerus]] 00:11, 28 February 2017 (UTC)
 +
 +
:::: I vote for going for "myn" - it is how it gets identified under ISO 639-2 anyway (in its group) AND add bibliographic notes(which one day will get migrated). Otherwise I am worried about precedents as well. Maybe keep notes and when we see more of these cases, we can discuss moving to 639-3?[[User:Anniemod|Annie]] 00:26, 28 February 2017 (UTC)
 +
 +
::::: OK, I have created an FR. I will put it on hold for a bit in case there are other opinions. [[User:Ahasuerus|Ahasuerus]] 01:55, 28 February 2017 (UTC)
 +
 +
:::::: Done. [[User:Ahasuerus|Ahasuerus]] 21:18, 2 March 2017 (UTC)
 +
 +
== 2017-02-23 site slowdown ==
 +
 +
Response time was pretty bad around 7:45pm server time. There were a lot of additional users hitting the site all at the same time. I am not sure what had brought them here, but they appear to be gone now. Perhaps the [https://en.wikipedia.org/wiki/Slashdot_effect slashdot effect]. [[User:Ahasuerus|Ahasuerus]] 01:03, 24 February 2017 (UTC)
 +
 +
== Year -> Date ==
 +
 +
All occurrences of the word "Year" have been replaced with the word "Date". The changes affect all bibliographic pages, edit pages, mouse-over Help bubbles and Wiki-based Help. They do ''not'' affect the way field names appear on moderator review pages due to certain technical issues which would take a huge amount of time address. Also, "Award Years" are still "Award Years" -- until someone starts giving monthly or daily awards :-)
 +
 +
If I missed anything, please let me know. [[User:Ahasuerus|Ahasuerus]] 01:32, 24 February 2017 (UTC)
 +
 +
: Additional occurrences of "Year" have been identified and corrected. [[User:Ahasuerus|Ahasuerus]] 15:30, 24 February 2017 (UTC)
 +
::What about those in Advanced Search drop downs? [[User:Uzume|Uzume]] 01:53, 25 February 2017 (UTC)
 +
 +
::: I think we are fine there. "Title Year" and "Publication Year" require YYYY values. Similarly, "Title Month" and "Publication Month" require YYYY-MM values. [[User:Ahasuerus|Ahasuerus]] 02:06, 25 February 2017 (UTC)
 +
::::I was thinking more of the sort fields like title search sort by year and pub search sort by year. [[User:Uzume|Uzume]] 02:37, 25 February 2017 (UTC)
 +
 +
::::: Good point - fixed. [[User:Ahasuerus|Ahasuerus]] 19:57, 25 February 2017 (UTC)
 +
 +
::::BTW, pub search results still show things like "Sep 1999" in "Date" column whereas title search shows things like "1997-00-00" in the "Date" column. [[User:Uzume|Uzume]] 02:37, 25 February 2017 (UTC)
 +
 +
::::: Thanks, I'll create a bug report. I have tweaked a few pages to make dates more consistent (see earlier discussions and patch notes), but there is a fair amount of inconsistency left. [[User:Ahasuerus|Ahasuerus]] 19:57, 25 February 2017 (UTC)
 +
 +
:::::: Fixed. [[User:Ahasuerus|Ahasuerus]] 20:53, 25 February 2017 (UTC)
 +
 +
== Redesign of the Verification Status Panel ==
 +
 +
I think we need to revisit the design of verification status panel to account for inactive verifiers. There are presently five primary verification slots and for many pubs these are completely filled. The majority of primary verifiers in many cases are inactive. I propose that inactive primary verifiers should be removed as primary verifiers and moved to side under the heading of Inactive primary, thus opening up the slots for the active verifiers.
 +
 +
I would also like to propose that a “manually entered” change history page be linked to each verified pub. With up to six verifiers, including transient, it’s difficult to know who to notify if a change is made, and the post clutters up the verifier’s talk page.--[[User:Rkihara|Rkihara]] 18:32, 24 February 2017 (UTC)
 +
 +
: There is a [https://sourceforge.net/p/isfdb/feature-requests/897/ Feature Request] to "Re-do primary verifications":
 +
 +
:* The verification page is to be split into three sections
 +
:* The top section of the Web page will be for the currently signed-in user's primary/transient verification. If the user has not verified the pub, this section will give the user the following choices:
 +
:** Primary-verify the pub
 +
:** Transient-verify the pub
 +
:* If the user has already verified the pub, he will be given the following choices:
 +
:** Remove his verification
 +
:** Change the verification type from Primary to Transient or vice versa depending on the type of the current verification
 +
:* The middle section will display other users' primary verifications. The currently signed-in user will not be able to change them.
 +
:* The bottom section will display secondary verifications. They will be displayed using the current table using the current layout and the current functionality.
 +
 +
: This FR was listed on my short list of things to implement in 2017 and received more support than the rest of the FRs. [[User:Ahasuerus|Ahasuerus]] 18:46, 24 February 2017 (UTC)
 +
 +
::Looking forward to it!--[[User:Rkihara|Rkihara]] 19:35, 24 February 2017 (UTC)
 +
 +
== "Interviewed Author" added to Advanced Title Search ==
 +
 +
"Interviewed Author" has been added to Advanced Title Search as a selection criterion. [[User:Ahasuerus|Ahasuerus]] 20:36, 25 February 2017 (UTC)
 +
 +
== Publication formatting ==
 +
 +
The way publication dates appear in our standard publication listings -- the table format used by the title page, the publication series page, Advanced Search results, etc -- has been changed to be consistent with the standard title table. In order to accommodate this change, the "Date" column was made slightly wider at the expense of the "Author/Editor" column. If the table layout doesn't look quite right, please do a full page reload (Control-F5 in most browsers) which should refresh the formatting information used by your browser. [[User:Ahasuerus|Ahasuerus]] 20:59, 25 February 2017 (UTC)
 +
 +
== Advanced Title Search - simultaneous use of Synopsis and Notes ==
 +
 +
Advanced Title Search has been modified to support specifying Synopsis and Notes search values in the same search. [[User:Ahasuerus|Ahasuerus]] 21:12, 25 February 2017 (UTC)
 +
 +
== Enhancing Advanced Search ==
 +
 +
Advanced Search is a useful tool, but it has a couple of issues.
 +
 +
First, some searches take a long time to complete. Not only is it an annoyance for the user who requested the search, but it can also affect other users if the Advanced Search locks certain parts of the database until it's done.
 +
 +
Second, the ability to specify a combination of AND and OR search conditions can be ambiguous. For example, if you select:
 +
 +
# "Title contains 'Star'", AND
 +
# "Title Type is exactly NOVEL", OR
 +
# "Title Tag contains 'juvenile'"
 +
 +
in Advanced Title Search, should the system search for:
 +
 +
* (condition 1 AND condition 2) OR (condition 3)
 +
 +
or
 +
 +
* (condition 1) AND (condition 2 OR condition 3)
 +
 +
? (For the record, it uses the latter algorithm.)
 +
 +
It so happens that these two problems have the same root cause, namely the fact that we allow a combination of "AND" and "OR" conditions. If we were to limit each Advanced Search type to either all "ANDs" '''or''' all "ORs", we could resolve these two problems relatively easily. You would still be able to search for "condition A AND condition B AND condition C" ''or'' for "condition A OR condition B OR condition C", but not for "condition A '''AND''' condition B '''OR''' condition C".
 +
 +
Given the ambiguous nature of mixed AND/OR searches (see above), I don't think it would be much of a sacrifice. Does anyone rely on mixed AND/OR searches? [[User:Ahasuerus|Ahasuerus]] 00:30, 27 February 2017 (UTC)
 +
 +
: OR gets a precedence to AND? That's interesting... I would have assumed the opposite. Or is it just always executing right to left, considering them the same order, regardless of what the operator is  (so C1 OR (C2 AND C3) as opposed to (C1 OR C2) AND C3)? [[User:Anniemod|Annie]] 18:42, 27 February 2017 (UTC)
 +
 +
:: I am not sure it's consistent. I'd have to dissect the SQL queries that the Advanced Search software generates. [[User:Ahasuerus|Ahasuerus]] 20:22, 27 February 2017 (UTC)
 +
 +
: Having a mixed case is useful in some cases (as long as you know how it will work after all - once  you know the precedence, you can write your query properly after all. If the price for performance improvement is losing the mixed case, then sure, cut it. But if there is a way to keep it (only getting called when there is a real mix of And/or), I'd like to still have the option. [[User:Anniemod|Annie]] 18:42, 27 February 2017 (UTC)
 +
 +
::I too think mixed is very useful (but tricky to program I am sure). I think fixing the code to be consistent and documenting the order would be good. I recommend just always doing: ((A op1 B) op2 C). If someone wants a different order you can craft it the other way around this way you can have any of (A AND B) AND C, (A OR B) OR C, (A AND B) OR C, (A OR B) AND C. Meaning, just make the first operator the precedence and document that (it is simple and gives you both options, albeit with some work for the search/query crafter). [[User:Uzume|Uzume]] 21:33, 27 February 2017 (UTC)
 +
 +
::: Unfortunately, the way our version of the MySQL optimizer works, mixed AND/OR cases can take a very long time to complete. Some of the odder queries that I have run on the development server took well over a minute (sic) to finish. They may be somewhat faster on the live server for obvious reasons, but even 20-30 seconds is way too long for a query to run, especially given its potential impact on other users.
 +
 +
::: (Puts his technical hat on.) The only way (that I know of) to address the issue with complex ORs in MySQL 5.x is to switch to UNIONs. I'll have to toy with the query generator to see if we can do it while still supporting mixed AND/OR queries. [[User:Ahasuerus|Ahasuerus]] 22:06, 27 February 2017 (UTC)
 +
 +
:::: How about splitting them - inspect the query before you run it -- if it is mixed, then go the long way; if it is not (go the easy way  - provided that we do not do that already). Add a warning at the top of the page that mixed queries are going to be slow and be done... Now - if the concern is that such a query can slow everyone else down, then maybe make it available only to logged in users? Or disable it until a better solution is found. [[User:Anniemod|Annie]] 22:13, 27 February 2017 (UTC)
 +
 +
:::::I don't use OR, but I could use NOT.--[[User:Rkihara|Rkihara]] 23:11, 27 February 2017 (UTC)
 +
 +
:::::: Way back "AND NOT" was supported along with "AND" and "OR". However, it could cause serious performance problems, so we disabled it. It happened a long time ago and the Advanced Search software has changed quite a bit since then. The [https://sourceforge.net/p/isfdb/feature-requests/3/ Feature Request] to fix and re-enable "AND NOT" was creates in 2007. Depending on the outcome of the current discussion, we may be able to re-enable "AND NOT". [[User:Ahasuerus|Ahasuerus]] 23:27, 27 February 2017 (UTC)
 +
 +
::::::: AND NOT would be very, very useful. More so than mixed AND/OR. But possibly if you found a performance-friendly way to implement one, it would allow the other, too.
 +
 +
::::::: As a last-ditch option, maybe there could be a page for logged-in users to submit a complex search query to be run during daily update time (scheduled slowdown time)? --[[User:Vasha77|Vasha]] 12:58, 28 February 2017 (UTC)
 +
 +
== Rie Sheridan (Rose) ==
 +
 +
I would like to switch Rie Sheridan's canonical name to Rie Sheridan Rose, which she's been using since 2008 (and recently republished the old stories under the new name), since we now have more titles under the new name. --[[User:Vasha77|Vasha]] 12:30, 28 February 2017 (UTC)
 +
 +
: Sounds reasonable. Unfortunately, there is no way to do a mass change, so it will have to be done one title at a time. [[User:Ahasuerus|Ahasuerus]] 19:55, 28 February 2017 (UTC)
 +
 +
::Working on it now. --[[User:Vasha77|Vasha]] 21:11, 28 February 2017 (UTC)
 +
 +
== "Shortfiction" section title ==
 +
 +
Shouldn't the section title "Shortfiction" on summary pages be changed to "Short Fiction"? There's already "Short Fiction Series". --[[User:Vasha77|Vasha]] 13:09, 28 February 2017 (UTC)
 +
 +
: I agree. I also think that we should change "shortfiction" to "short fiction" on Publication pages. [[User:Ahasuerus|Ahasuerus]] 14:49, 28 February 2017 (UTC)
 +
 +
:: Makes sense to me. [[User:Albinoflea|Albinoflea]] 16:58, 28 February 2017 (UTC)
 +
 +
::: I vaguely remember asking for a change of this earlier so I still support it :) [[User:Anniemod|Annie]] 19:09, 28 February 2017 (UTC)
 +
 +
:::: Done. [[User:Ahasuerus|Ahasuerus]] 19:54, 28 February 2017 (UTC)

Revision as of 01:57, 13 May 2017

This is an archive page for the Community Portal. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from January - June 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 · 47 · 48 · 49 · 50 · 51 · 52 · 53 · 54 · 55


Non-genre Shirley Jackson

Would it be all right to delete Shirley Jackson's non-genre stories? True, her supernatural fiction is very important, but it's only a small part of her work. Her summary page would have three times as much below the line as above if I simply marked things "non-genre". --Vasha 04:36, 29 December 2016 (UTC)

OK for me. Hauck 08:21, 29 December 2016 (UTC)
By the way, "After You, My Dear Alphonse" and "The Renegade" are in The Supernatural Index but this surely must be a mistake; you can read extensive summaries of them respectively here and here. Do Ashley & Contento ever make mistakes like that? --Vasha 19:06, 29 December 2016 (UTC)
Well, no bibliography is perfect, as Robert Reginald, an occasional ISFDB editor, used to say. I have been exchanging e-mail corrections with Bill Contento and Dave Langford (the SFE3 maintainer) for years. Ahasuerus 19:35, 29 December 2016 (UTC)
I'm inclined to think of her as well "above the threshold". (When L.W. Currey is selling three of her genre books for ≥ $1,000, that kind of demonstrates her importance.) One of the things about the non-genre listings is that it tells genre collectors of an author which books to *not* bother getting, and for the important authors, that's really useful information. Chavey 17:12, 30 December 2016 (UTC)
I was also thinking that it would be kind of nice to have the non-genre work listed for reasons similar to what Darrah said -- so people could look at a story and see definitely that it was non-genre, rather than wondering if we just hadn't listed it yet. The problem is that if you apply that logic to all authors who have a lot of non-genre work, you start listing everything ever. The whole concept of a threshold was to prevent that. Is one of the nebulous criteria for this line that the author is important? Even if they‘re not about 50% genre? I have made a rough survey of Shirley Jackson's fiction, and came up with about 42 short stories and 4 novels in the genre, about 120 stories and 2 novels out of it. That's 35% or so. --Vasha 19:04, 30 December 2016 (UTC)
Someone correct me if I'm wrong, but I believe for non-genre works of "above the threshold" authors, we do include book-length works (i.e., non-fiction books, novels, collections, and, I suppose, possibly chapbooks), but not short works contained in other publications. --MartyD 11:30, 5 January 2017 (UTC)
I agree. Hauck 09:05, 6 January 2017 (UTC)
To quote the Rules of Acquisition, "This includes any non-genre works published as standalone books as well as non-genre short fiction...". The policy specifically allows non-genre short fiction. I think it comes down to what do you mean by "other publications". If you mean a non-genre magazine or a non-genre anthology, then I would agree as those publications don't belong. However, if it's collection of an above-the-threshold genre author, than the non-genre short fiction should be included as well per our policy. Given your question is under this topic, I assume it was based on the edit by Vasha77 that you have on hold? If so, than if it's agreed that Shirley Jackson is above the threshold, that edit is valid per our current inclusion policy. -- JLaTondre (talk) 15:35, 7 January 2017 (UTC)
Makes sense. That was the submission, and I have accepted it. Thanks for the feedback. --MartyD 11:56, 9 January 2017 (UTC)

Zac Brewer

Zac Brewer has changed his name since coming out as trans. He has many publications as Heather Brewer, only a few as Zac. But not only is he no longer using the name Heather, but he has also changed online references to him as much he could, and set up all-new websites under the new name. Therefore, I think we should make Zac Brewer the primary name and Heather the pseudonym. What do you say? --Vasha 23:17, 29 December 2016 (UTC)

It's a fairly common scenario: an author becomes known as A, then changes his or her primary working name to B for some reason. In most cases it has to do with marketing, but we also have a number of authors who changed their working names after a gender change, e.g. Hank Stine/Jean Marie Stine or Amos Salmonson/Jessica Amanda Salmonson.
As per Help, "For authors who publish under multiple names, the canonical name is the most recognized name for that author". Swapping VTs is somewhat time-consuming, so in most cases we wait until the new name acquires a critical mass of titles. Then again, sometimes old, abandoned names linger for years because no one is willing to spend the time necessary to rearrange everything, e.g. see Megan Lindholm who has been better known as "Robin Hobb" for the last 20 years. In the grand scheme of things it's not too important as long as all titles appear on the same Summary page. Ahasuerus 00:26, 30 December 2016 (UTC)
Actually, I think it is important to recognize the new identity when someone has come out as trans. And it really wouldn't take very much work just to variant all the HB books to ZB. Plus, we hope he has a long career; so it will be much harder to do this change in the future! --Vasha 00:34, 30 December 2016 (UTC)
Well, our "canonical name" values are not judgements re: SF authors' identity. Similarly, our "working language" values are not judgements re: SF authors' primary culture. We don't get paid enough to tackle such weighty issues :-) We just record author names as they appear in books and magazines and then pick the most recognizable name as the canonical value. It's similar to the way we determine canonical titles -- "the canonical title is usually the first title for that work, but may be a later title if that title is much better known". Ahasuerus 00:57, 30 December 2016 (UTC)
It is true that currently, someone is likely to pick up a book with "Heather Brewer" on the cover and consult the database to find out what else this author has written. What's the harm, though, if the answer they get is "this author also uses the name Zac Brewer, consult that page for other works"? At some point, presumably, the name Zac will become recognizable, though he just announced the change a few months ago. I just don't see the point of waiting some nebulous length of time... is it recognizable in two years, three, what? Might as well make the change now -- it's not like the page for Heather will be deleted entirely. --Vasha 01:08, 30 December 2016 (UTC)
If you feel strongly about it, you can always reverse the direction of the pseudonym and the variants. :) Bibliographies tend to be just that -- records of existing records. Annie 01:16, 30 December 2016 (UTC)
That is so -- and the Heather Brewer records will still exist, if we variant them to Zac Brewer (as we should). --Vasha 01:25, 30 December 2016 (UTC)
My only concern with making the change now -- rather than waiting until there is a critical mass of "Zac Brewer" books -- is that it assumes that there *will* be more "Zac Brewer" (SF) books. What if the author switches to mysteries or stops writing altogether? At some point we'd have to decide that it's been "long enough" and go back and change the VT/pseudonym direction again. It's not a huge deal either way, but it seems safer to wait. Ahasuerus 01:27, 30 December 2016 (UTC)
Z. Brewer has currently one short story and one novel, but has announced two more novels coming out next year (both speculative). --Vasha 01:40, 30 December 2016 (UTC)
It's encouraging that they will be speculative -- the fact that the only novel that has been published as by Zac Brewer so far contained no known speculative elements had me a bit nervous. Barring another industry meltdown like the one we had in 2008-2009, we should be most likely safe making the proposed change. Ahasuerus 02:05, 30 December 2016 (UTC)
OK, how about waiting until one of those books actually appears, then? --Vasha 04:06, 30 December 2016 (UTC)
Sounds like a plan! Ahasuerus 04:10, 30 December 2016 (UTC)

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

I think the biggest problem with following the author's preferences in a certain subset of cases is that there are other, equally compelling, subsets of cases. For example, take an author who has gone through a very painful divorce and wants to have as little to do with her old married name as possible. Or take an author who has been forced to use a pseudonym because the books published under the original name didn't sell well and whose resurrected career is still hanging in the balance. She may well see anything that promotes the pseudonym as helping her career to survive and vice versa.
It's all understandable and justifiable, but if we start making exceptions for "worthy causes", it will be very hard to preserve any kind of objective standard. Ahasuerus 18:22, 31 December 2016 (UTC)
I would go so far as to say the ISFDB "cause" is a "worthy" one and thought we might recognize others and attempt to not ignore such, for our project our standards must come first. For the reasons mentioned above, the most common name is still the older one (until a large body of in-genre work changes that). Uzume 03:21, 6 January 2017 (UTC)

Improving our support for diacritics in author names

As per this discussion, our software considers character pairs like "Ó"/"O", "í"/"i" and "é"/"e" identical. For this reason the fact that we have a "Salvador Dali" author record on file means that we can't have a "Salvador Dalí" author record -- the software thinks that they are the same name and duplicate names are not allowed. Note that this only applies to non-English West European characters. Other similar characters are not affected, which is why we have a record for "Stanislaw Lem" as well as a record for "Stanisław Lem".

The current software behavior is inconsistent and frequently forces us to use compromises and workarounds when dealing with diacritics in author names. It's been suggested that we change the software to allow both "Salvador Dali" and "Salvador Dalí" at the same time. (The plan is to keep the lookup algorithm case-insensitive so that when an editor submits "robert a. heinlein", the software will realize that we already have "Robert A. Heinlein" on file and use that record instead of creating a new one for "robert a. heinlein".)

I have run a few experiments and tweaked some parts of the software on the development server. At this point I am 90% sure that we can change the code to support both "Salvador Dalí" and "Salvador Dali". A search on either "Salvador Dalí" or "Salvador Dali" will find both author records. It won't be a trivial change and it will require careful testing, but I think it's doable.

The only downside that I have found so far is that there will be a slight overhead when displaying submissions containing many authors. The current algorithm hardly takes any time at all to look up an author record when displaying a submission. With the new way of doing things it will take something like 0.04 second per author, which will mean 1+ second delays when displaying a submission with 30+ authors. Luckily, it doesn't happen very often and the trade-off appears to be worth it.

If there are no objections, I will start working on the proposed change once I finish processing the latest round of Fixer submissions. Ahasuerus 22:23, 30 December 2016 (UTC)

That will be a nightmare in terms of pseudonyms. Let's take this diacritically-challenged author. He's usually given as "Hervé" (even in the vast anglo-saxon world like here) but, sometimes he's been changed to "Herve" (like here). If I understand it correctly, the proposed change would mean that we'll have two authors (Hervé and Herve) that will have to be pseudonymistically linked with the correct varianting done. Hervé (or is it Herve) Hauck 19:15, 31 December 2016 (UTC)
That's right. It will be similar to the way Stanisław Lem's Summary page currently looks: Polish titles use "Stanisław" while translations use a mix of "Stanisław", "Stanislaw", "Stanislas" and " Stanislav" depending on the publisher's/translator's whim. Ahasuerus 20:11, 31 December 2016 (UTC)
Imagine the work and the look of the author's page (note that I'm already quite displeased by such pseudonyms settings (Liu Cixin being a pseudonym of Cixin Liu). Hervé (or is it Herve) Hauck 19:15, 31 December 2016 (UTC)
The "Liu Cixin"/"Cixin Liu" situation is temporary. The canonical name will be soon changed to "刘慈欣" and the two Romanized forms of the name will become pseudonyms. Ahasuerus 20:06, 31 December 2016 (UTC)
I'm completely against this move.Hervé (or is it Herve) Hauck 19:15, 31 December 2016 (UTC)
Another thing to consider is that some authors have nearly-identical names, the only difference being an accented character. A while back we ran into a couple of Dutch (?) authors whose names were something like "Peter Frai" and "Peter Fraï" (I don't recall the details), which we had to disambiguate with a "(I)". Ahasuerus 20:26, 31 December 2016 (UTC)
As a general observation, one of the long-term goals of this project has been to get as close to the "exactly as stated" standard as possible. We have found that everything works much better when we capture things as they appear in publications and handle the rest using variants/pseudonyms/etc. It's not always feasible, e.g. we wouldn't want to create a pseudonym/variant for all-lowercase and all-uppercase forms of author names/titles, but it has been the general direction of the project. Ahasuerus 20:34, 31 December 2016 (UTC)
To keep on this general level, that goal is commendable but, as I wrote before, I was at first strongly convinced by this approach, but very quickly I was confronted to such concepts as "case regularizations", "publisher regularizations", "initial regularizations", "space regularizations", "Ranks, suffixes, prefixes regularizations", "ellipsis regularizations" (to name a few, with some that appear here) that are not really conform to our "exactly as stated" professed standard. I was even called to order for entering a title as it was ("It's a typo" was the answer). That's why I frankly doubt that "exactly as stated" IS in reality the general direction of the project. The real debate to have is in fact to decide which of our idiosyncrasies we keep. Hauck 10:47, 1 January 2017 (UTC)
It's certainly true that we have added many caveats to the "exactly as stated" standard over the years. Some of them were due to software limitations. For example, the software supported only one publisher per publication, so we needed to have special data entry rules for publications associated with multiple publishers (like various book club reprints.) Similarly, there was no separate field for imprints, so we needed to have rules for squeezing imprints and publishers in the same field. The list goes on.
In addition, we wanted to avoid data fragmentation, which is also related to software limitations. Since there is no "variant publisher" mechanism, entering publisher names exactly as they appear on title (copyright?) pages would have created a lot of "publisher names" per actual publisher, making it hard to see what the publisher has actually published. Hence the "publisher regularization" rules.
Finally, some regularization rules were created (and sometimes enforced in the software) in order to facilitate searching and other types of user experience. The "ellipsis regularization" rules were among them.
Having said that, I should also point out that we have been slowly enhancing our software to allow closer compliance with the "exactly as stated" standard without adversely affecting user experience. For example, take the process of adding "transliterated titles/names" in 2015-2016. Users who don't know Kanji/Cyrillic/etc can still search on transliterated titles/names and see them in mouse-over bubbles while the primary data is now entered "exactly as stated", i.e. using the original script/alphabet.
It's a slow process, but if we continue pushing the software in that direction, we should be able to eliminate more and more "regularization" rules over time. Ahasuerus 22:36, 1 January 2017 (UTC)
I like the idea of having the author name as is but I do agree that the author pages are getting crazy. Maybe we need another user preference that hides all those "as by" from the author summary page. Incidentally Annie 11:46, 1 January 2017 (UTC)
Let me make sure that I understand the proposal correctly. Would this user preference suppress the display of "as by Pavel Vejinov", "as by Pavel Vešinov" and "as by Pawel Weshinow" on Павел Вежинов's Summary page but keep the translated titles intact? What about Robert A. Heinlein's Summary page? Would it make the line which currently reads "Life-Line (1939) [as by Robert Heinlein]" disappear since the canonical title is "Life-Line", i. e. the same as the variant? Ahasuerus 22:02, 1 January 2017 (UTC)
Yes for the Вежинов - all non-Latin-1 authors pages look overloaded now because of the too many English/German/French/Dutch titles we have. And we add more Russian/Bulgarian/Japanese titles, the Latin-1 ones will get in that direction (which is what Herve is also talking about I think). Annie 22:39, 1 January 2017 (UTC)
If the issue is that author pages can become too busy due to translations, wouldn't the user's response be to go to User Preferences and either turn off all translations or select a subset of languages that he or she is interested in? Ahasuerus 23:13, 1 January 2017 (UTC)
But I want to see all the translations - I just do not want to see whatever pseudonym was used for the specific variant. Annie 23:42, 1 January 2017 (UTC)
Plus if you only leave English and German for example on Lem's page (or any non-Latin language one), every single line will have "as by". Which makes the page really crowded. Same if I leave only Russian and Bulgarian on a Latin-1. Annie 23:57, 1 January 2017 (UTC)
Well, it depends on the number of translations. Коста Сивов, a Bulgarian author, hasn't been translated -- as far as we can tell -- so his page looks pristine. On the other hand, someone like Lem will have a lot of "as by"s. I have never thought of it as an issue, but I guess suppressing the "as by" component would make the page easier to read. I don't think it should be hard to implement -- please go ahead and create an FR. Ahasuerus 01:17, 2 January 2017 (UTC)
Right, I need to go figure out what my account used to be there :) Annie 01:54, 2 January 2017 (UTC)
Shouldn't be a problem -- our SourceForge settings let anonymous users create FRs and Bug reports. Ahasuerus 01:57, 2 January 2017 (UTC)
Ah, did not realize that. Thanks! :) Annie 02:00, 2 January 2017 (UTC)
One more note - people coming to the DB for the first time (no account, so no preferences) - they do not have languages filtered. And the page just looks... very busy. If I see that, chances are, I will go and look elsewhere for a list of Lem's works for example... This problem had gotten worse since we added the language support for authors. Annie 01:54, 2 January 2017 (UTC)
Users who are not logged in see the following message on Summary and Series pages:
  • Showing all/no translations. [Cookies-based button which lets them turn translations on and off] Registered users can choose which translations are shown.
One click of a button and all translations go "poof"! Ahasuerus 02:00, 2 January 2017 (UTC)
Again - the problem are not the translations but the "as by" on every single line in some cases. Annie 18:07, 4 January 2017 (UTC)
I guess we could add another cookies-based button to let unregistered users eliminate all "as by"'s, but I think that level of customization really ought to be handled via User Preferences. Ahasuerus 18:48, 4 January 2017 (UTC)
I am not sure for the Heinlein case - if the title, language and year are the same, we can as well not show it (we have the variant because of the name change). However - if it has titles under both variants, we need a way to filter under the other variant only so I suspect we need to have it there. Or have a "show all variants" link. Annie 22:39, 1 January 2017 (UTC)
- I would also love to be able to see a list of works under a pseudonym without the need to go for "show all works". Annie 11:46, 1 January 2017 (UTC)
Could you please clarify? Each pseudonym should have a "or view all titles by this pseudonym" link on the "See: [canonical name]" line. Are you seeing something else? Ahasuerus 21:55, 1 January 2017 (UTC)
They are there but in the merging/searching format as opposed to just a list (as on the main author page). So I cannot find out a glance which books from a series are available under the pseudonym (and for some languages this will mean under the language) Annie 22:39, 1 January 2017 (UTC)
Ah, I see. I think I looked into this issue a while back but gave up due to various complexities associated with creating a "pseudonym-only" bibliography page. I am sure it's possible given enough man-hours. If you find this functionality useful, please go ahead and create an FR. Ahasuerus 23:08, 1 January 2017 (UTC)

(unident)I think this will be good in the long run. In the short term, it will be a pain as everything is adjusted to meet the updated standard, but supporting Unicode (I'm assuming that's what's being changed, since it would allow for the differences in letters) is the way things should go. I'm not bothered by the pseudonym listings on author pages. ···日本穣 · 投稿 · Talk to Nihonjoe 17:35, 1 January 2017 (UTC)

Actually, the proposed change can be implemented without converting to Unicode. The database and the software are not quite ready for the conversion at this time, although it *is* on the list for a number of reason. For example, it will improve searches which include non-Latin characters. The way things work now, a search on "Асен" finds Асен Милчев, but a search on "асен" doesn't because the software has no way of telling that the Cyrillic "A" should be treated the same as the Cyrillic "а" for search purposes. Ahasuerus 21:50, 1 January 2017 (UTC)
I look forward to that day. There is a similar problem in searching for items in Japanese via kana aliases. As I understand it the only way around this currently is to search for all strings and/or create many transliterations. Uzume 11:24, 4 January 2017 (UTC)
Did you mean romaji instead of kana?--Rkihara 16:51, 4 January 2017 (UTC)
No, I mean kana. Though kana is used directly in Japanese so is kanji and even natives do not always know how to pronounce such depending on context, etc. Thus, kana is used as a means for a pronunciation guide and for indexing. As such, it is thus a form of transliteration. Romanji is just a set of Latin-based transliterations of Japanese. I am sure Russian speakers would appreciate Cyrillic transliterations of Japanese names and titles too (e.g., see wikipedia:Cyrillization of Japanese). To get back on topic, I just meant to search via kana one needs transliterations for such. Uzume 15:18, 5 January 2017 (UTC)

Disappearing cleanup reports

The number of cleanup reports dropped significantly in the last few hours. For example, Japanese Titles with a Latin Author Name, which had over thirty items on it (including several for which I could find no Japanese) earlier today, is no longer there. Did something happen to all the cleanup reports? ···日本穣 · 投稿 · Talk to Nihonjoe 06:20, 2 January 2017 (UTC)

Hmm...now they all seem to be back. Maybe I caught it when it was updating all of them? ···日本穣 · 投稿 · Talk to Nihonjoe 06:21, 2 January 2017 (UTC)
Yup. Regeneration time starts at 1 am eastern US time and finishes about 30-40 minutes later. Annie 06:28, 2 January 2017 (UTC)
Been there—done that. It is disturbing looking at and whittling down the list to see it suddenly empty and then refill but apparently that is our reality. Uzume 03:24, 4 January 2017 (UTC)
Yes, that's how the cleanup system was designed to work. However, there is nothing preventing us from improving it if we can come up with a better approach! :) Ahasuerus 04:30, 4 January 2017 (UTC)
I would have to look into how the report results are implemented and stored but we could have it so new results overwrite the old results only upon completion, thereby never leaving a period of time when there are no results available (and thus removing the experienced disconnect). Uzume 11:18, 4 January 2017 (UTC)
They are missing for 30 minutes at the most (for the last reports). I suspect that there are a lot of other places where developer time will be better spent... And as someone that works on these a lot - I prefer them disappearing and then showing up slowly than having the old one (and working off it) and having it all replaced in a second on me. This way I know when to back off and wait for the new ones and do not end up with a "so what had I gone through now" when new values get added. But it may be just me. :) Annie 20:09, 4 January 2017 (UTC)

Preface to Mary Shelley's "Frankenstein"

The preface to Mary Shelley's "Frankenstein" ends with the "signature": "Marlow, September 1817". This has been incorrectly interpreted in several of our records (11 publications) as if the name "Marlow" were a pseudonym for the true author, who is otherwise unstated (although usually believed to be Percy Shelley). In fact, the Shelleys had moved to the town of "Marlow" in March 1817, where they lived until March 1818, and this "sign off" is simply a statement of where and when the preface was written (as happens with many other similar essays). I have put some additional detail in the Percy Shelley preface record. Unless there is objection, I would like to change this title and this title from an authorship of "Percy Bysshe Shelley (as by Marlow)" to an authorship of "Percy Bysshe Shelley (as by uncredited)". This affects verifications by me, MLB, Chronsf, Gzuckier, Syzygy, Mhhutchins, Don Erikson, and Bluesman (3 times). Chavey 05:20, 3 January 2017 (UTC)

That sounds like the right move to me. Good find. Uzume 03:27, 4 January 2017 (UTC)
Completed. Chavey 13:30, 7 January 2017 (UTC)

Container Titles without a Language - 2017-01-03 update

We are down to 82 Omnibus titles without a language. I believe all of them are English titles.

I have modified the report to include collections. The updated data will be available tomorrow morning. There are about 2,000 language-challenged collections, almost all of them English. My cursory review found only one non-English collection by Harlan Ellison, but there may be a few more. If a volunteer editor(s) could review the list looking for other non-English collections, it would be great. Once we are confident that only English collections remain outstanding, I will set all language-deficient omnibuses and collections to English programmatically. Anthologies will be next. Ahasuerus 02:48, 4 January 2017 (UTC)

I wonder if all the various reports are really still necessary after things have stabilized (which might not yet be applicable here but certainly should be for some older things). As an example, eventually could this "report" not just degrade to a pre-canned search link? Uzume 03:30, 4 January 2017 (UTC)
If it is empty, it will not show up. Annie 03:43, 4 January 2017 (UTC)
That's right. You can also access our full list of cleanup reports, although I don't recall if non-moderators can see all 195 of them. Which reminds me that I need to make more of them available to non-moderators. Ahasuerus 04:29, 4 January 2017 (UTC)
Being run will catch any new titles that somehow end up in that situation - and having them nightly means that noone needs to remember to check the pre-canned link... Just thinking aloud Annie 03:43, 4 January 2017 (UTC)
True. Unfortunately, some reports lock certain sections of the database which means that the server appears to be unresponsive while they are running. It would be beneficial to take a closer look at the timings and determine if we can improve certain things. Ahasuerus 04:29, 4 January 2017 (UTC)
Make some of them weekly or monthly once they are down to one or 2 entries every few days/weeks? Annie 04:33, 4 January 2017 (UTC)
Sounds like a good idea! Ahasuerus 04:44, 4 January 2017 (UTC)
I can look through all the omnibuses, chapbooks and collections this evening and clear the non-English ones out so that everything else can get assigned automatically. Annie 18:17, 4 January 2017 (UTC)
All OMNIBUS editions in the language challenged report over here are English and can be set automatically. Collections check - tonight. Annie 18:46, 4 January 2017 (UTC)
Or during lunch. 2 non-English found and already updated; everything else is in the clear and can be set to English. Annie 19:28, 4 January 2017 (UTC)
Great, thanks! Ahasuerus 19:32, 4 January 2017 (UTC)
Without the 4 I had open and never submitted that is - done now. :) Will do one more visual pass just in case but we are good to go. Annie 19:34, 4 January 2017 (UTC)

Show printing on title list page

A proposal for the list of works on a page. Look at this page. The verified edition is 4th printing. When I look at the page, I see two entries with the same price and the same ISBN and publisher (or more than 2 for the some more popular works). The only way to find out if I need to clone or can verify one of the existing ones will be to open each one to see if my printing is there by any chance... Any way to add "printing" somewhere more visible on the list? Annie 18:16, 4 January 2017 (UTC)

It sounds like the functionality requested in FR 794. I don't think there were any objections when it was proposed. Ahasuerus 18:51, 4 January 2017 (UTC)
Yes, did not find it when I looked. Annie 19:10, 4 January 2017 (UTC)

Server downtime 2017-01-04 @8pm

The server will be brought down for automated language assignment at 8pm server time. Estimated downtime less than 5 minutes. Ahasuerus 00:38, 5 January 2017 (UTC)

The server is back up. Over 12,000 titles have had a language code auto-assigned. Next I will update the cleanup report to include anthologies and chapbooks. Ahasuerus 01:06, 5 January 2017 (UTC)
Yey! We are getting there :) Annie 01:15, 5 January 2017 (UTC)
Indeed! The cleanup report has been updated and the results will become available tomorrow morning. I expect that it will find a bit over 1,000 anthologies and around 1,000 chapbooks. Ahasuerus 02:00, 5 January 2017 (UTC)
All remaining OMNIBUS editions on the list are English. I'll clear the non-English CHAPBOOks next (probably over the weekend). Ready to list languageless EDITOR records after that? :) Annie 18:27, 6 January 2017 (UTC)
I thought the omnibuses were taken care of in the last patch. Did you, by chance, mean anthologies? Ahasuerus 21:21, 6 January 2017 (UTC)
*sigh* I do need more coffee apparently. Anthologies - yes. All 903 of them. Sorry. Annie 21:23, 6 January 2017 (UTC)
No worries and thanks for working on them! I will update and re-run the script later tonight. Ahasuerus 21:53, 6 January 2017 (UTC)
They are a good distraction when I get tired of making Finnish and Italian variants and fixing weird language assignments for the multi-language report :) Annie 22:51, 6 January 2017 (UTC)
Chapbooks are all clear for automatic English assignment. A few suspicious ones were verified to be English indeed and a couple got their language assigned differently. Annie 18:07, 8 January 2017 (UTC)
Thanks, I will take care of them in another hour or two. Ahasuerus 00:10, 9 January 2017 (UTC)
Would you also update the report to pull EDITOR records with no language? That should get a lot of short stories taken care of when they are cleared. Annie 00:42, 9 January 2017 (UTC)
That's right. The first patch will take care of chapbooks. The second one, which I expect will be installed around 9pm, will add EDITOR titles. I need to re-run nightly processing on the development server first, which takes a while. Ahasuerus 00:52, 9 January 2017 (UTC)
Done. Ahasuerus 01:47, 9 January 2017 (UTC)
All EDITOR records are English. So they need some languages assignment :) Annie 07:31, 9 January 2017 (UTC)
Thanks! I have started a new section to discuss the rest of the title types. Ahasuerus 16:14, 9 January 2017 (UTC)

Common name

What should be the common name (and main pseudonym) for this guy: here. Born in what was the Russian Empire, technically Lithuanian and lived in the USA for the last 20 years or so of his life. Do we use his Russian name or his Lithuanian name or his English name as his main name (and as his legal name come to think of it)... PS: Here he is Annie 00:00, 6 January 2017 (UTC)

We typically use the most common in-genre name, e.g. "Vladimir Nabokov" rather than "Владимир Набоков". In this case the challenge is that the spelling used by the Italian publisher responsible for his only known genre credit is unorthodox: "Mistislav Dobuzhinsky" instead of "Mstislav Dobužinskij" (Italian) or "Mstislav Dobuzhinsky" (English). I would suggest using "Mstislav Dobuzhinsky" as his canonical name and creating a VT/pseudonym. That way anyone looking for "Mstislav Dobuzhinsky" will be able to find him. Ahasuerus 00:53, 6 January 2017 (UTC)
Yeah, this is why I asked - because of that weird spelling - if his first name was properly spelled, I would have left it alone and left it as is. That sounds like a good plan. :) Thanks! Annie 01:43, 6 January 2017 (UTC)

Server downtime - 2017-01-06 @8pm

The server will be brought down at 8pm server time for another language assignment iteration. Ahasuerus 00:08, 7 January 2017 (UTC)

The server is back up after 4,000+ auto-assignments. Ahasuerus 01:05, 7 January 2017 (UTC)

Queue for Changes to Verified Pubs?

There’s a lot of variation in what changes a Primary Verifier will allow to pass without notification. Because of this, I’ve been thinking that it might be a good idea for changes to verified pubs to be automatically put in a special queue for a short period, maybe 2-3 days to allow the top level, active Primary Verifier of the pub to put a hold or pass on the changes. If there are no active verifiers or holds, then the pub passes to the submission queue. Changes by Primary Verifiers to their verified pubs go straight to submissions.—Rkihara 17:31, 7 January 2017 (UTC)

A few questions to make sure that I under the proposed functionality:
  • How will the pub's primary verifiers be notified about the proposed change? (I assume that all of them will need to be notified since there is no way of telling who is active)
  • Will only one of the primary verifiers be able to put a submission on hold?
  • Will moderators have the ability to approve these types of submissions regardless of their status?
Ahasuerus 15:54, 8 January 2017 (UTC)
  • I thought that a pass/hold option could be added to the new verifier notification feature.
  • If there's no way of telling who's active any primary verifier will do.
  • The verifier placing the hold would have to clear it for it to enter the submission queue, where it is subject to moderator approval.
  • I'm now thinking the window to hold a verified pub should be no longer than 24 hours to keep things from backing up.--Rkihara 17:58, 8 January 2017 (UTC)
Verifier notification pages are updated when a submission is approved. What you are proposing is a mechanism that would prevent submissions from being approved -- for at least 24 hours -- if the affected pub has at least one primary verification associated with it.
I think the biggest problem that we have with changes to verified pubs is that there is no way of telling what the data looked like prior to the change. If there was, then a primary verifier could easily go back, check what was changed and revert the changes if needed.
Unfortunately, a "history" system would be difficult to implement. Simple fields like "price" and "ISBN" would be easy. Contents items would be the hardest part. Not only can a submission add or edit titles, but it's possible for some of the affected titles and authors not to be in the database at some future point. A history system would have to capture a snapshot of each publication record before and after the change. That's a lot of work. Ahasuerus 00:27, 9 January 2017 (UTC)
Could a manual summary of the changes made to a verified pub be made mandatory by blocking submission to the approval queue unless something is entered into a change history field? We're already leaving notes on the primary verifier's discussion page.--Rkihara 17:07, 9 January 2017 (UTC)
Just for the record, note that in your updates to Astounding/Analog you're leaving notes on the pages of clearly inactive verifiers (Hall3730‎, Davecat, Alibrarian, Boxen) and not on those of the "active" verifiers, which is, IMHO, not very useful. Hauck 17:32, 9 January 2017 (UTC)
That's true, I'm in the habit of leaving notes with Primary 1, since there were no other primaries when I started. That's why having a mandatory change field associated with the pub would be better. It doesn't matter who's active.
For the record, I originally filled in the missing data for lot of these magazines without checking off as the verifier, then worked with Swfritter double check his verifications. Now that I'm going through again, I'm taking a primary verifier slot if open. That makes me an active verifier, and I also recheck the contents.--Rkihara 18:08, 9 January 2017 (UTC)

(unindent) So basically make the "Moderator Note" field mandatory in the "Edit Publication" form if the publication has been primary-verified? And then make sure that the value of the Moderator Note field is displayed when the primary verifier views the submission that modified his or her verified pub? If so, then yes, it's doable and relatively easy to implement. Ahasuerus 18:35, 9 January 2017 (UTC)

I was thinking of a permanent change history field, along the lines of the bibliographic field. This would allow anyone to see what modifications have been made to a pub over time.--Rkihara 18:53, 9 January 2017 (UTC)
Oh, a new "change history" field in publication records? And make it mandatory to add something to it when editing a primary-verified publication? Ahasuerus 18:58, 9 January 2017 (UTC)
Yes, that's what I was thinking of. That way there's no time limit to review changes and they're all in one place, rather than scattered across the discussion pages of the primary verifiers. I suppose there could be problems with merges?--Rkihara 19:42, 9 January 2017 (UTC)
Unlike titles, publications can't be merged, so that shouldn't be an issue.
My tentative take on this issue is that it should be doable. We will have to add a "Change Summary" field to Edit Pub and then to Import/Export and Remove Title. The field will be required for primary-verified pubs and optional for all other pubs. It will be maintained as a change log, with each change record capturing the following information: publication ID, submission ID, editor ID, editor-entered change summary. Publications will have a "View Change History" link which will take the user to a new Web page displaying this information as a list or a table.
This approach will allow editors to explain not only the "what" of each change, but also the "why". Currently this information is either not entered at all or entered in the "Moderator Note" field, which is transient. Of course, the approving moderator will need to make sure that the explanation makes sense, but that's to be expected. Overall I like it. Ahasuerus 00:26, 10 January 2017 (UTC)
I'm against this move. Adding another mandatory field is, IMHO, 1) useless as the moderating process and the obtaining (is this word english?) of pertinent information seems a quite streamlined process for the moderators that actually moderate the bulk of the submissions, 2) superfluous as the principles of notification are quite clear and just need to be collectively enforced, 3) counterproductive as this will raise another barrier to the entrance of new contributors, the entering of data is complex enough that we should aim for simplicity and ease. Hauck 08:04, 10 January 2017 (UTC)
(I would use "the obtaining".) A half-baked, pre-coffee thought: How about a slightly different approach? If there were a way to "publish" a submission to the primary verifier(s) -- prior to acceptance -- and those verifiers had a way to look at the submission and either to indicate approval/disproval or simply comment on it, that might streamline a lot of communication (and research) for moderators and for editors. The submission could also have something to indicate it is so published. Original submission could make this an additional option (i.e., a submit + publish) -- perhaps a checkbox toggled on by default for anything with primary verifiers? And the moderator review screen could also provide a way to do it. Notification to the verifiers could use that same new change notification mechanism. --MartyD 12:46, 10 January 2017 (UTC)
One other negative aspect will be a general slowing down of the approval process, either creating possible edit conflicts or simply discouraging potential contributors. To be frank, I don't really see what is the problem with the present moderation and notification process (it seems to work OK, without mishaps and with a correct communication level) but perhaps am I to immerged in it to see it from outside. Hauck 13:15, 10 January 2017 (UTC)
Looking at the first line, the original problem was the variety of things that verifiers allowed to pass without notice. The new schemes seem to be treating all changes as the same for all verifiers, leaving the onus on them to filter. The desire to avoid stalling the system means that verifiers who log on every few days will be bypassed anyway. Another sideways look at this is to add flags of various sorts for verifiers (e.g. images) and the active indicator, which would be displayed with the list of verifiers so an editor knows who to notify? Doug H 14:35, 10 January 2017 (UTC)
The edit conflicts are already bad enough - that will multiply them badly. And the software is challenging enough when you start to add this to the nightmare. I understand people wanting to protect their data and so on but this is a shared platform. Can a snapshot of all OLD values be dropped into a column somewhere so the verifier(s) can go and explore changes if need be? I almost gave up the first few days (despite moderators being around and fairly active). Annie 15:49, 10 January 2017 (UTC)
The proposed change only ensures that you log your changes, which you are already supposed to do on the verifier's page anyway. This puts "all" of the change history in one place instead of scattering it across the pages of maybe five verifiers. How hard is it to do what you're already required to do?--Rkihara 17:26, 10 January 2017 (UTC)
When you change many records where I'm PV2 and where the PV1 is inactive and do not notify me, I'm not sure that's what is required. But it's probably peripherical.Hauck 18:17, 10 January 2017 (UTC)
More than one primary verifier is a recent change, so most change notices will be found in Primary 1. I put them in PV1 for that reason, one-stop checking, instead having to look in PV1-5. Since the system flags you that I have made a change, a look in PV1 will find the note. There is no hierarchy in verification, and if there should be it seems that the "last" PV is the one that should notified if there is a change-as they've had the most recent look at the pub.--Rkihara 19:51, 10 January 2017 (UTC)
If you think that you're allowed to ignore our standard (and stated visibly on every unactive verifier's page) procedure "Otherwise, please post notices and inquiries only on the talk pages of the other primary verifiers." because you don't feel so, it's useless for you to talk about (I cite you) to do what you're already required to do.Hauck 08:49, 11 January 2017 (UTC)
Okay, since I'm making myself a primary verifier, if not already, I'll post it on my page.--Rkihara 09:00, 11 January 2017 (UTC)
People were talking about holding submissions for PVs to review - that is what I had been commenting about. Notifications of any type are not a problem of course Annie 17:55, 10 January 2017 (UTC)
The idea of holding the submission was abandoned and the discussion has moved down another path.--Rkihara 18:10, 10 January 2017 (UTC)

(unindent) 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.

Actually, checking the code, I see that it would be doable, although disk space may be a concern. Let's see. We have 784,600 Edit Pub submissions. Since 27.5% of our pubs have been verified, let's assume that 30% of Edit Pub submissions affected verified pubs. That's 235,000 submissions. Assuming 7Kb per verified publication (a mix of novels, collections, magazines and anthologies), that's 1.6Gb over a 10-year period. That's not really too bad and won't affect our disk space situation too much, at least not in the foreseeable future. So maybe it's doable after all. Ahasuerus 17:27, 10 January 2017 (UTC)

Maybe as a first step, not a full snapshot that has full links in all directions but just a snapshot of the submission as it shows on the left side (or just the red lines if possible). It won't contain changes in something done directly on publishers and internal titles but at least will have a story of what was in the field and will be better than it is now. Annie 17:35, 10 January 2017 (UTC)
I expect that a full HTML snapshot should be easier to implement than a partial snapshot. 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)
Ultimately, we need version control basically -- that is the only thing that will cover all changes from all sides :) But building that in a system not designed for it can be challenging. Annie 17:35, 10 January 2017 (UTC)
In a way, publication snapshots will be a versioning system of sorts. 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. Ahasuerus 18:42, 10 January 2017 (UTC)

2017-01-08 downtime @8PM

The server will be unavailable between 8pm and 8:05pm server (US Eastern) time tonight. I expect that about 3,800 titles will have a language code auto-assigned to them. Ahasuerus 00:28, 9 January 2017 (UTC)

We are back up. Ahasuerus 01:03, 9 January 2017 (UTC)

2017-01-09 server downtime @ 11am

The server will be unavailable between 11am and 11:05am server (US Eastern) time. Approximately 4,500 titles will have a language code auto-assigned. Ahasuerus 15:40, 9 January 2017 (UTC)

The server is back up. Ahasuerus 16:03, 9 January 2017 (UTC)

Language assignment road map

Here is a breakdown of language-challenged titles as of this morning: 

+--------------+----------+
| title type   |    count |
+--------------+----------+
| COVERART     |    19627 |
| INTERIORART  |     8764 |
| ESSAY        |     7457 |
| INTERVIEW    |      218 |
| NOVEL        |    36248 |
| NONFICTION   |     4407 |
| POEM         |      751 |
| REVIEW       |     2065 |
| SHORTFICTION |     6356 |
+--------------+----------+

It looks like all (or almost all) of them are associated with NOVEL and NONFICTION publications. I plan to change the cleanup report to include NONFICTION titles next. Once they have been cleaned up, we can tackle NOVELs. We'll probably have to do it one letter at a time. The rest of the titles are in NOVEL and NONFICTION pubs and will be handled via auto-assignment. Ahasuerus 16:13, 9 January 2017 (UTC)

Can you pull a list of what publishers are those novels from? Maybe there are some that do not even need eyeballing them... Annie 16:28, 9 January 2017 (UTC)
Yes, publishers is something that I was going to look into. Also, it's very likely that the 10,400 novels whose title starts with a "The" are English. Ahasuerus 16:48, 9 January 2017 (UTC)
Yes, every single title I looked at that starts with "The " (with the space) was English. I think you can safely just set those to English. Also, look at prices - anything in pounds is English. US$ has a few Spanish ones just to annoy us but I do wonder if we cannot just set them to English and call it a day (and errors can be fixed later?). Annie 17:01, 9 January 2017 (UTC)
As I recall, a number of early non-US publications contained a US$ price due to the way the data was originally imported. A lot of them have been cleaned up/deleted over the years, so perhaps it's not a problem any more. I guess we'll have a better idea once we process the letter 'A'. Ahasuerus 17:41, 9 January 2017 (UTC)
"A" will be very English-slanted because of the indefinite article... I'd rather see "D" as a first letter (which will get the Die/Der/Das from German or "L" (for the French) which have better chances to actually have non-English :) Annie 17:44, 9 January 2017 (UTC)
Oh yes, that's a good point. Ahasuerus 17:52, 9 January 2017 (UTC)

(inindent) A quick question - if a short story is in 2 collections and both collections are set to English, will your automation set the story to English too (provided that all that is in the collection is either English or not set)? While doing the Italian variants I keep stumbling on whole anthologies and collections where the containers are set to English but not all the stories are (or none in some cases) (because one of the containers is Italian most likely). So when I get the Italian out from the equation, will your script catch these and assign English to them? Annie 17:58, 9 January 2017 (UTC)

The script checks how many languages each pub is associated with. If it's 0 or 2+, then it ignores the pub. If it's 1, then that language is assigned to all language-less titles in the pub.
If a publication contains Italian and English titles, then the script will skip it. If and when an editor changes the pub to contain only English or only Italian titles, it will be processed by the script the next time it runs. HTH! Ahasuerus 18:29, 9 January 2017 (UTC)
Yeah, it did (for one case). An example: Senhor Zumbeira's Leg. The Italian pub should be out when my recent edit is accepted which should mean that this one gets its language set next time the script run, correct? I am figuring out if I can leave these alone or if I should set them to English while I am around :) Annie 18:35, 9 January 2017 (UTC)
That's right. I have tested it on the development server and the script assigned "English" to this title. Will the wonders of modern science and technology ever cease?! Ahasuerus 18:46, 9 January 2017 (UTC)
Great! Thanks for the confirmation. I just did not want it to be left behind and then to have to deal with it in 3 months again. Annie 18:48, 9 January 2017 (UTC)

(unindent) The cleanup report has been updated to include NONFICTION titles. The data will become available tomorrow morning. Ahasuerus 00:46, 10 January 2017 (UTC)

4357 English ones, 1 Spanish, 1 French. Once my pending are cleared, you can get the 4357 assigned automatically. Annie 17:22, 10 January 2017 (UTC)
Actually 4356 English and one in Klingon but as we do not support Klingon, English is the only option for it :). Unless if you want to add a new language "Other" which can be used for this and for any multi-language work. Annie 17:25, 10 January 2017 (UTC)
Great, thanks! I will re-run the script shortly. Ahasuerus 18:46, 10 January 2017 (UTC)
For the record: All "The " And "A " were English (still thinking on how to deal with the novels). Annie 19:24, 10 January 2017 (UTC)
Thanks. I have updated the script to auto-assign "English" to all language-less NOVELs and SHORTFICTIONs whose title starts with "The ". It should gain us almost 20,000 auto-assigned titles, which is nice. I plan to run it on the live server shortly. I will then update the nightly script to include NOVELs whose titles start with a "D" -- we'll see what it will find. Ahasuerus 19:34, 10 January 2017 (UTC)
Why don't you add the ESSAYs to the automated English for titles starting with "The " as well? The ARTs probably will also be ok (Interviews and Reviews can be cross-language (English books reviewed in FRench and vice versa) but the arts and the essays should be safe? And even the poems... Annie 19:38, 10 January 2017 (UTC)
Another good point. I have updated the script to include ESSAY and POEM titles.
I am a bit hesitant to handle COVERART and INTERIORART titles automatically because they seem to have a fair number of incorrectly assigned languages. Once we process the remaining 25,800 NOVEL titles, the auto-assignment process should take care of the rest. Ahasuerus 19:50, 10 January 2017 (UTC)
German novel with no language with a "The " cover with no language will end up as an English novel once the cover gets English and the Novel get assigned based on that. You are right, the arts will need to be after we deal with the Novels. What I was thinking of was a German novel with a set language with a cover with no language where the cover starts with a The which did not get assigned based on the novel because of something else in it blocking it. But these will wait a bit. Annie 19:57, 10 January 2017 (UTC)
Incorrect languages won't be fixed though - the only reason to be worried is the case above where a wrong assignment will also wrongly assign the title to a novel Annie 19:57, 10 January 2017 (UTC)
True enough. Ahasuerus 20:07, 10 January 2017 (UTC)
On the interviews (as we have only 218 language-challenged ones) - can you get them into the report tonight together with the "D" novels? Annie 19:57, 10 January 2017 (UTC)
Sure. There are only 108 interviews left. Here is where we stand right now:
+----------+--------------+
| count(*) | title_ttype  |
+----------+--------------+
|    13613 | COVERART     |
|     7274 | INTERIORART  |
|     4999 | ESSAY        |
|      108 | INTERVIEW    |
|    25827 | NOVEL        |
|      586 | POEM         |
|     1849 | REVIEW       |
|     4370 | SHORTFICTION |
+----------+--------------+
Ahasuerus 20:07, 10 January 2017 (UTC)
My edit to ask you for a new breakdown got blocked by you posting a breakdown :) Would you like to update the multi-language report to include (no-language, language) as a valid mismatched pair as well? Or wait until we clear the novels so the language-challenged "The " arts can be handled as well? Annie 20:11, 10 January 2017 (UTC)
Well, once we finish the language assignment process, it should be impossible to create new titles without a language. At that point I will simply change "Container Titles without a Language" to "Titles without a Language" and wait to see if any new offenders pop up. I will also handle the infamous "Undefined" language. Ahasuerus 20:18, 10 January 2017 (UTC)
Which reminds me - can we get a list of those "undefined" so we can see what these are and get them to their proper languages? Annie 20:24, 10 January 2017 (UTC)
I plan to add them to the "Titles without a Language" report (see above.) Ahasuerus 20:43, 10 January 2017 (UTC)
Somehow managed to miss the last sentence. Sorry :) Annie 21:24, 10 January 2017 (UTC)
BTW, I was surprised by the number of language-less REVIEWs left -- surely we didn't have that many reviews in NOVEL publications? -- so I spot-checked a few. It turns out that they were parents of pseudonymously published REVIEWs. I guess it means that we will have to handle them and other pub-less titles manually. Ahasuerus 20:18, 10 January 2017 (UTC)
Yeah, we have a lot of stories, essays and arts like that as well (I kept finding them and finally realized why they are still around even when the child is ok - they are not part of a container - I do not think it is just parents of pseudonymously published - it is any parent which is not part of a publication and that was created before the mandatory language was added to the DB). Annie 20:24, 10 January 2017 (UTC)

(unindent) OK, the report has been updated to include interviews and "D" novels. The data will become available tomorrow morning. Ahasuerus 22:28, 10 January 2017 (UTC)

"D" novels are cleared. 2090 English ones and 1 German one (pending its language assignment now). I will look through the interviews in a bit. If you want to add another letter, I can look at it tonight. Can we try L or I (or both) - I am chasing other language definitive articles basically... If they end up having the same number of non-English, we may be looking at a very English dominated lists (not that this is surprising) Annie 17:26, 11 January 2017 (UTC)
Interviews cleared as well (as soon as my pendings are cleared and I can finish the varianting and language assignments there). 3 needed unmerging to pull German versions out from the English ones (same name), 105 are pure English. A few inside of novels, all the others were parents that are not part of any publication. Annie 18:09, 11 January 2017 (UTC)
Approved, thanks! There are 795 eligible "I" novels and 878 "L" novels. They will be out next guinea pigs. Ahasuerus 18:33, 11 January 2017 (UTC)
Last set of variants also submitted after the last unmerge. Do you also want to throw the poems into the mix so I can slowly start clearing them as well? There aren't too many of them. Annie 18:39, 11 January 2017 (UTC)
Sure. I am running the changes on the development server and plan to deploy them within the next hour.
BTW, I have identified a scenario which may result in the creation of a language-less title even after we finish the auto-assignment project. When cloning a publication, the logic uses the language of the original pub's reference title when adding manually entered titles to the database. However, if you clone a publication without a reference title, the logic defaults to "no language". It should be a very rare occurrence since we have a cleanup report that finds publications without a reference title, but it's possible. Ahasuerus 22:48, 11 January 2017 (UTC)
So once we are done with the major language assigning, let's have a report that find languageless titles - which will be mostly empty but will catch these as well. However - that also will mean that neither the editor, nor the moderator realized that the language is missing... So even of the software can do these, I hope we won't see a lot of them remaining lagging around. I tend to go and check all of my submissions once approved to make sure something like that does not need adjusting... Annie 23:47, 11 January 2017 (UTC)
We have been slowly expanding "Titles without a Language" to cover additional title types. The day we finish the auto-assigning project will also be the day the logic will be expanded to cover all title types -- which is what you are asking for :-) Ahasuerus 23:51, 11 January 2017 (UTC)
Ah, yes. You are right. I lost track of all the reports for a little while here :)
All of the 227 items in the cleanup report "Art Titles by Non-English Authors without a Language" are artworks in English-language publications. (except for no. 102 "Science Fiction. Five Stories = Science Fiction. Fünf Geschichten" which is the cover of a bilingual pub. with no assigned language) --Vasha 23:38, 11 January 2017 (UTC)
In this case I had been assigning the smaller language - so in this case put German. :) Or in case of more than 2, whatever the publisher language is. And so on. :) Annie 23:47, 11 January 2017 (UTC)

(unindent) "L" and "I" novels cleared - 2 Italian, 3 Spanish and 1658 English one. As soon as my pending are accepted, the rest can get automatic language. Poems next. Annie 20:04, 12 January 2017 (UTC)

Poems cleared as well - all English. Most of them parents of variants or non-attached; a few are inside of novels and the rest are in the multilanguage works that are still not cleared. Annie 20:55, 12 January 2017 (UTC)
Great, thanks! Here is where we will be after the next iteration of auto-assignment:
+--------------+----------+
| title_ttype  | count(*) |
+--------------+----------+
| COVERART     |    11881 |
| INTERIORART  |     6960 |
| ESSAY        |     4906 |
| NOVEL        |    22037 |
| REVIEW       |     1849 |
| SHORTFICTION |     4337 |
+--------------+----------+
Would you like to tackle reviews next? Or perhaps some subset of novels? Here are the new counts (non-alpha characters excluded):
+--------+----------+
| letter | count(*) |
+--------+----------+
| A      |     2149 |
| B      |     1697 |
| C      |     1641 |
| E      |      828 |
| F      |     1140 |
| G      |      892 |
| H      |     1017 |
| J      |      361 |
| K      |      432 |
| M      |     1512 |
| N      |      634 |
| O      |      590 |
| P      |     1096 |
| Q      |      103 |
| R      |     1056 |
| S      |     3030 |
| T      |     1582 |
| U      |      248 |
| V      |      412 |
| W      |     1273 |
| X      |       29 |
| Y      |       78 |
| Z      |      119 |
+--------+----------+
Ahasuerus 22:24, 12 January 2017 (UTC)
Reviews and the Q, X, Y and Z novels (the smallest groups so they are off our plate)? Annie 22:30, 12 January 2017 (UTC)
Sure, can do. Ahasuerus 22:57, 12 January 2017 (UTC)
If you have a trivial way to find the non-alpha ones as well, I'd take them - if not, they will come with the last letters. Annie 22:30, 12 January 2017 (UTC)
There are approximately 100 non-alpha language-less NOVEL titles. Once the alpha titles have been processed, I'll change the logic to include all language-less NOVELs. At that point all non-alpha novels will appear on the report. As Marty would say, "Magic!" :-) Ahasuerus 22:57, 12 January 2017 (UTC)
Which is why I said "trivial" :) They will show up in an All report so not too worried :) Annie 23:05, 12 January 2017 (UTC)
And we will probably need to split these short stories and essays into groups as well I suspect- most of them will be parents that won't get assignments from anywhere else and they are too many for a single list... Annie 22:30, 12 January 2017 (UTC)
I hope that a fair number of introductions, afterwords and excerpts will be auto-assigned, but there is no easy way of telling until we get there. Ahasuerus 22:57, 12 January 2017 (UTC)
If the ones I cleared so far are any indication, I would not hold my breath. Annie 23:05, 12 January 2017 (UTC)
Come to think of it, there is a way to tell by creatively abusing the development server. If I change the language of all unassigned NOVELs to English and re-run the auto-assignment script, it should give us a rough idea of what will remain outstanding. Let me give it a try...
OK, here are the post-NOVEL/post-REVIEW counts:
+--------------+----------+
| title_ttype  | count(*) |
+--------------+----------+
| COVERART     |     2219 |
| INTERIORART  |     5442 |
| ESSAY        |     4342 |
| SHORTFICTION |     4203 |
+--------------+----------+
Not perfect, but I think it should be manageable.
In any event, I have updated the cleanup report to include reviews and the X,Y,Z novels. The data will be available tomorrow morning (or late at night for those on the West Coast.) Ahasuerus 23:59, 12 January 2017 (UTC)
Sounds about right on the textual ones. I wonder where all those INTERIORART are and why the automation does not handle them - we cannot have 5000+ parents out there, can we? Annie 00:08, 13 January 2017 (UTC)
Spot-checking suggests that some are parents while others appear in mixed language art books. Ahasuerus 00:58, 13 January 2017 (UTC)
Ah, those. I had been leaving them alone for now while clearing the rest of the multilanguage ones. Thanks for checking! Annie 01:06, 13 January 2017 (UTC)
X-Z novels are cleared - all are English. The Reviews will take a bit longer to clear :) Annie 18:41, 13 January 2017 (UTC)

(unindent) And that is all for the reviews - one need splitting (already submitted) and then varianting so the two pieces can get English and German as languages, everything else remaining is English. I did submit a few small changes (capitalization and 2 not-cleared Russian name). So the current list (all reviews plus X-Z novels) are clear for automatic English. Sorry it took me that long - real life intervened. Would you want to get me the U-W novels next? Annie 23:18, 26 January 2017 (UTC)

Excellent, many thanks! Sure, I will add U, V and W to the cleanup report shortly. Ahasuerus 03:29, 27 January 2017 (UTC)
The cleanup report has been updated. The new data will be available in another 80ish minutes. Ahasuerus 05:08, 27 January 2017 (UTC)
Now that was boring... All U-W language-challenged novels are English. Can I see the the G-K novels next? Annie 18:35, 27 January 2017 (UTC)
Sure, coming right up... Ahasuerus 19:26, 27 January 2017 (UTC)
Done. Ahasuerus 00:03, 28 January 2017 (UTC)

(unindent) G-K novels are cleared - all English :) Annie 19:15, 1 February 2017 (UTC)

Thanks! I will update and re-run the scripts shortly. I will also start a new Community Portal section since this one has become unwieldy. Ahasuerus 01:39, 2 February 2017 (UTC)

So is the chapbook that appeared at the bottom of the list: Standard Hollywood Depravity. How was that added without a language at all? I did not think that we can have this case anymore? Annie 19:15, 1 February 2017 (UTC)

<checks submission history> I see. The original NewPub submission created a proper English NOVEL title. I then went back and created an EditPub submission to change the publication and the title from NOVEL to CHAPBOOK. I also added a new SHORTFICTION title to the pub. Due to the fact that these changes were made in the same submission, the auto-assignment algorithm couldn't find a reference title, which is why it failed to assign a language. It may be possible to tweak the auto-assignment algorithm to be more robust, but even if it proves challenging, I expect that the number of exceptions should be manageable. Ahasuerus 01:39, 2 February 2017 (UTC)
Ah I see, that makes sense. I agree - we should not see a lot of these. Annie 04:20, 2 February 2017 (UTC)

Once you get those all set to English, can I see the "A" novels? Annie 19:15, 1 February 2017 (UTC)

Will do. Ahasuerus 01:39, 2 February 2017 (UTC)

Author Note field

Two items I just came across made me think that having a Author Note field would be useful to record information; see this submission which gives helpful source information for added info and this conversation about an author's language.

This type of information seems similar to how we document publications; I know we used to keep extra stuff like this in the Bibliographic Comments section of the Wiki, but it seems like we're moving towards including those things into the DB when they can be accommodated.

I didn't see a FR for this when I searched, but I could have missed it. Albinoflea 22:49, 9 January 2017 (UTC)

That would be FR 307, Move "Author" and "Bio" pages from the Wiki to the database.
The current plan is to:
  • finish the process of migrating Series/Publication/Publisher notes to the database proper
  • decide whether author records should have a single "Note" field or two fields, "Note" and "Biography"
  • implement the change
Ahasuerus 00:08, 10 January 2017 (UTC)
Ah... yes, that sounds familiar now. I must not have scrolled back far enough... Do we know how many existing Author Biography pages live in the Wiki? Albinoflea 01:34, 10 January 2017 (UTC)
As of last March:
  • Bibliographic Comments pages: 1,636
  • Bibliographic Comments Talk pages: 33
  • Biography pages: 1,896
  • Biography Talk pages: 1
Ahasuerus 02:05, 10 January 2017 (UTC)

2017-01-10 server downtime - 2pm and 3pm

The server will be down for the next iteration of language auto-assignment between 2pm and 2:05pm server (Eastern US) time. Ahasuerus 18:47, 10 January 2017 (UTC)

The server is back up. Ahasuerus 19:03, 10 January 2017 (UTC)
There will be another downtime at 3pm. It should only last 2 minutes and gain us about 20,000 auto-assigned titles. We are getting there! Ahasuerus 19:39, 10 January 2017 (UTC)
Everything should be back up. Ahasuerus 20:02, 10 January 2017 (UTC)

2017-01-11 downtime - 5pm

The server will be down for the next iteration of language auto-assignment between 5:30pm and 5:32pm server (Eastern US) time. Ahasuerus 22:05, 11 January 2017 (UTC)

The server is back up. Ahasuerus 22:32, 11 January 2017 (UTC)

Change in the publication view

Based on a small discussion here: here, here is a proposal for changing in the publications view. At the moment, when the variant of a title has a different language than the original, the software renders that as a translation in the publication content view. When the variant is an Interior Art, that does not look accurate. So the proposal is to treat is a variant and not as a translation so in Bilbon viimeinen laulu:
instead of
Bilbon viimeinen laulu • interior artwork by Pauline Baynes (trans. of Bilbo's Last Song 1990)
we will get
Bilbon viimeinen laulu • interior artwork by Pauline Baynes (variant of Bilbo's Last Song 1990)

Anyone with any opinion on this? Agreement? Disagreement? I do not care? Something else? Annie 00:08, 12 January 2017 (UTC)

That makes sense to me. Albinoflea 19:58, 12 January 2017 (UTC)
Sounds good. --Vasha 22:46, 12 January 2017 (UTC)
That's appropriate. While I know there are some instances of art with English text that has been converted to the same art with, say, German text, and hence should properly be viewed as a "translation", I believe that is rare, and could be handled by a note. Chavey 02:12, 13 January 2017 (UTC)
I'd think that shouldn't be so even in such a case: the moment we view something as INTERIORART we decide to drop the focus on the language; and a note like 'Involved text was translated by Gandalf Gremlin' would still be an option.
It'd be possible to view a comic or even a cartoon as SHORTFICTION (with illustrations), but I think there were good reasons to not pursue this line of thought. Stonecreek 07:05, 13 January 2017 (UTC)

(unindent) Looks like a consensus to me! The software change has been made and deployed. Ahasuerus 21:12, 13 January 2017 (UTC)

Poems not attached to publications

Before I submit deletions for the ones I am finding (a lot of them have no language assigned - which is how I am finding them), is there any reason to keep in the DB poems like this one: A Description of Morning when they are not part of a publication and are not a parent for a variant that is part of a publication. Thoughts? Annie 20:36, 12 January 2017 (UTC)

If it's a valid genre work, it should remain. It's possible to know a work is genre, but not have information regarding the specific publication. Ideally, there would at least be a note with some information why the record is present, but doesn't always happen. -- JLaTondre (talk) 21:08, 12 January 2017 (UTC)
I am not sure that some of the ones I saw were valid genre works but I'll leave them alone. I think we should at least have a "first published in" note for these... Annie 21:14, 12 January 2017 (UTC)
"A Description of Morning" is available on-line and is clearly non-genre. I deleted it. -- JLaTondre (talk) 01:40, 13 January 2017 (UTC)

2017-01-12 server downtime - 5:45pm

The server will be down between 5:45pm and 5:47pm server (Eastern US) time. Ahasuerus 22:25, 12 January 2017 (UTC)

Done. Ahasuerus 22:47, 12 January 2017 (UTC)

Fictionalized biography of Edgar Allan Poe

A question about inclusion / exclusion. We include non-fiction biographies of important authors. I couldn't find anything that seemed to address the issue of a fictionalized biography. The book "Dark Glory", by Dorothy Dow, is such a work. "A Companion to Poe Studies" says the book "confines its fiction principally to imaginary conversations and to narrative commentary and interpolation". At least one source I found lists the book as non-fiction, although it really isn't. Does this book belong in here? Chavey 02:07, 13 January 2017 (UTC)

If the truly non-fiction equivalent is "in", I'd say this should be, too. It's still about Poe. --MartyD 02:48, 13 January 2017 (UTC)
And I included such an item for H. G. Wells, because of his importance. Stonecreek 07:07, 13 January 2017 (UTC)
Thanks for the advice. Poe is certainly at a fairly high level of importance. I'll add the book. Chavey 05:40, 14 January 2017 (UTC)

Spanish speaking editor up for a challenge?

The stories here Fábulas need to be replaced with their Spanish variants. My google-fu does not seem to work very well for that one so can someone that speaks Spanish take a look and see if they can find the names of the stories and who translated each (if possible)? I don't even mind doing the varianting and replacing if someone can get the data and post it... Annie 20:15, 13 January 2017 (UTC)

I do not believe that the complete table of contents is online anywhere. [www.nexos.com.mx/?p=25143 Here] is one of the translated stories, but using its title as a clue didn't allow me to find the rest. If no one else speaks up, I will put in an ILL request; there's a copy at a public library in my state. --Vasha 21:20, 13 January 2017 (UTC)
Thus the challenge part of my question :) Thanks! Annie 21:31, 13 January 2017 (UTC)
It's a later edition from Lectorum, but I think this is the same collection. The prologue is dated Buenos Aires, 1983 (although it is credited to both Borges and Alifano). I will enter this edition and match the English and Spanish titles. We can see how its contents correlate with that other pub we have. --MartyD 12:30, 16 January 2017 (UTC)
The contents matched 1-for-1, so I swapped out all of the English titles for the Spanish. --MartyD 14:11, 16 January 2017 (UTC)

Tor.com / Request: Add The Witch Who Came In from the Cold

The February 3 issue of Tor.com, which I have added, has a review of the first three episodes of the Serial Box serial The Witch Who Came In from the Cold (not yet in the DB). Now I am not sure how to add that serial; but their previous production Bookburners provides a model. Would anyone like to enter The Witch... into the DB and link the review to it? (Currently I have the Tor.com article as ESSAY not REVIEW). --Vasha 02:44, 16 January 2017 (UTC)

Am I missing something? Why are we including tor.com's non-fiction? It is a blog, not a webzine. We include their fiction under the SFWA recognized market exception since it's by notable writers and/or frequently later collected in publications. I don't see the point of including their Batman, Star Trek, etc. rewatches and other non-fiction. Unless there is a downloadable version that I'm not aware of, we should be sticking to only the fiction. -- JLaTondre (talk) 11:35, 16 January 2017 (UTC)
that would make things easier... I have been entering everything on the assumption that it was being called a magazine. Are there other cases of something that is not a magazine where the fiction is entered? --Vasha 14:18, 16 January 2017 (UTC)
I know I have accepted some of these additions on the mistaken understanding that it's both the website and emailed newsletter. Upon further review, I see now the newsletter is something different. --MartyD 14:27, 16 January 2017 (UTC)
What exactly is the "periodical" that publishes the fiction? I think that ought to be clarified. There should be someplace in the DB to note that. If the stories are not being published in any periodical, how are they to be catalogued?
Here's a thought. These stories are always published individually as e-chapbooks. We do catalogue them in that form, as by the publisher Tor.com (which also puts out novellas). Stories appear simultaneously on the blog and in chapbook form. If we decide that the Tor.com blog is not a webzine, then maybe we should not list the appearances of stories there, but only their chapbook publication. --Vasha 20:19, 16 January 2017 (UTC)
Tor.com's fiction was original entered (per my impression) as a webzine as we have no way to handle a blog and it was the closest thing. While they may also release them as ebooks now, that wasn't the case originally. For example, the first story we have indexed is After the Coup which was published on tor.com in 2008, but not released as an ebook until 2010. Treating the original posts as a webzine seems reasonable to me given our explicit criteria that we will carry SFWA recognized markets, but extending that to all their blog posts doesn't. A note can be added to the Tor.com to explain how and why we are handling these somewhat differently than normal. I've put your current edits on hold pending further community feedback. If no one else argues for inclusion, I will go through and remove the non-fiction. Please don't take this negatively, we do appreciate your attempt to make the database more complete and accurate. -- JLaTondre (talk) 22:05, 16 January 2017 (UTC)
I vote for the deleting of the non-fiction. If we want to list every essay about SF available on the web, we're going to be flooded by the material (I'll be even able to enter the 600+ entries in my blog, great!). Hauck 18:32, 17 January 2017 (UTC)
I agree - tor.com do not have a webzine the way Strange Horizons does for example (their non-fiction is part of their webzine). As much as I like completeness, allowing these in opens the door for allowing reviews from non-genre media for genre books in and that will spiral out very fast. Annie 18:47, 17 January 2017 (UTC)

(outdent) As there have been no objections, I have removed the non-fiction. I have also added a note to the series record. -- JLaTondre (talk) 03:23, 19 January 2017 (UTC)

Thanks... sorry to make extra work for you. After I finish adding the contents to this year's Clarkesworld, I will put in the remaining Tor.com short fiction. --Vasha 03:46, 19 January 2017 (UTC)

2017-01-16 downtime - 6:30pm

A big patch will be installed at 6:30pm server (US Eastern) time. The server will be down for a few minutes. Patch notes to follow. Ahasuerus 23:12, 16 January 2017 (UTC)

The patch has been installed and the server is back up. As per FR 163, the "storylen" field has been split into 4 fields:
  • a "Story length" field proper
  • a "Content" field for omnibus designations like "1-4". Please note that you no longer need to enter "/" as the first character.
  • a check-box for "juvenile"
  • a check-box for "novelization"
The "New publication" page has been adjusted to support entering these values directly, without having to wait for submission approval. There are new cleanup reports for parent/variant mismatches associated with the new fields. Lots of other data entry forms - "Make Variant", "Add Variant", "Merge Titles", "Unmerge Titles", etc -- have been adjusted. The "juvenile"/"novelization"/Content data has been moved from the old "story length" field to the new fields.
If you see anything unusual, please report your findings here. I will update Help shortly. Ahasuerus 23:43, 16 January 2017 (UTC)
Hurrah! thanks for this. --Vasha 23:44, 16 January 2017 (UTC)
BTW, I would like non-genre to be at the top of the list; I use it much more often than the others. I don't care about the order of the others. --Vasha 23:50, 16 January 2017 (UTC)
Sure, that's easy to do. Unless there are objections in the next 24-48 hours, I will make the change on Wednesday. Ahasuerus 00:05, 17 January 2017 (UTC)
Done. Ahasuerus 20:41, 18 January 2017 (UTC)

(unindent) AddVariant is missing the checkboxes. Will it inhetrit all the new values from the mother work (juvenile/novelization)? Annie 20:33, 18 January 2017 (UTC

That's right. When you create a new title record using Add Variant or Make Variant, the new record inherits the original record's juvenile/non-genre/graphic/novelization/content values. Ahasuerus 20:40, 18 January 2017 (UTC)

Further "story length" changes

The "Story Length" field is now a drop-down list in all data entry forms. Moderator review pages still refer to its values as "ss", "nv", and "nt", but I expect to change everything to "short story", "novella" and "novelette" in the near future.

The name of the field has been changed to "Length" to make it consistent with the way it appears in Edit Publication. More streamlining of various Web pages to follow. Ahasuerus 18:03, 17 January 2017 (UTC)

Query to other editors; would you like to see the options rearranged in the order ss, nt, nv (I would) or are you too used to the existing order? --Vasha 18:43, 17 January 2017 (UTC)
novelette is the least common format (at least with the things I seem to read) so I would rather have it at the bottom. But I can live with them reordered Annie 18:48, 17 January 2017 (UTC)
I have been entering dozens and dozens of anthologies and magazine issues lately; I'd say they are 95 percent short stories, 5 percent novelettes, and almost no novellas. Where are you finding so many novellas? --Vasha 19:06, 17 January 2017 (UTC)
Separate chapbooks from a few publishers for the most part (favorite length so I am seeking them out usually). Novellas are hard for anthologies because of their length - so I am not that surprised at your stats. As I said - I can live with it either way. Annie 19:13, 17 January 2017 (UTC)

(unindent) As of Saturday morning: 

+----------+----------------+
| count    | storylen       |
+----------+----------------+
|    39561 | nt             |
|    19226 | nv             |
|   181918 | ss             |
+----------+----------------+

19:30, 17 January 2017 (UTC)

The latest batch of "story length" changes was installed a few minutes ago. It rearranged the "Add Variant" page to look the way Edit Title looks. It also changed all occurrences of "Story Length" and "Storylen" to "Length" to make it consistent. Ahasuerus 20:26, 18 January 2017 (UTC)
In reply to Vasha above, I prefer to see the existing order in the drop-down menu (novella, short story, novelette) because with short story in the middle it neatly separates the two similar words, and inclines editors away from unintentional mistakes. Am I alone in always thinking this was a cunning design trick? ;) PeteYoung 20:50, 18 January 2017 (UTC)
Hm, that's a good point. You're probably right. --Vasha 23:09, 18 January 2017 (UTC)
I always have to look up the definitions of novella and novelette, because I can never remember which one is shorter. So I, at least, would strongly prefer that they be listed in length order. Chavey 04:02, 21 January 2017 (UTC)
The longer the name, the shorter the story :) I used to mess up novel and novella when I was still learning English. So had to find a way to differentiate the two and the length of the word helped. Then I learned that there is a novellete as well - and the rule for the lengths saved the day. Annie 05:06, 21 January 2017 (UTC)
Thanks! Chavey 05:25, 22 January 2017 (UTC)

2017-01-17 server outage - 4:45pm

The server will be briefly unavailable around 4:45pm server time. It should be back up within a minute or two. Ahasuerus 21:32, 17 January 2017 (UTC)

Sorry, that was supposed to be "4:45pm server time". Ahasuerus 21:40, 17 January 2017 (UTC)
The server is back up. A couple of cleanup reports have been adjusted to work with the new "story length" logic. Some very old titles with messed up story length values have been corrected. Ahasuerus 21:47, 17 January 2017 (UTC)

German editor with some time?

The PV is not active anymore so cannot ask him... And I cannot find the German content online. So can someone find the German titles of Im höchsten Grade phantastisch? Thanks! Annie 19:09, 18 January 2017 (UTC)

The available sources seem only to list the nonfiction as such, not the contents. I'll see if I can get hold of a copy, but in any case this'll take some time. Stonecreek 12:18, 22 January 2017 (UTC)
Done. Stonecreek 18:24, 13 February 2017 (UTC)
Wonderful - thanks! Annie 18:31, 13 February 2017 (UTC)
Glad to have been of some use! (Back when this book and others were added & verified the use of the English language was state of the art: there wasn't any other option). Stonecreek 19:20, 13 February 2017 (UTC)

Removal of non-genre and graphic check-boxes for REVIEWs and INTERVIEWs

The "non-genre" and "graphic format" check-boxes are no longer displayed when editing REVIEWs/INTERVIEWs. Ahasuerus 18:56, 19 January 2017 (UTC)

And here I was looking forward to seeing some comic book reviews of speculative fiction works (*laugh*). Uzume 00:10, 27 January 2017 (UTC)

Coming patch and pending submissions

I have a fairly big software patch almost ready to go. It will wrap up most of the outstanding issues with the "story length" field, including finally changing "shortstory" to "short story". I plan to install it on Monday at 9pm server time. Unfortunately, the patch will invalidate any pending submissions which affect "story length" values; they will need to be rejected. Please plan accordingly. Ahasuerus 00:01, 22 January 2017 (UTC)

Change Empire Star to novella?

We currently have Empire Star by Samuel R. Delany as a novel (presumably because the halves of Ace Doubles are conventionally considered novels) but it's less than 27K words. If people want to change it to a novella, I will be happy to do the work. The same with another by the same author, The Ballad of Beta-2, which is 30K words. --Vasha 22:10, 23 January 2017 (UTC)

This has been raised before in relation to other works, although not specifically Empire Star and The Ballad of Beta-2 as far as I know. Yes, they are technically novellas as are probably the majority of 'Ace Double' titles that we currently have listed as novels, but to make such a change would first mean having to notify the many PVs of your intention that would change all their publications. You'd also need to change many records into CHAPBOOKS, create title records for each one and merge. And then there are the translations to consider. The simpler and far less labour-intensive option is just to make a note in each of the title records (1, 2) about the word count. I can't speak for other editors but that would be my recommendation. PeteYoung 04:16, 24 January 2017 (UTC)
Well, as is often the case of many-published works, many of the PV1s aren't active anymore, and I'd think that most of the other PVs didn't think about the works length when verifying. So, as the community portal is exactly for communal decisions, I'd say it'd be okay to change the titles to novellas if there isn't a neglection within the next few days. Our rules of categorization are to the point and we shouldn't deviate without a reason from it (the reason for the category 'novel' here, I think, is that the publisher called them so; this is quite often only a marketing ploy: novels sell better than novellas). Stonecreek 04:49, 24 January 2017 (UTC)
Okay, I propose a new operating procedure. As the titles seem to get only copied and verified by editors without taking a second look at the actual length, there seems to be the practicable way of asking the first active PV and possibly making the change following the okay. Stonecreek 17:29, 13 March 2017 (UTC)
No, there's now a trend to proceed this way (only notifying one PV on nebulous criteria -active?, 1st?- and leaving the others "unnotified") that is presently particularly annoying (it's used now for additions of letters to magazines). All (active) PVs are equals and should be notified, we can't have a process based on the hope that every PV follow all what happens on the "recent changes" page and that they remember that they PVed this pub. I'm perfectly aware that Empire Star and a lot of other texts that I PVed are novellas listed as novels but I followed the usage of the times. Hauck 18:38, 13 March 2017 (UTC)
I didn't say I'd not notify them, but only for asking the first active PV. Most PVs seem to be only seldomly active (or not at all anymore), so that's a way that things never get done. This would mean an active active PV (somebody who's around at least a few times a week) and a note here on the community portal. Stonecreek 19:59, 13 March 2017 (UTC)
That's what bother my old egalitarian self. Why only ask "the first active PV" (as if his/her opinion carried more weight) and just who is "the first active PV" anyway? (the one with the lowest number?, the one with the lowest number multiplicated by the difference between his last posting and today divided by the square root of his/her number of contributions?) or more simply what is an "active PV"?.Hauck 07:57, 14 March 2017 (UTC)
But it is just a notification - what's the problem in posting on all PVs pages - even if one does not wait for them to respond? Someone that is here once a month is more likely to read their page than to wade through all the chatter we produce on the Community Portal. By all means - not waiting for everyone to agree is fine (especially with an agreement in the Community Portal) but not notifying everyone is poor etiquette (I think). Annie 20:05, 13 March 2017 (UTC)
That's exactly what I meant! Thanks, Annie. Stonecreek 20:06, 13 March 2017 (UTC)
Indeed you did - I think I blanked through part of your last message - sorry. Not enough coffee again aparently :) I was mostly agreeing with Herve above that not notifying is not very nice. Annie 20:34, 13 March 2017 (UTC)
If I took a look at my user page after an absence, I'd like seeing messages about major changes to my verified works (I've often regretted that the DB isn't a wiki with page histories). But I consider that to be nice rather than essential. And admittedly, notifying a bunch of people is work. --Vasha 20:41, 13 March 2017 (UTC)
SFE3 calls these two works novels. I don't have the other secondary sources to check, but I'm willing to bet that most or all of them do the same... Right? That right there is a reason for leaving them as they are; however, since ISFDB aspires to be a resource for casual browsers who aren't familiar with the publishing practices of the 1960s, adding a note is a necessity. --Vasha 20:41, 13 March 2017 (UTC)
If you prefer to let these two as they are, that's fine with me. Chistian Stonecreek 21:18, 13 March 2017 (UTC)
How's this for a note? "This story is 27,000 words long, thus it is a novella by the ISFDB's usual classification. However, it is listed as a novel due both to the publisher having originally marketed it as such (in common with many other very short "novels" at that time), and standard bibliographies listing it as a novel." --Vasha 21:31, 13 March 2017 (UTC)
Drop "ISFDB" from that sentence -- this is the standard classification, not something invented for this DB. Or say something like: "that standard classification (link to it) blah blah". Otherwise the note sounds as if we just came up with some categories here... However - we had been entering a lot of the old dos novels into their correct types (even if the publisher called them novels) so isn't this against the current practice? I would say that we should list it as what it is, not what it is marketed as. Just saying. Annie 21:37, 13 March 2017 (UTC)
Also possible would be to change it and enter a note that this is a NOVELLA, but the publisher and some sources list it as a NOVEL. (It's really the most common publisher's trick to call a text a novel which ain't one). Stonecreek 21:44, 13 March 2017 (UTC)
I don't work much with older publications, so I'd rather leave it to those who do to decide this matter. Those of you who've entered dozens or hundreds of 1960s novels and dos-a-dos publications into the database, what do you want to do? --Vasha 21:57, 13 March 2017 (UTC)
On a related note, with the new "My Changed Primary Verifications" feature, would we necessarily need to notify all of the PVs? The system will advise them when there has been such a change. Just curious, since I thought that was the main purpose of that new feature. ···日本穣 · 投稿 · Talk to Nihonjoe 22:17, 13 March 2017 (UTC)
A notification usually specify what had been changed; the new feature merely highlight which fields are changed (and there is no way to see what was the old value). So yes, notifications are still pretty much part of the etiquette. :) Annie 22:38, 13 March 2017 (UTC)

2017-01-23 downtime at 9pm

As per the note above, the server will be brought down at 9pm server time. It should be back up within 5 minutes. Please note that any outstanding submissions which aim to change "story length" values will be rejected at 8:59pm. Ahasuerus 00:50, 24 January 2017 (UTC)

The server is back up. It was a beefy patch, so if you see anything unusual, please post your findings here. Ahasuerus 02:02, 24 January 2017 (UTC)
Um, yeah... I just tried to submit a new magazine issue and got "Error: invalid shortfiction length". --Vasha 02:10, 24 January 2017 (UTC)
Could you please try to do a full page reload using Control-F5? There was a change to one of the JavaScript files and your browser may still have the old version loaded. Sorry about the hassle! Ahasuerus 02:13, 24 January 2017 (UTC)
In the meantime, here are the patch notes:
  • "shortfiction" has been changed to "short fiction"
  • All occurrences of 'ss', 'nt' and 'nv' have been changed to fully spelled out words
  • A lot of changes under the hood
Ahasuerus 02:16, 24 January 2017 (UTC)
I restarted my browser and now submission works. --Vasha 02:21, 24 January 2017 (UTC)
Great! Ahasuerus 02:36, 24 January 2017 (UTC)

2017-01-26 server downtime at 11:55pm

The server will be unavailable between 11:55pm and 11:58pn server time. Ahasuerus 04:26, 27 January 2017 (UTC)

The server is back up. Ahasuerus 04:57, 27 January 2017 (UTC)

2017-01-27 server downtime at 3:10pm

The server will be down between 2:40pm and 2:43pm server tome. Ahasuerus 19:30, 27 January 2017 (UTC)

The downtime has been rescheduled for 3:10pm due to technical issues. Sorry about the inconvenience. Ahasuerus 19:58, 27 January 2017 (UTC)
The server is back up. Ahasuerus 20:12, 27 January 2017 (UTC)

Spaces after author names

In accordance with this conversation, I have a question for editors: currently, in some places there is an extra space after author names before punctuation like commas and brackets, like [as by Allen M. Steele ]. To me, this looks really weird and offputting & I'd recommend removing it; however, Ahasuerus says that some people find it useful. If that's you, please speak up in favor of keeping the space. --Vasha 22:13, 28 January 2017 (UTC)

As I recall, the argument that was advanced against this change the last time the question came up was that having an extra space made copy-and-paste operations easier under Windows. Ahasuerus 23:30, 28 January 2017 (UTC)
Hearing no objection, so ordered. Ahasuerus 00:07, 8 February 2017 (UTC)

Lisa Tuttle

Lisa Tuttle's 2 short stories: "The Curious Affair of the Deodand" and "The Curious Affair of the Dead Wives" are in the same "Jesperson and Lane" series as her novel The Curious Affair of the Somnabulist and the Psychic Thief. —The preceding unsigned comment was added by RogerSSS (talk • contribs) .

I've added them to the series, but you could do that yourself-- the place for indicating the series is in the title record, right below the date. --Vasha 07:59, 30 January 2017 (UTC)

Is Book Smugglers a blog?

It seems to me that the Book Smugglers' online presence is in the same situation as what we discussed for Tor.com -- they're a blog rather than a webzine. Would you folks agree? And they're only semipro, so no exception made to index their fiction. I've only been adding their online fiction if they also published it as a chapbook (they didn't always). --Vasha 22:03, 1 February 2017 (UTC)

As the current rules stand - yes, they are a blog so not eligible except if there is a special case for it. Annie 22:49, 1 February 2017 (UTC)

2017-02-01 server downtime at 9pm

The server will be unavailable between 9pm and 9:05pm server time. Ahasuerus 01:40, 2 February 2017 (UTC)

The server is back up. Ahasuerus 02:02, 2 February 2017 (UTC)

Language-less NOVEL cleanup

Here is where we stand as of this evening: 

+------+----------+
|letter|    count |
+------+----------+
| S    |     3001 |
| T    |     1571 |
+------|----------+

The letter 'A' will be added to the report shortly. Ahasuerus 02:21, 2 February 2017 (UTC)

Done. Ahasuerus 04:12, 2 February 2017 (UTC)
"A" novels cleared. 2 French ones (waiting to be approved) and a few changes in cases when I spotted them which will get English this way when that gets approved... The rest can get automatic English. How about N-Q next? Annie 19:06, 2 February 2017 (UTC)
Approved and thanks. N-Q coming right up. Ahasuerus 21:22, 2 February 2017 (UTC)
N-Q novels cleared. Nothing non-English in the whole list (some interesting titles though - the genre authors have weird imaginations...). So off with their heads... uhm - I mean, let's assign them all the language automatically. :) E-F next? Annie 18:11, 3 February 2017 (UTC)
N-Q done. The cleanup report has been adjusted to include E and F . Also, I have removed the processed letters from the table above. Ahasuerus 02:38, 4 February 2017 (UTC)
E-F novels are cleared - all English. Even the French sounding one that gave me some hope that something is not English in this group... How about B next? Annie 16:28, 6 February 2017 (UTC)
E and F have been auto-assigned. Thanks for working on them! B should be available tomorrow morning. Ahasuerus 17:03, 6 February 2017 (UTC)
All language-challenged "B" novels are English... :) "C" next? Annie 17:57, 8 February 2017 (UTC)
Done. "C" is in progress and will be available tomorrow morning. Ahasuerus 18:47, 8 February 2017 (UTC)
"C" are all cleared. English all the way down the list (a few formatting issues fixed while I was looking at them anyway). M and R next? Annie 16:17, 9 February 2017 (UTC)
Done and thanks again. M and R will be available tomorrow morning. Ahasuerus 18:38, 9 February 2017 (UTC)
M and R are cleared. S next? Once they are cleared as well, then it will be all remaining novels (T plus the non-letter ones).Annie 18:16, 10 February 2017 (UTC)
Done and done! Ahasuerus 18:54, 10 February 2017 (UTC)
S novels are all clear for automatic English. Time for all novels :) Annie 17:05, 13 February 2017 (UTC)
Done and done. Once the rest of the novels are wrapped up, we will have the following title types left:
+--------------+----------+
| COVERART     |     2961 |
| INTERIORART  |     5546 |
| ESSAY        |     4331 |
| SHORTFICTION |     4154 |
+--------------+----------+
The actual counts will be somewhat lower due to auto-assignment. Also, we could break them down further. For example, a search on ESSAYs which contain words like "Introduction" "Foreword" and "Note" results in approximately 600 hits . Ahasuerus 20:17, 13 February 2017 (UTC)
Except that I had found some Introductions and Forewords that had the English word but were not English (mainly Italian). I'd fix the name on some when I know how but still... So I am not comfortable relying on that alone to call them English... If you mean to allow splitting, a 4K or so list is workable so no need to split those. I think I would take short fiction next - later this week I suspect :) Then Essay, then cover art and at the end we can start chipping at the interior art. Annie 20:47, 13 February 2017 (UTC)
Right, I am aware of non-English "Introductions". I was thinking that making the lists shorter would facilitate the review process, but if lists with 4K titles are manageable, then so much the better! Ahasuerus 21:01, 13 February 2017 (UTC)
Sorry, I had been just playing with all those languages and my mind is going to automatic languages immediately :) I prefer smaller ranges (1-2K) but 4K is manageable for a few ranges :) Annie 21:32, 13 February 2017 (UTC)
And any chance I can get the 234 Undefined ones at some point so I can vet which ones are not English and we can have the rest reset to English? :) Annie 20:45, 3 February 2017 (UTC)
I plan to change all empty string ("Undefined") values to nulls, which will make them appear on the cleanup report. Ahasuerus 02:38, 4 February 2017 (UTC)
Ah, that makes sense (and will probably make it easier to fit with the current process. Thanks! Annie 02:57, 4 February 2017 (UTC)
Any idea when? :) Annie 19:10, 10 February 2017 (UTC)
I plan to convert them right before I change the nightly cleanup report to include all NOVELs. Ahasuerus 19:49, 10 February 2017 (UTC)
Should have waited one more day and it would have become clear. :) Always happens this way it looks like. Thanks! Annie 20:54, 10 February 2017 (UTC)

(unindent) Too many dots so time to unindent. All novels are cleared and require a bit of magic to turn them back into pumpkins... :) Next - short stories.

Excellent! Here is where we stand with the NOVEL titles out of the way:
+--------------+----------+
| COVERART     |     2186 |
| INTERIORART  |     5411 |
| ESSAY        |     4283 |
| SHORTFICTION |     4143 |
+--------------+----------+
Given the much reduced number of COVERART titles, do you still want to attack short fiction first? Ahasuerus 22:26, 14 February 2017 (UTC)
Art gives me a headache with the number of splits and multi-language books - so I would rather deal with the text elements and clear all of them first and then brave the arts :) Annie 22:32, 14 February 2017 (UTC)
Sounds good -- short fiction coming right up! Ahasuerus 22:48, 14 February 2017 (UTC)
Short fiction cleared - there are 4 language assignments pending I think and a few deletions in my pending queue - anything else on the language-challenged list is English. I think I want COVERART next or I will end up with 7500 art titles left at the end... and that won't be much fun :) Annie 00:15, 18 February 2017 (UTC)
2,100 COVERART titles coming right up! Ahasuerus 02:10, 18 February 2017 (UTC)
Can you get me some approximate numbers for interior art and essays groups if we split each of them in 3 groups (A-H, I-S, the rest)? The 4K group is manageable but a bit overwhelming (plus this way I will be able to alternate text and art)... Annie 02:55, 18 February 2017 (UTC)
Sure thing. I'll post the breakdown tomorrow. Ahasuerus 05:05, 18 February 2017 (UTC)

By the way - did you ever reset the Undefined ones to null? If you did - then all of them are one of the internal types we still have hanging - as today's list did not bring any surprise visitors. Annie 21:55, 14 February 2017 (UTC)

That's right, "undefined" is no more! Ahasuerus 22:26, 14 February 2017 (UTC)
So internal ones only - I do wonder if most/all were not parents or something. Oh well - not that it matters if they never show up again. And I see that Italian finally pulled in front of the unassigned. :) Annie 22:32, 14 February 2017 (UTC)

2017-02-02 downtime - 5pm

The server will be unavailable between 5pm and 5:03pm server time. Ahasuerus 21:23, 2 February 2017 (UTC)

The server is back up. Ahasuerus 22:01, 2 February 2017 (UTC)

Verified pubs without story lengths

This is not addressed to any one person... As I have been going through double-checking 2016 short fiction, I have been finding numerous (several dozen) anthologies and magazine issues marked "primary verified" which do not have the lengths of the stories indicated. That seems like a problem. Personally I would never mark a publication verified unless I had determined the story lengths. --Vasha 01:28, 3 February 2017 (UTC)

Well, ideally all verified publications would have "length" values assigned. However, determining accurate word counts can be a time-consuming process due to different fonts, different page layouts, borderline cases, etc. If a potential primary verifier has the time to verify everything else about the pub but not each story's length, it may be better to have a primary verification on file even without the length values. Ahasuerus 01:48, 3 February 2017 (UTC)
I agree that print books are a real headache-- I have spent way more time than I wanted to painstakingly counting in those. For ebooks, you have to know how to use Calibre or some such program in order to count. But at least there is no excuse for not cutting-and-pasting online stories into a word counting app!
Maybe someone could chime in with some tips for how to get word counts from ebooks in various formats? --Vasha 01:53, 3 February 2017 (UTC)
If I have the book and I am not sure at the lengths, I would still verify it and then try to figure out lengths if I can after that. Some people verify without even adding content so your mileage may vary. :) The verification says "I have the book/magazine" so if someone needs clarifications, I can pull it out and answer questions. As for the e-books and sizes - depends on the publications - a LOT of them add number of words in their stories (good people); in other cases once you know the number of words of one of the story, you can figure out the rest based on kindle pages (for example). And sometimes just sending a note to a publisher or an author will get you that one example to allow to untangle the rest. Annie 02:32, 3 February 2017 (UTC)

Note that as per this discussion, some caution on the subject must be in order so any excess of zeal must be avoided. Hauck 07:21, 3 February 2017 (UTC)

Add Gender to Author Edit?

I think things have cooled off enough to return to the subject of adding gender identification to the author edit. Hopefully we can have a calm discussion and a straight up/down vote on it.--Rkihara 19:35, 3 February 2017 (UTC)

I'd say that it will be useful to have it as long as in addition to the standard two, we also have N/A (for pseudonyms and the like) and a category for the cases where someone is shifting or not declaring or whatever other reason for not knowing it (LibraryThing uses "other/contested/unknown" for example for this case) which is different from the value when no gender had been selected yet. Plus a policy about what the genre of an author should be set to if they changed it in their lifetime so we do not end up in a constant editing mode when editors disagree :) Annie 19:42, 3 February 2017 (UTC)
I am not seeing why this would be a particularly useful addition, actually. Yeah, sometimes people want to look up only stories by women or nonbinary people or whatever, but there are other sources for that. Having to figure out what gender people prefer to be referred to as, or figure out if they've changed, can be a headache. More importantly, if you get it wrong, people feel it as a personal insult, more so than if you get other biographical data wrong! --Vasha 20:17, 3 February 2017 (UTC)
Some of these issues were discussed back in 2013, but the discussion became too heated and then petered out for lack of consensus. Ahasuerus 20:34, 3 February 2017 (UTC)
So are you against adding gender to the DB at all or against allowing gender to be set to something different than male/female for a person? If the latter, how do you propose to assign gender in non-binary cases? Annie 20:22, 3 February 2017 (UTC)
No, I actually like having it not specified now. Yes, we should either list the exact gender that the person gives for themself or just neither of the above if we implement it. But I think the risk of getting it wrong is great, and the usefulness less by comparison.
If we don't change the gender listed on a trans person's page it will seem like a refusal to recognize them, even if it's only the result of slowness. We already had a discussion where people didn't want to change a trans person's name until they had been using the new name for "a while" however long that was. Will the same apply to gender? --Vasha 20:36, 3 February 2017 (UTC)
That was a discussion of an author's canonical name, not a legal name. Canonical names are "the most recognized name for that author. The canonical name may be a pseudonym". For this reason our canonical name for Alice Sheldon is "James Tiptree, Jr." Ahasuerus 20:59, 3 February 2017 (UTC)
OK fair enough, but there's still the headache of figuring out what gender people prefer. As I say, I'm against adding it at all, but I think that if we do, there should be a standard saying that it should not be specified on a living author's page unless determined from the author's website, a recent self-contributed bio, or some such. --Vasha 21:08, 3 February 2017 (UTC)
Let me quote something that I wrote during the last iteration of this discussion:
  • When choosing new fields for a record, I consider a few factors. Fist of all, the data, if known, must be unambiguous. One could argue that the place where a person was "raised" is more important than the place where he or she was born -- and many authors list the former rather than the latter in their autobiographies -- but it's too nebulous for us to use because in many cases there is more than one location.
  • The second factor is relevance, e.g. DOB and DOD are relevant because they show when the author worked, help identify posthumous books, inform the reader that no, there probably won't be a sequel, and so on.
  • The third factor is how easy it will be to find the information. For example, a few years ago I considered proposing that we add a "Place of Death" field to the Author record, but concluded that it would likely remain largely empty for lack of data.
If identifying gender becomes, as you write, "a headache" -- e.g. see this 2003 post by Poppy Z. Brite (who later changed the name to Billy Martin) -- then it affects the first and the third criteria and the proposed addition goes to the back of the line (in my mind.) Ahasuerus 21:24, 3 February 2017 (UTC)
Having a standard that the source of gender information must be the author themself would help make it unambiguous. Not quite as good as DOB which never changes. Aren't you phasing out the wiki pages? They're the natural place to put notes about sources of info. --Vasha 21:43, 3 February 2017 (UTC)
Not the ones for authors (yet). I am sure that when we decide to tackle those, there will be space in the DB for the data from them. Annie 22:06, 3 February 2017 (UTC)
That's right. Once we wrap up the migration of series-, publisher- and publication-related notes to the database proper, we will tackle "Biblio" and "Bio" pages. We still need to decide whether they will be migrated to a single "Note" field or whether we will have a separate field for "Biography" (where does "Son of Anne McCaffrey" go?). All "Note" fields support the {{BREAK}} syntax, so any supporting information will appear on a separate page. Ahasuerus 22:56, 3 February 2017 (UTC)
Im absolutely against this field's addition. My position is simple: the only person that can specify at a given time the gender of an individual is solely him/her/itself and no-one else. As we can't guarantee this, we should not provide this data regardless of the bibliographic benefits that may derived from it.Hauck 10:01, 4 February 2017 (UTC)
I think it would be a useful addition. For authors who were born gender A and then legally changed it to gender B, we'll have to decide whether we should enter the last known gender like we do with legal names. Ahasuerus 20:23, 3 February 2017 (UTC)
Leaning on legal changes is not good practice when it comes to trans people. They can transition and live as their preferred gender long, long before the paperwork is straightened out (if it ever is). You just have to list a person's gender as what they say it is. --Vasha 20:46, 3 February 2017 (UTC)
That's another interesting question. The last time it came up, Darrah mentioned some of the difficulties involved:
  • There are some author databases that keep track of gender (e.g. LibraryThing), and at least some of them distinguish between "male" and "presumed male". The question becomes, how much do we need to know about them before we assign them a gender? Do we have to have seen them? (Is that good enough?) Do we have to have a reasonably authoritative reference that uses a gendered pronoun for them? Can we assume someone named "Susan" is female? As a gender researcher, I also would like to see this data in the database. (So far, when I've identified a non-obvious gender identity, I've made sure to add something to the author's wiki bio age that includes a gendered pronoun.) But to add this feature (which I support), we need to decide what rules apply when we specify a gender as "known".
Ahasuerus 21:06, 3 February 2017 (UTC)
Thanks for that! --Vasha 21:10, 3 February 2017 (UTC)
The issue of joint/house pseudonyms is an interesting one and I don't think it has come up before. Obviously, something like Editors of Chicago Tribune would have to be "N/A". However, we'd have to decide what to do about The Brothers Grimm, Ilona Andrews (a husband and wife team), and house names all of whose authors are/were of the same gender. My first reaction is that it's safer to use "N/A" for all pseudonyms due to their shifting nature. Either way, it shouldn't stop us from implementing gender support. Ahasuerus 20:23, 3 February 2017 (UTC)
I wouldn't think this to be a useful addition, too. Gender can be a theme in some works, and though there may be a dominancy in some author's work, I'd think it would serve more towards to put an author into a certain basket, obscuring the many other facets the author's works may show (and should, if the work is more than one-dimensional). I'd think that the tag function is enough to mark the works of gender interest. (I am aware that it is a biographical marking rather than a bibliographical one but people tend to confuse those things; also, we than also could - or should? - index if an author is married or lives in a stable relationship, has children, his major occupation could be of interest etc.) Stonecreek 10:35, 4 February 2017 (UTC)
Also, I fear that there wiil be some rush to assign gender to every author's entry, culminating into many false or questionable edits (similar to some of the novelizations markings we had: there were some that are part of a movie or tv series universe, but weren't novelizations of any film). Stonecreek 10:43, 4 February 2017 (UTC)
I imagine it would be moderately useful, if and only if there were a foolproof way of adding the data. As there obviously isn't, and as this would raise too many problems, I think it is not really worth the trouble. Linguist 15:47, 4 February 2017 (UTC).
Most bibliographies such as Austlit and Gale do list gender, and we already list info that is irrelevant to a purely bibliographical listing: DOB, DOD, Birthplace, Art, Artist, Interior Illustrations, etc.... There is no foolproof way to enter "any" data into the database, errors of gender should be no more common than errors of language, or DOB/DOD and Birthplace. The ISFDB is being more widely used as a research tool and gender information is useful in a lot of research. I solicited comment from Eric Leif Davin, author of Partners in Wonder, on the value of gender identification when I first proposed this. Unfortunately, this seemed to throw more fuel on the fire.
I have a personal interest in this, since I feel female writers bring a sensibility to the genre that most male writers lack. I know that Chavey is interested from an academic standpoint.--Rkihara 18:32, 4 February 2017 (UTC)
I appreciate Ahaseuras quoting my concerns about the challenges this would lead to from our previous discussion. I have recently been involved in a couple of Wikipedia edit wars over a person's gender, including one editor who insisted that Charlie Jane Anders should be listed as male, because they had found an (old) posting in which she referred to herself as male. Those two edit wars make me a little more skittish about assigning gender, at least for living authors. If we did include "gender", I think we would need to have a mandatory field for the source of that info. What would we accept as sufficient justification? "They use an obviously male name"? "I've seen them at a convention, and they were obviously female"? "I found an online post where they refer to themselves as male"? "I heard them talk about transition to female, so they must be female now." In each of these cases, I know counter-examples. Rkihara suggests that it's ok to have some mistakes in the database, but I don't think current authors will get as annoyed at us over an error on their date of birth, or being confused about assigning their "Richard Smith" art to the wrong "Richard Smith". Getting someone's gender wrong will, something like 50% of the time, get that person angry. And I think we should try to avoid that. So, while I previously supported the idea of including gender info, I think now that I can only support Vasha's suggestion about limiting gender information to deceased authors. And that would have to be enforced by the software. Chavey 19:50, 4 February 2017 (UTC)
Agreed, because while an individual or team can gather information according to whatever standards they think is right, a crowd-sourced project like this one is bound to run into trouble. That's how this DB differs from other bibliographies. --Vasha 15:18, 5 February 2017 (UTC)
I agree that it's good idea to list the source of a gender assignment, but I feel restricting it to dead authors is a little too restrictive. I think that Chavey's estimate that 50% of people would be upset if we misidentified their gender is way too high. I think that most would be upset only if we refused to correct their gender identification when notified. Authors who use initials for their given names are probably used to this. People that are trans or self-identify as another gender are aware that it takes a while for this to be widely known, and would probably be glad to see this information more widely disseminated.--Rkihara 21:10, 5 February 2017 (UTC)

(unindent) Given the variety of opinions on this subject, reaching consensus may be difficult. Still, let me suggest a slightly different approach, which may, with luck, alleviate some of the concerns. How about the following values for the proposed drop-down list:

  • Blank (default)
  • Female
  • Male
  • N/A (joint pseudonyms, house names, etc)
  • "See Note"

Since our underlying standard for author fields is "publicly available information", we could put whatever information is available publicly in the Note field and make it as detailed as necessary. Of course, we'll have to add support for author-specific notes first. Ahasuerus 21:54, 5 February 2017 (UTC)

Publicly available when and where - is Wikipedia enough at the time it is added to the DB here? And which DBs update their records properly so we can rely on them for verification? Will all the moderators take it upon themselves to verify every attempt to change the gender before approving? What happens in 10 years when someone changes a gender? Who is going to bring the data uptodate when someone changes their gender 5 years after theirs is assigned here? And I am worried about new editors, not reading through the rules for gender and deciding to help by assigning genders on all authors they have books from. The more I think about it, the more it feels like Pandora's box and I would rather not have it in the record than deal with the possible issues... Annie 03:02, 6 February 2017 (UTC)
Still strongly opposed to this idea. Hauck 08:18, 6 February 2017 (UTC)
If it's a straight up/down vote, I'd say no. If you want a single reason - we've got enough to do filling gaps rather than creating new ones. Doug H 13:50, 9 February 2017 (UTC)

(unident) Outcome: Consensus not reached. An FR will not be created. Ahasuerus 18:40, 17 February 2017 (UTC)

2017-02-03 server downtime - 3pm

The server will be unavailable between 3pm and 3:02pm server time. Ahasuerus 19:41, 3 February 2017 (UTC)

Everything is back up. Ahasuerus 20:02, 3 February 2017 (UTC)

Bill Longley

Raising a glass to 'our' Bill Longley, who died three years ago today. We're carrying on with the good work, Bill! PeteYoung 05:19, 5 February 2017 (UTC)

Cheers! (I miss him.) Chavey 09:06, 5 February 2017 (UTC)
Me too: I wouldn't be as much involved hadn't he shown his loads of encouragement and good humour. Stonecreek 09:59, 5 February 2017 (UTC)
Ars longa, vita brevis... Ahasuerus 17:31, 5 February 2017 (UTC)
I had been meeting him virtually a lot the last few weeks while I had been transferring wiki data to the DB. Haven't realized that he is not with us anymore :( Annie 03:05, 6 February 2017 (UTC)
Cheers! Miss you! Rudam 10:27, 6 February 2017 (UTC)

2017-02-06 downtime - 12 noon

The server will be unavailable between 12 noon and 12:02pm server time. Ahasuerus 16:41, 6 February 2017 (UTC)

The server is back up. Ahasuerus 17:02, 6 February 2017 (UTC)

2017-02-08 server downtime - 1:40pm

The server will be unavailable between 1:40pm and 1:42pm server time. Ahasuerus 18:23, 8 February 2017 (UTC)

Everything should be back up. Ahasuerus 18:42, 8 February 2017 (UTC)

Two different John L. Campbell?

Can someone look at John L. Campbell? I am pretty sure that we are dealing with two different people with the same name but will appreciate a second set of eyes. The author data belongs to the novelist from the 21st century. Annie 19:31, 8 February 2017 (UTC)

I'd say you're right. The photo on the website link belongs to a guy who is 40 years at most, likely younger. Presuming that it is from the 2010s he can't be the one who published stories in the Sixties. Stonecreek 19:44, 8 February 2017 (UTC)
Yeah and he does not claim the stories anywhere on the site either. Or anywhere else. Is that enough to get a disambiguation going? Annie 19:51, 8 February 2017 (UTC)
I'd say so! Good catch! Stonecreek 19:52, 8 February 2017 (UTC)
I've submitted the change in the authors for the two old stories (I think I got it anyway...). There is always a first time for everything. :) Annie 21:29, 8 February 2017 (UTC)

2017-02-09 server downtime at 1:15pm

The server will be unavailable between 1:15pm and 1:17pm server time. Ahasuerus 18:02, 9 February 2017 (UTC)

The server is back up. Ahasuerus 18:16, 9 February 2017 (UTC)

Author formatting?

I believe I have removed all extra spaces between author names. However, there is another discrepancy which we may want to address. Some of our pages separate co-author/co-editor names using commas while other pages use "and"s. For example, this publication page says:

  • Editors: C. L. Werner, Steve Lyons, Rob Sanders, Ben Counter

but the associated title page says:

  • Editors: Ben Counter and Rob Sanders and Steve Lyons and C. L. Werner

I think it would be desirable to use the same separator throughout the site. Personally, I prefer commas to "and"s. Ahasuerus 21:11, 9 February 2017 (UTC)

Good point. And yes, I think commas look better too.
Another suggestion: should the category "Shortfiction" on author summary pages be "Short Fiction"? --Vasha 21:55, 9 February 2017 (UTC)
I like the and on the author pages (being in bold and with the long enough string to split the authors)- it gives me an easy split in the long chain of names on some of the titles and makes the page easier to read. At the same time, I prefer the comma on publications and title pages. Annie 22:46, 9 February 2017 (UTC)
Agree with Annie on that, and also in tables of contents the and is good. --Vasha 14:36, 11 February 2017 (UTC)

2017-02-10 server downtime at 1:45pm

The server will be unavailable between 1:45pm and 1:47pm server time. Ahasuerus 18:30, 10 February 2017 (UTC)

Everything is back up. Ahasuerus 18:46, 10 February 2017 (UTC)

Changing the main search box?

As per feedback from an MIT librarian, the default search box, which requires the user to select a value from a drop-down list, may be sub-optimal. He proposed creating separate boxes for authors, fiction titles and series since they are the most common types of searches. I have seen similar comments on Usenet, so I am posting it here to see if there is broader support for this proposal. Ahasuerus 22:11, 10 February 2017 (UTC)

As an editor, I'm always having to scroll down to get to the Editing Tools (more so on publication pages due to the Other Sites - I have most turned off, but like to keep a few). I'd hate to see those even further down because the search box is now bigger. But I'd also like the convenience of making those searches easier. Maybe it's time for a whole revamp of the left menu? Have different menu layouts for logged in and non logged in viewers? If logged in, move the Editing Tools further up? -- JLaTondre (talk) 22:28, 10 February 2017 (UTC)
What JLaTondre said - more boxes at the top will make it even harder for editing. Plus - I seem to be doing a lot of searching for ISBNs - especially when I am adding so I would vote for that being somewhat more visible as well... I wonder if two different left bars (editing/browsing) controlled by a user preference and a switch at the top are not in order (so logged in users that are mainly searching do not see the editing either)... Annie 00:17, 11 February 2017 (UTC)
Well, we have FR 345, which reads:
  • Convert the Navigation Bar [on the left] to drop-down menus across the top of each page.
I was a bit shaky on the technical side when I created the FR in 2012, but I think I can do it now. My recovery is progressing slower than expected, so more demanding/larger projects are currently on the back burner anyway.
A less drastic change would be to move the "3 by 5 catalog card" image to the top of the page, which will free up additional real estate within the search box. Ahasuerus 01:45, 11 February 2017 (UTC)
If we are going to change that menu anyway, any chance to have the Editing tools on a publication a bit higher ( see Dark Star - the "Other sites" box is a great thing (for someone I guess anyway) but it is in the way a lot... Annie 01:58, 11 February 2017 (UTC)
Three responses:
  • I say yes, if it's possible to have different menu bars logged in or not, the editing tools should be above the other sites.
  • I am pondering the idea of dropdown menus at the top, and I think, on the whole, I don't like it. I think it would greatly slow down access for things that I click on a zillion times while editing.
  • If you must have preset boxes for multiple categories, maybe just name and fiction title? Too many choices is confusing. I don't quite see how the design would work, with the search at the side, since the bar is too narrow to have labels next to the boxes.
--Vasha 07:08, 11 February 2017 (UTC)
I agree with the opinion that the LHS is already rather packed, and multiple search boxes would add to the clutter. A few observations/thoughts I have about the box:
  • Name is not obviously author name. (I realize "author" doesn't do justice to the nature of the possible roles). --MartyD 13:20, 11 February 2017 (UTC)
How about "Person"? Ahasuerus 19:16, 11 February 2017 (UTC)
In a Fiction DB? If I see "Person", I would think "character in a book". Author may not be exact but will not be confusing at least. Annie 19:40, 11 February 2017 (UTC)
  • I frequently do multiple searches for the same category of thing. If that's not Name, I have to (remember to) repeat the picking each time. Something that records my last choice (could be a cookie -- doesn't have to be in the DB; no preference management needed) and starts the drop-down out with that choice selected might help some searchers.
  • The flat list is sort of haphazard. If the items in there were grouped (e.g., title things, publication things, etc.), it might be possible to make a different layout that takes advantage of those groups, perhaps via tabs + checkboxes/radios, pull-rights, or some other sort of progressive disclosure.
Ideally, search would just look at everything and then tell the searcher about all categories of matches. Tough to do that efficiently in the current set-up. --MartyD 13:20, 11 February 2017 (UTC)
It wouldn't be too hard to add a new type of regular search to cover "All" record types, where "All" would be defined as authors, titles, series, publishers, publication series, ISBNs, tags and awards. That's what Amazon, IMDB, MAL and many other popular databases do. We could then display the results either as page sections (a la IMDB and MAL) or as tabs.
However, as Marty suggested above, performance would be a concern. An author search on "heinlein" takes less than 0.1 second. A title search on "heinlein" takes approximately a second. An "All" search of all record types would probably take under 2 seconds. (That's assuming that we make the search smart enough to identify search strings that look like ISBNs and handle them differently.) Is it an acceptable trade-off? I guess we could create a User Preference to control whether "All" is the default selection in the regular search box. Ahasuerus 19:36, 11 February 2017 (UTC)
Personally, I like the IMDB.com search bar. It gives more options, and takes up less space. It's essentially the same as the Amazon search bar. (Hence it's something users are familiar with.) I prefer Amazon's listing of the categories on the left better than IMDB's listing it on the right, but I prefer IMDB's organization of the results better. Chavey 13:43, 11 February 2017 (UTC)

(unindent) I agree that scrolling is a pain when editing records. How about an alternative approach? The most popular action when a logged-in user is viewing an ISFDB record is "Edit [record]". We already display the ISFDB record number on the first line of each page, e.g.: 

Publication: Grave Markers: Volume Two         ISFDB Publication Record # 604827

How about adding an "Edit" link to the right of the number if you are logged in? And perhaps dropping the word "ISFDB" since it's redundant? The same line would then look as follows: 

Publication: Grave Markers: Volume Two         Publication Record # 604827 [Edit]

Would that make life easier? Ahasuerus 18:15, 11 February 2017 (UTC)

YES! That would be awesome. Annie 19:37, 11 February 2017 (UTC)
Very good idea. --Vasha 21:56, 11 February 2017 (UTC)
Very much like the idea of having an edit link there.
One of my personal issues with search is that I frequently accidentally perform the wrong type of search because of the trackpad on my laptop, and then have to type and select my search all over again. It would be great if when a search returned zero results, there was a way to automatically re-submit the search as a different type. Aside from the "A search for 'xxxxx' found 0 matches" notification, there's nothing else on the screen, so there should be plenty of screen real estate to work with. Albinoflea 03:39, 13 February 2017 (UTC)
How about re-displaying the search string in the search box when you land on the "search results" page? That way you could select the correct value from the drop-down list and resubmit the search without having to re-type the search string. Ahasuerus 04:35, 13 February 2017 (UTC)
I'd love that. Annie 04:49, 13 February 2017 (UTC)
Likewise. Albinoflea 21:59, 14 February 2017 (UTC)

DC Metaverse/Sandman Cleanup

This series, this pub, this title, this series, this title and this title are all interrelated and should be part of a common series. I would recommend " The Dreaming" to match this link. The Sandman and Books of Magic should remain as part of that series. Death should have her own series with this title and this title. How would I go about getting this resolved? TAWeiss 19:55, 12 February 2017 (UTC)

I have added the titles that you listed (plus the Sandman series) to The Dreaming series. I have also turned it into a sub-series of DC Comics Metaverse. Could you please check the results to make sure that everything looks OK? Ahasuerus 19:22, 13 February 2017 (UTC)
Death of the Endless series which includes this title and this title should also be part of the DC Metaverse. TAWeiss 22:51, 26 February 2017 (UTC)
Done! Ahasuerus 23:25, 26 February 2017 (UTC)

2017-02-13 server downtime at 2:15pm

The server will be unavailable between 2:15pm and 2:17pm server time. Ahasuerus 19:10, 13 February 2017 (UTC)

Everything is back up. Ahasuerus 19:17, 13 February 2017 (UTC)

Kim Stanley Robinson's Mars Trilogy

I'd like to propose a change of the name for the series, from Mars Trilogy to Mars (Kim Stanley Robinson). The reason for this is that this ain't a trilogy no more, since it was expanded by a collection and diverse short pieces. If there are any objections or ideas for a better name, please let me know. Christian Stonecreek 18:57, 14 February 2017 (UTC)

It is kinda the accepted name, despite it not being a trilogy though. If I come to look for the series, I would type Mars Trilogy. So I would not be very happy to lose the name from our listings. Annie 19:26, 14 February 2017 (UTC)
Yeah... I agree with Annie, it is still most frequently referred to as the Mars Trilogy when it is discussed and written about... in fact, of his three trilogies this is the one with the least amount of confusion over its name.
As for the collection The Martians, which in one edition or another collects all of the short pieces that aren't just excerpts from the novels, most of them don't relate to the main Trilogy story; many are set in a totally different timeline (e.g. Exploring Fossil Canyon, Green Mars (novella), A Martian Romance) or an alternate timeline with the same characters (e.g. Michel in Antarctica, etc.)... the way I usually like to think of it, if the Mars Trilogy was a DVD, The Martians would essentially be the DVD extras portion of the disc. Albinoflea 20:03, 14 February 2017 (UTC)
You are right with the collection, but should it be part of the series in the first place? A collection consisting of six pieces from Ray Bradbury's Martian Chronicles and seven other pieces wouldn't be considered as part of the series either. Stonecreek 20:26, 14 February 2017 (UTC)
That's an interesting question... the collection was assigned as part of the Trilogy before I started editing here, and I've always taken it for granted. Advertisements and interviews from around the time it was completed frequently list it as "a companion volume" to the Mars Trilogy... the NY Times called it "a series of variations on the grand theme of the trilogy". Whatever it is, it's an edge case. Albinoflea 21:25, 14 February 2017 (UTC)
Perhaps this collection -- as well stories like "Michel in Antarctica" -- could be put in a "Mars Trilogy Metaverse", which would also include the original Mars Trilogy as a sub-series? Ahasuerus 21:48, 14 February 2017 (UTC)

2017-02-14 server downtime - 5:40pm

The server will be unavailable between 5:40pm and 5:42pm server time. Ahasuerus 22:29, 14 February 2017 (UTC)

Everything is back up. Ahasuerus 22:41, 14 February 2017 (UTC)

People relationships

There are some authors and artists that are related to others (Anne and Todd McCaffrey, J.R.R. and Christopher Tolkien, for example). Would it be a useful or desired feature to have a bio field where that relationship could be indicated? Perhaps, when editing a person's entry, have a field to enter the name/ID of the person to whom they are related, and then a drop-down for what that relationship is (parent, sibling, child, spouse, etc.)

Thoughts? ···日本穣 · 投稿 · Talk to Nihonjoe 01:27, 15 February 2017 (UTC)

At this time author-specific biographical and bibliographical information is recorded in the Wiki, e.g. see Nicolaus Bornhorn's Summary page which links to his Wiki-based Bio page.
There is a Feature Request to "Move "Author" and "Bio" pages from the Wiki to the database". I plan to implement it once we finish migrating all publisher-, publication- and series/magazine-specific pages to the database. We will need to decide whether we want to have two author fields -- one for biographical data and the other one for bibliographical data -- or whether we want to use just one "Note" field. Ahasuerus 01:38, 15 February 2017 (UTC)

Search tweaks

The search box has been changed to keep the entered search value when displaying the search results page. For example, if you do a search on "farewell," and forget to change the search type from "Name" to "Fiction Titles", the resulting empty search results page will re-display "farewell," in the search box.

We can also change the software to preserve the last search type if it is deemed desirable. Opinions? Ahasuerus 21:50, 15 February 2017 (UTC)

I'd love that. This way if the search failure is because of a typo (or because of non-Latin capitalization issue), I can change a single letter and try again immediately Annie 21:52, 15 February 2017 (UTC)
I love that, and I would love the "type" idea, too. ···日本穣 · 投稿 · Talk to Nihonjoe 22:56, 15 February 2017 (UTC)
Definitely, repeat last search type. --Vasha 23:47, 15 February 2017 (UTC)
Yes. --MartyD 13:36, 16 February 2017 (UTC)
Yes please. Albinoflea 16:58, 16 February 2017 (UTC)
The ayes have it! The software has been modified to remember the last search type.
P.S. Please note that these changes affect the search results page only. If a search finds a single matching record, the software will automatically redirect you to that record's page. If that happens, the last entered search value/type will not be preserved. Ahasuerus 18:04, 16 February 2017 (UTC)
A little fix needed - when you search for a non-Latin name, when you go to the next page, instead of the name, you have the codes of the characters. Search for "вежинов" (with the small letter). I would like to be able to just change the first letter to capital to search again but you cannot really do that (short of knowing the codes of course). Same happens with Japanese characters (and anythinh non-Latin1 I suspect (you can use 之 to see it with a Japanese character)Annie 00:13, 28 February 2017 (UTC)
Thanks, I'll take a look. Ahasuerus 00:21, 28 February 2017 (UTC)
OK, I think I got it now -- too much escaping is a bad thing! Ahasuerus 00:39, 28 February 2017 (UTC)
Confirmed for the regular ones. But there is an issue with finding this title. Thanks! :) Annie 00:52, 28 February 2017 (UTC)
Well, the problem with this title is that it's actually "& & # x 2 3 ; x 2 5 C D ;" (spaces added to defeat HTML). If you search for this exact string, you will find the record, but, of course, that's not something a regular user would think of doing. I am not sure we can fully support these "fake HTML" cases until we convert everything to Unicode. Ahasuerus 01:39, 28 February 2017 (UTC)
Suspected as much - just could not resist trying that one. Thanks for checking! Annie 01:45, 28 February 2017 (UTC)

"Edit" links added to all biblio pages

Based on the outcome of this discussion, "[Edit]" links have been added to all bibliographic (author, title, publisher, series, publication series, publication) pages. The only exception is the three award pages (award, award type, and award category) which do not display ISFDB record numbers at this time. Ahasuerus 21:29, 16 February 2017 (UTC)

Awesome! Thanks for the quick fix:) Annie 21:59, 16 February 2017 (UTC)

Stoker preliminary nominees to be added

I've added Award records for most of the Stoker preliminary ballot, but have decided to leave it to someone else to add the last of the nominated works that aren't yet in the DB. Here they are:

  • NOVEL (Superior Achievement in a First Novel): The Monster Underneath by Matthew Franks; Until Death by Kari Kilgore; The Sanguinarian ID by L. M. Labat; The Black Goat Motorcycle Club by Jason Murphy; Wasteland Gods by Jonathan Woodrow; The Eighth by Stephanie Wytovich
  • NOVEL (Superior Achievement in a Young Adult Novel): Keep Away from Psycho Joe by Michael Brent Kelley
  • NONFICTION (Superior Achievement in Non-Fiction): Haunted: On Ghosts, Witches, Vampires, Zombies and Other Monsters of the Natural and Supernatural by Leo Braudy; A Season with the Witch by J. W. Ocker; Guillermo del Toro’s “The Devil’s Backbone” and “Pan’s Labyrinth”: Studies in the Horror Film by Danel P. Olson; The Turn to Gruesomeness in American Horror Films, 1931-1936 by Jon Towlson
  • COLLECTION (Superior Achievement in a Poetry Collection): Brothel by Stephanie M. Wytovich; Small Spirits by Marge Simon; Two Drinks Away from Chaos by Aurelio Rico Lopez, III; Voices from Empty Rooms by Lisa Lepovetsky; Field Guide to the End of the World by Jeannine Hall Gailey; Children of God: Poems, Dreams, and Nightmares from the Family of God Cult by Craig DiLouie and Jonathan Moon; The Seven Yards of Sorrow by David E. Cowen; Corona Obscura: Poems Dark and Elemental by Michael R. Collings

--Vasha 20:46, 17 February 2017 (UTC)

All done. --Vasha 04:01, 24 February 2017 (UTC)

Weird pseudionyming

In this one, shouldn't Gregory Francis be pseudonymed under the canonical name and not under another pseudonym? Or am I missing a case where this is the way to do it? Annie 22:28, 17 February 2017 (UTC)

You are right, it's a joint pseudonym of two writers and should be "pseudonymed" to their canonical names. I have made the change. Thanks for spotting it! Ahasuerus 22:33, 17 February 2017 (UTC)
I would have fixed it - I just wanted to make sure we do not have an obscure rule for that somewhere :) I think the canonical name of this guy got shifted from the full name to the shorter one at some point or the two names were connected after the joined pseudonym was already in place without completely cleaning the field - I am cleaning a lot of leftover story variants as well for him. Annie 22:53, 17 February 2017 (UTC)

2017-02-17 server downtime

The server will be unavailable between 8:30pm and 8:32pm server time. Ahasuerus 01:17, 18 February 2017 (UTC)

Everything is back up. Ahasuerus 01:31, 18 February 2017 (UTC)

Language-less titles - the final push

Language-less ESSAY titles by first letter:

None. None at all.

Language-less INTERIORART titles by first letter: 

+------+----------+
| I    |      133 |
| J    |       36 |
| K    |       15 |
| L    |      138 |
| M    |      207 |
| N    |       93 |
| O    |      139 |
| P    |      195 |
| Q    |       25 |
| R    |      127 |
| S    |      445 |
| T    |     1865 |
| U    |       60 |
| V    |       32 |
| W    |      217 |
| X    |        2 |
| Y    |       13 |
| Z    |        6 |
+------+----------+

Ahasuerus 21:05, 18 February 2017 (UTC)

All cover art entries are cleared. My pending queue has a few language assignments and a few unmerges (which will end up with new languages again). Once they are approved (and the language of the unmerges fixed), all remaining will be English :)
Essays A-H next? Annie 23:11, 19 February 2017 (UTC)
All relevant submissions have been approved. The first ESSAY batch will become available in the morning. Ahasuerus 00:29, 20 February 2017 (UTC)
Thanks! I just sent in the last batch of followups - after the unmerges. :) Annie 00:51, 20 February 2017 (UTC)
Looks good. I have auto-assigned the remaining COVERART titles, so we are down to ESSAYs and INTERIOART. Ahasuerus 02:09, 20 February 2017 (UTC)
A-H (plus special characters and numbers) essays are either English or in my pending list with a different language or for deletion ( a lot of them are for deletion). A-H (plus special characters and numbers) in Interior art next? Annie 04:06, 21 February 2017 (UTC)
Nice! INTERIORART/A-H coming up next. Ahasuerus 04:33, 21 February 2017 (UTC)
Thanks! With a bit of luck (and no work travel on my part), we should be done with assignments before the end of the month. :) Annie 05:34, 21 February 2017 (UTC)

(unindent) All Interior art records that start with A-H or special characters had been inspected and are neatly separated in 3 groups:

  • Group one - Legitimate non-English single titles - Title Edits submitted for all of those.
  • Group two - Mixed language versions - unmerge submitted for all
  • Group three - All of the rest - aka the majority - the English ones.

At this point it is safe to assign English to the whole group - and further update where needed will deal with it:

  • Group one - if the automatic language goes before my update, my update will fix the language; if my update goes first, the automation won't update this title (as it now has a language)
  • Group 2 - I have the log to work back through so setting these to English will not lose them (regardless if it happens before or after the unmerge) and will not cause any permanent issues (more titles added to the multi-language report is the side effect but I am clearing that one anyway)
  • Group 3 - all clear

All of the above explains why I was not that eager to deal with 5K of these on their own :)

So... next group please - Essays I-S? :) In the meantime I am dealing with the followups of the unmerges. If you want to wait for me to finish the followups, just add the new essays and leave the old interior arts in for tomorrow -- but making them English won't harm anything (as shown above). Thanks! Annie 23:09, 21 February 2017 (UTC)

Understood and thanks. All INTERIORART titles through H have been auto-assigned. The remaining ESSAYs will be making their triumphant appearance on the cleanup report by 1:30am server time. I can almost smell victory! :-) Ahasuerus 23:31, 21 February 2017 (UTC)
I will be happy to be done with this project - it had become a bit tedious in the last few batches (not enough to get me to stop so close to the end so no worries). Then I can go and kick some wiki entries out of the way again for a bit - does not seem like anyone is making any progress there. ;) Annie 23:41, 21 February 2017 (UTC)
Seems like the essays did not make it to the report tonight. :) oh well - tomorrow night then. Annie 07:06, 22 February 2017 (UTC)
So was it intentional or did just the report get a bit confused? Annie 17:51, 22 February 2017 (UTC)
Sorry, that was a snafu on my side. I had two patches almost ready to go, but only the first one was actually installed yesterday. All fixed now; the data will become available in the morning. (And yes, I can see how reviewing and fixing approximately 100,000 titles can get just a tad tedious ;-) Ahasuerus 18:02, 22 February 2017 (UTC)
No worries - figured out that either you never pushed it or messed up the filtering (I had been a bit creative on ranges in some types after all). Just wanted to make sure that there is no other reasons I am missing. I had not been counting how many I had been clearing - would not be surprised if it is close to 100 000 indeed. At least I did not send 99K or thereabouts "set to English" updates into the moderators' queue during the process - that would have caused a revolt. Annie 18:27, 22 February 2017 (UTC)
We had approximately 100,000 "bad" titles after the first, completely automated, round of auto-assignments. We are down to 6,000 as of this afternoon. Some auto-assigned titles were processed without a manual review, but my guess is that approximately half were reviewed. That's a lot of titles to eyeball! Ahasuerus 18:44, 22 February 2017 (UTC)
The automatic assignment of essays and stories starting with "The " did catch a few non-English by the way (stories and essays still unmerged from their English parents) - but they all popped up on the multi-language report so we are good there. And a lot of titles had weird languages - working through some of them had been fun. It could have been a lot worse - if we had a lot more international titles from the days when languages were not supported (or not mandatory or whatever the case was). Annie 19:07, 22 February 2017 (UTC)

(unindent) Two essays are waiting for their language in my pending list (one German, one French in the process of unmerging) and a few more German ones are trapped in two German non-fiction collections (I am leaving them there merged for now until we find the content so we do not end up doing triple work). Anything remaining is English. And we are officially done with the text titles. :) Next: the remaining Interior Art titles (all 3700 or so of them). Annie 20:40, 23 February 2017 (UTC)

Excellent! All eligible ESSAY titles have been auto-assigned and the remaining INTERIORART titles have been primed. I have also moved all language-specific title reports to the main "Title" section, but I should make a separate post about it in case things get lost here. Ahasuerus 00:13, 24 February 2017 (UTC)

(unindent) And done. I expect a few new titles to hit the multi-language report (I did not split a few because I saw more work to be done on their publications so I would rather see it on the other report and finalize them; anything that would not hit the report or does not need to is cleaned) but at this point all language-challenged titles that are not in my pending list can be assigned to English. And that means that the "To be Assigned" group lost the game and takes its rightful place at the end of the queue. It will take a few days to untangle some of the multi-language ones but outside of that the languages for titles cleanup project is officially finished. :) Annie 20:37, 27 February 2017 (UTC)

Fantastic (no pun intended)! Many thanks! Everything has been assigned, so the cleanup report should be, well, "clean" tomorrow. I am going to revert the change to the "Titles without Pubs" report next and post a note on the Moderator Noticeboard next. Ahasuerus 21:10, 27 February 2017 (UTC)
Yep. Thanks again for pulling the two types out of the report temporarily. Annie 21:15, 27 February 2017 (UTC)

Import coverart titles from publications?

Currently edit/importcontent.cgi (and presumably the .py too but I did not look at the source) allows one to import titles from another publication (i.e., so called "Option 1" which is nice improvement from before; I recall when there were no options but importing via a list of title numbers). That said, it does not seem to allow me to import coverart titles from one pub to another? Do we have a feature request for this? If not, can we get one? I did not test if edit/exportcontent.cgi would move covertart titles but I expect it won't (someone let me know if I am wrong). Thank you. Uzume 04:33, 19 February 2017 (UTC)

I think I have seen that already done (but you'd have to be careful when there's an already existing coverart title). If you just want the art transferred you only need to copy the address into the Image URL field. Stonecreek 11:52, 19 February 2017 (UTC)
The bulk import (Option 1) does not import the coverart. However, the individual import (Option 2) does allow you to import a coverart title. Option 1 already has a check box for including page numbers. A cover art check box could be added similar to the clone pub screen. -- JLaTondre (talk) 13:22, 19 February 2017 (UTC)
Yes, that makes a lot of sense, adding a checkbox if the cover artist is the same when bulk importing titles. --Vasha 14:37, 19 February 2017 (UTC)
Sounds like a good idea. If there are no objections, we can create an FR. Ahasuerus 15:55, 19 February 2017 (UTC)
Does the bulk import already bring over interior art? --Vasha 19:21, 19 February 2017 (UTC)
Yes, it does. We could add another check box for INTERIORART titles. Ahasuerus 19:28, 19 February 2017 (UTC)
Yes there should be for the same reason that Clone has it. (And the order of the boxes the same as Clone, no confusion.) --Vasha 20:38, 19 February 2017 (UTC)
If Clone has this I would consider this a bug (albeit minor) and not an FR. Uzume 00:54, 20 February 2017 (UTC)

(unindent) OK, FR 980 has been created. I don't think it's a bug since the current behavior is intended. Ahasuerus 02:14, 20 February 2017 (UTC)

The current behavior is intentionally different from clone behavior? Why? Uzume 03:06, 20 February 2017 (UTC)
Originally, there were no check-boxes. Then we added a "page numbers" check-box to the import process. Then we added an "intermediate" page with a couple of check-boxes to the cloning process. Then we added more check-boxes to the "intermediate" cloning page. Now we are considering adding more check-boxes to the Import/Export page. The fact that similar functionality is already available on the cloning side of things doesn't mean that there is a bug with the import/export software as it currently exists. Ahasuerus 15:49, 24 February 2017 (UTC)

Wiki sidebar tools

I just recently noticed (though it probably was true sometime back) that the left sidebar tools in the wiki are now at the bottom of the page past all the rest of the content. What I mean it is the sidebars labelled "navigation", "search" and "toolbox". This makes it very cumbersome to use these links on long pages (like this one for example but it is still true on short pages like the main page). I am using Chrome if that makes a difference. Has anyone else noticed this and have any feedback, comments, or solutions? Thanks. Uzume 16:39, 19 February 2017 (UTC)

I have tried it with Chrome, Firefox and IE. All three show the sidebar tools on the left. Have you tried clearing the cache and other browser detox tricks? Ahasuerus 16:51, 19 February 2017 (UTC)
I have this with Chrome too. It's also occurs in incognito mode with no extensions active so seems pretty clear it's a browser issue. It's been that way for awhile. Using the developer tools, I can see Chrome and Firefox parsing the <div> indents differently which causes the issue in Chrome. I suspect that the wiki software is probably doing something that isn't 100% standard and Chrome is no longer tolerating it. -- JLaTondre (talk) 01:53, 20 February 2017 (UTC)
I appreciate the response. I might have to look into that and see where things are going awry (we might be able to fix it in a template or some such). Uzume 02:46, 20 February 2017 (UTC)

Author birthdays and dates of death on the front page

This one was discussed (perhaps not for the first time) when I noted that the list of birthdays on the home page was growing too long to be interesting. Instead of limiting that list to only "genre" authors, whatever they are, maybe limit it to authors with a certain minimum number of titles? --Vasha 21:30, 19 February 2017 (UTC)

That's a good point. Perhaps the algorithm could be fine-tuned to select:
  • award-winning and award-nominated authors
  • authors with more than N titles
The only difficulty is that identifying these authors in the database will be time-consuming, so we'll need to move the data retrieval process to the nightly job. It shouldn't be a big deal since dates of birth/death change once every 24 hours by definition. Ahasuerus 23:48, 20 February 2017 (UTC)
That will give a huge preference to artists - they tend to have more works and the good ones have more awards. Not that I mind having some artists in the list but if the list gets nominated by artists, it will kinda be against what the DB is all about. Annie 23:57, 20 February 2017 (UTC)
True. It would also be biased against newer authors who haven't published many books/stories yet.
I guess we could add different weights or thresholds for different title types: you need to have more than N art titles to be included, but you only need M other types of titles to be included. Ahasuerus 00:05, 21 February 2017 (UTC)
We need to be careful for authors like Chiang for example - he has a limited number of stories and nothing else (no novels for example) but he does belong to the list. Although awards will probably get him into the list anyway. Annie 00:13, 21 February 2017 (UTC)
Even one award will be enough to get included, so he will be fine. Besides, he has close to 40 canonical titles, which will be plenty. Ahasuerus 00:22, 21 February 2017 (UTC)
And new and upcoming authors will almost never make it - or international ones - where we do not have a lot of their works... Maybe have a different rule for ISFDB minority language writers so they can make it in and for authors that had had works only in the last 2-3 years? Annie 00:13, 21 February 2017 (UTC)
The problem here is that we have no way of telling whether a non-English author has only one title on file because he never wrote any other SF or because we haven't cataloged his work yet. Ahasuerus 00:22, 21 February 2017 (UTC)
True but excluding them altogether is not a good idea either. Not sure how to solve that. Maybe reserve some number of spots for non-major language authors? At least this way we will have some non-major language presence...Annie 00:26, 21 February 2017 (UTC)
I guess the first question to answer is whether we are trying to make the front page a fair representation of:
  • what's in our database, or
  • worldwide speculative fiction
The former would be hard enough, but the latter would be much much harder.
Another thought. Do we really need to display these lists on the front page? They are nice to have, but they are hardly our "core competency". We could move them to a separate page and link them from the front page. If we do that, we don't need to limit the current lists. Or perhaps leave the top 10 authors on the front page and link to the full lists? Ahasuerus 00:50, 21 February 2017 (UTC)
You have something here. If we have only top 10 (or even top 5) and then a complete list (language-filterable if you feel like coding), that will be awesome. What I do not like is having a list that seems like a representation of the site that contains only English authors. Annie 01:13, 21 February 2017 (UTC)
I think the easiest way to do it would be to limit the front page lists to award winners. As long as there is at least one eligible award winner, that is -- February 29 can be problematic :-) Ahasuerus 01:39, 21 February 2017 (UTC)
Award winning authors who were born or died on Feb. 29th: Janet Kagan, Patricia A. McKillip, Tim Powers, Howard Tayler, Sharon Webb. That's 5 :-) Chavey 03:04, 22 February 2017 (UTC)
Winners and nominees for when the winners are not enough or none are there? There may be someone often nominated and never won still standing... As for Fen 29 - just make a date exception - if it is Feb 29, get top 5? :) Annie 01:43, 21 February 2017 (UTC)
I'll need to run a few queries to be sure, but I expect that we will need just two cases:
  • all winners and nominees whose birthday/anniversary of death is today
  • if none are found, the 5 people with the most titles
Ahasuerus 03:01, 21 February 2017 (UTC)
I cannot imagine a case (short of no one's birthday being on this day) that won't work with those two (if you do not want to give precedence to winners compared to nominations). Annie 03:14, 21 February 2017 (UTC)
As this would likely mean that for most days only award winners will be shown: likely neither Jules Verne, H. G. Wells, Franz Kafka or other of their contemporaries would show up. Stonecreek 04:56, 21 February 2017 (UTC)
Actually, H. G. Wells has a number of title-based nominations, so he would appear on the short list. Jules Verne has 5 nominations, but they are not title-based. We could check for matching author names, though.
However, I agree with your larger point. Any kind of automated algorithm which reduces the number of displayed authors from 30-50 to 5-10 is bound to be biased in some way. We can try to predict/identify the bias and adjust for it, but I don't think we can eliminate it. Awards are an imperfect indicator because there are many more awards now than in the past, but at least they are somewhat objective. Ahasuerus 05:14, 21 February 2017 (UTC)
I think we do need a list of creators who are displayed. This list would be filled with the criterions already stated: award winners/nominees and number of published works. I'd say both criterions should fill the list, and I propose a number of 7 works of fiction (or more?). That should be enough to downsize the number of displayed creators considerably. But we also need a list of eminent authors that would be left out: Mary Shelley or other early practitioners of the fantastic who didn't publish much but impacted on the genre. This list ideally should be moderated or would be accessible by moderators. Any comments? Stonecreek 12:04, 21 February 2017 (UTC)
Sorry to be blunt, but I've always felt that this list was completely useless. What's its intended purpose? (apart to have the pleasure to see your name in it once a year). Probably the relatives of the authors listed don't need us to remind their birthday. Hauck 19:08, 21 February 2017 (UTC)
A very good question indeed! I don't think there was any significance to the choice of author birthdays and dates of deaths when Al added them to the front page. As I recall, we just needed to put something on the front page that would change daily. Forthcoming books and author birthdays seemed like a good choice. The rotating banner was added around the same time.
As I mentioned earlier, I don't think they are our "core competency", but they have apparently become associated with the ISFDB -- or at least I have seen a lot of online discussions which mentioned them. It's really the only reason why I would be hesitant to replace them with a link to a separate Web page. Ahasuerus 19:27, 21 February 2017 (UTC)
Well for me (and I'd think for many others) it has the charm to find some new authors and even more to remind me of nearly forgotten ones. I do think it is one of the things that makes ISFDB special (and regardless of any condensed list, I'll try to maintain the long list: if there are more than 500 names on it per day, I'll likely revise this). Stonecreek 20:04, 21 February 2017 (UTC)

(unindent)Well, if the problem is what to put on this page, perhaps a kind of "last entered publications" may be a good candidate to 1) showcase our contributor's efforts and 2) give a casual browser the full breadth of our coverage. Hauck 08:12, 22 February 2017 (UTC)

Now, that's an interesting thought! We already have this publicly available list of recent edits which links to updated records. Perhaps we could:
  • limit the birthday column on the left to 10 award-winning and "marque" authors
  • display a link to the full list of people who were born and died on this date
  • replace the list of authors who died on this date (some users find it too morbid) with a list of the last 10 edited records
? Ahasuerus 15:20, 22 February 2017 (UTC)

(unindent) One thing to consider is that the front page already handles a similar issue. Every month, hundreds of new publications are released. However, the front page displays just 22 pubs, a small subset of what's out there. They are called "Selected Forthcoming Books". The list of all forthcoming books is linked to from the front page.

How does the software that generates the front page decide which 22 publications to display? It limits the list to "marque" authors. How does it choose the "marque" authors? Every time a user views an author's bibliography, the software increments that author's "view" counter. That counter is then used to generate the Most Viewed Authors Since 2005 report. Every January a bureaucrat runs a special script which checks all of the "view" counters. The top 2% of all authors are marked "marque". The process is completely automated, objective and doesn't require additional human input.

Since the "marque" data is already available and used on the same Web page, we can leverage it to select the subset of authors to be displayed and then link to a separate Web page which will display the full lists. Ahasuerus 15:10, 21 February 2017 (UTC)

But would this also take regard of important French, Bulgarian or other non-English language authors? Also, it could be of interest to display artists like Hieronymus Bosch who certainly inspired great amounts of fantasists. Nevertheless your idea is certainly better than the long list we have right now. Stonecreek 16:26, 21 February 2017 (UTC)
If it counts each click regardless of who does it, it may propel the Bulgarian ones to the top of the list in some days - the way I am checking on them sometimes :) More seriously though - maybe a mix between a currated list and the marque authors will do. The more international works and updates we get, the more clicks we will get on the non-English ones. Plus the full list will contain them all. Annie 16:34, 21 February 2017 (UTC)
I am concerned that a manually maintained/curated list of "important authors" would cause even more problems than the (in)famous "threshold". We all have different ideas about who is important and who isn't, so there is liable to be controversy. In addition, it will require continuing tweaking as the field evolves. Consequently I believe that objective criteria would serve us much better.
I can think of three objective criteria which we could use singly or jointly to promote "selected authors" and "selected forthcoming publications" to the front page:
  • The "marque" flag as discussed above; it's already being used by the "forthcoming publications" display logic
  • Awards and nominations
  • Total title count or possibly the count of certain title types, e.g. fiction
The last two criteria are computationally intensive, so we will need to move the selection logic to a nighty job to be run at midnight. As I mentioned previously, it won't be a big deal programming-wise. I also think that we should use the same selection criteria to select both authors and publications in order to be consistent. Ahasuerus 18:08, 21 February 2017 (UTC)
You're right about the differences about who is important, so I'd be okay with the criteria mentioned. Stonecreek 20:04, 21 February 2017 (UTC)
This is kind of tangential, but if the DOB/DOD is going to be reworked, it seems to me that the DOD should be sort by DOD, rather than the DOB as it is now.--Rkihara 20:06, 21 February 2017 (UTC)
I am not sure I care too much about this but to throw a little something on this fire, how about quantifying authors by counts of translations of their works (as a measure of how widespread the body of their work is since we cannot count books sold, etc.). That is an interesting (if not the only) measure of how important an author is seen by the the world at large (it is not without its flaws though of course). Uzume 01:11, 25 February 2017 (UTC)
I like the idea in principle but we have so little non-majority languages (statistics books added that it will be almost meaningless at this point :( Annie 01:16, 25 February 2017 (UTC)
True, it is sad we have more Ancient Greek titles than Arabic ones. Uzume 01:39, 25 February 2017 (UTC)
The old requirement to enter transliterated/romanized author names was a big disincentive for anyone who might have been interested in entering works in non-Latin languages. Hopefully things will pick up as we improve our multilingual support. Ahasuerus 02:09, 25 February 2017 (UTC)

Title date display

Based on the outcome of a recent discussion, the Title page and the "Diff Publications" page have been changed to display the exact date of the title in the "Year" field. (There is another FR to change the word "Year" to "Date".)

Next question: is the current abbreviated date format used in the "Date" column of the Title page OK or should we use the YYYY-MM-DD format instead? For example, this page displays "Mar 2016" instead of "2016-03-15", "Feb 2017" instead of "2017-02-21", etc. Ahasuerus 15:57, 21 February 2017 (UTC)

For me, "Feb 2017" (the month of a certain year) is absolutely sufficient. Stonecreek 16:28, 21 February 2017 (UTC)
If we have the actual date recorded, I'd rather see it without the need to go down into the publication. Annie 16:54, 21 February 2017 (UTC)
I do like it this way with the month in words. But making it "2 Mar 2010" would be just as good. --Vasha 18:19, 21 February 2017 (UTC)
Why not have both using hover over popup concept taken from transliterations? One issue to consider about date formats is internationalization and localization. "2010-03-02" is the same as "2010年3月2日", "2 Mar 2010", and "March 2, 2010". This is why I prefer the YYYY-MM-DD (ISO 8601) format we use elsewhere. I would prefer a unified style (why do we use "2 Mar 2010" on author birth and death dates but "2010-03-02" on title and publication record dates?) if we do not support a user selection option. Uzume 01:25, 25 February 2017 (UTC)
User-selected date display is certainly a good idea to consider. --Vasha 14:23, 3 April 2017 (UTC)
There are some inconsistencies in the way dates are displayed. Originally, the idea was that we would always use the international standard (YYYY-MM-DD) format instead of country-specific formats. Later on there was some concern that "YYYY-MM-DD" wasn't as user-friendly as something like "2 Mar 2010". There were a few attempts to make the display side of things more user-friendly, but the changes were inconsistent. A few years later I standardized the way the data entry forms behave, but I didn't touch the display side because I wasn't sure there was a consensus. It's definitely something to revisit at some point.
Also, although it's possible to add support for user-defined date formats, we will have to consider the potential for confusion when editors discuss dates on Wiki pages. Ahasuerus 19:55, 3 April 2017 (UTC)

My Errored Out Edits

This was a feature introduced in 2015 as described here. This also seems to be about the only documentation of the feature. A recent submission ended up there, and some of the edits were incorporated, but others were not. In this case, it seems that the notes and contents updated, but not the reviews. As a non-moderator - what is my role at this point? Do I resubmit the reviews? Do I wait? Doug H 19:05, 21 February 2017 (UTC)

Please add the Contents titles which didn't make it. It looks like the MIA titles include all of the regular reviews and 2 reviews-entered-as-regular-titles, right?
On the technical side, I will be able to copy the errored out submission to the development server tomorrow morning. I will then try to re-approve it and see if it generates an error. Once we know what it is, we should be able to fix it. Ahasuerus 19:14, 21 February 2017 (UTC)
That seems correct. I'll resubmit later this evening. Doug H 21:59, 21 February 2017 (UTC)
Resubmission worked. Looks like the second last ESSAY review partially succeeded - title/author but not the page number. So I'm good on this. Priority is your call. Doug H 13:28, 23 February 2017 (UTC)
Sounds good. I am in the process of restoring the latest backup file on the development server. We'll see what happens when I try to re-approve the submission. Ahasuerus 15:26, 23 February 2017 (UTC)
After installing the latest version of the database on the development server, I tried to restore the data to the state that it had been in prior to the submission approval. I then re-approved the submission. Unfortunately, it didn't result in an error, so we are still in the dark :-( Ahasuerus 21:59, 23 February 2017 (UTC)
Ah, but at least I can pretend it's not my fault. Doug H 16:34, 24 February 2017 (UTC)

Order of story length options

Undecided discussion from here: should the dropdown list of lengths be in increasing order (I vote yes) or stay as it is? --Vasha 21:08, 21 February 2017 (UTC)

I like the current order - less likely to confuse the two similar sounding ones. Annie 21:14, 21 February 2017 (UTC)
Me too. Keep it as it is please! PeteYoung 22:38, 21 February 2017 (UTC)
If excerpt got added to the mix it would be hard to "sort" that with the others. Uzume 01:28, 25 February 2017 (UTC)
I think you may be the only one who likes the idea of "excerpt" being a length... everyone else was talking about it being a title type --Vasha 01:32, 25 February 2017 (UTC)
That was not how I read things (but perhaps I misunderstood). As title type seems that seems problematic. As a length, I could usefully apply it to essays and nonfiction (like an excerpt of a bibliographic index, etc.) as well. Uzume 01:43, 25 February 2017 (UTC)
I read it as a support for Length as well :) However to your latest point here: length may be showing when you are working on publications for all titles but if you read the documentation and look at editTitle for example, it is valid only for Short Fiction. Making it usable elsewhere will probably require a lot of other changes (even if it becomes a length). Annie 01:46, 25 February 2017 (UTC)
Taking this discussion back up to the proper section now --Vasha 02:10, 25 February 2017 (UTC)

Sheree R./Sheree Renée Thomas

Would there be any objection if I switch the canonical name from Sheree R. (four publications) to Sheree Renée (eighteen publications)? I'm just about to add a collection with dozens more items under the latter name. --Vasha 03:27, 23 February 2017 (UTC)

Sounds like a plan. It looks like she changed her working name ca. 2005, but we never updated her canonical name. Ahasuerus 03:35, 23 February 2017 (UTC)
Looking at my email with her, that's the name she's used in all professional settings since at least 2011. Chavey 03:58, 23 February 2017 (UTC)
OK, pseudonym and variant changes submitted. Ready to delete empty titles as soon as those are approved. --Vasha 04:00, 23 February 2017 (UTC)

Cleanup reports: Language-specific title reports

As you know, we are nearing the completion of the language assignment project or at least its title part. Language-less titles are becoming extinct and will be gone in a few days. It will still be possible to create a title without a language during certain unusual Edit/Clone Publication operations, but it will be rare. For this reason, I have folded the "Titles: Languages" section of the cleanup reports menu into the general "Titles" section. Ahasuerus 00:17, 24 February 2017 (UTC)

You may want to change the text in that report from "No eligible container titles without a language found." to "No titles without a language found." Annie 00:41, 24 February 2017 (UTC)
Will do, thanks! Ahasuerus 01:05, 24 February 2017 (UTC)
Fixed. Ahasuerus 01:44, 24 February 2017 (UTC)

Authors without a language

Once all of the title records have had a language assigned, the next step will be assigning a "working language" to our author records. Here is the current breakdown: 

+------+----------+
| A    |     2292 |
| B    |     5692 |
| C    |     4931 |
| D    |     3249 |
| E    |     1366 |
| F    |     2431 |
| G    |     3402 |
| H    |     4754 |
| I    |      401 |
| J    |     1624 |
| K    |     2816 |
| L    |     3382 |
| M    |     6200 |
| N    |     1197 |
| O    |     1039 |
| P    |     3253 |
| Q    |      159 |
| R    |     3386 |
| S    |     6670 |
| T    |     2360 |
| U    |      194 |
| V    |      816 |
| W    |     3764 |
| X    |       23 |
| Y    |      388 |
| Z    |      381 |
+------+----------+

I think it would be best to create a cleanup report covering a few letters to see how many are not English. If the answer is "0", then we can safely auto-assign "English" to all of them. Ahasuerus 00:39, 24 February 2017 (UTC)

The answer will be "at least 1 in whatever letter we select" just to make things hard. At least if the titles were any indication... Can we get the pseudonyms cleared before we start manual inspections at least? :) Annie 00:43, 24 February 2017 (UTC)
Oh yes, pseudonyms. Sorry, it slipped my mind. I will see what I can whip up. Ahasuerus 01:04, 24 February 2017 (UTC)
There are 3,326 pseudonyms without a language even though their canonical name has a language code on file. Of that number, 233 canonical names are not English. I'll create an auto-assignment script later today. Ahasuerus 16:28, 24 February 2017 (UTC)
All eligible pseudonyms have been auto-assigned. The table above has been updated with the new counts. Ahasuerus 18:14, 24 February 2017 (UTC)
I hoped we will win more than that. On the other hand, 3,326 that won't need manual check are better than 0. Shall we proceed with a cleanup report of a letter or two and see how it looks like? I might as well keep my "language-challenged elements rescue" hat a bit longer and help with this one... :) Annie 18:37, 24 February 2017 (UTC)
Sounds like a good plan. Should we start with the letters X, Y and Z? :) Ahasuerus 18:49, 24 February 2017 (UTC)
Why not even though we will do our best not to have even a quasi-war with our French, right? :) Annie 18:53, 24 February 2017 (UTC)
The XYZ affair was mostly about (attempted) bribery. Strangely, no one has tried to bribe us yet! :) Ahasuerus 19:07, 24 February 2017 (UTC)

(unindent) OK, a new cleanup report has been created. The "XYZ" authors will become available tomorrow morning. A cursory review suggests that a significant number of artists' last names will be of Japanese or Chinese origin. Of course, as we all know, it can be hard to tell what the working language is when dealing with artists. Ahasuerus 20:29, 24 February 2017 (UTC)

Quick question - is there a moderator report somewhere that identifies pseudonyms and main authors that are assigned different languages? And if not, can a check be done to see if we have any couples like that? Annie 21:00, 24 February 2017 (UTC)
Not at this time. Checking the database I see that we have 53 [edit: 47] mismatches. Some of them are legitimate, e.g. Gregory Kern, but most aren't. Let me see if I can create a new cleanup report and allow moderators to ignore mismatches. Ahasuerus 21:12, 24 February 2017 (UTC)
Yeah, there will be a few legitimate but sometimes when the pseudonyms are connected late, there are real mismatches. Annie 21:42, 24 February 2017 (UTC)
OK, the report has been coded and deployed. It's moderator-only since (a) it lets you ignore records and (b) there are only 47 mismatches to process. Ahasuerus 22:21, 24 February 2017 (UTC)
Expected that to be the case. Someone will hopefully get to it. :) Annie 22:29, 24 February 2017 (UTC)

(unindent) So... what are we doing with authors like this one and his co-author. He wrote in Kaqchikel which we do not support... Annie 21:18, 27 February 2017 (UTC)

Well, it is easy to implement a new language; we have added quite a few over the last 5+ years. My only concern is that Kaqchikel is not listed by ISO 639-2 (available on this Library of Congress Web page) and we use a subset of their language codes. It's basically our way of outsourcing various issues associated with languages -- e.g. is X a language or a dialect? -- to an established third party. I guess we were liable to run into a language not listed by ISO 639-2 sooner or later... Ahasuerus 23:41, 27 February 2017 (UTC)
Actually it is there - just hiding. It is one of the Mayan languages - so maybe we should implement "myn" (ISO 639-2 collected them all together due to their obscurity) and we will need to add a note to which one it is into the author page? Annie 23:48, 27 February 2017 (UTC)
PS: ISO 639-3 has it on its own but I do not think we need to go that deep... Annie 23:49, 27 February 2017 (UTC)
Thanks for looking it up! This is a pretty good example of the challenges created by choosing between ISO 639-2 and ISO 639-3. Kaqchikel (aka "Cakchiquel" and a bunch of other spellings) is not directly supported by ISO 639-2. On the other hand, ISO 639-3 has codes for "Cakchiquel, Central", "Cakchiquel-Quiché Mixed Language", "Cakchiquel, Santa María De Jesús", "Cakchiquel, Southern", "Cakchiquel, Acatenango Southwestern", etc. In other words, ISO 639-2 is too little and ISO 639-3 is too much for our purposes. Perhaps there is a happy medium that I am unaware of because I am only peripherally familiar with these standards.
In this particular case I am tempted to go with "cak" for "Cakchiquel" (listed as "Cakchiquel, Central" by ISO 639-3), but I am concerned about establishing a precedent. Ahasuerus 00:11, 28 February 2017 (UTC)
I vote for going for "myn" - it is how it gets identified under ISO 639-2 anyway (in its group) AND add bibliographic notes(which one day will get migrated). Otherwise I am worried about precedents as well. Maybe keep notes and when we see more of these cases, we can discuss moving to 639-3?Annie 00:26, 28 February 2017 (UTC)
OK, I have created an FR. I will put it on hold for a bit in case there are other opinions. Ahasuerus 01:55, 28 February 2017 (UTC)
Done. Ahasuerus 21:18, 2 March 2017 (UTC)

2017-02-23 site slowdown

Response time was pretty bad around 7:45pm server time. There were a lot of additional users hitting the site all at the same time. I am not sure what had brought them here, but they appear to be gone now. Perhaps the slashdot effect. Ahasuerus 01:03, 24 February 2017 (UTC)

Year -> Date

All occurrences of the word "Year" have been replaced with the word "Date". The changes affect all bibliographic pages, edit pages, mouse-over Help bubbles and Wiki-based Help. They do not affect the way field names appear on moderator review pages due to certain technical issues which would take a huge amount of time address. Also, "Award Years" are still "Award Years" -- until someone starts giving monthly or daily awards :-)

If I missed anything, please let me know. Ahasuerus 01:32, 24 February 2017 (UTC)

Additional occurrences of "Year" have been identified and corrected. Ahasuerus 15:30, 24 February 2017 (UTC)
What about those in Advanced Search drop downs? Uzume 01:53, 25 February 2017 (UTC)
I think we are fine there. "Title Year" and "Publication Year" require YYYY values. Similarly, "Title Month" and "Publication Month" require YYYY-MM values. Ahasuerus 02:06, 25 February 2017 (UTC)
I was thinking more of the sort fields like title search sort by year and pub search sort by year. Uzume 02:37, 25 February 2017 (UTC)
Good point - fixed. Ahasuerus 19:57, 25 February 2017 (UTC)
BTW, pub search results still show things like "Sep 1999" in "Date" column whereas title search shows things like "1997-00-00" in the "Date" column. Uzume 02:37, 25 February 2017 (UTC)
Thanks, I'll create a bug report. I have tweaked a few pages to make dates more consistent (see earlier discussions and patch notes), but there is a fair amount of inconsistency left. Ahasuerus 19:57, 25 February 2017 (UTC)
Fixed. Ahasuerus 20:53, 25 February 2017 (UTC)

Redesign of the Verification Status Panel

I think we need to revisit the design of verification status panel to account for inactive verifiers. There are presently five primary verification slots and for many pubs these are completely filled. The majority of primary verifiers in many cases are inactive. I propose that inactive primary verifiers should be removed as primary verifiers and moved to side under the heading of Inactive primary, thus opening up the slots for the active verifiers.

I would also like to propose that a “manually entered” change history page be linked to each verified pub. With up to six verifiers, including transient, it’s difficult to know who to notify if a change is made, and the post clutters up the verifier’s talk page.--Rkihara 18:32, 24 February 2017 (UTC)

There is a Feature Request to "Re-do primary verifications":
  • The verification page is to be split into three sections
  • The top section of the Web page will be for the currently signed-in user's primary/transient verification. If the user has not verified the pub, this section will give the user the following choices:
    • Primary-verify the pub
    • Transient-verify the pub
  • If the user has already verified the pub, he will be given the following choices:
    • Remove his verification
    • Change the verification type from Primary to Transient or vice versa depending on the type of the current verification
  • The middle section will display other users' primary verifications. The currently signed-in user will not be able to change them.
  • The bottom section will display secondary verifications. They will be displayed using the current table using the current layout and the current functionality.
This FR was listed on my short list of things to implement in 2017 and received more support than the rest of the FRs. Ahasuerus 18:46, 24 February 2017 (UTC)
Looking forward to it!--Rkihara 19:35, 24 February 2017 (UTC)

"Interviewed Author" added to Advanced Title Search

"Interviewed Author" has been added to Advanced Title Search as a selection criterion. Ahasuerus 20:36, 25 February 2017 (UTC)

Publication formatting

The way publication dates appear in our standard publication listings -- the table format used by the title page, the publication series page, Advanced Search results, etc -- has been changed to be consistent with the standard title table. In order to accommodate this change, the "Date" column was made slightly wider at the expense of the "Author/Editor" column. If the table layout doesn't look quite right, please do a full page reload (Control-F5 in most browsers) which should refresh the formatting information used by your browser. Ahasuerus 20:59, 25 February 2017 (UTC)

Advanced Title Search - simultaneous use of Synopsis and Notes

Advanced Title Search has been modified to support specifying Synopsis and Notes search values in the same search. Ahasuerus 21:12, 25 February 2017 (UTC)

Enhancing Advanced Search

Advanced Search is a useful tool, but it has a couple of issues.

First, some searches take a long time to complete. Not only is it an annoyance for the user who requested the search, but it can also affect other users if the Advanced Search locks certain parts of the database until it's done.

Second, the ability to specify a combination of AND and OR search conditions can be ambiguous. For example, if you select:

  1. "Title contains 'Star'", AND
  2. "Title Type is exactly NOVEL", OR
  3. "Title Tag contains 'juvenile'"

in Advanced Title Search, should the system search for:

  • (condition 1 AND condition 2) OR (condition 3)

or

  • (condition 1) AND (condition 2 OR condition 3)

? (For the record, it uses the latter algorithm.)

It so happens that these two problems have the same root cause, namely the fact that we allow a combination of "AND" and "OR" conditions. If we were to limit each Advanced Search type to either all "ANDs" or all "ORs", we could resolve these two problems relatively easily. You would still be able to search for "condition A AND condition B AND condition C" or for "condition A OR condition B OR condition C", but not for "condition A AND condition B OR condition C".

Given the ambiguous nature of mixed AND/OR searches (see above), I don't think it would be much of a sacrifice. Does anyone rely on mixed AND/OR searches? Ahasuerus 00:30, 27 February 2017 (UTC)

OR gets a precedence to AND? That's interesting... I would have assumed the opposite. Or is it just always executing right to left, considering them the same order, regardless of what the operator is (so C1 OR (C2 AND C3) as opposed to (C1 OR C2) AND C3)? Annie 18:42, 27 February 2017 (UTC)
I am not sure it's consistent. I'd have to dissect the SQL queries that the Advanced Search software generates. Ahasuerus 20:22, 27 February 2017 (UTC)
Having a mixed case is useful in some cases (as long as you know how it will work after all - once you know the precedence, you can write your query properly after all. If the price for performance improvement is losing the mixed case, then sure, cut it. But if there is a way to keep it (only getting called when there is a real mix of And/or), I'd like to still have the option. Annie 18:42, 27 February 2017 (UTC)
I too think mixed is very useful (but tricky to program I am sure). I think fixing the code to be consistent and documenting the order would be good. I recommend just always doing: ((A op1 B) op2 C). If someone wants a different order you can craft it the other way around this way you can have any of (A AND B) AND C, (A OR B) OR C, (A AND B) OR C, (A OR B) AND C. Meaning, just make the first operator the precedence and document that (it is simple and gives you both options, albeit with some work for the search/query crafter). Uzume 21:33, 27 February 2017 (UTC)
Unfortunately, the way our version of the MySQL optimizer works, mixed AND/OR cases can take a very long time to complete. Some of the odder queries that I have run on the development server took well over a minute (sic) to finish. They may be somewhat faster on the live server for obvious reasons, but even 20-30 seconds is way too long for a query to run, especially given its potential impact on other users.
(Puts his technical hat on.) The only way (that I know of) to address the issue with complex ORs in MySQL 5.x is to switch to UNIONs. I'll have to toy with the query generator to see if we can do it while still supporting mixed AND/OR queries. Ahasuerus 22:06, 27 February 2017 (UTC)
How about splitting them - inspect the query before you run it -- if it is mixed, then go the long way; if it is not (go the easy way - provided that we do not do that already). Add a warning at the top of the page that mixed queries are going to be slow and be done... Now - if the concern is that such a query can slow everyone else down, then maybe make it available only to logged in users? Or disable it until a better solution is found. Annie 22:13, 27 February 2017 (UTC)
I don't use OR, but I could use NOT.--Rkihara 23:11, 27 February 2017 (UTC)
Way back "AND NOT" was supported along with "AND" and "OR". However, it could cause serious performance problems, so we disabled it. It happened a long time ago and the Advanced Search software has changed quite a bit since then. The Feature Request to fix and re-enable "AND NOT" was creates in 2007. Depending on the outcome of the current discussion, we may be able to re-enable "AND NOT". Ahasuerus 23:27, 27 February 2017 (UTC)
AND NOT would be very, very useful. More so than mixed AND/OR. But possibly if you found a performance-friendly way to implement one, it would allow the other, too.
As a last-ditch option, maybe there could be a page for logged-in users to submit a complex search query to be run during daily update time (scheduled slowdown time)? --Vasha 12:58, 28 February 2017 (UTC)

Rie Sheridan (Rose)

I would like to switch Rie Sheridan's canonical name to Rie Sheridan Rose, which she's been using since 2008 (and recently republished the old stories under the new name), since we now have more titles under the new name. --Vasha 12:30, 28 February 2017 (UTC)

Sounds reasonable. Unfortunately, there is no way to do a mass change, so it will have to be done one title at a time. Ahasuerus 19:55, 28 February 2017 (UTC)
Working on it now. --Vasha 21:11, 28 February 2017 (UTC)

"Shortfiction" section title

Shouldn't the section title "Shortfiction" on summary pages be changed to "Short Fiction"? There's already "Short Fiction Series". --Vasha 13:09, 28 February 2017 (UTC)

I agree. I also think that we should change "shortfiction" to "short fiction" on Publication pages. Ahasuerus 14:49, 28 February 2017 (UTC)
Makes sense to me. Albinoflea 16:58, 28 February 2017 (UTC)
I vaguely remember asking for a change of this earlier so I still support it :) Annie 19:09, 28 February 2017 (UTC)
Done. Ahasuerus 19:54, 28 February 2017 (UTC)
Retrieved from "https://isfdb.org/wiki/index.php?title=ISFDB:Community_Portal/Archive/Archive42&oldid=479014"