ISFDB:Community Portal/Archive/Archive42

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

---

Expanded archive listing

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)