ISFDB:Community Portal/Archive/Archive36

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

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

Archives of the community portal for April - June 2015

Suggestion for a correct sort order of multiple printings of a publication when date is not known

I just came across another case where I couldn't find any information about the publication date for the second printing of a book and had to enter it as "0000-00-00" - which results in wrong ordering in the list of publications ("0000" for the second printing comes first). Suggestion: add the possibility to sort the publications by adding a "fake" date to the "0000-00-00" - similar to the possibility to sort the titles in collections. Example:

Year of first printing, known because of reviews, publisher's catalogue etc: "2009-03-01"
Year of second printing not known, but obviously has to be later than the first printing: "0000-00-00|2009-03-02"

The software would use the date behind the pipe for sorting if it exists and as a result, the second printing would appear below the first printing in the list.

Jens Hitspacebar 16:35, 28 February 2015 (UTC)

Ah, well, I just realized it's a DATE field, where the pipe solution will not work. Hm. Jens Hitspacebar 17:32, 28 February 2015 (UTC)

Not quite the same thing, but FR 491 covers similar territory. Ahasuerus 19:00, 28 February 2015 (UTC)

I think that FR would be an improvement. It would still be nice to be able to break ties, though. I'm thinking of books like The Magicians of Caprona, where all we know about the first US & UK editions are "1980-00-00", but for other reasons we know which one came first. Or "0000-00-00" dates, where the 3rd printing certainly came before the 4th printing. Some editors have used fake dates to force the order, which is unsatisfying, but some way of specifying an ordering would be helpful. Chavey 22:24, 28 February 2015 (UTC)

Maybe a second DATE field called "date_sort" would be a solution. It'd be optional and only accepted if a publication date has been entered as well. If set it would be used for sorting. Jens Hitspacebar

If we're adding fields, why not a printing field? If I'm searching for a match to a publication I have in hand and I don't know the date, I'm checking for publisher and price. Then I check contents to see what printing it was. If printing were part of the publication list, I'd be happy. The sort date works for a single published format, but when you get multiple and don't know dates, you're really just guessing anyway. Doug 13:51, 2 March 2015 (UTC)

I'm also interested by a "printing rank" field. Hauck 14:36, 2 March 2015 (UTC)

This was discussed a few years ago and I thought we had an FR to add a "printing" field, but apparently not. Now we do :-)

Once we add it, the software should be able to do more intelligent sorting by date/publisher/printing. It will not be perfect because printing sequences may span multiple publishers, but it should be better than what we have now. Ahasuerus 15:08, 5 March 2015 (UTC)

I had always held on requesting a printing field as while it would greatly help in the grouping and sorting there were several aspects that seemed to cause pain:

We are not consistent in publisher names and/or publishers change their name slightly while continuing a printing series. The fix for this would be to add a field that points to the record for the first printing. If someone does not fill in the field then we'd use publisher name matching.
There are a large number of publications where we can only assume it's the first printing. They don't have number lines. Would we enter these as "1" in the printing field but also include a note that the printing is not stated? Perhaps we could enter "0" or some other special code.
I'm guessing we would enter "9" if a publication says "Ninth printing, October 1978" and that it would get grouped under the first printing but with a date?
How do we handle publications that use a letter line? For a while DAW Books had a word line where they spelled out "First" "Second" "Third". Would these get normalized to Arabic numbered 1, 2, 3?
How do we handle what appear to be later printings but they don't have a printing number?

--Marc Kupper|talk 15:57, 4 April 2015 (UTC)

All reasons why implementing a printing field would create more problems than the need to sort publications. An undated publication is an undated publication. Why make an attempt to sort it within those which are dated, or even within those at the end of the list which are not dated? My wish for a software change would be easier to implement: all unmonthed (only year dated) publications should be sorted as month-dated "13" unstead of "0" as it is currently. That way a later printing dated the same year as the first (but unmonthed) would not be displayed before the day and/or month dated first printing as it does now. Mhhutchins 17:53, 4 April 2015 (UTC)

Birth Place: Ireland

In general, we try to abide by our rule that "For locations whose names have changed over time, use the name as of the time of the author's birth". For Ireland, we use "Kingdom of Ireland" up to 1801, then "Ireland, UK" up to Dec. 5, 1922. From there, Northern Ireland has been consistent, but the larger part of the country has had three names: "Irish Free State" from 1922-12-06 to 1937-12-28; "Ireland" from 1937-12-29 to 1949-04-17; then "Republic of Ireland" from 1949-04-18 to the present. However, we clump together those three names as "Ireland". Should we? Chavey 10:23, 31 March 2015 (UTC)

I guess there is a larger issue here. Technically, Mexico is "United Mexican States", France is "the French republic", Germany is "the Federal Republic of Germany", Russia is "the Russian Federation", Italy is "the Italian Republic", etc. I think that using these official names would be an overkill. Ahasuerus 17:01, 31 March 2015 (UTC)

Thanks for the confirmation. I had just spent a bunch of time getting pre-UK date places listed properly, and was then worried that I was being OCD about the pre-1807 names, while ignoring later name changes. I think it's still pretty subjective, for example, to distinguish between Kingdom of Ireland/United Kingdom/Ireland, but not to distinguish between Irish Free State/Ireland/Republic of Ireland, and I certainly wouldn't want to have to write up rules to say when we pay attention to name changes, but I'm comfortable with going with "intuition" on when a change is significant enough to be reflected in our Birthplace names. Chavey 17:37, 2 April 2015 (UTC)

Stable links to the BLIC

I was surprised to read on Help:Using HTML in Note Fields#Links that to to the British Library Integrated Catalogue (BLIC) are not stable. If you are looking at a BLIC record such as this one for Science fiction by Mark Bould there's a "send to_v" in the upper right corner. If you right click on that and copy the link address then you will get a stable URL to that record. You can also use the E-mail feature and will get an e-mail that contains a stable link. Here are examples of links for the same record using Send-to and e-mailing the record to myself:

Send to	http://catalogue.bl.uk/primo_library/libweb/action/dlDisplay.do?vid=BLVU1&afterPDS=true&institution=BL&docId=BLL01016048936#
E-mail	http://catalogue.bl.uk/primo_library/libweb/action/dlDisplay.do?vid=BLVU1&afterPDS=true&institution=BL&docId=BLL01016048936

I'm wondering if this method is too mysterious to be documented? --Marc Kupper|talk 20:15, 4 April 2015 (UTC)

On October 9, 2013, I made a post to the Community Portal that all links to BLIC in the database should be removed because they were session-based and disappeared within 24 hours. After much research and testing, on November 12, 2013, I made another post which explains how one could create a stable link to an BLIC record. There are still hundreds of records in the database which use the first method and still need to be fixed. The primary verifiers of those records, and any conscientious ISFDB editor who comes upon them, should fix them using the method I described there. I will repeat that method here:

This link searches for BLIC record number 1234567890:

http://catalogue.bl.uk/primo_library/libweb/action/search.do?&vl(freeText0)=1234567890&fn=search

Just replace "1234567890" with the BLIC system number and it links properly. For example in this record, I entered "011983806".

http://catalogue.bl.uk/primo_library/libweb/action/search.do?&vl(freeText0)=011983806&fn=search

Of course, this requires that an editor first search for the record on BLIC, and once finding it, create an HTML link in the ISFDB record's Note field to the record on the BLIC website. At least one editor (SFJuggler) uses the method I describe, and I've seen no problems with the links in his notes.

I believe the method I describe is simpler and more direct than the one that Marc describes, though I'm not doubting that both works. I will add the method I discovered to the HTML help page. Thanks for pointing it out, Marc. Feel free to provide any alternative methods. Mhhutchins 16:46, 5 April 2015 (UTC)

Your method seems simple enough though your example confused me a little. From A Feast Unknown you linked to System number 011983806 is for A feast unknown : volume IX of 'The memoirs of Lord Grandrith', edited by Philip José Farmer. A better record seems to be BLIC BLL01016118308.

One difference is Mhhutchins' method shows a search result and you then click the title and then the "details" tab to view the record. I was showing how to get a direct link to the record's details. In looking into this more I found a simpler process. The UIN is shown as the last line in the details of every record. You append the UIN to http://catalogue.bl.uk/primo_library/libweb/action/dlDisplay.do?docId= to construct a direct link.

For exaple, If I search the BLIC for 'falling free bujold' it shows the search results for one record. I click 'details' in the search results. This shows the details in a frame with a scroll bar. Scroll to the bottom and the UIN is BLL01010084954.

http://catalogue.bl.uk/primo_library/libweb/action/dlDisplay.do?docId=BLL01010084954 is the direct link to the record for the publication Falling Free.

It appears you can convert a BLIC system number into an UIN by prefixing the system number with 'BLL01'.

Here are examples of linking using the system number vs. the UIN:

and:

--Marc Kupper|talk 23:05, 5 April 2015 (UTC)

Windy City Pulp Convention

I'll be attending the Windy City Pulp Convention this month. If anyone from the ISFDB is attending, would you like to meet for a drink?--Rkihara 20:04, 9 April 2015 (UTC)

Minor changes to the Tag Editor page

When I started working on making the Legal Name field a repeating field (similar to tags, email addresses, Web pages, etc.) I found that different fields behaved in subtly different ways. This is a bad thing for a number of reasons, so I have been working on standardizing the way these fields are handled.

As a side effect of my work, the Tag Editor page looks somewhat different now. Specifically, the field label is a little different and the field length is different. In addition, mouse-over help has been added. Ahasuerus 22:21, 9 April 2015 (UTC)

What kinds of awards do we want to support?

(copied from the Help desk)

The Moonbeam Children's Books Awards (Goodreads, Facebook, Wikipedia) are an interesting case, which may have wider implications. The organization that runs these awards charges authors an entry fee, $85 at this time, if they want their book to be considered. So are they really awards as we know them or are they a promotional tool? As one Web site puts it:

Every year, a round of book contest award programs is launched, and every year, on may publishing fora, a battle is waged regarding if the contests are worth it. [...] There are a lot of companies making a lot of money off contests. Most are “beauty pageants” where every entrant wins a nominal prize just for entering their book(s). Who really wins? Well, the promoter of said prize contest. Figure that at least 2500 people will pony up $50 or more for their book to be considered. Post a website and print 2500 certificates, and you’ve made an 85% profit. And most of the people participating are perfectly happy.

Do we really want to list "contest awards", especially if authors have to pay a fee in order to submit their entries? Ahasuerus 01:39, 13 April 2015 (UTC)

I don't think any effort should be made to add these "contest awards" to the database. Mhhutchins 02:02, 13 April 2015 (UTC)

Ditto. PeteYoung 02:08, 13 April 2015 (UTC)

I also agree on not considering these kind of awards. Stonecreek 03:07, 13 April 2015 (UTC)

Likewise, although for $85 I would be willing to reconsider... Albinoflea 21:02, 15 April 2015 (UTC)

New report: My Errored Out Edits

The top section of the navigation bar has been changed to let editors view a list of their errored our submissions. The report is currently empty for most editors (a good thing), but JLochhas, Hitspacebar, Linguist and Mhhutchins should be able to see their errored out submissions. In addition, the layout of some "My Recent" reports has been tweaked not to display the rightmost column if it doesn't make sense in the context of the report. Ahasuerus 02:11, 13 April 2015 (UTC)

Once resolved, is it possible to remove an item from the list? There doesn't seem to be a problem with the submission I made (I must have corrected it subsequently), but unlike the other lists, it should be removable to avoid having to determine every time you check the list whether the problem has been fixed. Maybe have an "issue resolved" link? Thanks. Mhhutchins 03:21, 13 April 2015 (UTC)

I can see how it would be useful. On the other hand, I am concerned that a naive editor may "resolve" an errored out submission and make debugging more difficult. I'll have to think about it; perhaps a compromise solution is possible. Ahasuerus 03:39, 13 April 2015 (UTC)

I'm not even sure if the report for non-moderating editors would be useful at all. Would they even know how to "fix" any problems that was caused by the errored-out submission? Perhaps I'm misunderstanding the report's purpose. Mhhutchins 04:42, 13 April 2015 (UTC)

Well, the main purpose of these reports is to tell our editors what happened to their submissions. If a submission errors out, it doesn't show up on any of the three main reports ("Pending", "Approved" or "Rejected"). Without this additional report the submitting editor would have no way of telling what happened to his errored out submissions.

In addition, this report lets editors view their errored out submissions as they existed prior to approval, which can be useful when a submission is partially approved. Ahasuerus 06:32, 13 April 2015 (UTC)

ISFDB downtime -- 7pm server time

I will be taking the site down at 7pm server (US Central) time. Everything should be back up within 5-10 minutes. Ahasuerus 23:38, 16 April 2015 (UTC)

We are back up. Ahasuerus 00:04, 17 April 2015 (UTC)

New author field - Transliterated Legal Name

The patch that will be installed at 7pm server time will add the much-discussed repeatable "Transliterated Legal Name" field to author records. It should be reasonably self-explanatory, but more detailed patch notes will be posted later tonight. Ahasuerus 23:59, 16 April 2015 (UTC)

See Alexander Bogdanov for an example of an author with two transliterated legal names. I am currently rerunning the nightly job, which will add a new cleanup report for "Non-Cyrillic Authors with Latin Legal Names". Ahasuerus 00:05, 17 April 2015 (UTC)

The nightly job has been rerun and moderators can now access the new cleanup report, "Authors with a non-Latin Working Language and a Latin Legal Name". Non-moderators can use Advanced Search to find "/"s or "("s in Legal Names and then submit edits moving transliterated values to the new field.

A few related issues have come up while working on these changes. First, it looks like some editors have been copy-and-pasting Cyrillic names from Wikipedia. It turns out that Wikipedia adds special diacritics to indicate which syllable should be emphasized, e.g. see http://en.wikipedia.org/wiki/Yevgeny_Baratynsky . We shouldn't be entering these diacritics because they mess up searches. I believe I have identified and corrected all of these problem names.

Second, it turns out that the way email addresses and Web page URLs were filed into the database had a problem. Normally, emails and URLs should not contain Unicode characters (although they are supported by Web browsers) and our software tried to enforce this restriction. However, the implementation was very poor: an edit containing an email address or a Web page URLs with embedded Unicode characters would be accepted, but the value would be silently discarded without letting either the submitting editor or the approving moderator know about it. Ugh. For now I have changed the software to accept email addresses and Web page URLs as submitted. There is another feature request to implement more intelligent URL validation at data entry time.

Third, I have discovered a problem with the way Advanced Search handles "OR" searches when multiply occurring fields like transliterated legal names, Web page URLs and Pseudonyms are involved. It's not an easy problem to fix, so I created a Bug report and left it alone for now.

Finally, please keep in mind that this is our guinea pig for implementing multiply occurring fields for transliterated values. If everything works well, we can expand it to cover Publishers, Pub Series, Titles and eventually canonical author names. Feedback is more than welcome! Ahasuerus 03:56, 17 April 2015 (UTC)

About your last statement: God, I hope not. There are editors who can barely understand how to fill out the fields of the software as it is currently. I can't imagine having to explain the use of multiple recurring fields. Mhhutchins 04:48, 17 April 2015 (UTC)

Well, a multiply occurring field is simply a field that has an "Add ..." button associated with it. We already have quite a few of them, e.g. "Add Author", "Add Artist" or "Add Web Page".

The idea behind adding a separate field for transliterated values is to simplify things rather than to complicate them. Currently, some editors enter non-Latin values using the original alphabet, some use transliterations and some use a mix. For example, consider this publisher. The Cyrillic name is "АРМАДА", but we list it as "АРМАДА (Armada)" with the Romanized form of the name appended in parentheses. Even worse, its only publication series, "Фантастический боевик", is listed as "Фантастический боевик (Fantasticheskij boevik, en. Funtastic Thriller)." [sic!]. Once the proposed "Transliterated ..." fields have been added to Publisher and Publication Series records, we will be able to move all Romanized names to these new fields and everything should look much cleaner. Ahasuerus 15:21, 17 April 2015 (UTC)

Thanks for this new, welcome field. If I understood the help notice right, does this mean that it can also be used for, say, Cyrillic or Japanese transcriptions of latin-alphabet author names ? Linguist 15:24, 17 April 2015 (UTC).

This is probably something that we will want to discuss on the Rules and Standards page in greater depth, but I can think of two types of scenarios which call for entering non-Latin names in the new field.

The first scenario covers languages which use multiple alphabets or, more generally, multiple "scripts" since not all scripts are alphabets. For example, Japanese names are usually written using kanji, but in certain cases hiragana or katakana are used. In Serbian, both Latin and Cyrillic are used. In Azerbaijani, the Perso-Arabic script, Cyrillic and Latin have all been used at different points in time (and place.)

When dealing with these cases, my preferred approach would be to enter the "primary" script of the author's working language in the Legal Name field and enter the other non-Latin forms of the legal name in the new Transliterated Legal Name field. Sometimes which script is primary is reasonably clear, e.g. in Japanese it's kanji and in Serbian it's Cyrillic. Other times it can be complicated, e.g. the primary script of the Azerbaijani language depends on whether the person lives in ex-Soviet Azerbaijan or the Azerbaijani-speaking region of Iran.

The second scenario covers people who have lived in different countries which use different alphabets/scripts. For example, Alexander Lomm was born in Russia to Czech parents who moved back to Czechoslovakia when he was still a boy. He published a number of SF works in Russian between 1959 and 1976 and then switched to Czech. We currently list Russian as his "working language", so I entered the Russian version of his legal name in the Legal Name field while relegating the Czech version and the streamlined Romanized version to the Transliterated Legal Name field.

Somewhat similarly, Pavel Amnuel, who lived in the Soviet Azerbaijan in the 1940s-1950s and wrote in Russian, emigrated to Israel in 1990 although he continues to write in Russian. I would enter the Russian version of his legal name in the Legal Name field and the Romanized/Hebrew versions in the Transliterated Legal Name field.

I can't think of any other scenarios at the moment, but perhaps I am missing something. Ahasuerus 18:00, 17 April 2015 (UTC)

(unindent) A cleanup patch has been installed. There should be no user-experienced impact except that moderators will find a few additional authors on the new cleanup report. Ahasuerus 00:50, 20 April 2015 (UTC)

Publisher's name modification

Hello. Would it be possible for a moderator to modify this publisher's name to Vértice (with an accent), which is the correct spelling ? The accent is not necessary when the word is written in capitals (as on the book covers), but should appear when it is written in lower case. TIA, Linguist 14:02, 17 April 2015 (UTC).

Done. Hauck 14:40, 17 April 2015 (UTC)

Thanks ! Linguist 15:27, 17 April 2015 (UTC).

Link to "view issue grid" for parent series missing if it has sub-series

I'm making myself familiar with series and sub-series right now and came across something odd which looks like a small site navigation bug to me: if you look at the page for the series Whispers, only the sub-series have a "View Issue Grid" link. There's no link to view the issue grid of the whole "Whispers" series and all its sub-series. However, if you go to the issue grid of one of the sub-series, e.g. to that of Supplement to Whispers, there's a link called "Sub-series of Whispers (View Issue Grid)" which leads to an issue grid of the parent series "Whispers" including all sub-series. Shouldn't that link appear on the page for the parent series as well? Hitspacebar 21:24, 26 April 2015 (UTC)

I agree that there should be a link to view the issue grid of the superseries. I'd noticed that in the past but discovered the same work-around that you did. It would be nice to have a direct link. Mhhutchins 22:11, 26 April 2015 (UTC)

The way the series display logic worked 5 (?) years ago when I added support for series grids, there was no easy way to determine whether a series which consisted of sub-series included EDITOR titles. However, I rewrote the series display logic late last year and the new algorithm may allow this determination. I'll see what I can do. Ahasuerus 22:17, 26 April 2015 (UTC)

But there is already a page, and therefore and algorithm, for the issue grid of the parent series and all its sub-series (http://www.isfdb.org/cgi-bin/seriesgrid.cgi?31508 in the example above]. Wouldn't it suffice to simply add a link to this issue grid on the parent series page? Hitspacebar 08:09, 27 April 2015 (UTC)

The grid link is only displayed if there are EDITOR (i.e. Magazine/Fanzine) titles in the series. There is no grid link on non-magazine series pages like Chronicles of Amber (Corwin) or its super-series Amber.

The trick is to determine whether there are EDITOR records in any sub-series before the logic starts displaying them. In the past it was not possible because the series display code displayed one series at a time and called itself recursively for sub-series. However, the recent rewrite added a fair amount of pre-processing, which, e.g., lets it display all tags associated with titles in sub-series -- see the Amber example. Ahasuerus 16:06, 27 April 2015 (UTC)

Looking at the superseries for Whispers, I can see why it might not work to have a link to the "issue" grid, if some of the publications aren't actually issues, i.e. they're book publications. That grid is awfully confusing, even to a veteran. I don't know what a casual user would think of it. Maybe in situations where there's a mix of EDITOR and non-EDITOR records, we shouldn't display a grid at all? Mhhutchins 17:11, 27 April 2015 (UTC)

Mixed magazine-book series have always been a thorny problem since it's a question of degree. If a series contains 1 magazine issue and 100 books, presumably we don't want a grid link to appear. But if it contains 100 magazine issues and 1 book, then we do. Where do we draw the line?.. Ahasuerus 17:40, 27 April 2015 (UTC)

(unindent) The requested change has been made. Ahasuerus 20:19, 27 April 2015 (UTC)

Server performance - 2015-04-26

I am aware of the fact that the server has been sporadically slow today. As far as I can tell, it's not something that we have any control over. If this pattern continues tomorrow morning, I will ask Al to have the hosting company bounce the server. Ahasuerus 22:38, 26 April 2015 (UTC)

Webzines

I've recently encountered a situation where a webzine submission was rejected. What's the deal with webzines? How is one to know which are acceptable and which not? Pkeets 04:46, 29 April 2015 (UTC)

As the Rules of Acquisition explain:

Works published in a web-based publication (webzine) and available only as an HTML readable file are not eligible for inclusion with the following exceptions:

It is published by a market which makes the author eligible for SFWA membership (listed here).
It has been shortlisted for a major award. (This last may include works which are self-published by an author on their own website. Otherwise such works are not eligible.)

That means that a webzine which is downloadable in any number of ebook formats (MOBI, EPUB, PDF, etc.) is eligible for the database. Of course, it goes without saying that the publication must also qualify under the definitions of speculative fiction. If you can provide a link to the webzine, I can give you a definite response as to whether it is eligible. Thanks. Mhhutchins 06:00, 29 April 2015 (UTC)

What is the list that gives webzines "of interest"? Some of these are on the pro list and some not. I believe some may have had nominations for the Nebula Award. Pkeets 12:08, 29 April 2015 (UTC)

That's a set of links to webzines which may be of interest to spec-fic readers. That's all. Being on that list doesn't mean they are eligible for inclusion in the ISFDB. (There's even a proviso stated in the header of the list.) Again, I can only help you if you can give me a specific webzine. Mhhutchins 17:10, 29 April 2015 (UTC)

Okay, Silver Blade is the one that came up. I understand they pay pro rates for poetry, but not fiction. Will that make a difference? Pkeets 18:59, 29 April 2015 (UTC)

If you're referring to this webzine, it doesn't appear to be downloadable as an ebook, so we go to the two exceptions stated in the inclusion policy. 1) It is not listed as a market for which SFWA accepts membership eligibility, and 2) it has not been shortlisted for a major award. So the webzine is not eligible for the ISFDB. Stories by authors published there can be listed and linked on the authors' bibliographic comments pages. Mhhutchins 19:38, 29 April 2015 (UTC)

Okay, thanks. Pkeets 00:40, 30 April 2015 (UTC)

Re warning messages

I have just noticed that no frantic yellow warning message appears when you combine NOVEL + ss in a submission. As forgetting to change an entry type in OMNIBUSes or ANTHOLOGIes appears to be not that uncommon, I was wondering about the relevance of such a warning — or am I the only one to make that mistake ? Linguist 13:31, 30 April 2015 (UTC).

The software allows the possibility of SHORTFICTION contents in a NOVEL-typed publication record, usually a bonus story or an excerpt. Is that what you're referring to? Mhhutchins 17:36, 30 April 2015 (UTC)

No, I mean when you create, say, an OMNIBUS, the entry type is automatically set to NOVEL. When entering a short story, I forgot to change the setting, and just added “short story” to indicate the length. I noticed the mistake when submitting the pub, but I was surprised that the discrepancy didn't trigger a warning signal. Linguist 19:23, 30 April 2015 (UTC).

I don't think such a rare occurrence warrants a warning. Most editors would see the error once they've checked the resulting record. (I may be assuming too much that editors actually check their records after they've been accepted.) I've just done an Advanced Search for NOVEL-typed records with a "ss" storylen and the results were nil. I did other searches: with a "nv" storylen and there were about 50 records returned, with a "nt" there was only one record, and with "sf" there were about 25. I'll correct all of them. It's possible that some of them are not really NOVELs but SHORTFICTION, but more likely that the storylen was the error. Thanks for bringing the possibility of NOVEL-typed records having a storylen designation. Mhhutchins 19:48, 30 April 2015 (UTC)

Sorry about the extra work :o) ! Linguist 19:56, 30 April 2015 (UTC).

Turns out that of the 80 or so records, about a dozen were actually SHORTFICTION, so corrections in the pub records had to be made. In the other cases, it was just a matter of removing the storylen designation. Mhhutchins 21:46, 30 April 2015 (UTC)

Language for synopses

Re this record: is there a stated or implied policy that text on the website be in English? Mhhutchins 20:26, 1 May 2015 (UTC)

Apart from the language policy, there might be another problem: the German synopsis is an exact copy of the synopis of the amazon.de record. Is that allowed? Isn't the synopsis copyrighted to amazon? Hitspacebar 20:36, 1 May 2015 (UTC)

It is my understanding of our language policy (posted earlier this year) that:

...all of our supporting data, i.e. Notes and Synopses, is already in English. The rule of thumb is that we enter actual data "as is" but any data about our data is in English since we had to pick a single language. 700 years ago it would have been Latin, 300 years ago it would have been French and 100 years from now it may be Chinese or something else, but for now it's English.

Ahasuerus 20:42, 1 May 2015 (UTC)

I think copying synopses from another source might be considered fair use as long as it's sourced. (They are usually generated by the publisher to be used in places like Amazon and OCLC.) There are too many editors adding these synopses without giving the source and that should be considered wrong. Mhhutchins 20:58, 1 May 2015 (UTC)

I added this rule to Template:TitleFields:Synopsis. Hope it's ok that way. Hitspacebar 08:46, 2 May 2015 (UTC)

Yes, that was OK. Having it documented makes a better case to present to new editors. Thanks. Mhhutchins 03:20, 3 May 2015 (UTC)

Larry Kramer's novel eligible for the database?

Is there anyone familiar with Larry Kramer's novel The American People: Volume 1: Search for My Heart? I'm holding a submission to add it to the database, but I can find nothing online about any speculative or fantastic element to the book. The closest thing to speculation I've found is that it purports to be a "secret history" of America, and I'm not sure to what extent that makes it eligible. We have works by Pynchon, DeLillo, and other postmodern authors which play similar games. Any advice in this regard is appreciated. Thanks. Mhhutchins 03:42, 3 May 2015 (UTC)

The publisher says:

In this magisterial novel's sweeping first volume, which runs up to the 1950s, we meet prehistoric monkeys who spread a peculiar virus, a Native American shaman whose sexual explorations mutate into occult visions, and early English settlers who live as loving same-sex couples only to fall victim to the forces of bigotry. George Washington and Alexander Hamilton revel in unexpected intimacies, and John Wilkes Booth's motives for assassinating Abraham Lincoln are thoroughly revised. In the twentieth century, the nightmare of history deepens as a religious sect conspires with eugenicists, McCarthyites, and Ivy Leaguers to exterminate homosexuals, and the AIDS virus begins to spread.

So it appears to be a rather sweeping "secret history" rather than a few tweaks here and there that historical novelists have been known to make. Ahasuerus 03:49, 3 May 2015 (UTC)

Thanks for the explanation. I'll accept the submission. Mhhutchins 06:31, 3 May 2015 (UTC)

Klepsydra by Roessner

I believe that Michaela Roessner's story "The Klepsydra" is double listed, once with a sub-title. The story was sold to the (unpublished) Polyphony 7 anthology, and later published in F&SF. I presume this is the same story, and the 2 listings should be combined. RogerSSS 04:45, 4 May 2015 (UTC)

Yes, I'm pretty certain it's the same story. I've merged the two into one record. Thanks for bringing it to our attention. Mhhutchins

Arnason hwarhath

The Eleanor Arnason story "The Woman Who Fooled Death Five Times" is definitely a hwarhath story (the subtitle is "A Hwarhath Folk Tale") and should be listed under that series. RogerSSS 15:13, 4 May 2015 (UTC)

I've done the change, but note that it may be a faster process to submit the modifications yourself. Hauck 15:43, 4 May 2015 (UTC)

Graham Joyce Partial Eclipse

Does anyone own Graham Joyce's Partial Eclipse collection from Subterranean? isfdb lists "The Mountain Eats People" in this collection, my own Joyce short fiction list does not, but I don't have the book to confirm this. Locus does list "Mountain" in Partial Eclipse, but a spot check of library websites does not. Can anyone check this? RogerSSS 02:47, 6 May 2015 (UTC)

The OCLC record doesn't mention it, but they record their contents from the content page (unlike the ISFDB that requires editors to enter contents from each work's title page.) Unlike most of their listings, the publisher's website doesn't give the complete contents. "Mountain" isn't mentioned. The ISFDB listing looks like it came from Locus, so if they're wrong (which seems likely), then our listing is wrong. I'll remove the story, and give the contents based on the OCLC listing and note that Locus adds another story. Mhhutchins 03:00, 6 May 2015 (UTC)

The Locus listing says there are ten stories, but lists eleven, so the extra story is very likely an error. Mhhutchins 03:07, 6 May 2015 (UTC)

Ben Peek's "Verflucht"

The german title "Verflucht" (German series title: "Ära der Götter") this title seems to be a variant title from this. The same character, same towns. The only uncertainty is a remark on the copyright page in my book: Titel der australischen Originalausgabe:"Immolation", Tor Books, London 2014. This title appears only here. Can someone tell me, what is right? --Wolfram.winkler 20:02, 10 May 2015 (UTC)

It's a title change, see what the author says. Hauck 21:19, 10 May 2015 (UTC)

Thanks for the info--Wolfram.winkler 07:03, 11 May 2015 (UTC)

I have varianted "Verflucht" to "The Godless" for you. Christian Stonecreek 09:19, 11 May 2015 (UTC)

You're faster than me, thanks.--Wolfram.winkler 11:28, 11 May 2015 (UTC)

Jared Diamond

I am wondering why the author Jared Diamond is on ISFDB. We have one non-fiction book work for him that was not reviewed and does not seem remotely specfict. Did I miss something or should this publication, title, and author be deleted? --Marc Kupper|talk 02:43, 16 May 2015 (UTC)

I say nuke it. Mhhutchins 03:34, 16 May 2015 (UTC)

<AOL> Ahasuerus 03:55, 16 May 2015 (UTC)

nuked</AOL> --Marc Kupper|talk 13:50, 16 May 2015 (UTC)

Broken link to help page on "Title Editor" page

The help link next to the "Graphic Format" label on the "Title Editor" page (cgi-bin/edit/edittitle.cgi) is broken. It links to the non-existing page http://www.isfdb.org/wiki/index.php?title=Template:TitleFields:Graphic but should probably link to http://www.isfdb.org/wiki/index.php?title=Template:TitleFields:GraphicFormat. Jens Hitspacebar 11:17, 16 May 2015 (UTC)

Moreover, the content of the help on http://www.isfdb.org/wiki/index.php?title=Template:TitleFields:GraphicFormat ("Titles of all types can be "graphic" except REVIEWs and INTERVIEWs.") doesn't fit the mouse hint you get when you hover over the question mark for the "Graphic Format" field ("COVERART and INTERIORART titles can't be graphic."). Hitspacebar 11:26, 16 May 2015 (UTC)

I updated Template:TitleFields:GraphicFormat to also state COVERART & INTERIORART cannot be marked as graphic. Fixing the link and updating the tooltip to include REVIEW and INTERVIEW will require a code change. Ahasuerus will likely take care of that when he sees this. -- JLaTondre (talk) 12:48, 16 May 2015 (UTC)

I would think EDITOR should also be an exception unless we plan on allowing graphic format periodical publications, i.e. comic books, into the db. Even though it might be stretching the definition, I suppose an ESSAY can be presented in a graphic format, which means a book-length nonfiction graphical work would make NONFICTION eligible for the flag as well. Mhhutchins 17:19, 16 May 2015 (UTC)

Yes, ESSAYs can be presented in a graphic format. Here's a current example I just edited: The Religious Experience of Philip K. Dick. Jens Hitspacebar 17:57, 16 May 2015 (UTC)

(unindent) Fixed. Thanks for identifying and reporting the issue! Ahasuerus 04:00, 17 May 2015 (UTC)

Messages and Replies

1. You have a new message. This takes you directly to the bottom of your talk page which is not necessarily where the 'new' message is. To find it one has to click the 'history' tab to find out what's been changed. Is there a more obvious way of finding the 'new' message or is there no way the yellow warning can take you directly there ? --Mavmaramis 02:44, 17 May 2015 (UTC)

I am afraid that our Wiki's behavior, including the way "New Messages" functions, is largely outside of our control because we use third party Wiki software. There are some things that we can customize, but they are relatively few and far between compared to the behavior of the core ISFDB application, which we are much better positioned to control.

I don't have a good answer to your specific question, but perhaps some other, more Wiki-clueful editors may provide one. Ahasuerus 04:07, 17 May 2015 (UTC)

As Ahasuerus said, there is no way to directly get to the new message(s). If it's a new topic, it should be at the bottom of the page (assuming the message leaver properly used the "+" to add a new topic). If it's not obvious from scanning the page what is new, you can use the page history. Click the "history" tab at the top the page and then click the "Compare selected revisions" button on the new page. This will show the difference between the current version and the previous version. You can also select which version you are comparing if you need to go back more than one revision. -- JLaTondre (talk) 12:02, 17 May 2015 (UTC)

Actually, there is a way to go to the page in "diff since last viewed mode". The new messages link was originally set up to do that, but it has been changed to simply bring up the talk page instead. It could be changed back. I agree with the sentiment and would prefer the diff mode behavior be restored, although there must have been some other usability issue that motivated making the change. --MartyD 18:31, 18 May 2015 (UTC)

The method is to add "?diff=cur" to the link to the page. --MartyD 18:37, 18 May 2015 (UTC)

Oh! Yes, I think I remember it now. The problem was that "?diff=cur" only shows you the last change as opposed to all the changes since you last viewed the page. I seem to recall that it led new users to believe that they had only one new message that they needed to respond to. Ahasuerus 18:43, 18 May 2015 (UTC)

I believe the change was prompted by my request. In "diff" mode you couldn't respond, and its display often confused new editors. My suggestion to have the message warning link directly to the user's talk page was just a matter of simplicity. If that's not "standard" wiki procedure, feel free to change it back. And be prepared to deal with new editors who have problems discovering their user talk page. As someone who deals with new editors more than the average moderator, I know that since the change the response to user talk pages has improved tremendously. Mhhutchins 19:09, 18 May 2015 (UTC)

2. Replies. If you ask a primary verifier a question it might be days, weeks or even months before they reply. Does this flag up as "You have a new message" ? Otherwise one would have to (a) remember to whom the message was posted, when, and topic or (b) scroll back through "my contributions" in the hopes of finding it amongst the swath of other messages which are more than likely merely notes about additions to records or image uploads. --Mavmaramis 02:44, 17 May 2015 (UTC)

An editor's response on his or her Talk page won't result in you getting a "You have a new message" notification. If you see that an editor hasn't been active in some time, it may be prudent to say something like "When you see this message and respond here, could you please also leave a note on my Talk page so that I would know to check your talk page?" Ahasuerus 04:07, 17 May 2015 (UTC)

I found out how to use the "Watchlist" function which is probably the way to go. But you're right. All came about as I was wondering how, if someone posts a question on your talk page, they know you've responded. --Mavmaramis 08:00, 17 May 2015 (UTC)

Yes, the watchlist is how the wiki is designed to track updates to pages. It's not perfect, as it shows all changes to the page, but it lets you know when you should check the page. -- JLaTondre (talk) 12:02, 17 May 2015 (UTC)

The latest version used by the English Wikipedia has a message notification system which shows you the changes and allows for linking to specific discussions. It also allows for pinging an editor (using {{ping|username}}), which notifies the editor that someone wants their attention on a specific page. That would be a useful tool here, too. Nihonjoe 16:43, 28 May 2015 (UTC)

In or Out? Fictional Biography of Sir Thomas Mallory

This title is a historical fiction approach to a biography of Sir Thomas Malory. If it were a true biography, Malory is important enough that we would include this as a non-fiction book. WorldCat lists it as a "Biography", and with the little we actually know about his life it may be the closest to a biography we can expect, but it really isn't. As a novel, it has a minimal amount of spec fic in it (primarily in repeating a few of the legends as if they true stories), but to me it doesn't qualify as a genre novel. So I currently have it entered as a non-genre novel. But of course the book isn't by someone who is a spec fic author, so we would not normally include such books in here. So: (1) Does this book belong in? (2) If so, should it be listed as a non-genre novel, as I did? (See the title rec Synopsis for some additional description of the book.) Chavey 00:34, 19 May 2015 (UTC)

Disregarding the subject of the "biography", if you've determined that it's not speculative fiction, why should there be a record for it at all? Unless you believe that the author, someone I've never heard of, is "above the threshold", a non-genre novel by a non-genre author is obviously not eligible for the database. Mhhutchins 02:21, 19 May 2015 (UTC)

It looks like there are two issues here. First, even if this was a non-fiction biography, would we want to include it? What about Lucian of Samosata or Homer? Second, fictionalized biographies and, more generically, fiction about SF writers appear to be a gray area. Would a non-SF novel about Jules Verne or H. G. Wells be "in"? Ahasuerus 02:40, 19 May 2015 (UTC)

Out, there's already too much clutter. Hauck 05:49, 19 May 2015 (UTC)

ISFDB:Policy has "Speculative fiction is defined to exclude / ... works set in a future indistinguishable from the present". I'm wondering if we should add a bullet with "Works set in a period indistinguishable from factual narratives of that period. For example, works set in the U.S. Civil War era would be excluded unless they were alternate histories. Related to this is that works where the author uses country or region names not found on maps, but is otherwise indistinguishable from factual narratives, would also be excluded." --Marc Kupper|talk 21:48, 24 May 2015 (UTC)

The reason why the clause about "a future indistinguishable from the present" was added was to exclude techno-thrillers. At the time it was felt that a techno-thriller set in the year N+2 where "N=present" was not "really SF".

I assume that the second proposed clause covers Ruritarian fiction, right? What types of books do you mean to cover by the first clause? Ahasuerus 06:25, 25 May 2015 (UTC)

Six authors with murky legal names

We have 6 authors whose transliterated legal names are known while the original versions are unknown. The authors and their respective working languages are as follows:

Somadeva Bhatta - Sanskrit.
- Done (सोमदेव भट्ट); he seems to be mainly known as Sōmadēva, though. I have also modified the ṭ of the first transcription, which appeared (according to by browser anyway) as t + square. Linguist 13:40, 28 May 2015 (UTC).
Asakura Hisashi - Japanese.
- I suppose this should be 禅師大谷, Zenji Ōtani, but I just can't find an occurrence of it (which is not a good sign…). Better wait for confirmation (or not) from a native speaker. Linguist 14:39, 28 May 2015 (UTC).
Rokeya Sakhawat Hossain - Bengali
- Solved via the Bengali Wikipedia.--Dirk P Broer 16:24, 29 May 2015 (UTC)
Xia Jia - Chinese
- solved via www.renditions.org.--Dirk P Broer 16:27, 29 May 2015 (UTC)
La La - Chinese
- Solved via www.renditions.org.--Dirk P Broer 16:29, 29 May 2015 (UTC)
Swapanburo - Bengali
- My browser can't cope with this, but the answer seems to be there (legal name followed by pseudo between brackets).
  - Got it, thanks! Ahasuerus 15:18, 30 May 2015 (UTC)

If anyone happens to know the original forms of their legal names, please update the record(s). TIA! Ahasuerus 02:02, 28 May 2015 (UTC)

How about the original form of their canonical name? In the case of La La, it's 拉拉. If we ever get support for credits published in another alphabet, isn't it more important to have their canonical name than their legal name? Especially so in cases like La La who publishes only under that credit, at least according to SFE3. I believe the effort being made to establish the original form of an author's legal name isn't necessary when that's not how their work is actually published. Mhhutchins 04:29, 28 May 2015 (UTC)

I certainly agree that canonical names are more important than legal names, but that's exactly why I started with the legal name field as a test case/guinea pig. I figured that since it was less important, any problems/bugs introduced during the implementation process wouldn't have as much of an impact and, besides, we would likely learn useful lessons. And that's exactly what happened -- there was a minor bug that I still need to fix and I have identified two enhancements that need to be coded :-) Ahasuerus 16:11, 28 May 2015 (UTC)

Asakura's real name is 大谷善次 (Ōtani Zenji), not 禅師大谷 (which is in the wrong order (should be 大谷禅師), and has two incorrect characters in it). The kanji for his pseudonym are 浅倉久志 (Asakura Hisashi), just in case we ever have a way to enter those, too. He was also sometimes called 大浅倉 (Ōasakura). I updated the author page for him. Hope that helps. Nihonjoe 16:32, 28 May 2015 (UTC)

Approved, thanks! Ahasuerus 17:07, 28 May 2015 (UTC)

According to the Japanese Wikipedia he also uses numerous other pseudonyms as well (apparently the publisher thought it did not look good to have everything translated by a single person) such as: 深谷節 (Setsu Fukaya), 沢ゆり子 (Yuriko Sawa), 牟礼一郎 (Ichirō Mure), and 大谷圭二 (Keiji Ōtani). I find it interesting the first two of those are normally considered to be female names. The last one also seems to have its own authority entries (see http://viaf.org/viaf/251386820/). Uzume 14:53, 30 May 2015 (UTC)