ISFDB:Help desk/archives/archive 14

This is an archive page for the Help Desk. Please do not edit the contents. To start a new discussion, please click here.
This archive includes discussions from December 2010 - July 2011.

Archive Quick Links
Archives of old discussions from the Help desk.

---

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

---

Expanded archive listing

Ben Robinson

Hi, anyone know if there are actually 2 Ben Robinsons? It looks likely, but I suppose it is possible one Ben Robinson actually did the illustrations (vs writing) for the Star Trek manuals after decades of non-SF work. Jonschaper 03:05, 1 December 2010 (UTC)

According to an interview with him, he was "a toddler" in the 70's. So he's not the cover artist for those Analogs from the 60's. Chavey 20:14, 1 December 2010 (UTC)

It's not likely they're the same person. I've split the records, giving the recent author the full name and disambiguating the 1960s cover designer. Thanks. Mhhutchins 21:04, 1 December 2010 (UTC)

ISFDB Data vs. Amazon Data

Hello- I'm fairly new to the ISFDB, and wondered if my following generality is more or less fair:

ISFDB data is more accurate/complete than Amazon data for older, out of print, pre-ISBN, or obscure publications. Amazon data is more accurate/complete for newer publications, until the ISFDB community manually incorporates it to the ISFDB.

How do both data sources compare to LibraryThing.com which seems to have hybrid community/Amazon data? Thanks! —The preceding unsigned comment was added by Gideon (talk • contribs) .

Welcome to ISFDB! :-) It's hard to compare our data with Amazon's since we catalog individual stories/essays, which Amazon typically doesn't. On the other hand, they have Look Inside and other features that provide more information than we do.

Generally, in addition to your observations, our data is better/more complete than Amazon's when the item has been physically verified by an ISFDB user. Amazon's data mostly comes from publishers and Amazons gets it months prior to publication, which can be a mixed blessing -- see this discussion of automatically created ISFDB submissions for a review of the challenges that it can present.

As far as social cataloging sites like LibraryThing/OpenLibrary/GoodReads/Shelfari/Anobii/Google Books/etc go, they all have their strengths and weaknesses. It's hard to generalize, but they tend to share the following characteristics compared to ISFDB:

1. They cover a lot more genres than we do
2. They have full time developers whereas we have volunteer developers
3. They have a lot of tools designed to help users enter their libraries whereas all we have is the verification mechanism
4. They are not as interested in the details of individual printings as we are
5. Some of them have more robust APIs that we do

We hope to have better integration with all of these sites (as well as with major library catalogs) in the future, but the field is changing so fast that it's a bit of a moving target. Ahasuerus 20:59, 3 December 2010 (UTC)

Adding my tuppence-worth - we're better than Amazon on most publications when they're actually available. I do try and fix our Forthcoming books on the front-page when we look a bit inadequate. That might be from Look-Inside, Worldcat, Publisher's website, Author's own site, whatever - I do like our front-page pubs to be the best we can do. (If anybody wants to send me free copies if they can't handle such themselves, let me know and I'll enter them for you!) I suspect that we ought to improve the verification mechanism so that people can use it for "own library" processes. It still feels odd to go up to a book-seller tapping away at his computer while I browse their books and ask who verified such-and-such title on ISFDB to see if it was me. BLongley 02:01, 6 December 2010 (UTC)

Verifying a newly entered book

When I create a new book record from a book in my collection, and wish to verify it, it appears that I need to enter the book, submit that data, wait for a moderator to approve it, and then remember to verify it. I suspect that there's no shortcut to creating a verified record, but I thought I should check just in case I was missing something. Chavey 20:05, 4 December 2010 (UTC)

Yes, a record must be created before it can be verified, and I see no other way around it. Before I became a moderator, I used to keep a stack of books beside my desk until they'd all been verified. Moderators have been pretty fast about approving new submissions, so the books shouldn't clutter up your floor too long! Or you can create a note with a shortcut on your desktop in which you list pending edits. Mhhutchins 21:39, 4 December 2010 (UTC)

Thanks! I just wanted to verify (pun intended) that I wasn't missing something. Chavey 22:18, 4 December 2010 (UTC)

You're correct - and I don't think I'd want it any other way. A second pair of eyes on a submission is always good - nobody is perfect. Even as a moderator, that can immediately go and approve my own submission and then verify it, I find that the moderator warnings are a good reminder when I've messed up. And it's probably better to keep those warnings for the moderator rather than reject edits up front, which could put off new editors. Mods are used to fixing beginner errors like "paperback" rather than "tp" or "pb" and feeding back corrections. And we're trying to add more automatic checks on dodgy edits, even our own ones. Or make it easier to check edits. Or make it easier to approve a no-brainer and go onto the next. Long way to go yet... BLongley 01:41, 6 December 2010 (UTC)

Please Revert Edits

I messed up the entries for Christina Rossetti's "Goblin Market" merging them with Ru Emerson's poem of the same name, then made it worse trying to back it out. Please revert the edits if possible.--Rkihara 07:50, 5 December 2010 (UTC)

Unfortunately, once a submission has been approved, there is no easy way of determining what the data looked like before the changes were applied. However, I have a copy of the database as it existed on Saturday morning and I can look some things up.

I don't see *any* poems by Ru Emerson or anything else that would look like "Goblin Market". I do see two titles that look like they need to be VT'd:

Title: Goblin Market
Author: Christina Rossetti
Year: 1862
Type: POEM 

Publications:

• Dark of the Moon: Poems of Fantasy and the Macabre, (1947, ed. August Derleth, publ. Arkham House, $3.00, xvi+418pp, hc, anth) Cover: Frank Utpatel - [VERIFIED]
• The Book of Fairies, (Oct 2000, ed. Michael Hague, publ. William Morrow / HarperCollins, 0-688-10881-4, $19.95, 125pp, hc, anth) Cover: Michael Hague
• The Treasury of the Fantastic: Romanticism to Early Twentieth Century Literature, (Feb 2001, ed. David Sandner, Jacob Weisman, publ. North Atlantic Books/Tachyon Publications/Frog, Ltd., 1-58394-030-8, $27.50, 747pp, hc, anth) Cover: Michael Dashow
• The Book of Fairies, (Nov 2006, ed. Michael Hague, publ. HarperTrophy / HarperCollins, 0-06-089187-4, $9.99, 127pp, tp, anth) Cover: Michael Hague

Title: Goblin Market
Author: Christina Georgina Rossetti
Year: 1859 

Publications:

• Beyond the Looking Glass: Extraordinary Works of Fairy Tale and Fantasy, (1974, ed. Jonathan Cott, publ. Hart-Davis, MacGibbon, 0-246-10810-X, 519pp, anth)
• Beyond the Looking Glass: Extraordinary Works of Fairy Tale and Fantasy, (1978, ed. Jonathan Cott, publ. Simon & Schuster, 0-671-79017-X, tp, anth)
• The Evil Image: Two Centuries of Gothic Short Fiction and Poetry, (1981, ed. Patricia L. Skarda, Nora Crow Jaffe, publ. NAL Meridian, 0-452-00636-8, $8.95, 479pp, tp, anth)
• The Dedalus Book of British Fantasy: The 19th Century, (Sep 1991, ed. Brian Stableford, publ. Dedalus, 0-946626-78-2, £8.99, 416pp, tp, anth) Cover: Richard Dadd
• The Dedalus Book of British Fantasy: The 19th Century, (Sep 1991, ed. Brian Stableford, publ. Dedalus, 0-946626-84-7, £18.99, 416pp, hc, anth) Cover: Richard Dadd

Are you sure it was Ru Emerson whose records were affected? Ahasuerus 08:40, 5 December 2010 (UTC)

OK, I see what the problem is. You set up Rossetti's "Goblin Market" as a VT of Emerson's "Chile Quilas". All I had to do was to break the VT relationship and they are unrelated titles again.

However, I don't see the "Christina Georgina Rossetti" version of the title any more. When I check the pubs where it used to be, e.g. [http://www.isfdb.org/cgi-bin/pl.cgi?THDDLSBKFB1991 The Dedalus Book of British Fantasy: The 19th Century, it's no longer there. Do you happen to recall what you did with this record? I guess it doesn't matter all that much, I can just re-enter the data from the backup. Ahasuerus 08:47, 5 December 2010 (UTC)

There is now a Poem called The Dedalus Book of British Fantasy: The 19th Century which may be the missing record. BLongley 11:10, 5 December 2010 (UTC)

Thanks, it was one of the messed up titles. I believe I have fixed all of them now. Ahasuerus 16:18, 5 December 2010 (UTC)

Thanks for the help! Major screw-up on my part. I'm not a programmer so I don't really understand how the database is structured. I'd hoped that it was like the wiki and it was possible to revert the history.--Rkihara 18:20, 5 December 2010 (UTC)

I'm afraid it's not that simple - ISFDB software has sort of grown incrementally rather than being designed for such. And we're terrible at explaining how to undo things. I believe Ahasuerus's comment "All I had to do was to break the VT relationship" means that he made it a variant of title-id 0. And I suspect what you tried to do before calling for help was to unmerge the titles. Unmerging Content entries like Shortfiction or Poems or Essays leads to new titles named after the "Container Title" - and I can't think of any situation where such is a good move, so we should probably consider software improvements as well as Help improvements. I think we've got to the stage where nobody knows everything anymore, and it might be worth back-tracking a bit and documenting stuff we DO understand. BLongley 01:20, 6 December 2010 (UTC)

Yes, that's exactly what I did. I considered the alternate procedure to break the merge, but as there were several titles merged, "unmerge" seemed to be the most expedient. I realized I was in trouble after doing that and that further attempts to make it right on my own might be counterproductive.--Rkihara 07:20, 6 December 2010 (UTC)

Search for recently added pubs

Is there any way to search the database for recently added publication records, e.g. "all publications added since xxxx-xx-xx", or "all novels added since xxxx-xx-xx"? Chavey 15:12, 10 December 2010 (UTC)

I'm afraid not. The nearest I can think of is to look at "Recent Integrations" and find the "NewPub" records. BLongley 11:43, 11 December 2010 (UTC)

AuthorHeader Bug report

I entered some bibliographic data for the author C. C. Buck Coffman (about a possible identity of this person). I included the AuthorHeader template, but as you can easily verify, the link generated by that template does not take the user back to that author. Chavey 02:25, 12 December 2010 (UTC)

Template fixed. The links work now. Mhhutchins 15:24, 17 January 2011 (UTC)

I found an additional bug in that template. If you go to the "Author" page for F. E. Daniel, M.D., the return link generated croaks on the comma in his name. Chavey 18:42, 18 January 2011 (UTC)

Yes, that's a known bug, at least in the author link template. I got around that one by using the altName. I'll see if that work-around can be used here. Mhhutchins 05:50, 19 January 2011 (UTC)

Nope, can't be done here because the database creates the link from the record and it can't be changed by the user. Anybody else familiar with templates know how to fix this one please go ahead. I can unprotect it if you're not mod-level. Mhhutchins 05:56, 19 January 2011 (UTC)

I'll see if I can coax it into doing the right thing. --MartyD 01:20, 20 January 2011 (UTC)

Ok, it should be working properly now. When no altName is specified, it was using the bare page name as an unlabeled argument to the A template. I modified that to pass the properly encoded page name (PAGENAMEE <- second "E") as "altName" and use the unencoded page name (PAGENAME) as "name". Let me know if you see any examples that do not work correctly. --MartyD 02:11, 20 January 2011 (UTC)

Thanks, Marty! There's also a problem with the {{BioHeader}} template. It looks like the same bug. Try the return link for C. C. Buck Coffman from his biography page. Mhhutchins 05:05, 20 January 2011 (UTC)

Yes. Made the same fix for that and removed the note about the broken link from the Coffman bio page you cited. --MartyD 11:37, 20 January 2011 (UTC)

Thanks guys for all the fixes here! That's the incremental developmental I tell my students about all the time :-) Chavey 11:57, 20 January 2011 (UTC)

how is it possible to see "All Authors" page

Hi, I would like to be able to browse Authors, not just search. Something like we have for Magazines, where you can see all the magazines. I was thinking something like philsp.com has: [1]

I need it for my personal research project, about history of SF.

If this page does not exist now, can I request to generate some csv or something which has this info?

e.g.:

ASIMOV, ISAAC; (1920-1992)

Thanks, QShadow.

P.S. I am compiling this list from many sources so it may be that after my full list is ready I could contribute some new info to the ISFDB back. —The preceding unsigned comment was added by Qshadow (talk • contribs) .

Perhaps the Author Directory might help. But be warned: there are thousands of authors in the database. Mhhutchins 16:31, 12 December 2010 (UTC)

Note that our Author directory also includes artists, editors, reviewers and interviewers. And we won't always have birth-dates, and we'd only have the names in Lastname, Firstname format for legal names, not for Canonical names - although we do have the lastname recorded separately so you could download a backup of the database and run a query such as:

select a.author_lastname, a.author_canonical, a.author_birthdate, a.author_deathdate 
from authors a

If you're only after fiction authors then you can cut down by the type of publication or publication content, e.g. restrict it to Novels and Collections, or Short-fiction contents of Magazines and Fanzines. What type of "Author" are you looking for? BLongley 17:58, 12 December 2010 (UTC)

Thanks, Mhhutchins and BLongley, I didn't even expect to get such a good answers and so fast!

BLongley, I would like to get only Authors of Novels and Stories, and other works of fiction. But not Authors that only compiled collections, or artists, editors, reviewers and interviewers, etc... I do not care about the name format, I can process it later.

But I didn't understand how do I make the query, should I download the database for it, or can i make it through the web interface? I didn't find help about query in the ISFDB help pages... Thanks a lot Qshadow.

The example Bill (BLongley) gave you is a SQL query against the database. To do that, you need to download a copy of the database back-up, load it into a MySQL database, and query against it yourself -- there's no web interface that would let you do that against the "production" database. See ISFDB_Downloads for the back-ups. You can find ALL authors (100 at a time) through the web interface, but there is no way to filter by type of works produced. To get a list of all authors, use Advanced Search; in the "ISFDB Author Search Form" section, enter a percent sign ("%") as Term 1; keep Canonical Name as the choice, and Submit Query. Due to the number of authors, this isn't very practical: There are almost 74,800 records.... --MartyD 19:08, 12 December 2010 (UTC)

Thanks MartyD, I will try this! --Qshadow 21:26, 12 December 2010 (UTC)

OK, the query would be like:

select DISTINCT 
a.author_lastname, a.author_canonical,
a.author_birthdate, a.author_deathdate
from titles t, canonical_author ca, authors a
where ca.title_id = t.title_id
and a.author_id = ca.author_id
and t.title_ttype in ('NOVEL','SHORTFICTION')
order by 1

There's only 41,918 authors by that definition (as of the last backup I loaded). You might want to wait for the next backup, as I've found and fixed 25 dodgy "Last Names" like "(author)". BLongley 17:41, 14 December 2010 (UTC)

SQL Pseudonym help

For a given title, how do we find the actual author given a published pseudonym? Right now, my query essentially pulls up everyone who has ever used a pseudonym as the author. For example, John Campbell as Don A. Stuart wrote "Who Goes There?".

select * from canonical_author where title_id=48111

This returns an author id of 11819. When we, then, go to find out who actually wrote the work, we run into the following problem.

select * from pseudonyms where pseudonym=11819

It lists John W. Campbell and Arthur J. Burks as authors, even though we know Campbell was the real author.

How do we fix this? Davimre 18:32, 13 December 2010 (UTC)

Nothing is broken. Your query isn't asking the question you intend. You are asking which author wrote the title, then you are asking for all authors who used Don A. Stuart as a pseudonym. To find out who really wrote "Who Goes There?", you first need to find the canonical TITLE, and from there the author.

select * from titles where title_id = 48111;

will show you that title_parent is not null, but rather 860248. So 48111 is a variant title, with 860248 the parent/master/canonical entry. If you now do your canonical_author selection using title_id = 860248 instead, you'll get an author_id of 14, which is the id for John W. Campbell, Jr. --MartyD 12:04, 14 December 2010 (UTC)

Variant Titles The Already Exist

I noticed that we have entries for both The Paradox Men and Flight into Yesterday by Charles L. Harness. I remembered reading something about this at The Paradox Men (Magill’s Guide to Science Fiction and Fantasy Literature), which brings up a question. How do you make a variant title out of an entry that already exists? Looking at the "Add Variant Title" screen, it looks like it would make a new entry. Or am I missing something?--AlHazred 15:40, 16 December 2010 (UTC)

The titles have already been linked: see here. If they had not been, you would first have to determine which would be the canonical (or parent) title. Go to that title record and copy the record number in your browser's URL window (the numbers after "cgi?"). Then go back to the title record of the variant and choose the option "Make This Title a Variant Title or Pseudonymous Work" under the Editing Tools menu. The next window has a field labeled "Parent #". Enter the record number of the parent record then click on "Link to Existing Parent". A moderator would then determine if the relationship is valid and either approve or reject the submission. Thanks. BTW because the pages of the database are not part of the Wiki, you can't use the traditional Wiki method of the double-brackets to create links. You can link by entering either the complete URL or using a template. More help for using the ISFDB Wiki can be found here. Thanks again for contributing. Mhhutchins 18:46, 16 December 2010 (UTC)

Suffixes on legal names

Sorry if this is a dumb question, but why does Help:Screen:AuthorData say to omit "suffixes" from legal names? Aren't they part of the person's name? If we don't capture a suffix as part of the legal name, where would we record it if/when it is not part of the canonical name? Thanks. --MartyD 13:45, 20 December 2010 (UTC)

The question is with respect to suffixes such as "Jr.", "Sr.", "III" etc. It does not, presumably, apply to honorific suffixes such as "Ph.D.", "O.B.E.", "K.B.E.", "PJF", etc. For example, "Sir Arthur C. Clarke, K. B. E." is a pseudonym for "Arthur C. Clarke", which seems appropriate. And yet we currently have 116 legal names that include "Jr.", 5 with "Sr.", 19 with "III", and 4 with "IV". So either the Help:Screen:AuthorData needs to be updated, or there's a lot of cleanup to do on these names. Chavey 14:43, 20 December 2010 (UTC)

I don't consider them part of the legal name, although I suppose if the US practice is to include them on the birth certificate they'd have to be considered such. Here, they're informal additions - e.g. I could be considered a "Jr." as I have the same forename and surname as my dad, and he could be "Sr." - or he could be "'IV" and me "V" as the three generations before were the same. Or we could go back through all the Bill Longleys we could find in our ancestry and we'd be at least "VIII" and "IX". Anyway, I don't see a problem - if it's not in the Canonical name then it probably doesn't need to be in the legal name. BLongley 15:54, 20 December 2010 (UTC)

I checked with a friend of mine, John F. Winklemann, III. He said that the "III" is on his birth certificate, as is the "Jr." on his father's. (Of course his grandfather's birth certificate does not have "Sr."). So at least in some cases, this suffix is truly part of the legal name. Of course without checking the birth certificates, we couldn't really distinguish between those that are "legal" from those that are just used familiarly. By the way, in the U.S. it is traditional to only use these types of suffixes for immediate descendants, so my son "Peter Chavey", who was named after his great-great-great-grandfather "Peter Chavey" would never go by "Peter Chavey, Jr." or "Peter Chavey, II". The place where this came up recently was with "M. Chaney J. M. Chaney," a.k.a. "James McDonald Chaney, Jr.". His "canonical name", used in his publications is just the one with initials. But if someone were trying to do research on this person, it might be useful to have that extra information so as not to confuse this person with his father, who might have been alive when his books were published. So it seems to me that this type of suffix should be included. Chavey 16:09, 22 December 2010 (UTC)

If I recall correctly, one of the reasons why suffixes were not supposed to be entered in the Legal Name field was that a lot of them are not only informal (i.e. not recorded on birth certificates, passports, etc), but they also change over time. "John Smith, Jr." may drop the "Jr." part when his father dies and may eventually become "John Smith, Sr.". That's what Miss Manners recommends and it can get quite confusing when you have a "Sr.-Jr.-III-IV" chain going and then the "Sr." dies, causing the numbers to be reshuffled. This "rule" is not always observed, but it creates enough confusion that we thought it would be better to leave suffixes out of it. I guess we could clarify Help to state that suffixes should only be included when they are known to be a part of the legal name. Ahasuerus 16:38, 22 December 2010 (UTC)

I guess our classic example of promotion is Kurt Vonnegut, Jr. to Kurt Vonnegut. I've no idea who "Kurt Vonnegut, Sr." was. BLongley 23:41, 22 December 2010 (UTC)

Strange names in the database

Hi, I have downloaded and installed the database as suggested in the help pages (I even updated the help with some info), I selected UTF8 as the default character set.

but for some reason some of the authors names in my query are very strange, here are some examples (author_canonical):

"*s**c *s*m*v"

A. P. H*rb*rt

"Ōhara Mariko"

actually i think few thousands are like this.

What is the reason for this? Btw, i just noticed that when i post here it writes the name "Ōhara Mariko" correctly, but when i open the export file in excel i see

& # 332;hara

but without the spaces

Thanks, Qshadow 22:21, 21 December 2010 (UTC)

It looks like there are two separate issues here:

1. *s**c *s*m*v" and "A. P. H*rb*rt" are regular names that were used as facetious pseudonyms by John Sladek and John Brunner respectively -- see their bibliography pages
2. "Ōhara Mariko" is a name that uses Unicode characters. These characters, e.g. "& # 332", are interpreted correctly by the ISFDB software and by the Wiki software, but it looks like Excel doesn't perform this type of translation by default. There may be a way to get it to do it, though, since Microsoft Office is a huge beast and can do all kinds of things if you ask it nicely. Ahasuerus 00:38, 22 December 2010 (UTC)

Thanks, for the explanation, Ahasuerus, I will figure out the character set issue. but I have question about the pseudonyms,

How can I know if something is a pseudonym, and whose pseudonym is it? I tried to make a query using the pseudonyms table, but in ::this table i see only some numbers in the pseudonym field, what does this numbers mean? I expected to find there the actual pseudonyms...

Qshadow 11:09, 22 December 2010 (UTC)

The numbers in the PSEUDONYMS table are links into the AUTHORS table. Ignore the "pseudo_id" column, that's just a primary key. Data like this:

author_id	pseudonym
72		14732
72		18941
72		13426

Means that AUTHOR 72 A. Bertram Chandler used other names Bertram Chandler and George Whitely and George Whitley (among others). Both the author_id and pseudonym numbers are links to AUTHORS entries. BLongley 23:57, 22 December 2010 (UTC)

Clear, thanks a lot! Qshadow 19:16, 23 December 2010 (UTC)

I could be wrong here but the database is not stored in UTF-8. It is stored in the same character set the web server sends the pages marked as: ISO-8859-1 aka latin-1. Unicode is supported by means of allowing HTML entities (the ampersand-something-semicolon sequences). I am not sure what the ISFDB software might do with these but they are rendered in the web browser as per the HTML specifications. If you find HTML entities in the DB you can looks on the website for that record and then tell your browser to view source and you will see the same HTML entity sequence. I would be surprised if the web server and DB do much with them besides store and send them as needed (if there is anything that needs to count display characters like truncation however that can get tricky to do right). On the flip-side of things, the wiki pages are sent in UTF-8 and its DB entires likely are stored in UTF-8. This is why you can stuff Unicode characters in the a edit window here and it will save and display it right and you will not see HTML entities in the source (they are still support however and you write them manually and they will render if you like).

I was going to suggest the DB and CGI scripts perhaps be updated one day to natively support Unicode via UTF-8 but I doubt that is a high priority. Uzume 23:54, 23 January 2011 (UTC)

That's a pretty good summary. And yes, converting MySQL tables and all related scripts from Latin-1 to UTF-8 is a time consuming and painful task, which will need to be done at some point. Ahasuerus 00:25, 24 January 2011 (UTC)

SQL for award names (eg "Hugo" or "Nebula")

Hello- in examining the the 'awards' and 'title_awards' tables of the ISFDB MySQL DB, all the info about date, author, is accessible but 'award_name' appears to be addressing the title (name of the authors work), not the name of the actual award, which I can't find anywhere (eg "Hugo" or "Nebula"). Can anyone explain to me how I can build an Award Bibliography page similar to ISFDB's page below (Isaac Asimov):

http://www.isfdb.org/cgi-bin/eaw.cgi?5

Thanks! --Gideon 19:55, 22 December 2010 (UTC)

See Schema:awards for the "award_ttype" codes. BLongley 00:05, 23 December 2010 (UTC)

SQL to associate Interview/Review with approriate Author/Title

Another SQL question, which I hope I've followed protocol by posting it as a separate topic on this wiki page instead of including in my previous entry above (#121[2]):

I can see reviews and interviews listed in the titles table (they are just titles that have a ttype of 'review' or 'interview'), but no way of referencing these to actual titles, publications or their authors. What have I missed?

Thanks twice! --Gideon 20:08, 22 December 2010 (UTC)

Review relationships can be seen in the TITLE_RELATIONSHIPS table: e.g.

title_id	review_id
37795		671957

Means this title was reviewed here. BLongley 00:16, 23 December 2010 (UTC)

Michael Carroll

I believe Michael_Carroll is actually two people. The current ISFDB record contains both writing and art credits and links to the Michael Carroll (space artist) Wikipedia article. However, there is also a Michael Carroll (author) Wikipedia article which links back to the same ISFDB entry.

Unfortunately, I cannot access the official website given in the "(space artist)" article (it appears to be down). The official website given in the "(author)" article does have a bibliography that has all the novels and 2 of the 3 short stories. I'd guess that the nonfiction and essays belong to the "(space artist)", but that's just a guess.

Any help resolving this would be appreciated. --JLaTondre 12:59, 28 December 2010 (UTC)

Sorry, I (and everyone else, I suppose) missed this posting. I've done a little work separating the two, based on the research I've done. I used the method that IMDB uses in disambiguating common names by adding (II) to the name of the YA author, leaving the artist with the proper name (he has more records and it was easier this way). I did not change the pub records because of ISFDB policy of recording pubs as credited, but I did change the title records of the novels. This may cause some problems somewhere down the line, and the fix, I admit, is probably temporary. Does anyone else have another, better way of doing it? There's only one short story that's disputable and I've asked the verifier if there was any biographical information provided in the magazine publishing the story. (I also corrected the link from Wikipedia to the artist's website, and the link from the Wikipedia page of the author to the ISFDB.) Mhhutchins 16:42, 5 January 2011 (UTC)

Another bump in the road: I think this anthology may be edited by a THIRD Michael Carroll. It's published by an American Christian publisher (and the stories appear to be Christian SF). I don't think this would be either the Irish author or the American space artist. Mhhutchins 16:59, 5 January 2011 (UTC)

Hold your horses. That book was edited by the space artist. It's mentioned in his bio. Mhhutchins 17:11, 5 January 2011 (UTC)

And Ron has confirmed that the short story "The Terrible Lizards of Luna" was written by the space artist. Everything looks to be separated correctly between the two authors. Mhhutchins 17:14, 5 January 2011 (UTC)

Thanks! --JLaTondre 23:33, 6 January 2011 (UTC)

Setting up Pseudonyms

I've had to set up pseudonyms a couple of times, and had to fight with the process each time. (The "Help" pages on this don't seem to help me as much as I wish.) So I tried to write up the process, at least as I ended up doing it the last time (with moderator hand-holding), in my home page. If someone who's good at this could take a look at it, I would appreciate it. (E.g., is it really this complicated?) Thanks, Darrah Chavey 06:04, 5 January 2011 (UTC)

It's not as complicated as that usually - but yes, there are often titles under the wrong name. Usually people that have a middle initial sometimes, and sometimes not. E.g. Brian M. Stableford, Esther M. Friesner, Stephen R. Donaldson. In those cases, yes you unmerge the relevant pubs, fix the author name, then remerge them. In those examples the pseudonym association already exists so you don't need step 4. If it didn't exist - e.g. you find the first "Bob Heinlein" - then yes, you would have to "Make pseudonym". BLongley 14:16, 5 January 2011 (UTC)

Your step 3 is a little long-winded too - if you have the same title under both names already, then make variant with the title number (You may still have to adjust the dates to match as well afterwards). If you make variant by Author name, you'll end up with two titles under the Canonical name that will have to be merged. Only Make Variant by Author name if you don't have the Canonical record already. Hope this helps! BLongley 14:16, 5 January 2011 (UTC)

Awkward dating of a magazine issue

I'm wondering if this situation has come up before (ATT: Magazine Editors), and what would be the best way to handle it. I just updated the Spring 1993 issue of Aboriginal Science Fiction. Until recently, it was dated 1993-00-00, and that was fine except that there's a statement on the contents page that it was "published in December 1992." (Finally get a publisher who is honest and it throws a wrench in the works!) I know magazines are usually published (read "printed") a couple of months before the cover date, but 99.99% of the time we date the ISFDB record the same as the cover (or the masthead statement). It's strange to date this issue and its contents 1992-12-00, and the issue grid looks even stranger. Any advice as to how this should be handled? Has it happened in any other of the thousands of issues in the database? Mhhutchins 01:10, 7 January 2011 (UTC)

This happened to me with Pantechnicon #10, which wasn't published until 2010-01-30. I went with the publication date. I didn't think the grid impact was all that bad. My case was further complicated because the magazine was published back-to-back with the "Early 2010" TQF #32, for which the 2010 date is consistent. I couldn't see using two different dates for one physical publication. --MartyD 02:55, 7 January 2011 (UTC)

The cover date and masthead date are the off-sale date which is the one that should be used in order to be consistent with the many other thousands of magazines that have been done that way. Help is pretty clear about this, if anybody still thinks Help matters. Confusion over this issue resulted in many tens of thousands of magazine issues/stories being initially entered with 0 for the month which has resulted in a substantial amount of work to correct and is still a vexing problem. Contento lists the month for the Aboriginal issue date as March, 1993.--swfritter 15:33, 7 January 2011 (UTC)

Magazines that are not commonly distributed on newsstands quite often do not use the off-sale date but rather the actual date of publication. This includes many current magazines. Emags almost always have the actual date of publication. Press the big red Easy button. Use the cover/masthead date except in very very special cases. Some things can be done precisely; those that can't are not worth burning up brain cells.--swfritter 16:03, 7 January 2011 (UTC)

Unfortunately, the quarterly issues of this title do not have an off-sale date, only the cover date (the season), and the "published" date on the contents page. Should I move the date of these issues up a couple of months (to compensate for the off-sale date) or should I drop all monthly dating entirely and make the months "00"? I'd rather not do the latter, because the stories become only year dated (the problem you mentioned.) Nevertheless I can't see changing the date to four months later just because that's how Miller/Contento places it on their issue grid. In fact, they don't actually date the issue itself, only placing it under Spring, which covers the months of Mar-May. Contradictory to that, they placed the Spring 1994 and Spring 1996 issues under February, which is the stated date of publication in each issue. So for this title, they've mixed the two dating methods.

According to Locus #385 (February 1993), copies of the Spring 1993 were received in December 1992. And Contento bases 99% of his data on the books and magazines received by Locus. Either way we do it, it's a mess. But then, that's nothing new. The problem of magazine dating has been persistently with us, and the new grid display makes the problem more clear. Mhhutchins 00:10, 9 January 2011 (UTC)

In the same vein, let me know if you think I should do something different with the dating of that Pantechicon issue, despite its peculiar physical characteristics. I am happy to make it be however we want it set up. BTW, if the masthead/cover date should be used in preference to the stated publication date (when both are present), the help could be clearer on that point. The phrase ...because there is no good way to determine actual publication date... implied to me that if we have a good way to determine the actual publication date -- i.e., the magazine's stating it -- we should use it. --MartyD 11:33, 9 January 2011 (UTC)

Another Pseudonyms question

Hi, I am new to MySQL, so my question may be connected to wrong query, but here it goes:

I am trying to get a list of all Authors that have Pseudonyms, Ideally what i want to get should look like

Author_real_name, Pseudonym_name, bday

I do not know how to do it yet, but i tried at least to get all authors with pseudonyms, and their Real Author_id.

So I used:

SELECT DISTINCT
p.author_id, p.pseudonym, a.author_lastname, a.author_canonical,
a.author_birthdate, a.author_deathdate, a.author_views
FROM titles t, canonical_author ca, authors a, pseudonyms p
WHERE ca.title_id = t.title_id
AND a.author_id = ca.author_id
AND a.author_id = p.pseudonym
AND t.title_ttype in ('NOVEL','SHORTFICTION','SERIAL');

But i got a very strange results, for example how it can be that the same pseudonym (and actually the same person) can have different real author_id ?

Author_id   Pseudonym   lastname
1414	    1100	Sapir	Richard Ben Sapir	27-07-36	27-01-87	1081
1418	    1100	Sapir	Richard Ben Sapir	27-07-36	27-01-87	1081
3145	    1100	Sapir	Richard Ben Sapir	27-07-36	27-01-87	1081
3146	    1100	Sapir	Richard Ben Sapir	27-07-36	27-01-87	1081
3147	    1100	Sapir	Richard Ben Sapir	27-07-36	27-01-87	1081

What am I doing wrong here? Qshadow 22:02, 8 January 2011 (UTC)

You're not going wrong, just not understanding the results. That first result indicates that Author 1414 (Warren B. Murphy) wrote as Richard Ben Sapir, the second result indicates that Author 1418 (Will Murray) wrote as Richard Ben Sapir, the third says Author 3145 (Molly Cochran) wrote as Richard Ben Sapir, etc. You get this sort of problem with House Names and ghost-writers. BLongley 15:05, 9 January 2011 (UTC)

Thanks, for explanation, and for correcting my post layout ;) Is there any way of getting what I wanted initially:

Author_real_name, Pseudonym_name Qshadow 17:12, 9 January 2011 (UTC)

There's no guarantee that we KNOW the real name, of course. But you can see which names we have linked to other names: the difficult bit here is that the pseudonyms table has entries linking two different authors, so we need to link to the authors table twice, which needs the concept of "table aliases": for instance:

SELECT DISTINCT
a.author_canonical,
a2.author_canonical,
a.author_legalname
FROM titles t, canonical_author ca, authors a, pseudonyms p, authors a2
WHERE ca.title_id = t.title_id
AND a.author_id = ca.author_id
AND a.author_id = p.author_id
AND a2.author_id = p.pseudonym
AND t.title_ttype in ('NOVEL','SHORTFICTION','SERIAL');

That should give you the "Canonical name" we've arbitrarily assigned, the pseudonym, and the legal name if we know it. BLongley 19:16, 9 January 2011 (UTC)

I am learning MySQL on your examples. Thanks, Qshadow 17:12, 9 January 2011 (UTC)

Glad to know it's helping! I might have to use you as a reference in future, next time someone asks me about my mentoring skills.... BLongley 19:16, 9 January 2011 (UTC)

This "table alias" idea is exactly what I was looking for!

I would gladly give you my recommendations ;) Thanks! Qshadow 20:07, 9 January 2011 (UTC)

BLongley, one more thing, how can i get the full list of authors, where if an author has pseudonym it works like you showed, but if not it just prints blank or "NA" in the pseudonym cell. Qshadow 18:10, 10 January 2011 (UTC)

OK, the new concept here is "LEFT JOIN". It's similar to joins done in the "WHERE" clause but it allows you to take all rows from some tables even if there's no corresponding match in others. Try this (but be prepared to wait, it takes 3 minutes on my machine):

SELECT DISTINCT
a.author_canonical,
a2.author_canonical,
a.author_legalname
FROM titles t, canonical_author ca, authors a
LEFT JOIN (pseudonyms p, authors a2)
   ON (a.author_id = p.author_id
       AND a2.author_id = p.pseudonym)
WHERE ca.title_id = t.title_id
AND a.author_id = ca.author_id
AND t.title_ttype in ('NOVEL','SHORTFICTION','SERIAL');

The idea is that we want all Authors (from the first author table) if they've published some fiction (by restricting it to authors with relevant titles) - those are the "LEFT" tables. And even if they've NOT got a Pseudonym we'll have the data anyway - by making it a LEFT JOIN to the pseudonyms table. We won't need a second copy of the authors table unless there is a pseudonym, so the second authors table is considered to be on the right. BLongley 18:48, 10 January 2011 (UTC)

Thanks for the explanation i got it, and it works! Qshadow 19:43, 10 January 2011 (UTC)

Bdays found for authors with more than 3000 views

Hi, I thought that it is important to have Bdays for all Authors with high number of views (I chose 3000 as high enough). So I googled for all that do not have bdays in our database, here is the list:

Full Name	      Bday
Brian Aldiss	      18-08-25
Gabriel King	      26-07-45
Donna Jo Napoli      28-02-48
J. D. Robb	      10-10-50
L. J. Smith	      04-09-65
Alexander C. Irvine  22-03-69
Benjamin Rosenbaum   23-08-69
James Gunn	      00-00-1923
L. Timmel Duchamp    00-00-1950
Robin Hobb           00-00-1952
P. C. Cast           00-00-1960
Matthew Woodring Stover	00-00-1962
Elizabeth Haydon     00-00-1965
Tony Ballantyne      00-00-1972
Paolo Bacigalupi     00-00-1973

There are other authors that have views>3000, but i could not find their bday.

I am not adding it myself cause i want to make sure i that didn't make any error with this. I am still learning the rules. Qshadow 20:38, 10 January 2011 (UTC)

The data is not that difficult to enter. Search out the author, then click on "Edit Author Data" under "Editing Tools" on the left side. Enter the data into the form under "Birth Date" in format YYYY-MM-DD. Enter "Birth Place" if known without abbreviations except for some countries such as "USA, UK," e.g., City, State, Country. If you need to see examples open some of the more complete entries with the editing tool to see how they were filled in. Entering the data will give you more experience and your submission will be vetted by a moderator so don't worry about mistakes.--Rkihara 23:41, 10 January 2011 (UTC)

Keep in mind that we do have some of these authors' birthdays on file, it's just that they are entered under their canonical names. For example, James Gunn's canonical name is James E. Gunn, Brian Aldiss's is Brian W. Aldiss, etc. Ahasuerus 02:13, 11 January 2011 (UTC)

I don't think we want the birthdays listed for pseudonyms as well, or they'll appear twice on the "Authors born on this day" section. And for some it would be just plain wrong anyway: Gabriel King wasn't born 1945-07-26: M. John Harrison was, but the other user of the name (Jane Johnson), certainly wasn't. BLongley 15:29, 11 January 2011 (UTC)

Hi BLongley, you are right, but I added the data before I saw your answer, so just reject the entries that are wrong. There where many pseudonyms indeed, but about half or more where genuine. Qshadow 15:45, 11 January 2011 (UTC)

Wrong author name?

The name 'Sapper' looks wrong, maybe it should be deleted cause we already have Sapper without quotes and 'Sapper' is not marked as a pseudonym. Qshadow 16:06, 11 January 2011 (UTC)

According to at least two secondary sources (Vault of Evil and a Google Books search of Ashley & Contento's The Supernatural Index), this publication gave the pseudonym in quotes. If this is true then a variant must be made. Even though Sapper may be considered a pseudonym, it is not set up in the database as one. We don't make them pseudonyms if the author never publishes work in their own name (Saki, O. Henry) or if never for their fiction (William Tenn, Hal Clement). So 'Snapper' would be considered a pseudonym of Sapper for ISFDB purposes. Mhhutchins 16:38, 11 January 2011 (UTC)

I understand, but one of them should be marked second name or pseudonym of the other. I think that 'Sapper' should just point to Sapper. Like with other multiple names. Qshadow 16:54, 11 January 2011 (UTC)

Standardization of Author names

Hi, I am prolific today, sorry :) Qshadow 16:44, 11 January 2011 (UTC)

Prolific is good, don't worry about that! BLongley 17:06, 11 January 2011 (UTC)

I noticed lack of standard for names of authors mostly because two or more authors have identical names but in reality are two different persons, here are some examples:

Anita Allen [2]
Anita Allen[1]
Adam Smith
Adam Smith (1723)
Adrian Smith
Adrian Smith (author)
Ian Irvine
Ian Irvine (2)
Ian McDonald (1946-)
Ian McDonald

if you need more examples just search for [ or for ( in the names. I think we should keep the Author name clean, especially from bdays, and definitions like (author), for this we have other columns or queries. I am not sure about the numbers though, maybe it is something that we should decide. but even in this case it should be either [ or (

Qshadow 16:44, 11 January 2011 (UTC)

Well, the reasons for separating them vary and sometimes we don't have enough information to do it in one particular way: sometimes we know one's an author and one's an artist, or that one was born much later, or that they are from different countries. I suppose we could demand standards for suffixes - e.g. use "[]" or "()", leave a space before it - but the only really useful thing to ensure is that the "Last Name" starts with a letter rather than a bracket or a number - or they will not be viewable in the author directory. BLongley 17:06, 11 January 2011 (UTC)

Ok, I will try to explain why i think standard here is good:

1. the automatic parsers that you use expect the last word to be author_lastname, this caused problem with lastnames for these guys:

John Grant (?-)
Will Bradley (?-)
John Bailey (Australia)
John Williams (Australian)
Adrian Smith (author)
Alan Barclay (Canada)
John Carnell (II)
Chauncey Thomas (II)
Charles Maarowitz (translator)
Time Jones (UK)
Gr*g *g*n
Robin Hardy [3]
John Grant [I]
MIchael Williams (early 20th)

they got last names like: (?-); (Australia), etc...

2. When I tried to use automatic duplicate finder to find if we have dups, i noticed that without this standard it will be impossible. Also any other automatic action that you can come up with later (or someone else out there) will get this problem. If we had standard here, like you said space and number, the automatic parser could use it. Qshadow 19:03, 11 January 2011 (UTC)

I have to disagree there - the "automatic parser" works with what people (or bots) actually submit. And they almost never submit with suffixes - so it's down to the moderator to figure out what disambiguation needs to be applied, if any. It helps when we disambiguate ALL the people with the same name e.g. people will submit "Robin Hardy" and we'll have to figure out which it is as we'll get a "New Author" warning. Some times it's easier to let all submissions go to the default author: e.g. almost all our "John Grant" entries refer to the same person. Occasionally John (really a Paul) will come and tell us where we went wrong. BLongley 19:43, 11 January 2011 (UTC)

So the bottom line if we want to add something say bday to the Author names to differ between them, we should add this bday to ALL authors that have the same name, but are actually not same person. if we just add bday randomly it makes the database non standard and hard to process. In addition to being a reference, there is a lot of interesting statistics that this database could be used for in the future. Qshadow 19:03, 11 January 2011 (UTC)

Which is why I point out that if we DO have to add suffixes, we should get the Last Name right. We might know a birthday for one and a birth country for another - I don't think we can standardise that. But when 95% or more of the submissions go the right way, I think I'd prefer to let the current active author have his own name. BLongley 19:43, 11 January 2011 (UTC)

Ok, but at least we need to fix the errors with the last names of the authors above, can it be done by users or only by moderator? Qshadow 20:34, 11 January 2011 (UTC)

Users can fix Last Names via "Edit Author data". BLongley 13:33, 12 January 2011 (UTC)

First work per Author

Hi, I am trying to create a query, where each author will get the date of his first work, so I could know the time frame of the author. I googled and found MIN and GROUP BY operators, and created this query:

SELECT DISTINCT
a.author_lastname, a.author_canonical, t.title_title, MIN(p.pub_year)
FROM titles t, canonical_author ca, authors a, pubs p
WHERE ca.title_id = t.title_id
AND a.author_id = ca.author_id
AND p.pub_title = t.title_title
AND t.title_ttype in ('NOVEL','SHORTFICTION','SERIAL')
GROUP BY a.author_canonical

it doesn't work as expected. But from the examples that i saw on the web it looks like a correct query (more or less). Maybe it is because MIN can't work good with pub_year as it is a date in different formats? Or maybe it's wrong to use p.pub_year because is it not really the date the work was first written, so how can i find the real title year?

Thanks for your help every time, Qshadow 16:40, 12 January 2011 (UTC)

You're right that p.pub_year isn't a good choice - but mixing DISTINCT and GROUP BY is usually a bad move too. I think the nearest you'd get by adjusting the above would be the first date for each title for each author, which doesn't sound like what you want. I think what you want is the first fiction title and date for each author? If so what you need is to find the earliest title date for each author (no need to go for publication dates at all, if we have the title dates right), then you can look up the titles for that author with that date. The first bit is easy:

SELECT  a.author_id, MIN(date_format(t.title_copyright,'%Y%m%d')) "Earliest_Title_Date"
FROM titles t, canonical_author ca, authors a
WHERE ca.title_id = t.title_id
AND a.author_id = ca.author_id
AND t.title_ttype in ('NOVEL','SHORTFICTION','SERIAL')
AND date_format(t.title_copyright,'%Y%m%d') != "00000000"
GROUP BY a.author_id

Then you need to go look up all the titles for that author with that date: so we use the above as a sub-result and get the remaining details like so:

SELECT  a2.author_lastname
, a2.author_canonical
, t2.title_title
, date_format(t2.title_copyright,'%Y%m%d')
FROM titles t2, canonical_author ca2, authors a2,
  (SELECT  a.author_id author_id 
	, MIN(date_format(t.title_copyright,'%Y%m%d')) First_Title_Date
   FROM titles t, canonical_author ca, authors a
   WHERE ca.title_id = t.title_id
   AND a.author_id = ca.author_id
   AND t.title_ttype in ('NOVEL','SHORTFICTION','SERIAL')
   AND date_format(t.title_copyright,'%Y%m%d') != "00000000"
   GROUP BY a.author_id) FT
WHERE ca2.title_id = t2.title_id
AND a2.author_id = ca2.author_id
AND a2.author_id = FT.author_id
and date_format(t2.title_copyright,'%Y%m%d') = FT.First_Title_Date

Note that I've excluded all "unknown" dates (0000-00-00) as those would come first otherwise: and I've formatted the title_copyright (a misnamed column, we use it for published dates rather than copyright). You will still get some duplicates where we don't know which was the first of several titles with the same first date, and some duplicates where the first title was published under a pseudonym as well. But it's a start. BLongley 19:38, 12 January 2011 (UTC)

Thanks a lot, this was exactly what i needed! I didn't know that title_copyright is the title date, I was looking for it everywhere. Do you have maybe some help page that explains every field in the database? Qshadow 19:55, 12 January 2011 (UTC)

Database_Schema is the original documentation. It's a bit out of date, but we haven't dared mess with the underlying database much, so a lot of the info there is still good. BLongley 20:10, 12 January 2011 (UTC)

This is it, thanks again. Qshadow 09:06, 13 January 2011 (UTC)

Date for Elizabeth Bowen's "The Cheery Soul"

This has a date of 1942, but the Apr 52 F&SF lists its copyright as 1941. Is there a source for the 1942 date? Jonschaper 02:16, 19 January 2011 (UTC)

1942 is mentioned in A Reading of Elizabeth Bowen's "A Day in the Dark" by Lis Christensen, Irish University Review, Vol. 27, No. 2 (Autumn - Winter, 1997), pp. 299. The full article is available at JSTOR and costs $24 to read, but this information is on the first page of the article, which is freely available (lucky us!). I suspect that the story was copyrighted in November-December 1941 and appeared in early 1942. Ahasuerus 04:02, 19 January 2011 (UTC)

Some important authors that are missing from database

Hi, I have finished crude comparison of my personal database of authors, with the ISFDB database, I found that some authors that are more or less important for Sci-Fi (depends of course) are missing from the database. Here is the list of 79 authors with full Bday and Death date info, it is too big for me to add it manually, it is better to add them automatically, of course after you check that they are ok. There may be some authors that are included wrongly of course.

	Giordano Bruno	1548	1600
	Vasily Levshin	1746	1826
	Jan Barszczewski	1794	1851
	Vladimir Odoevsky	1804	1869
	Mary Putnam Jacobi	1842	1906
	Emilio Salgari	1862	1911
	Jerzy Zulawski	1874	1915
	Otto Witt	1875	1923
	Garet Garrett	1878	1954
	Claude Houghton	1889	1961
	John Hargrave	1894	1982
	Hermann Kasack	1896	1966
	John Desmond Bernal	1901	1971
	Premendra Mitra	1904	1988
	Thomas Gardner	1908	1963
	Sidney Sheldon	1917	2007
	Satyajit Ray	1921	1992
	Krzysztof Borun	1923	2000
	Margit Sandemo	1924	
	Denis Lindbohm	1927	2005
	Miha Remec	1928	Now
	Eno Raud	1928	1996
	Edgar Valter	1929	2006
	Alexis Gilliland	1931	Now
	Carlos Rasch	1932	Now
	Ioanna Chmielewska	1932	Now
	Sujatha Rangarajan	1935	2008
	James Glass	1937	?
	Jayant Narlikar	1938	Now
	Yuri Nikitin	1939	
	Yves Fremion	1940	Now
	Edmund Wnuk-Lipinski	1944	Now
	Fannie Flagg	1944	Now
	Luiz Eduardo De Oliveira	1944	
	Paulo Coelho	1947	Now
	Julia Nidecka	1947	
	Humayun Ahmed	1948	Now
	Michael Szameit	1950	Now
	William Dietrich	1951	Now
	Muhammed Zafar Iqbal	1952	Now
	Marek Oramus	1952	Now
	Eugeniusz Debski	1952	Now
	Wolfgang Hohlbein	1953	
	Marek Baraniecki	1954	Now
	Massimo Citi	1955	Now
	Marie Hermanson	1956	
	Pavel Shumil	1957	Now
	Soucy Gaetan	1958	
	Ninni Holmqvist	1958	
	Alexander Gromov	1959	
	Charles Cosgriff	1960	Now
	Andrzej Ziemianski	1960	Now
	Eric-Emmanuel Schmitt	1960	
	Giampietro Stocco	1961	Now
	Simon Rose	1961	
	Morioka Hiroyuki	1962	Now
	Siim Veskimees	1962	Now
	Artur Baniewicz	1963	Now
	Johan Theorin	1963	
	Rafal Ziemkiewicz	1964	Now
	Darko Macan	1966	Now
	Feliks Kres	1966	
	Kimmo Lehtonen	1967	Now
	Helge Kautz	1967	
	Laurent Botti	1968	Now
	Catherine O'Flynn	1970	
	Jacek Dukaj	1974	Now
	Randolph Lalonde	1974	Now
	Christoph Hardebusch	1974	
	Mathias Malzieu	1974	
	Zoran Krusvar	1977	Now
	Kristi Brooks	1980	Now
	Licia Troisi	1980	
	Adrian Jones	1986	Now
	Fedor Kaul	18?	19?
	Maria Ernestam	1959	
	Pasi Ilmari Jaaskelainen	1966	
	Matthew Stokoe	1965	
	Peter Zsoldos	1930	1997

Hope it helps, Regards Qshadow 18:07, 20 January 2011 (UTC)

The way the database is set up, we can't actually add authors by themselves. We have to add publications (books, etc.) by those authors, and then the database automatically generates the author records to go along with them. (And then we can add birth and death dates.) So in order to add these authors, we would need to have at least some of the books that they've published. Of course that's asking a lot of you, but if could give us data about their SF books, then we could enter them. Or, of course, you could enter the books yourself. You have lots of international authors in your list, and since I recognize a few of them, I suspect you're right that they're important authors. International authors are a place where the ISFDB is still weak, and your contributions would be quite helpful. Chavey 18:38, 20 January 2011 (UTC)

Based on 40 years of reading SF (not Sci-Fi), I could build a strong argument against the use of the word "important" but that would be a subjective assessment on both our parts. :) Alexis Gilliland is in the database with his professional name Alexis A. Gilliland. But for the reason Chavey states above, author records can not be created. The ball's in your court. (Fannie Flagg and Sidney Sheldon wrote SF. News to me.) Mhhutchins 18:55, 20 January 2011 (UTC)

I made my database by merging several sources (one of them Wikipedia page of Sci-Fi Authors (I call it Sci-Fi to differentiate from SF (as in Speculative Fiction see http://en.wikipedia.org/wiki/Speculative_fiction#Emergent_distinction_from_Science_Fiction_and_pre-existing_genre_schema ), anyway I was suspecting that some of the authors in the list got there by mistake (that's why I pasted it here for review first). BTW, Fannie Flagg did write semi fantasy novel e.g. "Can't Wait to Get to Heaven", though i am not sure if it buys her an entry to ISFDB ;) —The preceding unsigned comment was added by Qshadow (talk • contribs) .

I see that "Vladimir Odoevsky" is in the database as V. F. Odoevski, the name that was used when his SF story appeared in English. The rest have either never been translated into English or their work is obscure, e.g. Mary Putnam Jacobi. I am sure we'll get them in time, but it won't happen overnight. Ahasuerus 21:35, 20 January 2011 (UTC)

I added full author data about Vladimir Odoevsky to the author record for V. F. Odoevski, including the link to his Wikipedia page and an image from WikiMedia. A quick look at the image tag templates didn't seem to help me figure out which tag to use for an image whose copyright has expired, so I added the WikiMedia citation in the image summary. I also used the Wikipedia page as the authority for the spelling of his name in the "Legal Name" field, although of course that doesn't change the attribution name used for his SF work. Chavey 22:07, 20 January 2011 (UTC)

Are we, The ISFDB, responsible for SF now meaning Speculative Fiction rather than Science Fiction? If so, shame on us! Being an elitist, I don't read Sci-Fi or SF; only S-F. Besides that, the term SF has another ambiguous meaning - those are the initials of my name. Guess I need to update the Wikipedia article to clarify that.--swfritter 23:58, 20 January 2011 (UTC)

Sorry, Stephen. Most of us, as bibliographers, are forced to be the opposite of elite, which I understand would be exclusionary. Speculative fiction covers a much broader spectrum than science fiction. We have a great tent here, and are willing to allow the trunk of mainstream's elephant to sniff around whenever it takes a notion to. I personally wouldn't read 90% of what's in the database, so in that regard I could be called an elitist. But when it comes to what's entered into the database, I have to be objective. Anyone who has worked on Fixer's submissions can commiserate. Mhhutchins 00:19, 21 January 2011 (UTC)

I was only using "SF" as in the center of our initials, i.e. ISFDB. But I certainly understand how various people take such terminology very seriously. For example, after the SciFi channel changed their name to the SyFy channel, one of the big fans in our area insists on always referring to it as the "SyFylis" (pronounced "syphilis") channel. Chavey 01:42, 21 January 2011 (UTC)

Sense of Humor not allowed on the ISFDB? --swfritter 15:30, 21 January 2011 (UTC)

There is an outstanding feature request to improve Sense of Humor support, but it's not at the top of the list of priorities :-) Ahasuerus 18:14, 21 January 2011 (UTC)

Seriously, S-F (Science-Fiction) is a specific type of Speculative Fiction. The term SF can be confused with Speculative Fiction but the term S-F cannot. I don't remember any usage of the term Speculative-Fiction. To my mind the term Sci-Fi is (or has become) a pejorative term. That is probably part of the reason that the channel changed it's name.--swfritter 15:30, 21 January 2011 (UTC)

A number of people within fandom have regarded "Sci-fi" as pejorative for decades, but it doesn't seem to have affected the buying/viewing public. Based on the changes that the channel formerly known as the Sci-Fi Channel has been implementing lately, they are just trying to diversify into more profitable niches. Ahasuerus 18:51, 21 January 2011 (UTC)

I have no idea how anybody could have inferred that I meant quality assessments should be used to determine whether data should be entered.--swfritter 15:30, 21 January 2011 (UTC)

That's Internet-based communications for you! :) Ahasuerus 18:14, 21 January 2011 (UTC)

Maybe I should just pretend I am at work. I used to ponder over every email for hours before I sent it. And hours pondering over emails as I tried to decipher the "true meaning" of the text; but I got paid for that.--swfritter 18:27, 21 January 2011 (UTC)

Communications skills (and development skills and data entry skills and data organization skills and management skills etc) are very much transferable. Sometimes you get paid for using them and sometimes you get to pay for the privilege of using them :-) (Thankfully, running the ISFDB server is relatively inexpensive. Unless we bite the bullet and move to a dedicated system to address all the overnight downtime issues that we have been having lately, that is.) Ahasuerus 18:59, 21 January 2011 (UTC)

(Unindent) Ahasuerus mentioned that your "Vladimir Odoevsky" is in the database as V. F. Odoevski. In addition, your "Thomas Gardner" is our Thomas S. Gardner, your "James Glass" is our James C. Glass, and your "Pasi Ilmari Jaaskelainen" is our Pasi Jääskeläinen. This problem with some of our entries insisting on searchers using a middle name or initial to find them was part of the discussion I was asking about in the Community Portal, with respect to "Elizabeth A. Lynn", and does not seem to be something we have a good answer to yet. Chavey 02:29, 21 January 2011 (UTC)

Thanks Chavey, I really appreciate that you took the list seriously, I also agree that the search should be more robust, at least allowing search without middle names, and allowing search using English alphabet, since even when I searched for "Pasi Jaaskelainen", I got nothing, so the person who wants to find foreign authors needs to know and replace all English letters before search. This is not feasible for average internet user. Qshadow 09:23, 21 January 2011 (UTC)

A bigger issue is the degree to which we want to document world SpecFic. It requires a great deal of work on the part of moderators to process such submissions and the willingness of software developers to accommodate alphabetical issues; no foreign language experts have stepped forward to ease that moderator load. --swfritter 15:30, 21 January 2011 (UTC)

Well, we have a number of editors (and two moderators) who have been working on non-English European pubs, so our resources have improved significantly in this area. Software is a different story -- I am working on improving Unicode support (including a big fix that has been in the works for almost a month now and may go live this weekend), but it's been incremental and we really need to convert the database to native Unicode format before we can fully resolve the collation (including search) issues that Qshadow pointed out above. Ahasuerus 18:46, 21 January 2011 (UTC)

I think Swfritter may have missed out on the effort taken to find "foreign language experts". I'm torn between the "ISFDB is English mainly, with references back to original foreign language publications when necessary" and the more completist "All Speculative Fiction". I can only read English: it's nice to know when it was published in another language first, but a German title republished in French is of no use to me. When full language support is enabled I'm going to be even less useful. And then I can go back to reading a bit more, rather than moderating . BLongley 02:04, 22 January 2011 (UTC)

I don't think the average person using the ISFDB realizes how many thousands of hours some of the moderators have spent on the system. Despite that we still have woefully inadequate coverage of English language ebooks and current magazines.--swfritter 15:30, 21 January 2011 (UTC)

Yup, Fixer has identified tens of thousands of ISBNs that we are missing, and that's just for the titles that we already have on file! Ahasuerus 19:27, 21 January 2011 (UTC)

The initial mission statement was to enter every English language SpecFic pub and associated artwork. Now we are doing the same for every title ever published in the history of the world. No wonder we sometimes feel a little overwhelmed.--swfritter 14:57, 26 January 2011 (UTC)

The "tens of thousands" referred to above are just for English language non-audio ISBNs that we are missing. Here are the stats from Fixer's Worldcat session this morning:

• Total SF ISBNs for our current Title records found in Worldcat: 483449
• ISBNs already in ISFDB: 127,607 (including 21,125 foreign ISBNs)
• ISBNs not in ISFDB: 355,842 including:
  • English non-audio: 51,802
  • Non-English: 255,693
  • Audio/digital: 48,347

Unfortunately, Fixer is spending a great deal of his time trying to keep up with Amazon. Ideally, he should be mostly working on higher profile pubs that we don't have yet, but first I need a solid 100 hours to redesign Fixer to better handle multiple data sources. (Which I don't have at the moment.) Ahasuerus 15:31, 26 January 2011 (UTC)

I have substantially reduced my time because I would rather spend the time reading. The ISFDB is a collaborative effort. Like Wikipedia the quality and quantity of data is dependent on the willingness of people to do the work.--swfritter 15:30, 21 January 2011 (UTC)

Well, that and the skill set. I have taught myself enough Python and MySQL to be moderately productive on the software side, but I am no longer young (apparently few of us are) and learning new skills is more of a challenge than it used to be. And the technology keeps changing... Ahasuerus 19:27, 21 January 2011 (UTC)

Donations Page

The donations link in the navigation of the wiki does not work correctly.

Does isfdb accept paid advertising? —The preceding unsigned comment was added by Estar (talk • contribs) .

Nope, no advertising, paid or otherwise. The Donations page is something that is automatically created when the core Wiki software is installed, so I guess we should put some text there explaining that ISFDB doesn't do advertising. Ahasuerus 21:11, 21 January 2011 (UTC)

Well, we do sort of advertise - the home page with "Selected Upcoming Books" does look like a set of adverts. And while acting as "Speaker-to-LJ" I've come across people thinking that we could promote some titles over others. We don't and can't (as Editors or Moderators) and if there is a bias that only programmers can fix, I suggest we fix that as well. I'd be happy to accept donations for my bibliographic work here - I'm currently unemployed and not rich enough to retire yet - but I'm not going to take bribes to promote any work over another. BLongley 01:40, 22 January 2011 (UTC)

There are times when we have to limit the amount of information that we display on a single page. For example, it would be impractical to display all forthcoming books on the front page because there are hundreds of them. Any time we limit what information is displayed, we have to decide what criteria we use to choose what is dropped. At the moment, we are using the concept of "marque authors", i.e. the top 2% of our most frequently viewed authors. Here are the top 30:

+--------------------+
| author_canonical   |
+--------------------+
| Stephen King       |
| Ray Bradbury       |
| Isaac Asimov       |
| Arthur C. Clarke   |
| Poul Anderson      |
| Philip K. Dick     |
| Robert A. Heinlein |
| Ursula K. Le Guin  |
| Robert Silverberg  |
| Michael Moorcock   |
| R. L. Stine        |
| H. P. Lovecraft    |
| Richard Matheson   |
| Harlan Ellison     |
| Frederik Pohl      |
| Gene Wolfe         |
| Terry Pratchett    |
| Larry Niven        |
| Orson Scott Card   |
| Jack Vance         |
| Stephen Baxter     |
| Andre Norton       |
| Roger Zelazny      |
| Fritz Leiber       |
| H. G. Wells        |
| Brian W. Aldiss    |
| Vernor Vinge       |
| Neil Gaiman        |
| L. Ron Hubbard     |
| Anne McCaffrey     |
+--------------------+

and so on down to Ejler Jakobsson and Gregory Feeley. Like any other scheme, it's not perfect (pseudonyms are a pain), but it's reasonably bias-free. It's also dynamic: now that I have re-run the process that sets the "marque" flag, the front page shows books by "newly marqued" authors likeJohn Joseph Adams, Patricia Briggs, Joe Abercrombie, Jack Campbell, Tobias S. Buckell, and Graham Masterton, which were not there an hour ago. Ahasuerus 05:43, 23 January 2011 (UTC)

Perhaps a "Show random forthcoming books" might be a useful option. I don't find "marque authors" particularly useful myself, especially after they've died and and are not producing new works. BLongley 18:26, 23 January 2011 (UTC)

Except for "L. Ron Hubbard" of course. How does he keep doing it? ;-) BLongley 18:29, 23 January 2011 (UTC)

The problem with "random forthcoming books" is that you will likely get a lot of self-published stuff. As far as limiting the front page to truly "new" books, we can certainly change the logic to drop reprints, but then UK users won't be seeing "first UK editions" and US users won't be seeing "first US editions", which doesn't appear to be desirable. Ahasuerus 00:30, 24 January 2011 (UTC)

I see we have several SFBC editions on our main page at the moment. I think these should be excluded (along with any other "book club" publishers) lest we show bias towards cheap reprints over ones that make authors some more money. (I'll let Mike explain the needed exceptions for SFBC first editions.) BLongley 22:37, 10 February 2011 (UTC)

And this makes me think we should figure out how to deal with NONGENRE OMNIBUS. BLongley 23:38, 10 February 2011 (UTC)

Need some information

Hi: I'm Ravi from Madras, India. I once came across an article by I.A. debunking an earlier article which, he wrote had been accepted, believed, and even having readers raise queries.... much to his surprise. The original article, he wrote, was a spoof, about predicting using a water-cock. Unfortunately for me, the second and more important article was in a page torn from a magazine. It generally pointed out how credulous people could be. it gave some tantalising details about the original essay: like "I start a drop of water flowing..... and event X occurs". Could ANYBODY give me details about these tow. Regards Ravi

I assume that "I. A." is Isaac Asimov, right? If so, you may want to post an ID request on rec.arts.sf.written. The regulars are pretty good at that sort of things. (And welcome to ISFDB! :-) ) Ahasuerus 04:19, 22 January 2011 (UTC)

You are probably referring to from his Thiotimoline series (see also: Thiotimoline). I am not sure in which magazine the "debunk" article might have been printed in but methinks this material is also present in I. Asimov: A Memoir. If you absolutely need the magazine article I am guessing it is in Asimov's Science Fiction - 1992 somewhere. I hope this helps. Uzume 01:55, 24 January 2011 (UTC)

Strange problem with two Shadowrun novels

There is a very strange situation with two Shadowrun novels: Shadowplay (Shadowrun #9) and Nosferatu (Shadowrun #14) (verified second edition of Nosferatu). I have both of these novels, and the ISBN that I see for Shadowplay is credited as the ISBN for Nosferatu, here. This happens on Amazon, also. As I look at the books in my hands:

1. The Shadowplay ISBN is 0-451-45302-6. The catalog # is LE 5302, which matches. The copyright page says "First Printing, February, 1993" with the number line "10 9 8 7 6 5 4 3 2 1".

2. The Nosferatu ISBN is 0-451-45427-8. The catalog # is LE 5427, which matches. The copyright page says "First Printing, August, 1994" with the number line "10 9 8 7 6 5 4 3 2 1".

The information I see on the actual books directly contradicts the verified publications linked above, as well as the Amazon page. It also contradicts the information on the Shadowrun novels Wiki page.

Is anyone familiar with this situation and can anyone tell me what's going on here? Thanks :) AndonSage 05:00, 24 January 2011 (UTC)

From what you say, you have the pubs in hand. Go ahead and make submissions creating records for each of them. Just enter everything exactly as it appears in the book itself. Don't use the existing pub records as a guide. Then, once the submissions are accepted, do a Primary Verification for each so that anyone who questions the veracity of your pub data will know that it was physically verified. Thanks. Mhhutchins 06:02, 24 January 2011 (UTC)

I am not saying Amazon is right but the Amazon link you quoted confirms 0-451-45302-6 is Nosferatu as the existing ISFDB pub records also indicate. Amazon also confirms 0-451-45228-3 is Shadowplay as this ISFDB pub record indicates: Shadowplay. Finally Amazon states that 0-451-45427-8 is also Nosferatu but seemingly that is not denoted in the current ISFDB records. As Mhhutchins said feel free to add it or any other changes you know to be correct based on the publication you have available to you but pay close attention to detail. Thanks. Uzume 06:56, 24 January 2011 (UTC)

One other thing to look for is if the ISBN appears in multiple places on the book (spine, back cover, front cover, inside front cover, copyright page), check that they are all the same. Sometimes one location will have the wrong ISBN, while another location will have the correct one. --MartyD 11:29, 24 January 2011 (UTC)

What I'm wondering is if Roc accidentally printed the wrong ISBN on a book, or if an ISBN was re-used that shouldn't have been. My theory is that the original Nosferatu used ISBN 0-451-45302-6 and then that ISBN was used again for the reprint of Shadowplay. Part of the problem in figuring things out is that the 2003 Fanpro reprints use the exact same internal pages (including the copyright page) as the original printing, so there is no way to tell the edition from the copyright page. Also the front cover art is the same, as well as the back cover art/blurbs. I have two copies of two different Shadowrun novels: Headhunters (Shadowrun #27) (ISBN 0-451-45614-9 and ISBN 0-451-45630-0) and Beyond the Pale (Shadowrun #30) (ISBN 0-451-45674-2 and ISBN 0-451-45710-2). The only differences I can see between the different ISBNs is that 1) one edition has a Canadian price printed with the US price on the spine at the bottom and on the back cover in the lower left while the other edition does not, and 2) one edition has the bar code of the ISBN printed on the inside front cover while the other edition does not. As far as I can tell, from comparing the ISBNs to the List of Shadowrun novels wiki page, the editions with the Canadian price and the bar code are the original editions. But I don't know if I can extrapolate from only two novels for the entire series. That wiki page has incomplete information, too, which doesn't help the analysis. For example, only one ISBN is listed for Shadowplay, and it's not the 5302 one, although the FASA SKU shows 5302-1 (note that this is the only SKU with a "-1" on the end).

If we assume that the Canadian price and ISBN bar code are indicators of the original editions, then this would give credence to my theory, because my edition of Shadowplay (ISBN 0-451-45302-6) does not have a Canadian price, so it would be a reprint, and my edition of Nosferatu (ISBN 0-451-45427-8) also does not have a Canadian price, so it also would be a reprint, allowing for 0-451-45302-6 to be it's original ISBN. Additionally, I have checked other Shadowrun novels that I own against the current SKUs, and novels with no Canadian pricing are matching the current SKU (indicating reprint), while those with Canadian pricing are not matching (indicating original edition)

In any case, if we conclude that ISBN 0-451-45302-6 was used twice, once for the original Nosferatu and once for the reprint of Shadowplay, how should this situation be handled? AndonSage 12:01, 24 January 2011 (UTC)

It's happened before. It'll happen again. You'll find other pubs in the database that have the same ISBN, one record doesn't preclude the other one. Just enter what's stated in the book. Mhhutchins 15:33, 24 January 2011 (UTC)

For the Shadowrun novels 2003 reprints, which date should be used for the Year field? Should it be the date listed on the copyright page (which is the same as the original edition), with a note that states it's a 2003 reprint, or should it be 2003, which is the year the reprints were published, with a note regarding the original date? AndonSage 00:11, 25 January 2011 (UTC)

Is there a statement in the book itself of a publication date? Is there a number line? Just enter what's stated in the book. We can only reconcile the different printings when the record has been created. Don't worry about the record being in a final state before making the submission. It can always be updated with notes, etc. Mhhutchins 00:36, 25 January 2011 (UTC)

Both editions (original and reprint) use the exact same copyright page. So using Beyond the Pale (Shadowrun #30) as an example, both editions (ISBN 0-451-45674-2 and ISBN 0-451-45710-2) have 1) "First Printing, March, 1998" and 2) a number line of "10 9 8 7 6 5 4 3 2 1".

The reason I worry about getting the information entered correctly the first time is that I have 40 of the Shadowrun series novels, and I would rather not have to go back and fix things if I can get it right the first time :) AndonSage 00:56, 25 January 2011 (UTC)

Until the records are created, there's no way I can help you. Anybody else want to deal with this? Mhhutchins 01:16, 25 January 2011 (UTC)

I created a new record and updated another, so I'll see what happens and go from there. AndonSage 01:52, 25 January 2011 (UTC)

(unindent) I approved Never Trust an Elf and 2XS. One thing that occurred to me while reviewing the submissions is that perhaps it is better to treat these like reprints of an unknown date and use 0000-00-00 instead of 1992-mm-dd, assuming we really think the First Printing statement does not apply to the publication at hand. Although this case is further complicated by the presence of the number line, it's still similar to the situation we run into elsewhere with a "first printing" statement and no other edition identification, yet we know it's not the first printing (say, from the price). --MartyD 12:21, 25 January 2011 (UTC)

The reason I went with the date listed on the copyright page is that is the date, which, along with the ISBN, people would look for when searching for the book's entry here on ISFDB. Each reprint entry also includes a note regarding the assumed reprint. If you want me to change the way I'm entering the reprints, let me know, otherwise I'll continue the way I'm doing it now. AndonSage 22:51, 25 January 2011 (UTC)

Linking interviews to interviewees

I'm working on Paul Walker's Speaking of Science Fiction. One of the interviews is with James Schmitz (stated in the pub without the H.), so I followed the help screen new pub If the interview uses a non-canonical name which is already recorded in the ISFDB as a pseudonym or alternate name of the canonical name for this person, simply enter the version of the interviewee's name used in the interview.. The problem is, that the interview now shows only on the James Schmitz bibliography, and not on the James H. Schmitz bibliography. I can't make it a variant, because there is no interviewee field on this screen. Any suggestions? Thanks, --Willem H. 16:59, 25 January 2011 (UTC)

I believe interviewees should be entered as their canonical name, regardless of whether a pseudonymous version of their name exists in the database. And for the very reason you cite: it would only appear on the pseudonym's summary page. Even if we were able to make a variant of the interviewee field, it would still be a mess. Those instructions should be changed. An editor can always note the form of the name used in the interview in the record's note field. In most cases, the title gives it away anyway. You're going to find the same situation for the Horace Gold interview in the same collection. Mhhutchins 18:19, 25 January 2011 (UTC)

I agree with Michael totally on this. I've even written a script to find such and have spent some time moving such interviews. Thanks for finding the help page that needs fixing. BLongley 19:02, 25 January 2011 (UTC)

Another problem that this example shows is "Author Tags". It would be nice if we could move those to the canonical author too - but only the original tagger can do so cleanly and keep the "users of this tag" credits. BLongley 19:07, 25 January 2011 (UTC)

We have 3 related bugs for user-defined tags and votes -- they are not handled properly when titles are deleted, merged or turned into VTs. See Bugs 3165570, 2816243 and 1743289. Ahasuerus 22:43, 25 January 2011 (UTC)

And I thought it was something I couldn't find. Thank you very much! I'll change it again, and add notes. --Willem H. 19:30, 25 January 2011 (UTC)

Title searches

I'd like to be able to browse, or search, the names of title series. A standard search allows you to specify a search through Title Series, but does not allow a search that gives more than 100 results. An attempt to do so tells you to go to the "Advanced Search". But the Advanced Search does not appear to have the ability to search the names of a Title Series. Oddly enough, you can search the names of Publication Series (a much newer feature), but not in the names of Title Series. Is there a way to do this that I've overlooked? Chavey 21:23, 29 January 2011 (UTC)

This is not what you want, but you can use "%" and "_" wildcard characters in the search box to help control your list. "%" means match zero-or-more characters, and "_" means match exactly one character. Whatever search string you type implicitly gets "%" at the beginning and end. So something like "BA_% %" will get you things starting with "BA", plus some other stuff (there's no way I'm aware of to constrain it to starting with BA). --MartyD 00:12, 30 January 2011 (UTC)

Experimentation shows that the system doesn't use the Unix-style "$" and "^" to anchor to the beginning or end of a string. Reading up on SQL wildcards tells me that I should be able to use "[abc]" to match an "a", "b", or "c", but this doesn't seem to work in regular or advanced searches. Are there any other search tricks? Or any way to search (or browse) title series? Chavey 17:29, 30 January 2011 (UTC)

"$" and "^" come from regular expressions. They are available in many programming languages these days, but are quite opaque to the average user, which is why they haven't been implemented.

Internally, Advanced Search performs numerous and rather complex manipulations with the entered data before it sends the query to the database. We have to be careful not to cause system hangs like the one that we had a few weeks ago. That said, we should be able to add "Series" to the drop-down list in Advanced Title Search without too much trouble. FR 3168314 has been created. Ahasuerus 20:15, 30 January 2011 (UTC)

Thanks much! Chavey 04:36, 31 January 2011 (UTC)

Science fiction books published per year

I'm trying to get an estimate of how many science fiction books were published per year by country, excluding variants. How would I go about writing a SQL script for this? Much thanks Davimre 15:25, 1 February 2011 (UTC)

Variants probably won't affect the totals much if you're just after an estimate. "By Country" is difficult as we don't record that - but you could approximate it by looking at the currency symbol: e.g. for US books:

SELECT date_format(p.pub_year,'%Y'), count(distinct p.pub_title)
FROM pubs p
where p.pub_price LIKE '$%'
GROUP BY  date_format(p.pub_year,'%Y%')

Use LIKE '£%' OR LIKE 'L%' for UK books, etc. BLongley 17:05, 1 February 2011 (UTC)

Won't magazines also be included with that SQL?--swfritter 17:53, 1 February 2011 (UTC)

Good point. Probably needs

and p.pub_ctype IN ('ANTHOLOGY','CHAPTERBOOK','COLLECTION','NONFICTION','NOVEL','OMNIBUS')

as required - I'm not sure if Davimre wants Chapterbooks and Non-Fiction Books about SF as well. BLongley 18:05, 1 February 2011 (UTC)

Actually, looking at the results, I'm not sure I'd trust them. For US books, there still seems to be blips for 2004/5 when Dissembler was going a bit berserk, and 2010 when Fixer got involved. UK books seem to have escaped the Dissembler problems. BLongley 18:16, 1 February 2011 (UTC)

I'm interested simply in genre SF, so chapterbooks, anthologies and the like, but this can be sorted with the ctype. To get a little more specific, I'm primarily interested in the period from 1920-1980, which hopefully avoids the later issue described above.Davimre 19:21, 1 February 2011 (UTC)

OK, cutting down to that period is easy: add

and date_format(p.pub_year,'%Y%') >= '1920' 
and date_format(p.pub_year,'%Y%') < '1981'

But for the earlier period you'll need to look at "Shillings and Pence" format for UK books - e.g. "p.pub_price LIKE '%/%'". Hope this helps. BLongley 20:30, 1 February 2011 (UTC)

(unindent) For newer publications, you might be able to take advantage of the ISBN to help. Many ISBN language groups are in effect single-country. You could then count the distinct ISBNs, grouping by the pub year and ISBN language. See [3] for the group list. For the many-country groups, such as English and French, you might then use the currency to try to get a breakdown by country. --MartyD 01:48, 2 February 2011 (UTC)

A few other things to keep in mind. First, compared to Amazon.com, there appear to be fewer self/vanity-published books listed by Amazon UK. Fixer filters out the best known vanity publishers, but it's still a significant percentage of our haul of US-published book. If you are not interested in self/vanity-published books, then the US numbers may be misleading.

Second, Fixer has no easy way of telling Canadian books from US/UK ones at this time. This means that they sometimes get entered using US dollars or UK pounds. And, of course, our sample of books published in other languages is too small to be representative. Ahasuerus 06:37, 2 February 2011 (UTC)

At least a third of more than 20,000 pubs from before 1960 do not have price data.--swfritter 14:29, 2 February 2011 (UTC)

Magazine Contents List

Hi, I am trying to do something simple: I want to get a list of contents for say, Astounding Magazine, for all it's issues, like this

IssueDate   MagazineName   StoryTitle   Author

but i am lost as to how do I connect Magazine to Title? Is this possible? Thanks, Qshadow 20:24, 2 February 2011 (UTC).

See the examples in ISFDB:Help_desk/archives/archive_13#MySQL_query_question for some pointers. Your query would be a little different, but you can at least see how to work with the relationships. --MartyD 02:04, 3 February 2011 (UTC)

Thanks MartyD, the p.pub_ctype = 'MAGAZINE' was what i was looking for, but I now do not understand what for Magazine table is used? in your example it is not needed. Qshadow 13:17, 3 February 2011 (UTC)

The Magazine table is obsolete - it just has the names of the first 92 Magazine series that were entered. We use the Magazine (and fanzine) pages on the Wiki now. BLongley 13:53, 3 February 2011 (UTC)

Links in lists?

Is it possible to have a link in a list? I used the following in the notes for Dog Wizard and it didn't convert it to a link, it shows the whole line:
[http://www.locusmag.com/index/b213.htm#A2900.13 Locus]
Is there another way to get the link? Thanks :) AndonSage 13:29, 3 February 2011 (UTC)

In notes, you have to enter Links HTML style rather than Wiki-Style: so you'd need to enter it as

<a href="http://www.locusmag.com/index/b213.htm#A2900.13">Locus</a>

However, Locus pages aren't stable and I wouldn't recommend trying to link to a specific page there. BLongley 13:47, 3 February 2011 (UTC)

If you're using the online Locus Index to Science Fiction, just give Locus1 as your source. Mhhutchins 19:27, 3 February 2011 (UTC)

Thanks for the information, guys. AndonSage 07:32, 4 February 2011 (UTC)

Galaxy #9 (1995)

The revived Galaxy (ed. E. J. Gold) printed 8 issues before it folded in early 1995. However, it turns out that issue #9 was prepared and appeared on Galaxy's Web site, which is no longer available. I used Jacqueline Lichtenberg's Web site to mock up a pub record for Galaxy #9, but it's rather bare bones at the moment. Do we have better sources to flesh it out? Ahasuerus 04:06, 5 February 2011 (UTC)

Do you know what the URL for their website was back then? We might be able to find a copy on the Wayback machine. I checked "galaxysciencefiction.com", but their oldest record of that site is from Sept. 2000. It seems likely that we would need to find an older version of the website, and it's hard to imagine that the Wayback Machine missed them for that long, so I wonder if they had a different URL before that one? Chavey 01:20, 6 February 2011 (UTC)

This image from Galactic Central is somewhat legible, and discusses the transition to electronic format, but alas does not provide a URL. Albinoflea 21:39, 26 March 2011 (UTC)

Help with a short story

The short story Liberty Port in The Warrior should be associated with Liberty Port, but instead it's got it's own record. I don't know what I did wrong. Help please? AndonSage 13:25, 5 February 2011 (UTC)

I'm assuming you added Liberty Port to The Warrior? If so, you didn't do anything wrong, it's how the database works. When you add a title to a publication, it creates a new title record that you then have to merge to the existing one. If you go to David Drake's page, you can click on "Check for Duplicate Titles" link under the "Editing Tools:" heading on the left. That will bring up any candidate matches of which the two Liberty Port stories are listed at the bottom. Check the title links to verify they are the right ones, then select the two merge check boxes & hit the merge button. A page will be displayed that shows any conflicts; pick the data records to keep and hit the complete merge button to submit your change.

It would be a nice feature addition to be able to add a title record to a publication by its title number. But as far as I know, creating a duplicate record and merging it is the only way at the moment.

If this wasn't clear, ask questions. --JLaTondre 14:11, 5 February 2011 (UTC)

Thank you very much! Your directions were great :) I'm keeping a copy in case I need them in the future :) AndonSage 00:22, 6 February 2011 (UTC)

Help with SFBC reprints

I'm having a problem trying to find information on what I think are SFBC reprints; specifically, Anne McCaffrey's Tower and the Hive series. The SFBC number codes located on the back dusk jacket of the first four books in the series are not matching the ones already entered here, so I figure they are reprints. However, I can't find any information on these books, anywhere :( I have gone through the SFBC Book Listings years and all of the Locus years.

To give an example, the SFBC # listed for The Rowan is 17565. However, on my copy the # is 19848 (located at the bottom left corner on the back). The ISBN listed on my copy (located at the bottom right corner on the back) is that of the original hardcover edition.

The other codes on my books are 19850 for Damia, 19851 for Damia's Children and 19853 for Lyon's Pride. The fifth book in the series, The Tower and the Hive, is the only one that is the same as I have, with code 19652.

Can anyone tell me how I can find out the publishing data for these versions? Thanks. AndonSage 12:53, 7 February 2011 (UTC)

P.S. I see that Wings of Pegasus has original and reprint editions, so the information must be somewhere?

Any reprints which not listed in the Locus database are going to be hard to date. The SFBC Listings on this wiki wasn't created to list every SFBC printing. It's only when a title becomes a selection does it appear on the list (and afterwards only if it's a new edition). The abandonment of the Gutter Code system in 1987 was a great blow to bibliographers who attempt to date BCEs. Based on the SFBC ID numbers you've given, these could have been printed anytime between the late 90s and mid-2004, when a new 7-digit numbering system began. I would suggest that you create new records for the pubs, entering the dates as "0000-00-00". Place the trade publisher's ISBN in the note field and the SFBC ID# in the ISBN/Catalog # field. Sorry I couldn't have been of more help. Mhhutchins 19:43, 7 February 2011 (UTC)

It just occurred to me. Because the last book (The Tower and the Hive) was a selection in June 1999, there's a strong possibility the other books in the series were reprinted at the same time. The book club would do this when a later book in a series became a selection, going back and reprinting the earlier titles in the series. The closeness of the ID numbers is additional evidence that these were published in 1999. When you create the records, record in the notes that the publication date may have been in 1999, giving this as the reason. Seems like it's the best we can do for now. I was out of the club for a couple of decades and don't have access to their catalogs. Maybe someday we'll get an editor who kept all of the SFBC catalogs. Mhhutchins 19:50, 7 February 2011 (UTC)

Thanks for the suggestions. I've entered a record for the first book in the series, The Rowan, and I'll see if the approver has any notes for me, before I enter the remaining books. AndonSage 06:41, 8 February 2011 (UTC)

I accepted the submission for the new printing, but suggest that you mention that the image is for the first SFBC printing (the ID number is clearly visible on the image.) That way a future editor won't question either your ID number, or that you mention the trade ISBN is on the back of your printing's dustjacket, and it's not on the cover image. I would also suggest removing the price, because it's likely that in 9 years the price would have changed. In 1999 the average price for a SFBC printing was more than $10. Thanks. Mhhutchins 07:51, 8 February 2011 (UTC)

Or, now that the record is created, I can upload a new cover :) I removed the price, as suggested. Thanks for your help :) AndonSage 09:45, 8 February 2011 (UTC)

BTW, I forgot to zero out the date on Damia, but I can do that after the record is created and I upload a cover. AndonSage 09:57, 8 February 2011 (UTC)

Hopefully it doesn't look too unacceptable to either of you - I'm always wary of accepting edits for things other Moderators are more expert than I am in. But we sometimes get to see an empty queue now! :-) BLongley 23:45, 10 February 2011 (UTC)

Yep, they all look pretty good. Thanks to all involved. Mhhutchins

Help with book that doesn't seem to exist!

I recently purchased a used copy if Ithanalin's Restoration by Lawrence Watt-Evans. As far as I can tell, it's an SFBC edition, because it doesn't have a price on the dust jacket, and it has 56290 in a white box at the lower right corner of the back cover. It's also the usual SFBC size. However, I can't find information for this edition anywhere! The Ithanalin's Restoration title entry doesn't have an SFBC edition, I couldn't find it in the SFBC Book Listings, I couldn't find it at the Locus Index to Science Fiction, and it's not even mentioned as a published edition at the author's website page. I wrote to the author today, to see if he can remember when or if it was offered as an SFBC edition, but other than that, can anyone tell me where else I might look to get the publication date and pricing information for this book, please? Thanks :) AndonSage 17:30, 19 February 2011 (UTC)

Watt-Evans lists it on More About the Novels as one of the 5 SFBC editions of his books. On his list of Ethshar Books For Sale, he lists it as "Science Fiction Book Club hardcover, 2002", but doesn't say when (or if) it was a Book-of-the-Month club edition. I suspect it was an "alternate selection", hence didn't make as many of the SFBC lists, but it certainly is odd that it doesn't seem to be on any of those lists. Chavey 22:11, 19 February 2011 (UTC)

Based on the info that Chavey has uncovered, I'd recommend creating a new record for it using the usual methods for SFBC entry. Because those years of the SFBC list are based solely on Locus1 (e.g. see the warning at the top of the the 2002 list), there would not be a listing of it there. Once you've created the record, I'll add it to the list. During that time, the SFBC release averaged about a month after the trade release, so I'd recommend dating the record as 2003-00-00, but note that the date is an estimate. BTW, the parent companies of the BOMC and the SFBC at that time had recently been merged, yet they retained separate identities. It was around this time that members were able to "cross-order" selections from one club to the other. I'm 99% sure this Watt-Evans title was a SFBC selection and not a BOMC selection. Mhhutchins 16:15, 20 February 2011 (UTC)

LWE responded to my inquiry. "The SFBC edition came out at roughly the same time as the trade edition, give or take a month or two; I'm afraid I don't know exactly. I think it might have been January 2003, but I'm not at all sure. It was a featured alternate, as I recall. And I don't remember the price at all. If I had to guess, I'd say $14.95, but it's just a guess." I'll create the record with a date of 2003-01-00 and a price of $14.95. If anyone happens to know if that price doesn't sound right, we can change it. Thanks for the help :) AndonSage 20:27, 20 February 2011 (UTC)

I've accepted the submission, but suggest that you give the email from the author as the source for the unstated data. Thanks for finding this edition. Locus is a great source, but not infallible. Thanks again. Mhhutchins 20:52, 20 February 2011 (UTC)

Ok, I've changed the Notes. I'll verify the record if you're ok with the changes. AndonSage 21:00, 20 February 2011 (UTC)

Poppy Z. Brite's "Bayou de la Mere"

Hi, the above title appears both accented and unaccented for different printings of the same book. Jonschaper 22:39, 20 February 2011 (UTC)

I'm in the camp that believes accented letters (or any diacritical marks) should not be used as a basis for making a variant title. There are others that disagree. BTW, the two records are not for publications of the same book. One is for it's publication in the author's collection (not accented) and the other is for the original anthology in which the story first appeared. If it were up to me the two records would be merged, retaining the accent. Some other editors would push for a variant creation. Mhhutchins 22:58, 20 February 2011 (UTC)

After checking our current search functionality (which finds either) - I'd merge and keep the accent. Of course, I'm a monoglot that doesn't really understand the importance of accents in foreign languages. BLongley 23:24, 20 February 2011 (UTC)

[This publication record] not being kept in Notes

When I enter Notes, I use [This publication record] by either the Stated printing, or the number line, to indicate it's applicability to the current record. However, this information is not in the Notes after they have been accepted. I just checked all of my recent edits that were accepted by Bluesman. What's going on? AndonSage 00:47, 21 February 2011 (UTC)

I just checked a record change accepted by Kraang, and [This publication record] showed up. So it's only Bluesman where the problem is happening. Can someone let him know, please? Thanks! AndonSage 03:12, 21 February 2011 (UTC)

He'll see this, or if you want to make certain, place a note on his talk page. I personally don't fully understand what's meant by "This publication record". Isn't the fact that you're recording the notes in "this publication record" understood? I think you may have seen this term used by user Dragoondelight and picked it up from his notes. What exactly is your interpretation of the term? Mhhutchins 03:21, 21 February 2011 (UTC)

I'm not sure where I picked it up, but you're correct in that I got it from someone else. It's my understanding that it's used to indicate which edition a book might be. For example, a book with First Printing, January 2000, but Third printing by number line, would have [This publication record] next to the Third printing by number line to indicate the record is for the third printing, not the first printing. In any case, if I picked up a habit that's not needed, then it doesn't matter :) Although, I still wonder why it doesn't appear in my Notes only when Bluesman accepts the changes. Maybe he thinks it isn't needed and is editing the words out? AndonSage 04:18, 21 February 2011 (UTC)

If a book states "First Printing: January 2000" followed by an incomplete number line, I make the following note:

"First Printing: January 2000" stated on the copyright page with an incomplete number line: "0 9 8 7 6 5 4 3" indicating a third printing.

Everything that is actually printed (stated) in the book is in quotes. Everything else is my words. I've never seen a book that states Third printing by number line. Using quotes to record exactly what appears in the book can clear up any ambiguities. Stating [This Publication Record] thus becomes redundant. You wouldn't be recording data from another book unless you're giving the other book as a source for information in this record. Still, when you've used the statement, I've left it the way you entered it. Mhhutchins 22:46, 21 February 2011 (UTC)

You can look at Recent Edits, search for the title, and see if it was edited after your submission (order is most recent to oldest, so anything "above" your edit in the list is someone else's after yours). If so, ask that editor. The software does not strip out text in square brackets or that particular phrase. --MartyD 21:36, 21 February 2011 (UTC)

Question about entering a new novel

My copy of John F. Carr's Nov 2010 expanded version of Time Crime by H. Beam Piper arrived today, and I am wondering if the correct way to enter the record for a new edition of an existing story would be to use the "Add Publication to This Title" option? Thanks :) AndonSage 19:00, 24 February 2011 (UTC)

I suspect you mean Title 21832 rather than Publication 21832? Personally, I'd add it as a new title with the extra author, explaining that it was an expansion of the other title. BLongley 19:27, 24 February 2011 (UTC)

Whoops! Yes, I meant the Title, sorry. Funny how using a P instead of a T changes things :) Anyway, I don't see an "Add New Title" option, so would I use the "Add New Novel" option? And that will create a new title? AndonSage 19:38, 24 February 2011 (UTC)

Yes, use the "Add New Novel" function. Mhhutchins 20:54, 24 February 2011 (UTC)

The new record has been entered. Let me know if I did it correctly, or need to fix something. Thanks :) AndonSage 21:14, 25 February 2011 (UTC)

A question about the page numbering, is there a page numbered "1" or does the first regular numbered page ("24") follow the roman-numberaled "xxii"? I'm not sure what the note means by "ONE". Is that the chapter number? Thanks. Also, I'm going to make this a variant of the Piper novel of the same name, based on the new author credit, not because on the variant text (which the database currently doesn't support). Mhhutchins 21:32, 25 February 2011 (UTC)

I'm sorry, I missed your questions after my last post, and just now saw them. Anyway, to answer: 1) There is no page numbered "1". Page "24" follows the roman-numeraled "xxi" after two unnumbered pages. The roman-numeraled pages are a prologue, which begins on unnumbered page "v". The first numbered page is "vi". Unnumbered page "xxiii" (or perhaps "23") is a heading page, with the words "TIME CRIME" (over) "PART ONE" (over) "BY H. BEAM PIPER". 2) "ONE" is the chapter heading. Instead of, for example, "Chapter One," it's just "ONE." I hope this answers your questions. I'll check back to see if you have more. Also, I was wondering why this book doesn't show up on John F. Carr's page? Thanks :) AndonSage 06:53, 4 March 2011 (UTC)

It doesn't show up on Carr's page, because its title record is a variant of a title that is credited solely to Piper. The only way for it to show up on Carr's page is to break the variant relationship and have its record be considered a completely new novel. The database can't (so it doesn't) handle variants in text, only in variants of title or author credit (as this pub currently stands). If you think it is a new work that is sufficiently different from the Piper only work, you should break the variant relationship. Ask if you don't know how to do that. Mhhutchins 05:33, 5 March 2011 (UTC)

Thanks for answering. I just wondered why the book didn't show on Carr's page, not saying I didn't agree with the way it was handled. You guys are the experts :) AndonSage 06:39, 5 March 2011 (UTC)

Removing votes when they have been entered in error

Hi, I wondered whether it is possible to remove a vote on a book? You can of course change the mark - but only to a number between 1 and 10, not to a "null vote". If this is not possible, could this be looked into? —The preceding unsigned comment was added by DS (talk • contribs) .

Not at the moment, I am afraid. There is a feature request to "allow removing votes", though. Ahasuerus 02:39, 26 February 2011 (UTC)

Pseudonyms of Pseudonyms?

seems to currently be about 40 records in the pseudonyms table indicating that an author record is a psedonym of another author record that is already a pseudonym of some other record...

select distinct a.author_id, b.author_id 
from   pseudonyms a, pseudonyms b 
where b.author_id = a.pseudonym;

I'm guessing these are glitches and the psuedonym relations should be flattened? —The preceding unsigned comment was added by Hossman (talk • contribs) .

Pseudonyms should be only one layer deep. Pseudonym B of Pseudonym A of Author X should be changed to be a direct pseudonym of Author X. Likewise, any variant titles should be only one layer deep (Title T by Pseudonym B and Title T by Pseudonym A should each be variants of Title T by Author X; Title T by Pseudonym B should not be a variant of Title T by Pseudonym A). Relatively recent changes in the editing software have added automatic reorganization of some of these things at the time the relationships are established/changed, but some things can still be done, and some improper things still linger. --MartyD 12:00, 27 February 2011 (UTC)

At least some of these cases are in the database because "author A", a real person, published a story as "author B", another real person. For example, Eric Frank Russell's "The Mechanical Mice" has appeared as by Maurice G. Hugi as well as by Maurice A. Hugi. Maurice G. Hugi was a real person who sometimes wrote as Brad Kent, a pseudonym that he shared with Dennis Hughes, hence the pseudonym chain. Ahasuerus 03:37, 28 February 2011 (UTC)

Inclusion question about a specific pub

The background: Sarah Monette recently held a charity fundraiser. Among the items for sale in that fundraiser was a slim collection of recently published short stories in her Kyle Murchison Booth series. (If it's relevant, three of the four stories in the collection already have entries in the ISFDB, and the fourth will once the contents of a particular magazine are added.) Monette had the collection printed herself using Lulu.com, and it was strictly limited to the number of copies (169) that were ordered during the fundraiser. One of those copies was ordered by me, I've received it, and I'm now wondering whether it's appropriate to create an ISFDB listing.

On the one hand, it certainly seems to fit the rules of acquisition, since it's a printed book published by a vanity press. While 169 copies is low, it's roughly similar to certain other speculative fiction limited editions. On the other hand, the collection's window of availability was very limited, and some of the basic publication information about is it unavailable (it doesn't seem to have an ISBN) or difficult to source (the price was $20.00 US, but the only way to prove that is by linking to a LiveJournal post). I've checked the archives of relevant pages for discussions of similar pubs, but if there are any I failed to find them.

I personally would lean toward including the collection, but I'm a neophyte when it comes to the relevant issues, so I'm looking for input from those with more experience. BrendanMoody 04:22, 1 March 2011 (UTC)

I think it's definitely "in". The number of copies printed never stopped genre bibliographers from listing books, e.g. "26" is a very popular number for certain limited editions -- for an extreme case see this edition. The fact that information about this book is hard to find just means that our data will be so much more valuable :-) Ahasuerus 05:09, 1 March 2011 (UTC)

I would agree with Ahasuerus. Consider The Shunned House, by H. P. Lovecraft. We list four editions, and of those the print runs were: (1) 50; (2) 100; (3) I don't know; (4) 67. We list them all. We have several "ReaderCon" Souvenir books listed, because they had stories in them, but of course they were only publicly available during the convention. The first edition of J. K. Rowling's The Tales of Beedle the Bard had a "print run" of 7 copies, and was only offered for sale once, at an auction. It's listed. The stamp collecting world lists stamp errors that were available for a very short time. The comic book collecting world lists "Thrill Comics #1" (a.k.a. "Whiz #1"), even though it was never available, but printed only for copyright purposes, and only 3 copies are known to exist. I would suggest that ISFDB should try to be as complete as the standard references in these other fields. Chavey 08:58, 2 March 2011 (UTC)

Question about an SFBC book

The current record for the SFBC book Annals of the Time Patrol has notes that say the Gutter Code is "O09", indicating a first printing. The copy I have has a gutter code of "O47". Locus1 only shows a May 1984 printing, which is what the record has. According to Gutter Codes, "O47" should be November 1984, but then again "O09" should be February 1984, not May 1984. So how can I determine a publication date for this book? Thanks :) AndonSage 07:05, 4 March 2011 (UTC)

See Help:How_to_enter_a_SFBC_publication, especially the Conclusion. The original date is based on when the title was an SFBC selection, not when it was printed (the gutter code indicates the printing date, not the "publication" date). For later editions, you don't need to determine the date, and you would just record the additional gutter code and location in the notes for the existing entry. You can also update the Annals of the Time Patrol entry in Publisher:SFBC_1980-1984#1984 to record the additional known reprint code in the Notes column (see other entries in the list for examples). --MartyD 12:16, 4 March 2011 (UTC)

I edited the current publication record, so it's just awaiting approval. Can I primary verify the record even if I only have the reprint "O47" edition? I also added the gutter code to the notes for the Annals of the Time Patrol entry in Publisher:SFBC_1980-1984#1984. One more question... what does "thus", or "first edition thus" mean? Thanks :) AndonSage 03:21, 5 March 2011 (UTC)

Yes, you can do a primary verification. In the notes indicate which of the gutter codes is verified. "First Edition Thus" means that this is the first time that all of the constituent parts of the book have been published together. For example, an omnibus of three previously published novels which have never been published in one volume before is considered "First Edition Thus". If these same three novels are later published together in one volume by another publisher, the term could not be used for this new book. If a collection adds a new story that wasn't in the first edition, the publication could be called "First Edition Thus". If a novel is revised and/or retitled the term could also be used. A book will never actually bear this statement. It's strictly a bibliographer's term. All that being said, a book that states "First Edition" is not always the true first edition of the book. It could mean the first edition published by this publisher. I always note edition statements in quotes to indicate that this is stated in the book. It's not a value judgement. When something isn't stated in the book itself, and I know it to be a fact, I follow it with a parenthetical "not stated" (see here and here for the two ways I use the statement). Mhhutchins 05:54, 5 March 2011 (UTC)

Thanks a lot for the explanation. I was thinking "Thus" was an anagram for something, hehe. Now it makes sense :) AndonSage 21:00, 5 March 2011 (UTC)

Is there unique title id for every title?

Hi, I wonder if there is such a thing as unique title ID, for every title, regardless of how many different publications this title had. Because for every title query that I make i get the same titles, but each time they have different title id, for example when i queried for all titles that where published in the Magazines, i got

 191159 Stuart Fleming "The Avenger"

but when i made a query for all titles, i got

 76952 Damon Knight "The Avenger"

but these two are actually the same title.

Is there any "master title id"?

P.S. I just want to point out that i am familiar with field t.title_parent = '0' which means that this is the canonical title, but this is not good, i need ID.

Thanks, Qshadow 15:02, 17 March 2011 (UTC)

There is no unique id for a title-name like "The Avenger" - as you can see, we give a different title-id to different works with the same name. BLongley 16:14, 17 March 2011 (UTC)

BLongley, i feel that you misunderstood me, Stuart Fleming is a Pseudonym of Damon Knight, and this IS the same work, only it is cataloged under different IDs.

http://www.isfdb.org/cgi-bin/title.cgi?76952

http://www.isfdb.org/cgi-bin/title.cgi?191159

Some titles have many different IDs, not just two. so what can I do?

For example I have created a table with titles from all magazines, and now i would like to create another table with all the rest titles (that were never published in a magazine), it seems it is not possible without such unique ID. Qshadow 21:28, 17 March 2011 (UTC)

The ISFDB considers those two different titles. It uses the parent id to link them. You can still do what you want, but you will need to do some more work in order to handle the variant relationship. --JLaTondre 23:07, 17 March 2011 (UTC)

The natural thing might be to collect all title id's, and then filter out all of those that are a variant title of something else (have a non-zero parent id). But that depends on what you're trying to collect. The "variant title" relationship captures both title variations and author name variations. So, for example, the book Frankenstein has been published as "Frankenstein", "Frankenstein, or The Modern Prometheus", "The Annotated Frankenstein", "The Mammoth Book of Frankenstein", etc. Presumably, you want only one title in your listing for those variants. As JLaTondre says, that's not built into the system. But you can generate the "canonical title" for a work by following the chain of variants up to a unique title record. Then, when you process "Weird Tales, May 1932", and find "Frankenstein (Part 1 of 8)", following that chain of variants up gets you to the same title record, so you know that it's appeared in a magazine. (Of course usually that "chain" of variants is just a single link, to a parent and not to a grandparent, etc. But I don't think that's guaranteed.) Chavey 00:57, 18 March 2011 (UTC)

The "make variant" code automatically re-points would-be grandchildren to the correct parent record, so grandchildren should never be created automatically. It may be possible to defeat the safeguards (I haven't checked in a while) and create them manually, but that would be against the spirit of the database schema. Ahasuerus 03:37, 18 March 2011 (UTC)

Thanks guys, so let's say I would like to build a query that will give me:

 MasterTitleID, CurrentTitleID, Title

where Master title id represents all the same works, even if they have different titles. How can I do this? I have been thinking and this is too difficult for my SQL skills. Of course after I could do such thing I could solve all the other problems (like generating all titles that where not published in any magazine, etc...). Please give me some ideas. Qshadow 13:53, 18 March 2011 (UTC)

Non-fiction, non-genre article

I would like to add an article that appeared in a non-fiction, non-genre journal (a science research journal). My reason for doing so is that (1) the author is well-known enough to justify inclusion of her non-genre work; (2) one of her short stories makes reference to the research described in that journal article (and I would include that fact in the notes for both works). It seems like the natural thing would be to make an entry for the journal, list it as non-genre, include the one article in its contents, and make a note that most of the journal contents are not included. However, I've read regularly that the system can't handle things like a "non-genre, non-fiction" sub-category, or a "non-genre, essay" sub-category. So will such an entry end up appearing "wrong"? And is this the right way to enter such a work? Chavey 23:01, 18 March 2011 (UTC)

Unfortunately, that's outside the scope of the database. It's inclusion would open the gate to many non-fiction pieces published in non-genre magazines by genre writers. Imagine all of the non-genre magazine records for Isaac Asimov's non-fiction pieces! And arguably very few writers would be considered bigger than Asimov in the field. Michael Bishop wrote an article for his local newspaper about a cancer clinic, the result of some of the research for his novel Who Made Stevie Crye?, and I would never consider that the database should include records for those newspaper issues. Why not just note the article in the note field of the fiction record? Mhhutchins 23:21, 18 March 2011 (UTC)

You've convince me. I'll handle it that way. Chavey 00:55, 19 March 2011 (UTC)

Review of a series?

How do we handle a review of a series, i.e. this? We have the series in the DB but can't link to that. We have the various titles, but should I link it to the first one? Should I enter multiple reviews, one per title? --MartyD 01:04, 24 March 2011 (UTC)

Since it can't be linked to the series, it should not be entered as a review. Make it into an essay, giving the name of the series in the essay title, so that if the software is updated sometime in the future, it can be located and linked. Mhhutchins 02:16, 24 March 2011 (UTC)

As four titles are specifically named, I'd enter those separately. The review doesn't really cover book five - "it’s absolutely impossible to review The Last Olympian without ruining the dramatic effect". BLongley 08:52, 24 March 2011 (UTC)

How many editions do I enter?

Since anybody can publish another edition of an out-of-copyright book, it will continue to be difficult to keep up on how many such editions there are. But here's a specific question regarding "Frankenstein, or, The Modern Prometheus". Centipede Press published special high-quality versions of this book in 2006 & 2007. There are 4 different versions of their publication, which vary based on such things as cloth bound (300 copies), silk-cloth bound with cloth slipcase (500 copies), leather bound with tray case (25 copies), leather and silk-cloth bound with tray case and extra prints (22 copies). This link shows them all, and they all have different appearances. Should this be listed as 4 editions, 2 editions (one in 2006 and one in 2007), or 1 edition? Chavey 22:19, 25 March 2011 (UTC)

The rule of thumb is "one publication = one verifiable physical item", so in this case I would enter then as 4 separate pubs. BTW, one common problem with limited editions is that many advertised varieties never see the light of day. Ahasuerus 23:00, 25 March 2011 (UTC)

And because of that last caveat, you should always note the source of your data, if you do not have the book in hand. Then, at least, we'll know where the info came from if it eventually turns out to be erroneous. That's why I like how Locus1 states "announced, not seen". (But this doesn't mean you can create a record for a book, an edition of one copy, with Neil Gaiman's hand-drawn caricature of Dream with a personal inscription, or I would have already done it.) Mhhutchins 23:42, 25 March 2011 (UTC)

Cloning 'flaw'?

Have been meaning to draw attention to this for awhile: cloning a pub automatically merges contents, etc. but creates a new cover artist record which has to be manually merged [assuming the artist is the same]. I'm not sure if there's a software 'fix' or not, but could there be a message that flashes informing editors of the extra step for the cover art if there is no 'fix'? --~ Bill, Bluesman 21:13, 3 April 2011 (UTC)

Tekwar series

Is there any dispute that Ron Goulart was the ghost-writer of most of Shatner's Tekwar series? Jonschaper 04:45, 6 April 2011 (UTC)

The current policy is that we only list pseudonyms if they have been disclosed elsewhere. In this case, Shatner has admitted that at least some of the novels were co-written with Goulart, e.g. he mentioned in his 2009 autobiography that "TekLab [was] written with the very talented Ron Goulart". I am not sure if the details of his contribution to each volume in the series has been made explicit, though. To use another example, Don D'Ammassa wrote in his Encyclopedia of Science Fiction that Goulart "has ghostwritten books in the Tek War series for William Shatner", but do we know if he was the ghost on all of them? BTW, Shatner's acknowledgments and dedications may help -- they are often fairly clear about who did (most of) the work without saying so explicitly.

Once we find sources, we can set these titles up as "by Ron Goulart and William Shatner (as by William Shatner)". Ahasuerus 06:44, 6 April 2011 (UTC)

The only source I have is the following from Ray Lovell: "[Ron Goulart] has ghost-written William Shatner's novels, Tekwar(1989), Teklords(1991), Teklab(1991), Tek Vengeance(1993), Tek Secrets(1993), Tek Justice(1993) - per Hawk's Authors' Pseudonyms II(1995)" Jonschaper 23:07, 6 April 2011 (UTC)

OK, I think it's good enough for our purposes. Ahasuerus 00:52, 7 April 2011 (UTC)

however it doesn't make it clear if Shatner did any writing, and I don't see any listings for "Tek Justice" outside of television so I have no idea how accurate it is. Jonschaper 23:07, 6 April 2011 (UTC)

Typically, it's best to set up ghostwriters as uncredited collaborators. There are two reasons for it. First, it can be hard to tell how extensive each party's contribution was -- e.g. just last month I discovered that Andrew Neiderman apparently used V. C. Andrews' plot outlines for a number of his "V. C. Andrews" novels. Second, if we don't credit the titular author in the canonical Title record, the title won't appear on his/her Summary page. If we were to do that in this case, a casual browser wouldn't be able to tell that Shatner was associated with the "Tek" series. Ahasuerus 00:52, 7 April 2011 (UTC)

"Ghosts and Gadgets"

Is there a way to find out whether the ghouls in Ghosts and Gadgets are real or of the Scooby-Doo variety? Ahasuerus 06:48, 6 April 2011 (UTC)

Well, using Amazon Look-Inside and reading the postscript, it seems that there is one real ghost and several fakes. BLongley 14:00, 6 April 2011 (UTC)

I'm actually slightly tempted to find this series - there's something very Charles Addams about the art at least. BLongley 22:39, 6 April 2011 (UTC)

Thanks, approved! Interestingly enough, the hardcover is only £7.99. Ahasuerus 00:57, 7 April 2011 (UTC)

Science Fiction Writers

Before I get too far into this, I thought I'd feel out the community here about the best way to proceed with this pub...

I'm in the middle of entering Science Fiction Writers: Critical Studies of the Major Authors from the Early Nineteenth Century to the Present Day, the 2nd edition from 1999.

I started by cloning the 1st edition; all the author articles that were featured in the 1st edition are also present in the 2nd edition, plus 23 new articles. If the title of the article and the author were the same between the two editions, I left them alone, but if the title changed (i.e. Arthur C. Clarke became Sir Arthur C. Clarke) or if the article's author changed (i.e. original version was written by x, but new version written by y) I attached a new title to the pub and then removed the original title once the original pub cloning was approved by a moderator.

Now, the Introduction to the Second Edition states:

Whenever possible, the original critics have revised and updated their essays as appropriate; when this has not proved possible, new critics have been enlisted. In some cases, the article's bibliography has been updated or a few paragraphs have been added, but in many cases, the original articles have been extensively revised or completely rewritten in order to take advantage of recent critical biographical discoveries and in order to provide a unified critical approach to their subjects. (I'll include this text as a pub note the next time I edit it.)

As a result, some articles from the 1982 edition have been revised by new authors, and both authors are now credited in the 1999 edition.

So my question for these articles is: should they be merged/made variants of the original articles from the 1st edition? Or should they be left as stand-alone titles in their own right? Clearly there is a relationship between the two, but I'm not sure how best to capture this relationship given the ISFDB editing guidelines I've come across.

Also, when the articles from the 1982 edition had the same title and author as the article in the 1999 edition, I left the original title attached to the pub and just updated the page number. Is that correct, or should I create new titles for all of the articles attached to this pub?

Thanks. Albinoflea 21:20, 6 April 2011 (UTC)

Well, I would be tempted to enter this as a completely separate book (i.e. not a clone of, nor merged with, the 1st edition). They sound sufficiently different to justify that. Then you might just merge those articles that were clearly repeats or updates of the articles in the 1st edition. I know I've seen that approach used at least once here before (but I don't remember where). But I wouldn't do this unless some of the other experienced editors jump in. Chavey 02:15, 7 April 2011 (UTC)

Something else to keep in mind: variants are based on a change in either title or author credit. Variants should not be created because of a change in text. If an article in the second has the same title and same author as the first, but you have sufficient evidence that it has been substantially revised, a new record should be created, without it being made into a variant of the earlier record. Disambiguation would be necessary in this case. As for the book itself, I agree with Chavey that the evidence you give would support creating an entirely new pub record, just like the editions of Anatomy of Wonder. In the end, you'll be the editor who will have to support any choices you make about the relationship between the two books and their contents. Mhhutchins 05:06, 7 April 2011 (UTC)

One more onerous task I can take off of my to do list. In what little time I spent thinking about it I thought it should not be cloned and the titles should be disambiguated. I also considered entering the more extensive discussions of certain works as reviews which is something that could be considered in the future. Extensive documentation of the first edition seems a bit superfluous --swfritter 23:21, 7 April 2011 (UTC)

I dread to think how much time we could waste on detailing Clute/Nicholls or Clute/Grant references in full. I presume we all know that we work from "The Encyclopedia of Science Fiction, 2nd edition" and that there is a 3rd forthcoming Real Soon Now?) With references like Tuck, we should eventually encompass all their data and can retire them as obsolete (although we might want to have secondary/tertiary Tuck Verifications just to make sure we've done it right.) Living, active, bibliographers are a pain we just have to cope with. ;-) BLongley 23:59, 7 April 2011 (UTC)

OK, for now then I'll create a new pub record for the book itself, and will not merge any of the articles that are in the pub with those from the previous version.

Also, for now, I'll leave the articles alone that have the same author and title in both editions, assuming that although there is likely textual variation I cannot quantify that variation until I can put my hands on a first edition copy and compare the two. (Probably next week sometime...) --Albinoflea 02:06, 10 April 2011 (UTC)

This is the record you unmerged from the first edition. According to most sources (and the cover you just uploaded), the editor of the second edition is Richard Bleiler, not E. F. Bleiler. Will this be corrected when you work on the essays? Thanks. Mhhutchins 02:37, 10 April 2011 (UTC)

Indeed, I had caught that myself, and of course you've just approved the edits, although I've now gone back in and linked to worldcat and added the blurb from the 2nd ed Introduction. Thanks,Albinoflea 03:24, 10 April 2011 (UTC)

Untitled Stories in Collection

We currently have Edgar Franklin's collection Mr. Hawkins' Humorous Inventions that was entered from Bleiler's Science-Fiction: The Early Years but without contents. Bleiler states "In the book, the stories are run on as untitled chapters" and then goes on to list the stories by the name of their original magazine appearance. This certainly sounds more like a collection than a fix-up or novel and I'd like to add the stories. Should I add the stories with their original magazine titles and a note that the titles don't appear in the book, or, should I make variants as "untitled" of each of the stories? Thanks. --Ron ~ Rtrace^Talk 11:43, 19 April 2011 (UTC)

The most difficult part of the process is physically verifying that a certain story corresponds to a certain chapter. Project Gutenberg has the book and that would help. Fictionmags also has Argosy listings. Given the accessibility issues with the original magazine appearances I think entering the the actual story titles would be acceptable as long as there is an explanation in pub notes. --swfritter 13:26, 19 April 2011 (UTC)

I went ahead and did this. Bleiler does have the contents by the Argosy titles. However, I was able to find this in Google Books, and was thus able to verify the titles were correct and get the page numbers. This was possible since the magazine titles are the name of the invention and thus are somewhere in the body of each chapter. I was going to say that I trusted Bleiler, but it turns out he has the title wrong, as does Clute/Nicholls. Bleiler78 has it right, though. Thanks. --Ron ~ Rtrace^Talk 00:12, 20 April 2011 (UTC)

I also cloned the Project Gutenberg edition. It has been downloaded a grand total of two times in six years.--swfritter 13:37, 20 April 2011 (UTC)

Peter Davison

In this pub there is a "Note on the Text" title credited to an author named "Peter Davison". The author record 10813 with that name as the canonical name, however, is definitely not the correct author. I suspect that this author 123665, named "Dr. Peter Davison", is the correct author record. How do I go about resolving this issue? Under normal circumstances I would create a variant title with a pseudonym, but since "Peter Davison" is already taken by a real author, I guess I can't make a pseudonym with that name... or can I? By the way, LibraryThing has an interesting disambiguation note (see bottom of the linked page) with information about the several authors named "Peter Davison". Patrick -- Herzbube ^Talk 20:36, 28 April 2011 (UTC)

As our existing Peter Davison isn't actually a Davison, and not actually an author, I renamed him aside as "Peter Davison (Doctor Who)". No doubt there'll be an autobiography at some point, but we can deal with that when it comes. BLongley 20:55, 28 April 2011 (UTC)

Thank you. I made "Peter Davison" a pseudonym of the canonical "Dr. Peter Davison". Patrick -- Herzbube ^Talk 21:36, 28 April 2011 (UTC)

Peter Phillips

Search for author "Peter Phillips" and you will find 3 authors (122471, 122468 and 122469). Their canonical name has been enhanced with the year they were born, apparently to distinguish them from each other. Is this really the way how distinguishing authors should be done? For instance, I have this pub which includes a story by one of these authors, and it looks really weird to see an author listed as "Peter Phillips (1920-)". Patrick -- Herzbube ^Talk 23:42, 30 April 2011 (UTC)

That's how databases work. Each name must be unique or all the records will be displayed on one author's page. Now if this were not a database, it would just be a matter of creating individual pages for each person, but then you'd lose all linking, searching, and sorting... Look at how another database handles Peter Phillips, all twenty of them. Mhhutchins 00:06, 1 May 2011 (UTC)

I don't think we've come up with a better way yet. Although it doesn't have to be by Year (see section immediately above for instance!) If all but one are inactive now, then it's fairly safe to allow the remaining one to keep the un-disambiguated name - but not totally safe. I've long wanted to put some sort of artist/author flag on our "authors" so I don't get search results polluted by the kinds I'm NOT searching for - and in my worse moods, I've wanted "mere" editors, interviewers and reviewers flagged too. Yes, I know many "authors" are actually in several of those categories, and when we sort out translators and editors of Collections I'll want even more options, but I think this is the best we can do at the moment. There may be a better way of dealing with multiple currently active authors with the same name - if there's more than one, I suggest we don't allow either to take the name over just yet. The two "Colin Harvey"'s are a pain, for instance, and we have some stuff not yet attributed one way or the other. :-/ BLongley 00:13, 1 May 2011 (UTC)

At one point, I tried to keep a name with two authors exactly as published and created a canonical name for each that included their birth year, and made their shared name into a pseudonym for each. It displayed ok when it came to pub contents, something like "A Story" (1990) • shortstory by John Smith (1941-) [as by John Smith] but the summary pages were a mess and searching didn't work well either. If you clicked on "John Smith", there was nothing displayed on that page, and all their stories were lumped together when you did a "Show All Titles" search. So I considered it a failed attempt to correct the situation, and haven't tried any other way since. Mhhutchins 00:30, 1 May 2011 (UTC)

I think there's a Feature Request (and if not, there should be) to improve "Show All Titles" to show the Canonical Author(s). It probably runs into the "how to deal with multiple authors?" question that hasn't been settled anywhere else, but it might be a start. BLongley 00:54, 1 May 2011 (UTC)

Regarding "that's how databases work"... well, it depends entirely on the database design (e.g. if the author name is an index that allows only unique values) and how the user interface manages duplicates if they are allowed. In the case of IMDB, the roman numerical suffix is slightly greyed out and offset from the main name, which is IMHO a rather elegant solution to indicate that the suffix is not really part of the name. Anyway, it was not my intention to stir up old discussions or criticize too much - if this is how things work (at the moment), then I'm fine with that. Patrick -- Herzbube ^Talk 01:20, 1 May 2011 (UTC)

Adding a New Magazine

I have begun entering issues of the magazine Dark Discoveries. How does the main entry for the magazine and the list get created? Do I need to create these? Thanks Sunnikay 18:09, 5 May 2011 (UTC)

Not sure what you mean by the "main entry" and list, but a start is to put the "EDITOR" records into a "Dark Discoveries" series. I've done that for you here. You might want to add Issue Numbers as well. If you mean on the Wiki Side, well, I've given it an entry in the Magazine List where you can add Editor or Publisher details or anything else that doesn't quite fit into the database proper. Have a look at other magazine pages to see what sort of info people like. BLongley 18:55, 5 May 2011 (UTC)

In fact, Magazines#Creating_a_new_magazine_wiki_page is probably more helpful than my comment. I've done steps 1-5. BLongley 19:03, 5 May 2011 (UTC)

Thanks!!! Sunnikay 19:41, 5 May 2011 (UTC)

You're welcome - we aim to please. And thanks for your comment "am a newbie to ISFDB, but have already found it addicting and amazingly fun to update" - please recommend us to your friends too! :-) You seem to have taken to it pretty well so far, please remember that most of our new editors find our help frankly user-vicious, so please do point out where we can improve. BLongley 22:23, 5 May 2011 (UTC)

How to add shortfiction to Neal Stephenson's bibliography?

hello, i'm new to this, but i'd like to contribute.

is there some place to simply "add a short work (shortfiction)"? should one simply create a new page in the wiki?

i couldn't find the "Short Works"[1] link on Neal Stephenson's bibliography[2],nor any place to add a short story, in particular: "Jipi and the Paranoid Chip", Forbes magazine, July 7, 1997 [3]. also available online [4].

1. Help:Screen:AuthorShortWorks
2. http://www.isfdb.org/cgi-bin/ea.cgi?429
3. http://en.wikipedia.org/wiki/Jipi_and_the_Paranoid_Chip
4. http://www.vanemden.com/books/neals/jipi.html

--Siznax 18:36, 7 May 2011 (UTC)

You can't enter short-fiction on its own, we need to know what it was published in. For Forbes magazine, this Help:Entering_non-genre_magazines might be what you need? BLongley 19:54, 7 May 2011 (UTC)

What in hell is Help:Screen:AuthorShortWorks? Is it from the early days of the ISFDB? If so, it needs to be deleted from the help pages as it no longer applies. Mhhutchins 03:27, 8 May 2011 (UTC)

It was a holdover from the ISFDB 1.0 days. It has been disincorporated. Ahasuerus 06:04, 8 May 2011 (UTC)

Back to the question, as Bill Longley advises, you must create a record for the book/magazine/publication in which the piece appeared, whether it's the first, last or only appearance of the story. Because Forbes is not a speculative fiction magazine it has to be entered using the guidelines for non-genre magazines that Bill has linked in his response. It's fairly straight-forward, and only a minimum number of fields need to be completed to create a magazine record: from the main page of the database choose "Add New Magazine" under the "Add New Data" menu, then enter the title and date of the issue in the Title field("Forbes, July 7, 1997"), in the Editor field enter "Editors of Forbes", skip the Tag field, in the Year field enter "1997-07-07", skip everything until the Contents section. Then add the page number on which the story begins, the name of the story, skip the date field (the system adds it automatically), entry type should be SHORTFICTION, choose length (probably "shortstory"), and then enter the author's name. Go to the bottom of the page and click "Submit Data". A moderator will look it over, either accept or reject or hold for more information (they will contact you through your user talk page if more information is needed.) That's it. If you have any more questions, don't hesitate to ask here or on the Moderators noticeboard. Mhhutchins 03:27, 8 May 2011 (UTC)

Title record not listed on author page

I am planning to merge this title and this title. Before I do so, I would like to point out, though, that the second title record does not appear on the author page (54) - search for "The Wheel of Time" and you will see that only one of the essay titles is listed there. Is this a bug in the software? I am asking because usually when I do merges I see both titles on the author page. I notice that both titles have exactly the same data (even down to the publication date), so it might be intentional that the software merges the titles on the author page. It's still unusual, though, and if this is indeed an intended feature, I personally find it not such a good idea because it makes it harder to spot duplicates that could be merged. Patrick -- Herzbube ^Talk 20:35, 21 May 2011 (UTC)

I see both essays on the author page, one 4 entries below the other. BLongley 20:48, 21 May 2011 (UTC)

They appear one above the other on my Silverberg summary page. Mhhutchins 21:04, 21 May 2011 (UTC)

Did you just add it to the database? If so, you need to refresh the page. If you used your browser's back button to return to the summary page, you're seeing your cache's version of the page, not the updated one. Mhhutchins 21:06, 21 May 2011 (UTC)

Ooh, it's moved! They're now both just above "Eartthsea" - which surely has one too many "T"s? BLongley 21:10, 21 May 2011 (UTC)

Oh yeah, refresh - the brand new browser feature I completely forgot about... *hit myself*. I'm glad, though, that I asked because it gave you a chance to publicly :-) find out about my typo... *hit myself again*. Topic closed, apologies for wasting your time. Patrick -- Herzbube ^Talk 21:31, 21 May 2011 (UTC)

It wasn't a waste, anything that teaches us a bit more about editing or moderating is good. I get called on my early mistakes several times a month still, and I'm supposed to be one of the "Major Contributors". BLongley 01:26, 22 May 2011 (UTC)

Setting up Magazine wiki pages

There is a help page for adding a new magazine, but it only seems to apply to adding a particular issue (and it seems to omit some necessary steps for magazine-specific concerns). However, there doesn't seem to be any help page for setting up, or maintaining, a "Magazine Wiki Page", such as linked to by Magazines. (Am I missing where it is?) At the end of the Magazine link above, there is a section on Creating a New Magazine Wiki Page, which feels like the beginnings of such a Help page. In following those directions, we start with a basic Magazine Template, which includes a table of "Bibliographic Tasks", i.e. a check-off of the steps that need to be done on a new magazine, or on any of the existing magazines, to complete a full bibliographic indexing of that magazine. Most of these steps are obvious; some not so much. I assume that the "Determine canonical names" and "Check for pseudonyms" mean that I should check each of the authors of fiction, artwork, essays, etc. to verify that they have been entered and linked to their canonical name (right?). Does "Cross check reviews" mean something similar, presumably including verifying that the title of the book as listed agrees with the title we have? What does the "Merge titles" refer to? Chavey 14:29, 22 May 2011 (UTC)

"Cross check reviews" means that you've linked any reviews in the magazine to pre-existing title records. When this task was written there was no automatic linking. "Merge Titles" is even simpler. Sometimes magazines reprint stories that were originally published elsewhere. By checking "Merge Titles" on the task list you're saying that you've made sure that each of the stories have been merged with any pre-existing records in the database. (Something you should also do when adding collections or anthologies that reprint stories.) Any way, it's not necessary that you do any of those tasks, or even that you create a wiki page. (Some of the magazine editors would shoot me for saying that.) Use the template to create a page, and then add a link to the Magazines list. You only need a basic knowledge of wiki write-up to do both. That may be why there's not a detailed help page.

It's more important that you place the magazine's editor records into a series so that a Magazine Issue Grid is created on the database side. Some people forget that the wiki is an adjunct to the database and a tool to do tasks that can't be accomplished on the database. With the creation of the Issue Grid, the wiki magazine pages have lost most of their importance. Before anybody starts slamming me, I said most. They still have some purpose. Mhhutchins 20:21, 22 May 2011 (UTC)

I just checked out the link to creating a new magazine and it looks to be basically all you really need to know. Is there anything specific about the set-up that you need help with? If you prefer, I could do the basic page set-up and linking, and then you can fill in the data. Mhhutchins 20:25, 22 May 2011 (UTC)

Adding ToC to anthologies/collections?

Is there a clever way to add titles to anthiologies/collections? Just entering ToC creates duplicate titles that need to be merged. Is there an option to enter title id straightaway? Similar to the way it's done when one makes variant of a title. Thanks! P-Brane 02:28, 27 May 2011 (UTC).

Ask and you shall receive -- Help:Screen:ImportContent! :-) Ahasuerus 02:35, 27 May 2011 (UTC)

I want more!:) I want to enter ToC for a new collection, there is no previous collection with data that I need to import. P-Brane 03:04, 27 May 2011 (UTC).

Oh, I see. I still can do it, but I will need to add title by title. Still better than nothing, I guess. P-Brane 03:06, 27 May 2011 (UTC).

The ability to enter Contents titles by ID was discussed at one point, but I don't think a formal feature request has been created yet. Ahasuerus 03:35, 27 May 2011 (UTC)

Changing an award name

How do I change the name of an award? Currently the Locus Awards are in the system as the "Locus Poll Awards". There ain't no such thing. (Aren't all awards a poll of the voters?) Mhhutchins 19:41, 2 June 2011 (UTC)

At the moment, you ask Ahasuerus nicely. I'm working on Award Name Editing for moderators, but it's not ready for prime-time yet - it seems we want to get away from two-letter-codes. And standardise categories within awards, and order them consistently, and decide which awards get placings rather than winner/nominations... there's quite a bit wrong really. BLongley 20:18, 2 June 2011 (UTC)

Award for a series

Is it possible? Second place in the 2010 Locus Award for best collection went to The Collected Stories of Roger Zelazny, Volumes 1-6. How do we enter this award in the db? Mhhutchins 20:10, 2 June 2011 (UTC)

Not currently possible, unless you give the same award to each individual volume. Which I wouldn't encourage long term - I've recently rejected awards for each individual episode of a serial, but at least there we have an overall title to link to instead. We do need to improve linking - title matches are to real titles, although as you've seen already, variants get confused. Author matches are text-based at present. And publisher matches and series matches are out at present. It seems that Artist matches could be improved too, as the Cover Art awards sometimes go into more detail than our COVERART record suggests. And foreign language support is pretty non-existent due to the variants problem. So much to do still, but keep the comments coming. BLongley 20:32, 2 June 2011 (UTC)

Check out the Locus Awards for 2003: look at the category for Best Young Adult Novel (BTW, the Locus Awards don't give "Best" in their categories. How do we correct that without editing every award in the category?) You'll see that Abarat isn't linked. Is there a way I could enter The Collected Stories of Roger Zelazny in the 2010 Collection category and not have to link it? Also how do I link Abarat to its title record? The only thing I can think to do is to add a new award to the title record, but then how do I delete the award that's currently listed? So many questions...it's like I'm a noob all over again. Also I just noticed that not every category has been added for any of the years I've checked for the Locus Awards, especially in the short fiction categories. Guess this is going to take longer than I thought. Mhhutchins 21:12, 2 June 2011 (UTC)

Re: "Best" - I'm afraid removing those en masse is still Ahasuerus only I'm afraid. I had the same problem with the Stoker Awards which are not for "Best" but for "Superior Achievement in". Only took me 50 edits or so... BLongley 23:08, 2 June 2011 (UTC)

The Locus Awards have 200 awards...each year! Mhhutchins 23:38, 2 June 2011 (UTC)

Yeah, we'll work on that - let's get the Award names fixed first, then we can do Categories within Awards, by which time people will be moaning about "Levels" within Categories. We did tell you you're all just Beta-Testing Award-Editing at the moment? BLongley 01:00, 3 June 2011 (UTC)

Sorry, I just noticed there are editing links on the pages for each year's awards. I can edit AND remove awards from there. (...and he walks aways after slapping his palm to his forehead...) Mhhutchins 21:15, 2 June 2011 (UTC)

Yes, to link "Abarat" you'd have to remove the award via the "Award by Year" page and then re-add it to the title. And you can add an "untitled" award for the Zelazny series via the "Award by Year" page. (In fact, from ANY "Award by Year" page.) It's really not intuitive is it? I think that "new award" should at least default the year and award type for you if you're on one of those pages. And we should probably have a new award link on Author pages for all those "Lifetime Achievement", "Grand Master", "Best Editor" awards and such. Although I'm really not sure we need to record "Best Buns (Either Gender)" BLongley 23:08, 2 June 2011 (UTC)

Re the individual awards: Locus gives them out for editors and artists (about 20 in each category per year). Those will have to wait until we add award links to Author pages. Mhhutchins 23:38, 2 June 2011 (UTC)

Actually, with the text-matching, it's not too bad entering a series of the same award to different people, with good use of the "back" button. I did 7 years of "Best New Talent" Ditmar submissions in under 15 minutes. (Whether they'll "take" on approval is another matter...) BLongley 23:57, 2 June 2011 (UTC)

I don't think I'll use the "untitled" award approach to entering the individually awarded Locus Awards. It might be too much of a hassle to convert them when/if we ever get the ability to link awards to individuals. Mhhutchins 00:12, 3 June 2011 (UTC)

I think it's a "when" for those, although only Ahasuerus can say for sure if we'll automatically correct all previous entries. I suspect at worst that an award given to the Canonical name will stay and ones given to pseudonyms might need manual fixing, like we currently have with interviews. Awards given to a particular person for a specific work, apart from the usual fiction awards, will probably need some manual rework - e.g. not all Coverart Awards link yet. I think there's enough of those that I'll look into an Award Linking option like our Review Linking option, rather than make people add and remove each one. BLongley 00:31, 3 June 2011 (UTC)

Serials, Supplements and the Miller/Contento Reference

Sorry to dump a bunch of questions in the same topic, but they've all arisen together. I recently acquired Miller's Alicia in Blunderland and wanted to ensure the the serial appearances of this novel were added, which originally appeared in Science Fiction Digest and its successor Fantasy Magazine. I am using the Miller/Contento reference as my source for these fanzines. The following questions have arisen:

1. Miller/Contento lists page numbers with a plus sign (e.g. "32pp+"). Should we understand that to mean that the page count does not include the covers? The one example where fanac.org has a scan of the issue, would seem to indicate this as the case (M/C has "20pp+", the scan has 24 pages). Should I take that "+" as indicating that I should add 4 pages for covers to the total?
2. Miller/Contento indicates that there was a serialization of the round-robin Cosmos that "were not printed in the issues themselves but as separately-bound supplements". How should these be added? Should they be included in the container of the Fanzine without page numbers or as separate titles? If they should be separate, what title type should be used? Fanzine? Chapterbook? Should they be put in the Fanzine series? I haven't added any of the installments yet. My own feeling is that it may be cleaner to keep them within the fanzine, with a note about their separate binding. That seems to be the way M/C is handling it.
3. It appears that the Cosmos round robin never was published as a standalone novel and there is currently a parent title record of type SERIAL. When I have previously encountered serializations that weren't published as books, I have made a parent title of type NOVEL or SHORTFICTION (example). While I thought I read somewhere that I was making variants as preferred, I can't seem to find it now. What it the preferred title type of a serial that only ever appeared as a serial?

Thanks. --Ron ~ Rtrace^Talk 00:32, 3 June 2011 (UTC)

1. Yes, you should add 4 to the page count of all Miller/Contento references with a plus sign.
2. I would include the supplements as part of the original record, since they probably weren't available separately. The page count field would be, for example, "64+12". And note the supplement in the note field, and the serial in the contents section of the record. Making the supplement as a separate CHAPTERBOOK record seems to be overkill. Besides, how would you connect the two separate records?
3. I've created a parent NOVEL record in cases like this (when a serial has never appeared as a book), but there was a dispute at the time about how these should be handled. I don't believe any firm decision was ever arrived at. This was back when we were converting all serials (and their parts) as variant title records of the original NOVEL record. The only way to connect serial parts for non-book serials was to create a false NOVEL record. I still believe this was the best way to do it. Others disagreed. Mhhutchins 03:57, 3 June 2011 (UTC)

Adding Books

I want to add some books to Anne Spencer Parry's page. I have a wiki login but it doesn't seem to work for the ISFDB. [Unsigned comment by Kholburn, 11:39 5 June 2011]

What error message do you get? "Login failed: Bad password" or "Login failed: Bad user name" or something else? BLongley 13:05, 5 June 2011 (UTC)

It appears that you do have a proper ISFDB login (as kholburn). The books you've added were added as plain text to a discussion page for Anne Spencer Parry, which is intended for discussion of whether certain books belong in her list, documentation for where we got information about her, etc. As you noticed, they didn't get into the "main" bibliography. To add these new books to the system, you need to (1) login as kholburn; (2) on the ISFDB Home Page, check out the navigation panel on the left (where the first two lines should say "Logged In As ... kholburn"), scroll down to the section "Add New Data", then click "Add New Novel"; (3) Proceed to add the formal data about these new books. Initially, the fields of requested data can be a little intimidating, and at the top there's a link for "Help:Screen:NewNovel" that gives you all of the details for what those fields mean. (You don't really need to follow the more general link on "Help:Screen:NewPub".) That help page itself is pretty long, and can be a bit intimidating, so I recommend a "Just in Time" approach to it -- read the sections that discuss fields that you don't quite understand what to enter (e.g. "Year", "Pub Format", and "Artist1)." And for your first entry, it's a good idea to go to the link at the top of that help page which suggests "If you are new to editing the ISFDB, please see Help:Getting Started.". (You can click on that link in the last sentence as well.) Meanwhile, if you have other questions, feel free to post them here. And welcome to the ISFDB Kim (I assume)! Chavey 13:21, 5 June 2011 (UTC)

If you created a user-id of "kholburn", note that the Wiki Software will have changed it to "Kholburn" with a capital K. BLongley 13:33, 5 June 2011 (UTC)

His Share of Glory: The Complete Short Science Fiction of C.M. Kornbluth

I've noticed that all of the references to this book end with "Cyril M. Kornbluth" instead of "C.M. Kornbluth" as is shown on covers (and in the case of my SFBC edition, the title page). Shouldn't the book names be fixed to end with C.M. Kornbluth? Or is there some reason they were entered with "Cyril" instead of "C." in the first place? AndonSage 19:58, 12 June 2011 (UTC)

I hope that it's now fixed. Hauck 20:38, 12 June 2011 (UTC)

Looks like it, thanks :) AndonSage 21:30, 12 June 2011 (UTC)

Deleting Stuff

I have been looking at the page for the SF artist Harry Turner and find that there are 1. Omissions from the list of artwork in issues in Nebula (an other publications have been missed completely), and 2. Some horror stories have been attributed to him but they were written by another Harry Turner. How do I add further data to the artwork list and delete the wrongly added short stories? —The preceding unsigned comment was added by Farrago (talk • contribs) .

You will have to create records for the publications which are missing. I've left a list of links in the Welcome section on your user talk page that will help you get started. You can't delete the stories as they are legitimately credited to Harry Turner, even though it may not be the same Harry Turner. If you're certain that the four short stories published in the Pan anthologies in 1971-1976 are not by the artist of the same name, we have to disambiguate the two people, probably by adding a birthdate to the story writer's name. I'll do some research to determine if there are any connections between the two. Thanks for bringing this to our attention. (Remember to sign all comments on the Wiki by adding four tildes [~~~~] at the end of your message.) Mhhutchins 16:02, 16 June 2011 (UTC)

It looks like all issues of Nebula are in the database, but some of them (the orange issues) haven't been primarily verified. These don't have the interior art credits that the others may have, so they may have the Harry Turner art credits that you know are missing from the database. You can update any record that is missing art credits by using the "Edit This Pub" link under the Editing Tools menu on the page for the issue's publication record. For example, if you have Issue #9, go to its record (here) and click on "Edit This Pub". Add the missing art credits under the Content section by clicking on "Add Title", complete the fields for that record, then continue adding titles until you've finished. Mhhutchins 16:18, 16 June 2011 (UTC)

Magazine title format

I want to fix up Dark Moon Digest but am unsure of how to handle the titling. This is a tp quarterly magazine. As best I can tell from the covers and this one Look Inside, the issues have no stated date beyond a copyright year. The covers have "Issue Number 1", etc. The one copyright page we can see says Dark Moon Digest -- Issue One over Copyright © Dark Moon Books 2010, and its TOC page says Dark Moon Digest over The Horror Fiction Quarterly over Issue Number 1. Amazon lists a pub date for each as the 1st day of the quarter (Oct 1, Jan 1, Apr 1). The entries are in there with the non-standard titles grabbed by Fixer. I'm not sure what "standard" should be, though. Help? --MartyD 13:51, 18 June 2011 (UTC)

I would suggest using the title "Dark Moon Digest: The Horror Fiction Quarterly". The long title is used on the table of contents page on the "Look Inside" you refer to, and is used in the main description on the magazine's Home Page. That home page starts with "The Horror Fiction Quarterly", then it has "Dark Moon Digest" in a separate box, with a larger font, but in the running text it refers to it as "Dark Moon Digest - The Horror Fiction Quarterly". You could use that title (with the dash), but we usually use the colon for 'sub-titles', and that seems to apply in this case. With respect to months of publication, I would suggest using the Amazon months. Issue #1 has a leading editorial, in the "Look Inside", that says "It seems somehow fitting that our first issue is coming off the press in October." And their home page says that issue #4 will be "available mid-June". This agrees pretty closely with the quarterly dates that Amazon lists. Chavey 17:26, 18 June 2011 (UTC)

I can't log in.

I have an account (I got an e-mail and responded to it etc., and on this page where I'm typing I see appropriate info (my user name and tabs for my contributions etc. But when I try to edit, I am asked to log in and then told that my user name is "bad." Kdammers 01:56, 23 June 2011 (UTC)

Was the user name that you chose, by chance, "kdammers"? If so, then keep in mind that the wiki software automatically capitalized the first letter at registration time, so you will have to log in as "Kdammers". Ahasuerus 02:27, 23 June 2011 (UTC)

Saga of Dray Prescot: The Havilfar Cycle

I recently added an omnibus for the Saga of Dray Prescot series, The Havilfar Cycle e-book. I wanted it to appear as the other omnibus titles, but it doesn't have the [only as by Alan Burt Akers ] like the other titles. When I originally added the e-book, it added the title, which I then had to change the title author from Alan Burt Akers to Kenneth Bulmer, because it wasn't showing up where it should with the other titles. When I compare the title I added to another title, like The Havilfar Cycle I, the one I added is missing the Variant Title by Alan Burt Akers.

So anyway, I don't know if I did something wrong, or haven't done something that needs to be done. Can anyone let me know what I need to do please? Thanks :) AndonSage 19:58, 4 July 2011 (UTC)

I don't know who did what, but the result is a pub record by Alan Burt Akers, linked to a title record by Kenneth Bulmer. Both records should have been by Alan Burt Akers, with the title record a variant of the Kenneth Bulmer record. Repair will take a few (sensitive) edits, if you want I can do this for you, or lead you through the steps. --Willem H. 20:34, 4 July 2011 (UTC)

The title record author was originally Alan Burt Akers (pseudonym of Kenneth Bulmer), created when I created the pub record, but it wasn't showing up with the rest of the series under Saga of Dray Prescot, so I changed the author to Kenneth Bulmer, and now it shows up where it should, but without all the data it needs. Basically, I needed two title records (one by Kenneth, and one a variant by Alan), but only had one, and don't know how to create a title record by itself. I'll fix it if you tell me what to do (or you can fix it if it's easier for you), but what I'd really like to know is how to do it correctly from the beginning, please :) AndonSage 21:20, 4 July 2011 (UTC)

Actually, following the R&S discussion, it appears these should both be Bulmer, with no variant. The title edit is half of that, and I assumed you were just waiting for that to go through. Now you need to change the author credit on the pub itself to match. If you need help, let me know. --MartyD 21:52, 4 July 2011 (UTC)

If you look at the entire series, all the other omnibus editions use Alan Burt Akers as the author for the Publications, and have two Title records, one for Kenneth Bulmer, and a Variant for Alan Burt Akers. I had to add another omnibus because the Havilfar Cycle in e-book form is books 6-11, not split into two parts (books 6-8 and books 9-11) like the paperback and hardcover editions. I'm totally lost as to what needs to be done now, so yes I need help please. AndonSage 22:58, 4 July 2011 (UTC)

If I read this correctly, the titlepage gives credit to Alan Burt Akers, and according to our rules it should be entered this way (titlepage is still leading, isn't it?). Perhaps you can check this, to determine which is the right credit. --Willem H. 06:16, 5 July 2011 (UTC)

All the publications are entered with the author as Alan Burt Akers, which is correct. It's the title records where we have two records for each book... the main title record by Kenneth Bulmer, and a variant title record by Alan Burt Akers. The problem right now is that I had to enter a new publication for The Havilfar Cycle, which created one title record, with Alan Burt Akers as the author. Since that title record wasn't showing up in with the rest of the titles in the series, I changed the author from Alan Burt Akers to Kenneth Bulmer. So now it shows up where it should. But we need another title record that's a variant for as by Alan Burt Akers to match the other title records. I don't know how to create a title record by itself. All I want is to get The Havilfar Cycle set up like all the other titles in the Saga of Dray Prescot series. And I need to know how to do this for the future, because I can see the same thing happening when The Pandahem Cycle (which is also 6 novels, plus 2 short stories) is released. AndonSage 06:48, 6 July 2011 (UTC)

I just noticed the Add Variant Title option when I was looking at The Havilfar Cycle title record, so I used that to add a variant with Alan Burt Akers as the author. Hopefully this was the way to do things? AndonSage 06:52, 6 July 2011 (UTC)

It all depends on what's on the title page. If the title page says only Akers, then we should have that as the title record for the publication (i.e., undo your original edit), and make the title by Akers a variant of the title by Bulmer. If the title page says only Bulmer, OR if the title page says Bulmer writing as Akers, then your title edit should stand, and the publication's author should be changed from Akers to Bulmer. And there would be no need for a variant. To get the title to appear in the series, you only had to add the series name to the title. But note that when you look at any author's bibliography page, series there show only works by them -- titles in the series are not displayed there (to see the series, you need to click on the series name; the series display will show all titles in the series, regardless of authorship). The way to get the omnibus to show up in the series on Bulmer's page is to make the Akers-credited title a variant of a new or existing Bulmer-credited title (Make This Title a Variant Title or Pseudonymous Work). It will take a couple of edits at this point, so let us know what the title page says, and I will fix it up. Thanks. --MartyD 11:23, 6 July 2011 (UTC)

The title page is available in the "Look Inside" feature in the Amazon record for the book. (They appear to have combined the Title Page with the Copyright Page.) On that page, only "Alan Burt Akers" appears (multiple times). "Kenneth Bulmer" appears to be pretty much limited to the cover. So the book should appear as a publication inside an "Alan Burt Akers" title record, then (as MartyD notes), you should then "Make This Title a Variant Title" of a new record, in which you change the author from Akers to Bulmer. Chavey 12:36, 6 July 2011 (UTC)

The title page from the Kindle book:

The Havilfar Cycle

The Saga of Dray Prescot Omnibus edition

Alan Burt Akers

Mushroom eBooks

The copyright page is listed in the first five lines of notes for the publication The Havilfar Cycle. AndonSage 19:27, 6 July 2011 (UTC)

(unindent) Ok, so that means the pub should have a title by Alan Burt Akers, and that should be a variant of a title by Kenneth Bulmer. All fixed. By the way, if you put a title into a series and then discover it should be a variant (as in this case), when you do the Make This Title a Variant operation, the series information will be moved to the new parent automatically. Sadly, none of the other information (notes, synopsis, web page. Wikipedia) is. I moved it to the parent, rather than leaving it duplicated. I suppose we should put in a feature request to have the variant title show that information from the parent when having none of its own.... --MartyD 23:29, 6 July 2011 (UTC)

Thanks for your help on this! Everything looks good now :) AndonSage 07:30, 7 July 2011 (UTC)

The Crystal Horde

Policy check: Should John Taine's The Crystal Horde be made a variant of White Lilly with the note that The Crystal Horde is a "substantial rewrite", or should they be left separate with notes? Jonschaper 03:06, 6 July 2011 (UTC)

I would leave them separate, with a note in each. Variants are for title and authorship variations, not text variations/derived works. So if the text is not "the same", they should be separate. I don't know if it's formally documented, but there's some leeway in what "the same" means (I guess it's a bit like the "certain threshold"). We tend to consider minor revisions to be the same work, and whether any particular revision crosses the line from minor to major is left to the ISFDB editor's judgment. I'd say the combination of "substantial" rewrite and different title take it out of the realm of minor.... --MartyD 10:07, 7 July 2011 (UTC)

The Time Vault

"The Time Vault" is a series of reprints in Perry Rhodan paperbacks, but the stories themselves are totally unconnected. In contrast, other series of independent stories like "Probability Zero" at least have a connecting theme. Should this "series" be elminated to avoid creating the false impression that the stories themselves are connected? Jonschaper 03:35, 6 July 2011 (UTC)

Sounds to me like they should be a publication series, not a title series. A publication series will often have unrelated books. As it was explained to me "If you can imagine reprinting those books in a different format, where they wouldn't be part of that series, then it's a publication series. If any reprint of the book would still be part of the same series, then it's a title series." Chavey 02:48, 7 July 2011 (UTC)

Unfortunately these are SHORTFICTION, not Publications. I think there's no value in making these a series at all. Same with "Shock Short". BLongley 12:12, 7 July 2011 (UTC)

I concur. Some of these stories may be a part of a "real" series, so putting them in an artificially created one would cause problems. Ahasuerus 13:13, 7 July 2011 (UTC)

Content page #: Title page or story page?

In a collection or omnibus, it can happen that the page number for the title page is 1 or 2 pages prior to the page number on which the story begins. Which page should be entered for the page number in the contents for that book? Oddly enough, the relevant Help screen doesn't specify this detail. Chavey 23:27, 9 July 2011 (UTC)

If the title doesn't appear on the first page of the story, but on a previous page, I use the page number on which the title appears. Other editors may do it differently since there's no specific standard. Mhhutchins 00:37, 10 July 2011 (UTC)

That's the approach that seems right to me. But, since we don't have a standard, if I'm a Primary2 verifier on a book that uses the other approach, I assume that means I should leave the page numbers as is, right? Chavey 08:13, 10 July 2011 (UTC)

In those cases, I tend to stick with the same page number that the ToC states, if it matches either. (ToC can be completely wrong so I always double-check.) You can also ask other Primary Verifiers if they mind it being changed, but personally I don't think it worth the effort. BLongley 17:16, 10 July 2011 (UTC)

Disentangling two authors

How do I disentangle two authors who have been mistakenly combined? Mary Helen Wright is not the same person as Helen Wright/Helen S. Wright. (Or if she is, I died in 1997 and didn't notice...) (This question was asked by new member "Arkessian" at 08:49, 11 July 2011.)

This is the right way to do it! The technical process to "disambiguate" two authors is a bit complicated for a new member of the ISFDB, so the easiest thing is to give us the details and we'll fix up the records for you. I assume that you are Helen S. Wright, and wrote A Matter of Oaths, and that "Mary Helen Wright" co-authored Great Adventures in Science. Did I get that right? What we need to do is to list the two authors using sufficiently different names that users can't confuse them, e.g. we can't continue to list "Mary Helen Wright" as "Helen Wright", even though that's the name she published under. (For example, we might use a 'name' such as "Helen Wright (1914-1997)".) Of course we can't list you under "Helen Wright" either, but since you publish as "Helen S. Wright", that's an easy solution. When you answer this, we invite you to share any of the basic data that you would like to have listed in your bibliography. The information we list, when publicly available, includes birthdate (not everyone wants that published), birth place, an author web page (when it exists), author photo (if copyright free), full legal name, and occasionally an email address (but most authors don't want that published). So let us know, and we'll fix things up. And thanks for letting us know about our mistake! Chavey 13:07, 11 July 2011 (UTC)

I've separated "Helen Wright (1914-1997)" as she's unlikely to write anything more. That leaves Helen S. Wright as the canonical you (where you can share whatever biographical details you're happy to make public) and your pseudonym "Helen Wright" (which should just redirect to the canonical you.) BLongley 14:36, 11 July 2011 (UTC)

Thanks!

Variant Title Problem

I think I may have made a bit of a hash of entering this anthology into the database. Checking for duplicates worked fine for 4 of the authors, but Moonless Night by Robert Chilson was a bit more complicated because the author is a pseudonym for Rob Chilson. I made it a Pseudonymous Work, but now on the Summary Bibliography there are two entries in the short fiction for variant titles of Moonless Night (1978) that look identical. One links to a copy of Galaxy and the other to Science-Fiction-Stories 84.

Can someone tell me where I went wrong? Deagol 19:33, 21 July 2011 (UTC)

I don't think anything went wrong, and certainly not you. One of the latest developments here is the addition of a language field to title records. Officially we don't support translated titles yet, but off the record they are allowed in. In this case, the second title record is the German translation, the first is the English original. Both are variants of the Rob Chilson title of course. Can you change the title of the German version to the translated title? --Willem H. 20:10, 21 July 2011 (UTC)

Yes, can do that. Would you suggest leaving the date as for the original publication, or changing it to that of the translation? I see pros and cons either way. Deagol 21:46, 21 July 2011 (UTC)

That's probably a new "Rules and Standards" discussion. :-/ I'd leave it as date of translation, but the recent language support improvements are so new that we haven't really figured out all the implications yet. Please comment on all the problems they have caused, we may fix them yet! BLongley 22:57, 21 July 2011 (UTC)

Ok, will do that. I guess I will have to unmerge the other titles as well so I can change them to the translated title.

The only problem caused is the apparent repetition of titles. One possible fix for the future would be to include the language in brackets for the non-English variant titles. Deagol 16:49, 22 July 2011 (UTC)

You seem to have encountered the new problem with unmerges. When unmerging container titles, we recently decided that the publication's authors were often more correct than the current authors. This seems to be less desirable when unmerging contents titles. :-( I'll look into that and see if I can fix it, but for now the old "add new and remove old" method may be best. BLongley 23:08, 22 July 2011 (UTC)

I've approved the unmerges, then fixed the authors. Title change and language and make variants still to be done - and I'll look into "language in brackets for the non-English variant content titles" as another software fix. BLongley 23:25, 22 July 2011 (UTC)

I think we need to encourage editors to not use the "unmerge titles" function to remove contents from pubs until the bugs have been worked out. The "remove titles from this pub" function works perfectly well. Mhhutchins 19:32, 23 July 2011 (UTC)

Agreed. And maybe a warning to Moderators that such innocent-looking edits aren't necessarily going to work as expected. Sorry about all the problems, I think I introduced some of them - but the latest fix should sort that and maybe fix Reviews and Interviews too. But Developer/Tester/Administrator time is pretty limited at present. :-/ BLongley 23:19, 24 July 2011 (UTC)

Anthologies Titles & Publication connection

Hi All, I have a list of my favorite anthologies by their title_id, and I am trying to make a query of their contents.

But i run in to trouble, because the contents are only listed per publication, so how do I connect between anthology title_id to the pub_id ? There can be several pub_id per one title_id, and some of the publication do not have any contents, so i probably need to get contents from all the pub_id of the same anthology and remove dups later.

So to sum up the question is how do I connect between anthology title_id to the pub_id ?

Thanks, Qshadow 12:59, 22 July 2011 (UTC)

The pub_content table will contain an ANTHOLOGY title for each pub that has the required title_id: and the other contents will have the same pub_id. So for Witches for instance, the query would look like:

select DISTINCT t2.title_id, t2.title_title
from titles t1, pub_content pc1, titles t2, pub_content pc2
where t1.title_id = 519989
and   pc1.title_id = t1.title_id
and   pc1.pub_id   = pc2.pub_id 
and   pc2.title_id = t2.title_id
and   t2.title_ttype != 'ANTHOLOGY'

Notice the DISTINCT, which removes duplicates (if they've been properly merged). Hope this helps. BLongley 15:03, 22 July 2011 (UTC)

It's working! Thanks a lot! Qshadow 16:17, 22 July 2011 (UTC)

Editor or Author for Anthology?

Hi, I am trying to determine for each anthology if it is a collection of works by different authors or is it anthology of just one author. How is it possible? I was thinking that maybe:

1. If the author of the anthology marked "Editor" on the title page - that means that inside will be collection of different author.
2. If the author of the anthology marked "Author" on the title page - that means that inside will be stories of just this Author.

Am i correct? And if yes how can i determine in the DB, which one is editor and which one is author? Thanks, Qshadow 09:42, 25 July 2011 (UTC)

We use different title types. Collection is used when all works are by the same author(s) and is usually (always?) credited to the author(s). Anthology is used when works are by different authors and are usually credited to the editor(s), if known. --MartyD 11:08, 25 July 2011 (UTC)

Thanks MartyD, I thought for some reason that i saw anthologies of one author only, but i was wrong. Qshadow 08:59, 26 July 2011 (UTC)

How to match between two tables of titles

Hi again, I have run into major problem, when I tried to match titles from magazine contents and titles that I got from query of anthologies contents, I noticed that although the same works are present in both lists, the do not match because their title_id is different, and sometimes their names are different too. E.g.:

From the Magazine table
Robert Williams "Refuge for Tonight" title_id=917052

From the Anthologies table
Robert Moore Williams "Refuge for Tonight" title_id=58536

Is there any way I can find out that this is the same work?

Thanks! Qshadow 18:03, 27 July 2011 (UTC)

Yes. If you look at the title_parent column, you'll see that one is a variant of the other:

select title_id, title_parent
from titles
where title_id in (917052, 58536)

title_id 	title_parent
58536		0
917052		58536

If there is a variation in title or author, then we'll record all published combinations and make all of them bar one variants of the "canonical" title which will have a parent of 0. BLongley 18:21, 27 July 2011 (UTC)

Thanks, I know that, but when I tried to use

AND t.title_parent = '0' 

in my query for magazine contents or for anthology contents, I was loosing all the works that where published under non canonical names.

Is there any way to make such query to return ALL the works from magazines with their "canonical" title, even when some of these works where published using non "canonical" title?

Regards, Qshadow 18:45, 27 July 2011 (UTC)

Yes. Where you're currently choosing "title_id", choose the "title_parent" instead if it's not 0. You can use the MYSQL CASE or IF functions: e.g.

select title_id, title_parent, if(title_parent=0,title_id,title_parent)  
from titles
where title_id in (917052, 58536)

title_id 	title_parent 	if(title_parent=0;title_id;title_parent) 
58536		0		58536
917052		58536		58536

This should be fine unless the chains of variants are messed up - there should only ever be one canonical title, not a chain of variant of variant of canonical titles. If you do find such chains let us know so we can correct them. BLongley 19:05, 27 July 2011 (UTC)

Wow, this was brilliant, everyday one learns something new... I will definitely tell you if i find any chains. Thanks, Qshadow 19:46, 27 July 2011 (UTC)

Original language of the sf work

Is it possible to know the original language of the work? I found the translation_id but the schema page says it is unused. So the bottom line, if I would like to get only the titles that where written by English speaking authors. What should I do? Regards, Qshadow 14:14, 31 July 2011 (UTC)

A little over a month ago we added a new field, title_language, to the "titles" table. It points to the "languages" table, which lists the 73 languages that we currently support. At the moment, only 7,000 records, i.e. 1% of the total number, have this field populated, but things will improve over time. Ahasuerus 15:29, 31 July 2011 (UTC)

1%? More like 0.01%. See the stats page. :-/ Yes, we're improving as we force people to choose a language for titles, but I think we'll need a few mass-updates at some point. BLongley 02:29, 2 August 2011 (UTC)

Thanks, it will help. but still we will not be able to tell if this is say a translation from foreign language to English, or original English title. So maybe add another field? Qshadow 20:02, 31 July 2011 (UTC)

All translations will be turned into variant titles in the foreseeable future, we just need to add a few software enhancements before we can start doing that. Ahasuerus 21:17, 31 July 2011 (UTC)

Also if you are just looking for items written originally in English.... that's the largest subset of the database.... rather more like looking for 'hay' in the proverbial 'haystack'. Kevin 03:00, 2 August 2011 (UTC)