ISFDB:Moderator noticeboard/Archive 09

The discussions are archived from the ISFDB:Moderator noticeboard page, please do not alter them. Links to other archive pages can be found at ISFDB:Moderator noticeboard/Archives.

Fixer's progress

Fixer is getting better at identifying and submitting forthcoming books, so I am leaving a few dozen submissions in the queue for general munching. It's probably best to put Fixer submissions on hold before you do anything with them since some require additional investigation on the Amazon site(s).

P.S. I am a little ambivalent about the number of audio books that we are importing. Perhaps we could make "Include audio books" an option on the "Forthcoming books" page. We will also need to add a "New/Add" flag as well as Title level (Series and tag?) info to that page at some point. Ahasuerus 05:16, 30 November 2009 (UTC)

I have added a new section to Help:Screen:Moderator on Moderating Automated Submissions. Additions, clarifications, and corrections are more than welcome! Ahasuerus 00:24, 1 December 2009 (UTC)

I have made a few changes to Fixer's logic to put >50% of new audio books in a special "suspense" file so they won't be showing up in the submission queue quite as often. Still, there are a *lot* of them out there.

Please keep in mind that Amazon.com and Amazon UK are increasingly duplicating each other's data. In the past, you could tell which books should come from Amazon.com and which ones from Amazon UK by checking their publisher, but there are more and more publishers printing books on both sides of the ocean, which makes it harder to tell US books from UK books. So far Fixer has been submitting "fantasy" books from Amazon.com and I had to correct a number of pubs which should have come from Amazon UK. Next, I will ask Fixer to submit "fantasy" books from Amazon UK, which should result in fewer submissions since many ISBNs will have been processed by then. Oh well, we'll see how it goes... Ahasuerus 05:39, 1 December 2009 (UTC)

Between the Strokes of Night

As three moderators are involved: The cover of Between the Strokes of Night refers to first or second printing.--ErnestoVeg 17:35, 30 November 2009 (UTC)

Sometimes an editor will add a cover image from a previous printing if the cover art is the same. It appears the only difference in this case is the printed price, so it should be OK to leave this image on the record. MHHutchins 18:07, 30 November 2009 (UTC)

Locked Out of Submissions Page

I'm unable to access the submissions page. When I try to sign in it changes my first initial from upper case to lower, then returns a "Bad User Name" message.--Rkihara 00:25, 1 December 2009 (UTC)

I have experimented with logging in and out (also using Fixer) and everything looks OK. Anyone else having problems? Let me check the database settings for your account... Ahasuerus 00:36, 1 December 2009 (UTC)

The account looks fine as far as I can tell, but then I am no Wiki expert. Perhaps deleting all auto-complete/password/cookie/etc information for ISFDB may help? How about using another browser, e.g. Internet Explorer instead of Firefox or vice versa? Ahasuerus 01:19, 1 December 2009 (UTC)

I blew everything out as you suggested and now it's working again. Thanks! I should have thought of that 8>(.--Rkihara 02:45, 1 December 2009 (UTC)

Great! As a general observation, if one can sign in on the Wiki side, then any problems signing in on the ISFDB side are highly likely to be on the client side since the two sides share the same database information. Ahasuerus 02:55, 1 December 2009 (UTC)

Preventing accidental cross-approvals?

One things that we have run into over the the last few years is moderators accidentally approving other moderators' submissions. It's rarely a big deal, but it can be confusing (where did my submission go?").

Would it be useful to prevent moderators from being able to approve/reject other moderators' submissions the way the software currently prevents other moderators from approving Held submissions? The software change would be fairly trivial since all the heavy lifting was done a few weeks ago when Hold/Unhold was reworked. We can always "hard reject" a submission if the submitting moderator becomes unavailable. Ahasuerus 04:16, 2 December 2009 (UTC)

Better might be to highlight them somehow. That would let someone help another moderator and would also let submissions be approved when the moderator becomes unavailable. Assuming the worst affect of an approval is a bit of confusion, actively preventing it does not seem warranted. --MartyD 11:13, 2 December 2009 (UTC)

I agree with Marty on this. Just a highlighted warning should be good enough. MHHutchins 15:14, 2 December 2009 (UTC)

Please no auto-approve. --swfritter 17:16, 2 December 2009 (UTC)

If you are referring to automatic approval of submissions created by moderators, then fear not, it's not in the cards :) Ahasuerus 17:42, 2 December 2009 (UTC)

I end up rejecting a fair number of my own submissions. I know there are a couple of cases where I have accidentally approved not so much because of inattention but because of a misguided mouse.--swfritter 17:16, 2 December 2009 (UTC)

It's been my experience that I sometimes click on the wrong submission by mistake because the New Submissions screen is still repainting and the links are moving around. And once you are on the Submission Review screen, you immediately approve it since you "know" you are looking at your own submission. Highlighting/color-coding submissions in the New Submissions screen wouldn't help with this problem, but making changes to the Submission Review screen to make submissions created by other moderators stand out may. Ahasuerus 17:42, 2 December 2009 (UTC)

I think colour coding on the Review screen (like for verified publications) is best. When I finally get around to magazines, I would like some other moderator to be able to vet my first n attempts! --clarkmci / j_clark 02:12, 9 December 2009 (UTC)

Variants

Could someone else take a look at two submissions [here] and [here]. I think these should be done differently but don't know how. I also have a third one on hold that (I think) takes the variant the wrong way. I accepted two submissions for the same variant relationship before realizing this. Trying to branch out but I should have known better. ~Bill, --Bluesman 20:48, 4 December 2009 (UTC)

According to Contento and other sources, "The Bottomless Pool" was written either by Robert Bloch alone or by Bloch in collaboration with Ralph Milne Farley even though it was originally published as by Farley alone. The second submission seems to be the way to go. Ahasuerus 23:26, 4 December 2009 (UTC)

And the other one? I'll approve the second one. Thanks! ~Bill, --Bluesman 23:45, 4 December 2009 (UTC)

Since we selected the collaborative Title record to be our canonical Title (usually a good idea for ghost written stories where the respective contributions of the parties is unknown), we need to make the two other Titles (one published as by Farley and the other one published as by Bloch) into variant titles of the canonical one. You have already approved the submission that created a VT for the Farley title and I have created another VT submission for the Bloch one, which I will let sit in the queue for a bit. Once we have approved my submission, we can reject the original one.

Does letting the submission sit for a bit kind of age it, as in fine wine?¿? ;-) I still don't know what to do with the 'backwards' variant (Bloch/Fiske) [here]. ~Bill, --Bluesman 00:14, 5 December 2009 (UTC)

That one was entered, as you said, backwards, and there were three other backwards variants in the database. I have reversed them and set up the right VTs, so it's safe the reject the one linked above. Ahasuerus 01:38, 5 December 2009 (UTC)

Thanks! ~Bill, --Bluesman 02:01, 5 December 2009 (UTC)

I've don't understood, but it is not a problem. I've entered the data for the first publication of the story and then I've linked the reprint as variant (as the author was different).

We will also need to add a note to the canonical title explaining that the story first appeared as by Farley and was later reprinted as by Bloch and that we don't know how much Farley contributed. Ahasuerus 00:06, 5 December 2009 (UTC)

There are many situation to adjust in Strange Stories, but after this discussion, I don't dare to submit directly the data.

(unindent)
Strange Stories, August 1939 (cfr: Contento and Monthly Terrors)
53 • A Sentence of Death [by Norman A. Daniels] • John L. Benton
95 • The Curse of the Crocodile [with Henry Kuttner] • Bertram W. Williams • ss
Strange Stories, December 1939
46 • Vigil • Hampton Wells • ss; forecast as by Manly Wade Wellman
85 • The Grip of Death [with Henry Kuttner] • Robert Bloch • ss
101 • Passing of Eric Holm [by August Derleth] • Will Garth • ss
Strange Stories, August 1940
79 • The Dead Shall Rise Up [by August Derleth] • Will Garth • vi; forecast as by Tally Mason.
There are also four stories by Tally Mason to adjust.--ErnestoVeg 08:29, 5 December 2009 (UTC)

Magazine Note for Newbie Moderators

Some of the new moderators may not have had much experience with magazines. One of the most important things is to make sure that a link to the magazine publications be placed on the wiki page of the magazine. Otherwise there is no simple way to access the magazine and, even worse, it is possible that someone may enter that same issue because they do not have any way of knowing that it has already be entered. Of increasing importance is that the Editor records of magazines be merged by year and placed into series. I will try to come up with some instructions for accomplishing that task but for right now it is important that the approving Moderator update the wiki or make sure that the submitter has. If a moderator should feel particularly uncomfortable about a magazine submission it might be a good idea to either pass it by or put it on hold, notify the submitter, and place an entry on the Moderator noticeboard asking for advice. Magazines can be a bit vexing; there are a lot of problems associated with them that will not occur anywhere as often with books.--swfritter 00:45, 5 December 2009 (UTC)

As one of the aforementioned, how does one find, never mind update the wiki for a magazine? Start with the basics as this is all new to me. ~Bill, --Bluesman 01:05, 5 December 2009 (UTC)

Click the "Magazines" link: under the "ISFDB Wiki" link in the "Other Pages" section of the menu. BLongley 14:50, 5 December 2009 (UTC)

Here is some Help which will be modified to include instructions on magazine series.--swfritter 16:38, 5 December 2009 (UTC)

The above Help now has a rough draft for modifying a magazine series - with pictures (although a little grainier than I would wish).--swfritter 17:23, 5 December 2009 (UTC)

ISSNs on magazines

Dumb newbie mod question, for which I wasn't able to find a clear answer. In some recent submissions an editor supplied catalog numbers (ISSNs?) for WEIRDTFALL1989, WEIRDTFALL1990, WEIRDTSPRING1991, and WEIRDTSUM1991 among a large number of other changes to those pubs. So I let them through, but what is the proper way for those numbers to be handled? Should they be "#"ed? "ISSN: "ed (or variation thereof)? Left as-is? I see in the database examples of all three (and #ISSN, #ISSN:, and ISSN variations) approaches, all in significant volume, "#nnnn-nnnn" being least popular. Thanks. --MartyD 12:12, 5 December 2009 (UTC)

They don't need to be #ed as they're not long enough to trigger an invalid ISBN warning. I think the last discussion ended with the feeling that it's pointless having them on each publication, it might as well be left on the Magazine Wiki page. So I guess the answer is "it doesn't really matter". BLongley 14:53, 5 December 2009 (UTC)

It was when I asked the same question that it was suggested that they don't belong there since they don't identify a specific issue. You will note that the individual magazine wiki template now has a place for them.--swfritter 15:13, 5 December 2009 (UTC)

Unmerge for Deep Breathing Exercises

Hi, I made a submission to unmerge The Best of Omni Science Fiction 4 from this. My intention is to change the title for this particular appearance of the story to "Deep-Breathing Exercises" (consistent with both the TOC and story's title page in this publication) then merge it with the other appearances of the hyphenated version of the title (I was unable to edit this directly with the publication). Please let me know if there's a better route to accomplish the same goal. Thanks Jonschaper 13:14, 5 December 2009 (UTC)

I think you should remove the title, not unmerge it, and then just add the hyphenated version (and merge that, as you say). I don't really know what unmerge will do in this case, so I've left it for someone who knows more than I do. --MartyD 14:52, 5 December 2009 (UTC)

Thanks, I'll go the route you suggest Jonschaper 23:27, 5 December 2009 (UTC)

Aniara

Before I start editing this pub and it's title, I have a few questions. There's no other verifier, so there's nobody I can ask personally. In the introduction Tord Hall states: "The 102 Songs of Aniara form a symbolic poem about our own age". I.m.o. this makes it one poem, and not (as stated in the notes with the title) a collection of 102 poems. It also means, that the publications can't be collections. A rough word count comes to 25.000 to 30.000 words, which makes the poem the length of a novella, and logically the pubs should then be chapterbooks. Am I missing anything? Thanks, Willem H. 14:08, 5 December 2009 (UTC)

A little research suggests it's not always been one long poem, the first 29 "cantos" were published separately as "Song of Doris and Mima" in the collection "Cikada". And it's not always 102 poems: apparently there's 1 prologue poem and two epilogues around the 100 canonical entries. So I'd leave it as a collection, without wanting to enter all 100/102/103 individual entries. BLongley 00:06, 6 December 2009 (UTC)

Ok, I can live with that. The songs (or cantos) have no individual titles, so entering them is no option (would look ridiculous). I'll just add some notes. Thanks, Willem H. 11:05, 6 December 2009 (UTC)

ErnestoVeg appears with wrong titles on my watch list

Morning! My watchlist showed Ernesto as changing books I and Willem H had dealt with at other editor's sites. EV shows his own site correctly, but both those sites reflect the 'title entry' of the previous entry to his. Is this a bug or glitch? Thanks, Harry. --Dragoondelight 15:22, 6 December 2009 (UTC)

I am afraid I am not sure I fully understand the issue. We are talking about the "my watchlist" option displayed at the top of each Wiki page, right? What it displays depends on how each user set up the "Watchlist" tab of his Wiki Preferences. Perhaps you and Ernesto have them set up differently? If not, could you provide an example? Ahasuerus 16:18, 6 December 2009 (UTC)

I believe the problem is, that Ernesto doesn't use the alt-plus (or add new comment) option, but adds a heading and text to the last subject in a wiki page. I have noticed this before, when on the recent changes page it looked like he added a comment to an existing discussion. Willem H. 16:39, 6 December 2009 (UTC)

I've understood. Sorry.--ErnestoVeg 18:05, 6 December 2009 (UTC)

I wasn't criticising your actions, just clarifying the problem. Sometimes there are more ways than one to reach the same goal. Willem H. 18:43, 6 December 2009 (UTC)

Now I understand. Thanks, Harry. --Dragoondelight 20:35, 6 December 2009 (UTC)

How do I fix double-included title?

Would someone remind me how to fix this problem, where the title is included twice in THHRFGSBKT2009? (I imported contents from one of the other pubs and ended up getting the novel again). I tried Remove Title, but the submission wants to remove both of them. I know I've seen this before, but I just don't remember the path to the fix. Do I change the pub type and then muck with the titles? Thanks. --MartyD 02:52, 12 December 2009 (UTC)

Clone one of the copies and after its approved delete the original. This should leave you with one copy. This is one I came up with two years ago after trying numerous other solutions, why it works and how these dups are created is beyond me :-)Kraang 03:06, 12 December 2009 (UTC)

That did it. Thanks. --MartyD 03:18, 12 December 2009 (UTC)

It works because the cloning process fails to find the second Title -- see Bug 2873841, "Clone pub doesn't clone multiple container titles". It's a bug which just happens to be a workaround for another bug :-\ Ahasuerus 04:43, 12 December 2009 (UTC)

A good reason not to allow importing of contents from a novel-type pub record. It should be restricted to collections, anthologies, etc. Or at least there should be a warning when attempted. Thank goodness that the second bug makes it easier to fix the first one (which really isn't a bug, because it does exactly what you want it do: import ALL of the contents, including the novel.) The same thing will happen if you import a title that's already in the pub you're adding titles to. MHHutchins 05:12, 12 December 2009 (UTC)

Nature's way of telling me to go look at that feature request for importing specific titles, eh? :-) --MartyD 10:49, 12 December 2009 (UTC)

Just fixing Import so that it doesn't add titles already present would be good. I've got bitten by that at least once today where I started with a collection that had a known introduction, found the contents and added those, then imported the TP contents to the HC. Had to delete the introduction from the HC (which removed both) and re-add it, then remerge. Clone seems like a nice workaround, will try it in future. BLongley 22:53, 12 December 2009 (UTC)

Whilst I'm thinking about it - does anyone use "Export"? I don't think I ever have. BLongley 22:53, 12 December 2009 (UTC)

I don't think I have ever used it, but I have seen editors make references to it. Ahasuerus 01:17, 13 December 2009 (UTC)

(unindent)Let me see if I can fix Remove Titles to remove one title only... Ahasuerus 01:28, 13 December 2009 (UTC)

Bluetyson ebook submissions

Most of them seem to be available at Fictionwise.--swfritter 16:54, 13 December 2009 (UTC)

Do we need to do anything special with them, beyond checking the ISBNs? Now that it's too late for me to ask, at least for some of them.... --MartyD 17:11, 13 December 2009 (UTC)

Keep checking the comments I have made on his page. They will probably apply to other entries. Most of the information you can check is at Fictionwise.--swfritter 17:22, 13 December 2009 (UTC)

I would not make too many changes to the submissions. These entries may result in some standards discussions.--swfritter 17:38, 13 December 2009 (UTC)

You will note I have put a number of submissions on hold. No major problems but a few things need to be resolved.--swfritter 18:27, 13 December 2009 (UTC)

I will try to be careful. If you'd rather I left them alone, just say so. The size of the list seemed to cry out for pitching in.... --MartyD 18:35, 13 December 2009 (UTC)

There isn't anything major here. The novels, with the exception of those that had extra content, were all fine. The major issue is page numbers and ISBN's/Formats. We are likely to get a growing number of ebook submissions and we are all going to have to get used to some of the idiosyncrasies. Once we hear from Bluetyson‎ a lot of things will be clarified.--swfritter 21:30, 13 December 2009 (UTC)

broken author-title link / false empty pseudonym bibliography?

Would someone explain why this title but this empty author arises, and the appropriate technique for fixing it? Thanks. --MartyD 15:20, 21 December 2009 (UTC)

Click Show All Titles for Patricia D. Cacek who needs to be made a pseudonym of P. D. Cacek at the author level. Since the title is pseudonymous it will only show up on the canoncial author's page.--swfritter 16:16, 21 December 2009 (UTC)

Import Contents and Container Titles

I imported contents into this nonfiction pub from a later edition. Somehow, I've ended up with two container title records. I've used the import contents tool in the past without indecent, and am wondering if the issue arose due to this being a nonfiction title. --Ron ~ Rtrace^Talk 14:28, 22 December 2009 (UTC)

That's entirely possible. I plan to make some page-related changes to the Import script later this week and may poke around a bit. And yes, it looks like the same title record is listed twice. If you view the main Title page, you will note that the first two pub links lead to the same page.Ahasuerus 15:26, 22 December 2009 (UTC)

Also, I'm unsure how to proceed to fix it. The obvious approach would be to use the remove titles from pub tool to delete one of the container titles. My worry (perhaps unfounded) is that the software may not be able to distinguish between the twice listed container title, and that it may delete both, leaving me with no container titles. Is there a preferred technique to cleaning up this sort of issue? Thanks. --Ron ~ Rtrace^Talk 14:28, 22 December 2009 (UTC)

When a Title record is listed twice in a pub, "Remove Title" removes both instances of the record. Once the container title has been removed, you will need to re-add it manually. It's on my list of things to address, hopefully later this week once they finish rewiring some parts of my wetware. (Note to self: stop this "getting older" business before it gets out of hand!) Ahasuerus 15:26, 22 December 2009 (UTC)

Thanks. It worked like a champ. Hope things go well (with the rewiring). --Ron ~ Rtrace^Talk 03:04, 23 December 2009 (UTC)

2009-12-24 submissions

Fixer has been slowly improving in various areas and today I let him out to play. He was understandably anxious to exercise his growing powers and he did generally OK. He still can't tell "Tor Books" from "Tor UK", but that's really Amazon's fault, so if you see a "Tor Books" submission with an empty (or irregular) price field, make sure to check Amazon UK.

There are still things that I would like to teach Fixer, but his current state is sufficient for most practical purposes, including keeping Forthcoming Books updated, so I will set him aside for now -- there are plenty of submissions in the queue to keep the most eager moderator busy. I'll see what I can do about the core software over the holidays now that I don't have "things" sticking out of various parts of my wet/hardware. Ahasuerus 07:34, 24 December 2009 (UTC)

I haven't been following Fixer's progress closely --clarkmci / j_clark 11:35, 24 December 2009 (UTC)

The thorniest issue that Fixer was facing earlier this year was determining the right Amazon location to get the raw data from. We didn't want UK pubs to be created with US prices and vice versa and there was no easy way of telling which country's data should be used for a random pub. (Australia and Canada are separate cases, but I have something in the works for Australia and perhaps for Canada.) When Dissembler was stopped in his tracks due to the new authentication requirements, I had to do most of this stuff manually, helping Fixer along every step of the way.

Once the immediate issue was addressed and we shifted to a catch-up mode of sorts, I went back and created a few publisher lists. First, we have a list of "genre publishers based in the US" and a separate list of "genre publishers based in the UK". Fixer sucks in pretty much everything that these folks publish, but I still have to do manual pre-screening of submissions since some publishers like Forge publish books of genre interest even if the books themselves may not be SF. Unfortunately, there is some US/UK overlap, e.g. Tor publishes books in both countries and Black Library, Solaris and Orbit have some overlap, but it works 95%+ of the time and the rest can be adjusted by moderators if they see strange prices like $11.32. I am afraid that there will be more overlap as we go forward :-(

Then we have two lists for "non-genre publishers based in the each country that are known to publish SF". Fixer queries them for "genre books only" where "genre books" are defined as having certain keywords like "fantasy" and "paranormal romance" associated with them. Fixer knows how to ask Amazon UK to send him a list of all books which, say, Faber published last month and that had the word "vampire" anywhere in the title, subject or browse nodes. By running these queries against all "known US-based" and "known UK-based" publishers, Fixer can get a vast majority of the eligible data from the right source.

Finally, there are SF books published by "publishers unknown to Fixer". These will be identified by simply scanning both stores for known SF keywords and there is no guarantee that the data will come from the right store and the price will be right. Hopefully, this will account for just a few percent of the total and, besides, Fixer will be constantly adding to his list of "known publishers".

The next step is to automate certain Fixer activities so that he could do a better job of auto-rejecting comics/manga without human intervention, but that's not as simple as it sounds since there are quite a few legitimate novels based on comic book characters like X-Men. And so it goes :) Ahasuerus 06:20, 25 December 2009 (UTC)

but a couple I looked at today/yesterday were set to be novels though the keyword "anthology" was in the MOD NOTES (eg Hartwell & Cramer Year's Best Fantasy 9). Is it possible to detect such a keyword & set the type accordingly, or is Amazon not consistent re collections vs anthologies & it'd be likely to give erratic results? Just wondering. --clarkmci / j_clark 11:35, 24 December 2009 (UTC)

Unfortunately, Amazon uses the keywords "Anthology" and "Anthologies" for anthologies and single author collections alike. Still, Novel is almost invariably (although not 100%) wrong, so I have asked Fixer to change the type to ANTHOLOGY when one of these keywords is present in the list of browse nodes. Good point, thanks! Ahasuerus 06:20, 25 December 2009 (UTC)

I think this is a mistake. When it's a new NOVEL that should be an ANTHOLOGY or COLLECTION, it can be corrected in one edit. When Fixer makes it an ANTHOLOGY that should be a COLLECTION or a NOVEL, it takes two edits to fix - publication record and title record. Either going back to NOVEL as a default or making single appearances of an ANTHOLOGY or COLLECTION fixable in one go would be better. Unless Fixer can get them far more often right than wrong - it needs to be more than 50% right to make it less work for us. BLongley 01:18, 28 December 2009 (UTC)

Hm, that's a good point, although I suspect that there are more Anthologies than single author Collections at Amazon. I haven't been keeping track of the success rate, has anybody? Ahasuerus 02:39, 28 December 2009 (UTC)

(Later) Also, I noticed an item that was as by L. E. Modesitt Jr. in Amazon: Fixer seems to get the title record to put Modesitt, Jr., but not the comma in the publication record, creating a new author with a "Stray" publication. Gail Z Martin (i.e. no period after the Z) I suspect will do the same (but it's late here, so I'll leave it for you - it's on hold - [1]). Both books matched an existing title with the correct author form so did/will do an Automerge. Just wondering if you had spotted this. --clarkmci / j_clark 13:07, 24 December 2009 (UTC)

That's a good point as well. Fixer performs a look-up against the last backup to see whether the new pub should be auto-merged with an existing Title. During the look-up, Fixer makes allowances for garbled punctuation and capitalization, so "gail,z martin" would match "Gail Z. Martin". If Fixer finds one (and only one) eligible Title, he modifies the submission to "auto-merge" with the found Title and changes the Publication type to match the Title type. What Fixer wasn't doing was changing the spelling of the authors' names in the pub to match the spelling of the authors' name in the pre-existing Title. Change made, thanks! Ahasuerus 06:20, 25 December 2009 (UTC)

Thanks for the background. I'm certainly interested in what Fixer can do for Australian publications. Please give me a "heads up" when it's ready. --clarkmci / j_clark 22:43, 26 December 2009 (UTC)

Will do! There is no "Amazon.au", but there are other Oz-based stores that specialize in SF and that Fixer can profitably munch on :) Ahasuerus 22:55, 27 December 2009 (UTC)

Forthcoming Books -- First Quarter of 2010 Done!

Fixer and I are happy to report that 2010-01, 2010-02 and 2010-03 have been processed and submitted. Fixer is running a cleanup pass, but the bulk of the data is in. I will be doing weekly passes to pick up any stragglers, but otherwise we should be done until late January, when Fixer will do April 2010. Many thanks to all approving moderators, it's been quite an adventure!

I will start running Fixer for earlier periods soon, but the runs should be fairly short. It would be obviously advantageous to start at the beginning of time to make sure that first editions are imported first, but it's not clear at what point in time Amazon's data becomes so bad as to be effectively unusable. Oh well, we'll experiment and see what happens! :) Ahasuerus 22:51, 27 December 2009 (UTC)

There's been quite a few juveniles in the recent batches. I wonder if it would be easy/make sense to ask Fixer to automatically set books less than 100pp to be CHAPTERBOOKs rather than NOVELs?

I'm not keen on that. Maybe 64 pages or less. 100 pages IS a NOVEL to a kid - and if we have to rework all YA or younger titles according to length rules, we've got a massive task ahead and one I'd not be joining. When there's an award nomination in a category that demands a length, fine, figure out if it's a NOVELLA or NOVELETTE or whatever, and create a CHAPTERBOOK entry when needed - but I didn't help re-enable CHAPTERBOOK support to recategorise every "short novel" we have. BLongley 01:49, 28 December 2009 (UTC)

I believe Bill has a point as an awful lot of juvenile books are under 40K words/100 pages. The main purpose behind re-adding CHAPTERBOOK Titles was to emphasize short works which have also appeared as standalone books so that they wouldn't drown in the Shortfiction section. That's all good and well in the adult world, but, when dealing with juveniles, what are the chances that these works also appear in magazines/anthologies as "Shortfiction"? If we were to create CHAPTERBOOK records for all of them, I suspect we would end up with a lot of Chapterbook/Shortfiction pairs which would only confuse matters. Ahasuerus 02:36, 28 December 2009 (UTC)

Request for Moderator Thoughts/Opinion

The following link is a listing for the Pulphouse Hardback Magazine publication.

http://www.isfdb.org/cgi-bin/title.cgi?35504

Currently this link has 4 HC editions listed for each of the seasons. I was looking into adding another variant to one of these entries (The deluxe leather boxed edition for Spring 1990).

My question is,

Is this entry containing 4 different volumes correct or should it have been entered as 4 separate items.

While all were published in the same year they are distinct entities all with different fiction.

Thanks,

-Devin

Magazines use somewhat peculiar/arcane rules and you have run into one of them.

Non-Magazine publications are supposed to have one (and only one) Title record of a matching type, e.g. a Novel pub contains one Novel Title, a Collection pub contains one Collection Title, etc. Magazine publications, on the other hand, do not have Magazine Titles; instead they have EDITOR Titles.

EDITOR Titles are organized in Series and it is these Series that enable us to go from a single issue to the whole series, e.g. when viewing Pulphouse: The Hardback Magazine Issue 9: Fall 1990, you can click on "View All Issues" and see the whole series, which can then lead you to individual issues.

If we had an EDITOR Title per every Magazine pub, we would end up with hundreds of Editor records per Series for Analog, Amazing and other long running magazines. That would be a bad thing since it would take a long tome to create the page and it would be so long as to be overwhelming and confusing. For this reason, we merge EDITOR records for magazines that appeared during the same year and had the same editor(s). That way there is only one line per each year/editor combination, which makes Series displays much tidier.

The downside to this arrangement is that Magazine pubs are not designed to handle multiple printings, which is apparently what we have in this case. In the past, when we ran into this problem (e.g. with Destines), we changed the publication types from Magazines to Anthologies since the latter let you enter multiple printings. I'll wait for other moderators to chime in with suggestions re: the best way to handle this case. Ahasuerus 05:02, 1 January 2010 (UTC)

I see. Well then the question I guess I should ask is if this should be considered a magazine or an anthology.

To my mind it seems to far more resemble and Anthology rather then a Magazine.

If the main reason for doing magazines grouped into years then this should like be reclassified. There are only a limited number of years of publication 1988 through 1992 or something like that and there are 4 Hardback editions each year. These editions don’t have advertising or anything like you would see in a normal magazine and seem to average maybe 20 to 30 stories each from different authors.

Thanks for the explanation.

Dmatlock 12:55, 1 January 2010 (UTC)

Anthology seems to make the most sense to me also. Any other idea I can think of gets too complicated. Most users would probably not perceive the volumes as magazines. Links to the title records could still be placed on the magazine wiki page with an explanation.--swfritter 14:36, 1 January 2010 (UTC)

The number of issues doesn't decide that it be typed one way or the other. There are magazine editor records for titles that only had one issue. I suspect this title was first entered as a magazine because it called itself a magazine. When it can be one way or the other, we usually let the publisher's description or mission statement tip the balance. Note how Postscripts finally gave up calling itself a magazine with issue #18. That's how it's reflected in the ISFDB records. Another editor has been working on Tales of the Unanticipated which started off as a magazine, but along the way became an "anthozine" or "magthology" (ugh!) I don't think the lack of advertising in Pulphouse makes it less a magazine (it did have a regular column), but if you feel strong enough please proceed to convert the records. (I don't feel strong enough either way, and I'm the only one who has verified one of the issues.) You'll have to first unmerge all of the editor records (you note how they were combined as magazine editor records should be). Then change the type of each issue from magazine to anthology. Might be a few more steps as well, but I'll lead you along the way if you're not familiar with the procedure. MHHutchins 04:20, 2 January 2010 (UTC)

Are there more deluxe leather editions than the one you mention? Any other editions? Is it worth the work for if there are not many other editions? Another option would be to leave the existing pubs as magazines and enter any other editions as reprint anthologies. I eventually plan to do the same with the Astounding full-issue reprints from Project Gutenberg.--swfritter 14:33, 2 January 2010 (UTC)

I don't think the fact that there was a variant edition of each issue of Pulphouse determine if it should be considered a magazine or anthology. Most issues of Postscripts had a variant edition (a few had two variants), and I was able to enter them as a different record. (See here). It does take some extra steps to clone the original record (you have to change the type to "anthology" first, and do some manipulation with the editor record), or you could easily enter your leather-bound edition as a "New Magazine" (or "New Anthology", if you prefer), and then merge the content records with the existing records. Either way, just let me know and I can help you through it. MHHutchins 16:49, 2 January 2010 (UTC)

Suspected Error on Verified Entry

I was looking at Requiem a tribute anthology to Robert Heinlein found at this link,

http://www.isfdb.org/cgi-bin/pl.cgi?RQMNWCLLCT1992

I believe there is an error with the 2nd content entry called Editor’s Forword (Requiem) on page 6. As the title suggest this is written by Yoji Kondo (aka Eric Kotani).

I have a primary reference for this copy and the single page forward is indicated to be Yoji Kondon but I didn’t want to edit something that has been verified.

What is the best way to proceed on this?

The best way is to click on the link of the editor which verified it, which leads you to his user page. From there click on "Discuss this page" (or the "Talk" tab, depending on which ISFDB skin you're using.) Then leave a message with your questions or concerns. Here's a link to the FAQ about verification which suggests waiting a week to hear back from the verifier before posting a request on the moderator noticeboard. MHHutchins 04:43, 2 January 2010 (UTC)

I also have one additional question on the same point. There is no publication date for this content entry. However, would it be safe to assume as it is an editor’s forward for this publication that the original date should be the same as the publication year of the book?

Actually it is dated. Any content record which doesn't display a date indicates that it's the same date as the publication which contains it. If you click on the content record you'll see the year of publication. If you edit the record, you'll see the month of publication. MHHutchins 04:43, 2 January 2010 (UTC)

Addition,

http://www.isfdb.org/cgi-bin/pl.cgi?CRCMPLRQDT1984

This is also a verified publication. My copy has the last numbered page as 319 but the verified reference has 320. How should this be reconciled? Thanks, Dmatlock 17:52, 1 January 2010 (UTC)

Same way as above. Notify the editor who has taken on the responsibility as a primary verifier of the pub. In both cases, these are pubs that I verified, so a note on my talk page would have sufficed. As did the note on this, the moderator's page. I've corrected both books (with an explanation of the page count of the latter). Thanks for finding the errors. MHHutchins 04:43, 2 January 2010 (UTC)

title / author pairing fix-up

I noticed while reviewing an edit having nothing to do with the relationship that some pubs under The Interpreter are credited to "Brian Aldiss" instead of to the title's "Brian W. Aldiss" (the former already a pseudonym of the latter). Is there any reason NOT to make a variant credited to "Brian Aldiss" and move the corresponding pubs to that title? --MartyD 11:22, 9 January 2010 (UTC)

It so happens that there is already a variant credited to "Brian Aldiss", we just need to move a couple of incorrectly entered pubs under that Title record. Also, it looks like this pub and this pub are the same book attributed to two different forms of Aldiss's name. Both are verified, but the verifiers are mostly inactive at the moment. Ahasuerus 22:07, 9 January 2010 (UTC)

More generally, we need to write a script that would find all pubs whose authors do not match the authors of the "reference" (or "main") Title. I am also beginning to think that we need a way to enforce consistency between pubs and their reference Titles in the software, but that could mean a fairly major rewrite. Ahasuerus 22:11, 9 January 2010 (UTC)

"Brithish Perry Rhodan" new series?

I'm unfamiliar with the Perry Rhodan series, so I put it on hold for one a moderator that knows more about it. Once someone leaves a note here I'll remove the hold. I've left a note on the submitters page. Kraang.Kraang 00:44, 12 January 2010 (UTC)

Changes to Moderator screens

Moderator screens should no longer display Notes and Synopsis fields as changed when:

they were not changed, and
there is an embedded carriage return in the text

The change to the underlying code was rather sweeping. If you run into any issues with carriage returns, please report them here. TIA! Ahasuerus 03:16, 18 January 2010 (UTC)

Curiously, the problem was on submission (which is what decides which fields have changed), so for anyone who's a non-moderator seeing this, if you notice a submission's XML containing a multi-line field that was not edited, let me know. --MartyD 11:30, 18 January 2010 (UTC)

Playground of the Mind Froms?/Excerpts?/Dates?

Morning everyone! I submitted a plethora of changes that need to be compared with the two previously verified versions. I do not know if it would be easier to make comparisons on the moderator screen or not, but I am making this so the other moderators know I anticipate some discussion. The Verifier moderators have been notified. Thanks, Harry. --Dragoondelight 15:32, 22 January 2010 (UTC)

Babylon 5 Script Books

Starting in 2005, J. Michael Straczynski started releasing books with the scripts to the TV Show Babylon 5. Would these be appropriate to add to the database? --WeAreGrey

Are the same as this. [2]? Thanks, Harry. --Dragoondelight 16:26, 30 January 2010 (UTC)

No, those books are episode guides. These contain the actual scripts. --WeAreGrey 16:40, 30 January 2010 (UTC)

Submit them. I'm pretty sure they're OK for the database, as there are other teleplays and screenplays in the database. Mhhutchins 17:37, 30 January 2010 (UTC)

We have to deal with scripts as there are books with both the short story and their screenplay versions. Or because an author has achieved a "certain level" of notoriety. Or it's because it's the only way we get to record a classic like Quatermass. BLongley 23:12, 30 January 2010 (UTC)

Coincidentally, Jane Killick doesn't seem to have made much money from her episode guides and now works as a local radio presenter in my area. Not even talking about SF. BLongley 23:12, 30 January 2010 (UTC)

Dome

I did some editing with this pub, and I'm not sure how to proceed. It was listed as by J. Michael Reaves (pseudonym) and Steve Perry, but the book is by Michael Reaves and Steve Perry. After my unmerging, edits and remerging it now has two entries for the novel, the right one and the wrong one. The J. Michael Reaves and Steve Perry novel should be removed, but is listed as the container title. Can this be deleted or does that have consequences? Willem H. 16:25, 7 February 2010 (UTC)

You can remove it safely. The display is not as smart as it is pretending to be. The "Container" is based on the NOVEL pub to NOVEL title match, and there can be only one container. If you remove that one, the same logic will be applied the next time, and you'll see that what was once not the container title is now the container title. --MartyD 17:20, 7 February 2010 (UTC)

Thanks. I wasn't sure about this. Submitted the removal. Willem H. 17:56, 7 February 2010 (UTC)

Date precision

Dumb question. --MartyD 13:37, 23 February 2010 (UTC)

Not as dumb as it may appear! :) Ahasuerus 16:31, 23 February 2010 (UTC)

See this submission. If we actually have a publishing DAY, do we prefer to keep that level of precision in the title dates, or do we want to ~~round~~ truncate to the month? --MartyD 13:37, 23 February 2010 (UTC)

Well, it depends. If the day of publication comes from Amazon.com, then we generally keep it until the book can be physically verified (which also helps our Forthcoming Books logic) and then overwrite it with what is stated in the book, which is typically just the month of publication. However, if the day of publication is explicitly stated by the publisher, e.g. if the story was originally published in a weekly magazine, then we use the day. In this case (a self-published anthology from Lulu.com) the publication date was apparently provided by the publisher, so I think we want to use the day of publication.

If you go back in time, there are other, more complicated cases, e.g. publication dates reconstructed based on publisher archives, but that's a different can of worms. Ahasuerus 16:31, 23 February 2010 (UTC)

Thanks. That followed my inclination, so I'm happy. :-) I preserved the day in the date. --MartyD 12:09, 24 February 2010 (UTC)

Artist edits in Pubs with Multiple Artists

Just so you are all aware:

If someone edits an existing pub having a multi-artist credit via a jointly-credited coverart title (i.e., the artist credit is displayed as "Ima Painter and Ivan Artist" rather than "Ima Painter , Ivan Artist" -- also in each artist's bibliography the title will show up as "with" the other artist) and changes one artist's name, the original name is removed from the joint credit, and the new name is added as a separate individual credit. The only way to fix a name on a jointly-credited coverart title and leave it jointly credited is to edit that coverart title via the Edit Title screen instead of the Edit Pub screen.

You cannot tell from the moderator screens that this is going to happen, you just have to know it.

To repair a case where a single joint credit was broken into multiple individual credits, edit the pub and remove all but one artist (best to check the artist bibliographies to see if any one of them happens to be shared among multiple pubs and, if so, keep that artist). Then edit the coverart title credited to the remaining artist, and add back the others you removed. This will produce jointly-credited coverart.

For background, see any/all of:

I have started a design discussion at ISFDB:Proposed_Design_Changes#Artist_and_Coverart_handling_in_Pub_Screen, for those of you who are interested. Thanks. --MartyD 12:59, 28 February 2010 (UTC)

Thanks for the advice on how to bring these records back into shape, i.e. one record with two artists credited. The method you describe on fixing it works pretty good. It gets tedious if there are several pub records with the same cover art. One of the artists' credit has to be removed from every pub, but once that's done, just editing the single cover art record adds the omitted artist back to every pub. Mhhutchins 17:52, 28 February 2010 (UTC)

One more warning about this. If someone tries to merge a jointly-credited coverart title with one of the coverart titles from a pub with the same artwork but where the artists have individual coverart titles (e.g., merging a title by Leo Dillon+Diane Dillon with a title by Leo Dillon, dropping the "Leo Dillon"), this will be ok, BUT... Before accepting the submission, edit the pub(s) on which the separate credits appear and remove the other artists (e.g., in my example, remove Diane Dillon). --MartyD 01:38, 5 March 2010 (UTC)

Forthcoming Books - April-June 2010

It's almost time to update Forthcoming Books for April-June. Unfortunately, my ISFDB time was very limited in January-February and March looks even worse, so I won't get a chance to implement the Fixer enhancements that were planned. This means that the next Fixer run (hopefully starting this weekend) will do a poor job of distinguishing between US and UK editions put out by obscure or "trans-Atlantic" publishers. Please be on the lookout for non-standard prices like $9.23/£4.42 and other indications that the book was originally published on the other side of the Atlantic. Ahasuerus 15:38, 6 March 2010 (UTC)

Please note that some cover images provided by Amazon.com may be "canned". For example, see The Shape of Fear, and Other Ghostly Tales submission. It may be safer to remove these images until we get more reliable scans. Ahasuerus 17:44, 7 March 2010 (UTC)

From the note on Amazon about pages being blurred or having markings from the original or the scanning process they might actually be using a generic cover with title text for all the books they publish. Dana Carson 04:02, 8 March 2010 (UTC)

Aha! More evidence of this nefarious conspiracy! Ahasuerus 04:27, 8 March 2010 (UTC)

I did notice they (the ones for Nabu Press) do vary from pub to pub, though, so I had kept them on the ones I did. Oh well.... --MartyD 12:31, 8 March 2010 (UTC)

(unindent) OK, the big boys have been processed. Here comes the hard part: small presses, juveniles, vanity publishers, paranormal romances, criticism, etc. Fixer started by submitted everything for March 2010 (cleanup from the last run) and there are quite a few submissions to process. Keep in mind that if there is no price or if the price is unusual, chances are that the other Amazon store has the right record. Ahasuerus 02:27, 10 March 2010 (UTC)

The books from General Books LLC are mostly 19th century, early 20th century public domain many of which can also be found on Google Books. These are also Print On Demand titles from General Books.--swfritter 15:14, 10 March 2010 (UTC)

The cover images may possibly be scans of the original editions? Note the statement for the Nabu press title - "This is a reproduction of a book published before 1923. This book may have occasional imperfections such as missing or blurred pages, poor pictures, errant marks, etc. that were either part of the original artifact, or were introduced by the scanning process." Unlike Project Gutenberg, Google Books and many of the others who are releasing these scanned books are making a minimal effort to clean them up when converting them to ebook or paper - so don't be surprised if there are pages missing or other significant problems. Basically a ripoff to charge money for these books without cleaning them up. At least the ebook editions are free at Google Books and Barnes & Noble.--swfritter 15:24, 10 March 2010 (UTC)

Just a heads up that I will be mostly on the road over the next 9-10 days and April is almost upon us, so I plan to convince Fixer to upload the rest of April over the next 24 hours. The queue will likely get long for a bit since there are about 700 outstanding entries in Fixer's database. Only about 15-20% of his internal records end up as ISFDB submissions, though, so no need to panic (yet!). Come September, the busiest month for publishers, then we will panic :-) Ahasuerus 04:25, 16 March 2010 (UTC)

Nihilesthete

I don't read horror in any shape or form, including the daily news, so really have no idea whether to accept [this] new publication or not. [Here] is the blurb that I read. Any input would be appreciated. The pub is currently in a holding pattern... Thanks! ~Bill, --Bluesman 17:55, 7 March 2010 (UTC)

Probably just under the wire for slipstream (which includes almost anything) and/or dark horror which often does not have explicit fantastic elements but which is often lumped with fantastic stories (Poe for instance). "Equipped with all sorts of prosthetic devices" perhaps qualifies it from the S-F hardware aspect. Depends upon whether the devices are currently available.--swfritter 01:56, 8 March 2010 (UTC)

Betascript Buffy publications

I need some (or perhaps a lot of) help with these Fixer-originated entries:

Under current rules, are they the kind of thing that should be in at all? If they are in, is the treatment of putting them in the "Buffy the Vampire Slayer" series correct, or should I make some sort of sub-series for them? Note: There must be something wrong with their ISBNs. Whether the ISBN-13 or ISBN-10 from Amazon is entered, our display renders the ISBN-13 with an extra 978 tacked onto the front. Maybe this is trying to tell me these aren't actually books? Any guidance appreciated. --MartyD 11:18, 13 March 2010 (UTC)

I believe at one point we talked about including/excluding scripts on the Rules pages, but I am afraid I don't recall the outcome :( Also, we have pubs for scripts attributed to Buffy the Vampire Slayer and Pulse (sic!), which will presumably need to be cleaned up based on whatever we decide to do with these three. Ahasuerus 18:59, 13 March 2010 (UTC)

As far as ISBNs go, all English language ISBNs should start with a "0" or a "1" while these ISBNs do not. It looks like they are Amazon's "ASINs" (Amazon Standard Identification Numbers) rather than regular ISBNs. Ahasuerus 19:03, 13 March 2010 (UTC)

The Abode of Life

The Abode of Life In the notes for the fourth printing, it states that the price was changed by someone to a new higher price. I have a copy of the fourth printing with the original price still intact. Should I submit a change to have the price on the current entry changed, or should I submit a new copy of the fourth printing with the original price? Thank You WeAreGray 13:55, 28 March 2010 (UTC)

Yes, submit your copy with the original price. You can clone this entry and change the price and remove the note about the blacked out original price. The reason for the blacked out original price was for the Canadian market which sold editions at a higher price. Some times they would print a different cover with a white maple leaf and the price in Canadian dollars and other times they would use this method. Eventually dual priced covers discontinued this practice.Kraang 00:23, 29 March 2010 (UTC)

The throne of Saturn

I have this book by Allen Drury (present in the ISFDB for a short story) in my collection of SF books, see here. It's an kind of space race story (here to Mars) and I was wondering if it could be in the scope of the project. Hauck 17:57, 31 March 2010 (UTC)

It looks IN to me - even a "Space Race" story between here and the moon would be Speculative enough for me. Maybe a Space Race like the Ansari X-Prize would be out as non-speculative, but probably in on "NONFICTION of interest" grounds. BLongley 19:11, 31 March 2010 (UTC)

I am surprised that it wasn't already in the database.--swfritter 12:52, 1 April 2010 (UTC)

Done. Hervé Hauck 16:13, 1 April 2010 (UTC)

Fixer and the country of origin

Just a note that due to some changes in Fixer's internal logic, a higher than usual percent of May submissions will come from the wrong country. They are usually easy to spot, e.g. an Ace book priced at £8.33. Ahasuerus 04:49, 5 April 2010 (UTC)

Robert Hunts

Does anyone know anything about Robert Hunt (artist) vs. Robert Hunt? I'm wondering if the covers under Robert Hunt are mis-credited and should be credited to "(artist)". I will do some research when I get a chance, but I figured I'd ask. --MartyD 10:49, 5 April 2010 (UTC)

The artist's website confirms some of the covers. I don't know if the reviewer and/or essayist is the same person though. BLongley 18:36, 5 April 2010 (UTC)

John W. Campbell: An Australian Tribute

Well this is still sitting there. Anyone willing to say what they think is wrong. I know I separated out the untitled essays, but the way they were entered makes it seem like they were part of each other, instead of being written in tribute to Campbell separately. Entering them as a series, which Ernesto used as a title, allows them to retain their impact. Actually, since there are only 300 copies that we know of I think some people would be startled at what was said. One compares Campbell to Stalin. Thanks, Harry. --Dragoondelight 15:22, 8 April 2010 (UTC)

The only thing I was concerned at was "Not Separately Titled". I'm sure there's a better way, or an official way, to note such entries, I just couldn't recall what, so left it. BLongley

Sorry, Harry, I didn't understand the submission, and I don't know if some of what you've proposed is officially ok or not, so I have left it for a more experienced moderator to handle. --MartyD 09:53, 9 April 2010 (UTC)

When someone dies and Locus prints submitted tributes under one banner title, I've titled each of the untitled pieces the same as the banner title. In this case I would title each of the essays as "John W. Campbell: An Australian Tribute". I'll accept the submission, Harry, if you agree that the titles of the untitled essays be changed, even if to "untitled (John W. Campbell: An Australian Tribute)" which is probably closer to standards. Mhhutchins 13:17, 9 April 2010 (UTC)

No problem there. "untitled (John W. Campbell: An Australian Tribute)". Thanks, Harry. --Dragoondelight 14:47, 9 April 2010 (UTC)

Another thing: I wouldn't type the discussion panel as an INTERVIEW, but as a multi-author ESSAY. If it's left as an interview, it would be displayed twice on the 12 (!) participating author's pages. Even once as a essay would be quite overbearing. Thanks. Mhhutchins 13:22, 9 April 2010 (UTC)

No problem. Not interview then. Can you kill that part and I will redo it as a multi essay. Would hate to run into multiple blanks, if it slides through. Would "John W. Campbell, A Symposium (Discussion)" work as a multi-essay title header? That would hopefully cause a user to reflect back to the beginning of that segment and see the names posted in full there, thus I need not create chairman pseudonym? Thanks, Harry. --Dragoondelight 14:47, 9 April 2010 (UTC)

Anything else? Personally this book is only good for people who are doing an in depth study of Campbell and are looking for peer evaluations. Most SF readers would find it dull. I could not stand the multi essay first part, because there was no interaction, yet I have no problem with multi-essay for a discussion. Thanks, Harry. --Dragoondelight 14:47, 9 April 2010 (UTC)

Unable to pick and choose when it comes to edits within a single update. So I'll approve it because it would be easier for you to make corrections than to reenter all of the good stuff. Mhhutchins 19:34, 9 April 2010 (UTC)

I went ahead and made the changes discussed. Please look it over and see if it looks OK. Thanks. Mhhutchins 19:44, 9 April 2010 (UTC)

Most Excellent. Sorry, for causing you all the trouble. Now it has definition. Sorry, for asking but I waited so long that I forgot exactly what I had done and some took some thought. Rethinking is hard. Thanks Greatly, Harry. --Dragoondelight 12:06, 10 April 2010 (UTC)

Tom Swift cleanup?

I was running various internal consistency checks earlier tonight and realized that our Tom Swift lists are badly out of whack. Tom Swift 1980s, His Tom Swift and Tom Swift (2) should all be merged into one series and duplicates eliminated. Numerous duplicates in Tom Swift (3) also need to be merged. I tried to clean up a few, but it takes a fair amount of time due to ubiquitous pseudonyms and VTs and I need to concentrate on Fixer this weekend before we fall behind again (I may not have much ISFDB time the rest of the month). Any volunteers, perchance? Ahasuerus 02:52, 11 April 2010 (UTC)

No time but this site does seem to be the place to start.--swfritter 15:32, 11 April 2010 (UTC)

I whacked them around a bit today and we seem to be in better shape now. Removed three series but had to add the one with the Hardy Boys. I'm a bit suspicious of the "Harriet Stratemeyer Adams" titles that weren't published under a pseudonym though, perhaps Dana Carson might like to confirm or fix those? BLongley 18:36, 11 April 2010 (UTC)

Looks much better, thanks! And yes, I also suspect that the "Harriet Stratemeyer Adams" pubs in "Tom Swift, Jr." should all be changed to "Victor Appleton II". Ahasuerus 19:19, 11 April 2010 (UTC)

My fault, I was still learning and used the real name instead of the name on the book. Will switch them around. Dana Carson 22:54, 11 April 2010 (UTC)

Gaiman graphic novels in?

Reviewed in this pub. If so, is the artist treated as co-author or interior artist?--swfritter 15:43, 11 April 2010 (UTC)

I consider Gaiman "above the threshold" and thus his graphic novels should be in the database IMHO. (Just as Harry Stine's nonfiction is included in the db, as is the nonfiction of many spec-fic writers.) That doesn't mean every publication of the title should be in the database, probably just the first edition and any notable reprints (first UK, etc.) There has been no discussion about how graphic novels by spec-fic writers "above the threshold". There are two approaches that could be considered: 1) entered as novels with only the writer given author credit and the artist given interiorart credit or 2) entered as chapterbooks with both artist and author given credit for the book with separate content records for the fiction (most often as shortfiction/novella) and interiorart. Mhhutchins 16:33, 11 April 2010 (UTC)

Even though I agree Neil's above the threshold, I haven't gone out of the way to include graphic novels which are just "collected comics". Or we'd have a lot more "Sandman" titles, and Alan Moore would have a lot of "Swamp Thing" entries. I do NOT delete graphic novel versions of books we would have anyway. And actually had to reinstate several Stardust publications after someone got over-zealous on the deletions, and deleted text-only versions and even audiobooks. I don't want them as Chapterbooks though, we've abused that category enough already. Novel is fine with me, and an Artist can be co-author if credited that way by normal title-page rules. If there isn't a title-page - consider why. Is it because it's just a collection of comics? BLongley 18:54, 11 April 2010 (UTC)

I also draw the line at comic book collections, but graphic novels are another kettle of fish. The problem gets more complicated when a graphic novel is released serially (g.v. Watchmen). In that case, I use the first combined edition. I see your point about chapterbooks, but until graphic novels are allowed into the database, there's no standard for entry for those by above the threshold spec-fic authors. Mhhutchins 19:35, 11 April 2010 (UTC)

I can cope with Watchmen even if it's a "complete collections" of a comic mini-series - I think the intent was always clear, and we haven't got the individual comics at least. V for Vendetta too. It's at the edge of a slippery slope though, so I haven't unboxed my comics yet. And I see we're a bit inconsistent. BLongley 20:16, 11 April 2010 (UTC)

My usual approach when I find them in book reviews is to document them in pub notes no matter who is involved in the production. In this case a phantom is created. Unless someone else wants to add the book, I think that will be my approach in this case.--swfritter 19:01, 11 April 2010 (UTC)

A review of something that isn't normally included can of course be entered as an "essay" type for completism, and avoid the phantom author creations that I hate - it doesn't have to be notes alone. So an Essay called "Review of Graphic Novel Death: The High Cost of Living, by Neil Gaiman and Chris Bachalo" would be fine by me to ensure complete contents without phantom authors. The same could be done for several regular columns: for instance, some Lowe Film Review Nick Lowe columns go in to more detail than others on films being reviewed. BLongley 19:23, 11 April 2010 (UTC)

I do the same thing with graphic novels that are by under the threshold authors. See the two reviews on page 107 of the March 2010 issue of IASFM. Mhhutchins 19:35, 11 April 2010 (UTC)

Sounds like a good approach. I consider notes and essays to be marginal solutions with the only reason for there existence being that they can later be used to track down the reviews when graphic novel data entry is implemented. I really wish we had movie stub titles with imdb links.--swfritter 20:43, 11 April 2010 (UTC)

Chapterbooks - have I got this staight?

This should be a chapterbook with the actual story added as content? I want to make sure I give no false information and the standards for chapterbooks seem to change daily.--swfritter 18:45, 11 April 2010 (UTC)

It's not so much the standards for Chapterbooks, but the standards for ebooks or audiobooks of single short stories. Chapterbook seems to be the most common solution. We did suggest single-content Collections or Magazines at one point, but that didn't win out. BLongley 18:58, 11 April 2010 (UTC)

Thanks. The last thing I wanted to do was give some out of date advice to a new editor.--swfritter 19:02, 11 April 2010 (UTC)

ebook rather than e-book for binding

I guess that says it all.--swfritter 18:49, 11 April 2010 (UTC)

From my last backup, using this SQL:

select pub_ptype, count(*)
from pubs
where pub_ptype like 'e%book'
group by pub_ptype

I get these results:

ebook	1316
E-book	46
Electronic book	1

I think we have a winner. BLongley 19:14, 11 April 2010 (UTC)

Hoo-Ray. We have a standard published in Help and people are sticking to it, mostly.--swfritter 19:19, 11 April 2010 (UTC)

Omnibus, Collection, or chapterbook

this pub.--swfritter 18:53, 11 April 2010 (UTC)

Collection. BLongley 19:00, 11 April 2010 (UTC)

Fixer: May 2010 submitted

May 2010 has been submitted, so the queue is quite long at the moment. Have you noticed that this skiffy stuff has been getting increasingly popular lately?.. Ahasuerus 22:41, 11 April 2010 (UTC)

For [3], where the information is currently too incomplete to be useful, I'd like to allow Fixer to try picking it up again next month. Is there any way to have that happen, or is the case that once an ISBN has been captured, it will never be picked up again? --MartyD 10:52, 19 April 2010 (UTC)

Might be worth leaving a note on Fixer's page. BLongley 18:11, 19 April 2010 (UTC)

That was my first thought as well, but if I delete the suspect ISBN from Fixer's database, he may re-submit again too soon. A more elaborate solution would be to enter something like "Fixer: hold until 2010-09-01" in the Reject field. Fixer would then have to parse all rejected submissions from the weekly backup file and put the affected ISBNs on hold until 2010-09-01 or whatever the date may be.

However, this solution would take a moderate amount of work to implement and maintain and I am not sure it's worth the effort. Also, if an Amazon record has incomplete information immediately prior to (and especially after) the announced publication date, then it's fairly unlikely that it will be updated. It may also be canceled or delayed. Ahasuerus 21:15, 19 April 2010 (UTC)

I did find several submissions had inadequate data - but for older titles, managed to process them fairly manually (although I will have to revisit Astrosaurs again). I must admit I was frankly disheartened by the lack of good new SF (by which I exclude most paranormal romance) BLongley 18:11, 19 April 2010 (UTC)

There are a LOT of specialty/small presses, vanity publishers and POD publishers these days. Their output dilutes the better known stuff published by the major publishing houses. It would be much worse if Fixer wasn't auto-suspending the 4-5 largest vanity publishers. Ahasuerus 21:15, 19 April 2010 (UTC)

and feel our homepage is over-run by reprint titles I didn't want to buy in the first place. BLongley 18:11, 19 April 2010 (UTC)

It would be easy to exclude reprints and Titles labeled "paranormal romance" from the home page. Ahasuerus 21:15, 19 April 2010 (UTC)

As an aside - my processing has led to a backlog of Author webpage updates, feel free to tackle any of those. It's the Anthologees (is that a word?) that are causing the headaches - I find one title, find their webpage, find they're in another anthology, add that, find a load more authors... repeat ad infinitum, it seems. BLongley 18:11, 19 April 2010 (UTC)

Yes, our Anthologees (nice neologism!) tend to be incestuous, especially in the urban fantasy/paranormal romance corner of the field. Ahasuerus 21:15, 19 April 2010 (UTC)

Fixer: 2009 cleanup

Fixer has submitted a number of books by "major SF (and only SF) publishers" published in 2009. For the most part they are legitimate, but I suspect that a few, especially the ones without a price, were canceled and never appeared. Caveat moderator! Ahasuerus 04:12, 22 April 2010 (UTC)

2008 has been submitted using the same "low hanging fruit" approach. There are a few known US/UK mess-ups, but at least I tried to pre-screen canceled ISBNs to the best of my ability. Ahasuerus 03:17, 23 April 2010 (UTC)

Help with funky characters

I could use some help with what to do, if anything, about the (Hungarian) title of this pub. The "ű" ended up encoded in the database as ű as far as I can tell. Should I just leave it or go back to the editor and see if there's an equivalent that can be represented raw, without encoding? --MartyD 11:30, 24 April 2010 (UTC)

And another one: this one uses "ő", which ends up encoded as ő. --MartyD 12:00, 24 April 2010 (UTC)

Unfortunately, even though I have done some work on adding foreign language support, most of it was done at the design level, i.e. how different Title and Pub records should be organized and displayed, so I don't fully understand how the software converts various encodings. Al did all the original work on Unicode characters (and spent a small fortune on aspirin) and I believe Marc has looked into them, so they may be in a better position to comment.

Ideally, we should convert the whole database to Unicode, but for now it's a mix of Latin-1 and Unicode (and the Swedish collation sequence, believe it or not!), so band-aids is the best we can do. As long as the data displays correctly and as long as you can search using these characters, I think we should be "OK" for the currently approved value of "OK". It would be nice if you could find, say, "Móra Ferenc Ifjúsági Könyvkiadó" by searching on "Konyvkiado", but it only works for a small subset of diacritic characters at the moment. Ahasuerus 18:45, 24 April 2010 (UTC)

I'll address the original question from MartyD. At present we are displaying pages using ISO-8859-1. This is an eight-bit character set that supports roughly 200 different characters. Data entry field on a web page allow you to enter Unicode character values which are outside the 200 available in ISO-8859-1. The browser sends the data to the web server using "ű" style encoding. Or rather, the web browser sends the "ű" URL-encoded as "%26%23369%3B". Most web servers decode the "%26%23369%3B" and give the web application (Python in our case) a string containing "ű".

Thus someone types a "ű", sees "ű" on the page, and the web server will get a string with six plain-ASCII characters or "ű". If the web server sends that string off to a web page the browser converts six characters "ű" into "ű" for display purposes. A challenge is what happens if our human had typed "ű" and not a "ű" in the input field? It turns out that the web server, and thus Python, can't see the difference. There's no clean workaround for that.

The good news is if someone types "ű", or "ű", into a search box that the code will search for "ű" and find it in the database. The bad news is that sorting, functions that try to compute the length of the field (should a string-length on "ű" return 1 or 6?), and functions that try to process one character at a time all get messed up. For example, in another thread we are discussing the ISFDB author directory. If we add support for drilling up, down, or sideways or into the directory the code needs to know that "ű" is one character. Up is a mess as if you back up one character you get a ";" meaning the code needs to then scan back to see if that semicolon is part of a "ű" style sequence.

As Ahasuerus suggested, a solution is to convert the database to Unicode. Unfortunately, Unicode is a 21 bit value though in most systems it's easier to deal with this 32-bits at a time. 99% of the characters we encounter are plain ASCII which fit in 7-bits. Rather than burning 32-bits per character we could use encoding such UTF-8 or UTF-16 but we would run into exactly the same problems we already have with the "ű" style encoding of Unicode in an ISO-8859-1 string.

FWIW, Microsoft implemented this ghastly thing known as wide characters which looks like Unicode but only works in 16-bits. There's no support for character values U+10000 to U+10FFFF. This fooled people into thinking they were working working with Unicode when in fact they are only working with the Basic Multilingual Plane (BMP) subset of Unicode. That's also known as plane-0 (zero). Microsoft wide characters can't handle characters in Unicode planes 1 to 16.

To sum it up. "ű" in the database is fine for now. Fortunately, we do not have much code that processes data one character at a time. It's possible MySQL is aware of the "ű" encoding or can be made aware of this in terms of sorting. --Marc Kupper|talk 03:38, 7 June 2010 (UTC)

Update on my comment about detecting if a human types "ű" or "ű". It's hard if you are using ISO-8859-1 encoding for the web page. If you are using UTF-8 encoding then the browser sends "%C5%%B1" for "ű" and "%26%23369%3B" for "ű". Converting the forms to use UTF-8 is not trivial.

I filed a bug report as editpub is overcoding the data for the edit form before sending it to the browser which means editors are seeing "ű" instead of "ű". --Marc Kupper|talk 04:58, 7 June 2010 (UTC)

Actually Microsoft's "Unicode" support is UTF16LE and does support supplementary planes via surrogate pairs. Originally they only implemented UCS-2 (which strictly is a fixed 16-bit encoding) but then later extended it to variable length 16-bit encoding compliant with UTF-16LE. I move for UTF-8 support to be implemented (eventually). On that note, what if anything will be done with translations? I added some discussion to Talk:Requirements:Translations. Uzume 22:18, 12 December 2010 (UTC)

Adventure House facsimile reprints

Fixer is entering author data as editor. Since there appear to be Look Insides for these titles the actual editor can be determined that way or by looking at the magazine that is facsimiled - if it is in the database.--swfritter 15:21, 5 May 2010 (UTC)

Also note that Fixer is classifying these as Novels. They should be anthologies.--swfritter 15:23, 5 May 2010 (UTC)

Unfortunately, the only way for Fixer to tell that a book is an anthology rather than a novel is for Amazon to list it as such. And we all know how reliable Amazon's data entry people and taggers are. Ahasuerus 03:28, 9 May 2010 (UTC)

Fixer does vamps - the 2009 edition

Boy, they sure published a lot of vampire shaggers in 2009! Well, to be fair, they also published a lot of werewolf shaggers, mermaid shaggers, cougar shaggers and even Navajo medicine man shaggers in 2009. It should keep us busy for a while.

In the meantime, I finally managed to get a hold of the SF records at the British Library in a format that Fixer can understand (MODS XML), so I will be working on that. Pre-submission cleanup of almost 500 paranormal romance records is more than one should have to do in one week... Ahasuerus 03:26, 9 May 2010 (UTC)

So is POST-submission clean-up... there's a lot of author web-site updates on the queue if anybody wants to go find the rest. Despite the onslaught, there's plenty more out there. :-( 16:33, 10 May 2010 (UTC)

New submission approval shortcut for moderators

I have added a link that allows moderators (and only moderators) to go from the post-submission screen directly to the review/approval screen for that submission. At the moment, it's only enabled for Edit Title, Edit Pub and New Pub, but if it looks useful, I will add the same link to the other post-submission screens. Ahasuerus 03:14, 10 May 2010 (UTC)

Eek, a bug! Working on it... Ahasuerus 03:25, 10 May 2010 (UTC)

OK, the bug has been squashed. Sorry about that, I had forgotten that the same script handles "Edit Title" and "Add Pub to Title". Ahasuerus 03:42, 10 May 2010 (UTC)

Just tested it out. Works great. Someone deserves a raise! Mhhutchins 05:50, 10 May 2010 (UTC)

Yes, double that man's pay immediately! BLongley 16:29, 10 May 2010 (UTC)

Very nice feature. I love that man! :-) --Willem H. 18:44, 10 May 2010 (UTC)

I take it as a sign that the feature is popular :-) As of tonight, 10 post-submission pages take moderators directly to the approval screen. I'll do the other 10 tomorrow. Ahasuerus 03:46, 11 May 2010 (UTC)

All post-submission pages have been updated, so if you find any that won't let you go directly to the approval screen, please let me know. Ahasuerus 03:39, 12 May 2010 (UTC)

Title merges don't have a link to the approval screen. Mhhutchins 00:40, 13 May 2010 (UTC)

Title merges/unmerges and author/publisher merges were hiding in a different area, but that only postponed the inevitable! Good catch, thanks! Ahasuerus 03:35, 13 May 2010 (UTC)

June 2010 submitted

Fixer is done with June 2010. He says he is exhausted, so he will take a break for a week or two. Ahasuerus 01:41, 16 May 2010 (UTC)

Fixer recovered earlier than expected and submitted the first batch of February 2010 items overnight. (He apologies for the comics, they sneaked in under the radar.) Once the rest of February has been submitted and processed, Fixer will do a final clean-up pass for January-June and then we will be all caught up for the year! Ahasuerus 16:19, 20 May 2010 (UTC)

February has been submitted. Ahasuerus 14:31, 22 May 2010 (UTC)

Inferred ISBNs

Editor Astromath is proposing to change a couple of pub entries from 5-digit catalog numbers to ISBNs derived from the publisher ID and 5-digit number on spines/covers. I.e., no leading zero AND no trailing check digit. Explaining the derivation in the notes. I don't have an objection -- seems in the proper spirit -- but Help:Screen:EditPub#ISBN doesn't explicitly cover this case. Thanks. --MartyD 10:25, 26 May 2010 (UTC)

I see no problem in anyone doing this if the ISBN link leads to one or more of the outside databases on our "Other Sites" list (Worldcat, Amazon, Abebooks, etc.) Otherwise what's the point? If it does link, you should clearly state in the notes what catalog number is actually printed in the book itself. I would object to him doing it for books that he can not physically verify, or books that were published before ISBNs became the standard (circa 1970). Also be aware that some publishers created pseudo-ISBNs in the early seventies similarly to the way you propose, but adding the price as part of the number. They turned out not to be valid ISBNs, and should be entered as catalog numbers (using the #). Mhhutchins 15:45, 26 May 2010 (UTC)

I'm fine with it too, so long as 1) the derived ISBN is of some use and 2) we still record what is actually printed in/on the book. There's a loose feature request to allow a second ISBN/Catalog Number field for oddities such as different ISBNs on DJ and book, or on cover and copyright page, so as long as we record all the relevant numbers they'll be of use someday. BLongley 17:48, 26 May 2010 (UTC)

Note that British SBNs (omits leading 0 only) and ISBNs go back a bit further (mid-60s), but even I don't force them into an ISBN if it goes nowhere. But I especially don't mess with serial numbers where the price is the last part of the catalogue number, which seems more common on US pubs - if in doubt, stick with what's there and don't guess a derivation. BLongley 17:48, 26 May 2010 (UTC)

I've been mulling this one over and am starting to lean away from inferred or derived ISBNs. ISFDB's mission is not "linking to book seller sites." The main up-side of derived links is that it's easier to check for images or things like the publication date. I added support to my personal book database for "SBN #value" which looks at the value and constructs an ISBN. At present it recognizes and parses six different SBN formats. I've been thinking of adding support for "Ace #####", "DAW Ux####", etc. If we added something like this to ISFDB it would reduce the need for putting derived ISBNs in the catalog # field. --Marc Kupper|talk 02:36, 7 June 2010 (UTC)

I think that the long term solution is to create a new field, "Catalog ID", and move all non-ISBN values out of the current composite field to the new field. It will let us capture both data elements, which will improve our search and linking capabilities. Of course, some books have more than one catalog ID (think DAW 25 years ago) and there is also that nagging issue of "corrected ISBNs", so ultimately we may need even more fields, but it will be a start. Ahasuerus 02:53, 7 June 2010 (UTC)

I've asked for a second or more ISBN/Catalog(ue)-ID field for ages - just add it/them, populate it/them with the same data as the first to start with, and see how people use it. I think we need two ISBN fields at least, for publications that really do have two different ISBNs (and I don't mean publications that list the ISBNs of other versions, I mean publications where cover doesn't match copyright page, or even where the same DJ has two different ISBNs. "Rebound by Sagebrush" and other such library bindings created from paperbacks come to mind). But you might as well see if people want LCCN and such recordable as well, although a lot of such possible fields will not link anywhere. (Which is possibly a blessing for the coders!). BLongley 20:52, 7 June 2010 (UTC)

It's not that there is opposition to adding more catalog/ISBN fields, it's just that the changes are not trivial, especially as they relate to the ISBN search logic. We haven't added a single new field to the core tables in the last 2+ years and although (we think) we know enough about the application to start doing it now, it's still a big step for us.

I am thinking that adding a new field for Catalog IDs would be a conservative first step since the impact should be limited. Once we confirm that we do indeed know what we are doing, it should be easier to add other fields for Publication Series, more ISBNs, etc, eventually leading to modifying other tables that will be affected by the proposed Foreign Languages project.

As far as adding more fields for LCCNs, WorldCat IDs, etc goes, we may end up with a repeatable field which lets you select from a drop-down list of supported external sources, but that will have to wait until other things are taken care of. Ahasuerus 03:10, 8 June 2010 (UTC)

I tried a second field for the Amazon link in my personal DB a while back and quickly reverted as I found I was double-entering data for many publications, particularly those with SBNs. I went back to one field and added logic to check for the "SBN " prefix and to build the links from that. It's a hack but the code first scans the catalog # field a "*" and if found it discards all of the string up to an including the * before then parsing to make a link. This allows me to enter "DAW UE1756; *0-87997-756-6" as that publication has both the catalog # and an ISBN. For an older pub without an ISBN it's "DAW UY1171; On Amazon at *0879971711" and here's what I did for a book that had an ISBN but it's not usable "Signet 451-W7671 $1.50; Spine 0-451-07671-0; On Amazon at *B000EVOQH8" (I did that one years ago and see that the ISBN is now valid though they call it a Roc edition). --Marc Kupper|talk 08:26, 7 June 2010 (UTC)

I find derived ISBNs very valuable when all we have is a stub title and no pubs - I can do a title search on Amazon and get a little data, with ISBN, but often crap publisher info with no pagination or format. But if I process that stub publication, then I have quick links to Worldcat and other sites where I can often pick up more data. Yes, it may turn out that the physical publication has never shown any indication of an (I)SBN but I prefer to have those stubs rather than nothing - and the Worldcat link at least, with the "other editions" possibilities that gives, makes it far faster to build up a new-to-us title's complete(ish) history. BLongley 20:33, 7 June 2010 (UTC)

Coup de Grace and Other Stories by Vance

This is one of the strangest things I've seen in the database yet. It looks like there are eleven different versions of this collection, but only one shows on the Jack Vance page. I'm tempted to delete the other ten, but I don't know the consequences (or if there are any). Any ideas? --Willem H. 15:21, 31 May 2010 (UTC)

They are all different (different tags). Each has no title record. I tried deleting one in a local copy of the database, and it worked ok (both deleted it and caused no other visible harm). --MartyD 16:32, 31 May 2010 (UTC)

Thanks! I'll erase them. --Willem H. 18:34, 31 May 2010 (UTC)

Bluetyson submissions

Apparently Clarkmci has not been able to process Bluetyson's submissions. I would assume he is not able to do so but it is collectively rude to leave submissions in the queue for that length of time. Both of the submissions involve standards. The ACE double submission is the most contentious one. It would seem that decisions should be made and Bluetyson be apprised of the logic behind those decisions. And also be given an apology.--swfritter 14:00, 1 June 2010 (UTC)

It appears that Bluetyson disappeared shortly after these submissions, otherwise I think he would have inquired into the reason why they're still being held. And of course, you're correct that Clarkmci should have either dealt with them himself, removed the hold on them, or even simply asked for other moderators' input. It also appears he's only popped in occasionally since then. Your response to the original inquiry on Bluetyson's page leads me to think you're in the best position to handle the question about the "Snows of Ganymede" merge. Rtrace and Dragoondelight should be able to handle the "Of PIGS and MEN" question (they've each verified the two differing titles.) Ahasuerus, can you remove the holds on these submissions? Mhhutchins 00:00, 2 June 2010 (UTC)

I could directly manipulate the database to remove the hold, but there is always a small chance of an unfortunate typo. Since these submissions are fairly straightforward, it's safer to do a "hard reject" on them and then re-create them manually. Ahasuerus 00:05, 2 June 2010 (UTC)

Currently the common practice seems to be that Ace double "novels" like The Snows of Ganymede are categorized as NOVEL even though they are obviously of novella length. Which means the only way to link the original magazine novella appearance is by making the magazine NOVELLA a variant of the NOVEL. I don't agree with this undocumented practice. The only rationale I can think of for this as that the user is more likely to look for the title in the novel section. Since the pub has been verified by four different editors I am not in a position to do the merge that seems appropriate to me.--swfritter 13:17, 2 June 2010 (UTC)

One possible solution: change the record of the magazine version from shortfiction/novella to serial/(Complete Novel). (Despite their length, we know many works are called "novels" in magazines that truly aren't.) Then make that record into a variant of the novel title record which is currently part of the Ace double. We can note in each of the title records the word count/length and explaining the reason for their status. It's obvious the database must show the connection between the two pubs, and to avoid any future attempt at merging the two titles. Mhhutchins 18:46, 2 June 2010 (UTC)

The only time I did that was for "Anthem" and I found that so annoying I changed it back. I would have changed the other appearances but did not because the Libertarians told me not to; the Ayn Rand contingent is a particularly scary group. Unlike "The Snows of Ganymede" it was also first published as and marketed as a novel. The (Complete Novel) or novella variant method (the one I prefer) are essentially conflict-avoiding solutions that I find barely acceptable but nevertheless OK; the ACE novels are a special case so this method impacts few other situations. So I would make the novella a variant of the novel but perhaps append (magazine version) to the title. This would follow more closely the method currently being used.--swfritter 19:19, 2 June 2010 (UTC)

Against my better judgment I made the proposed change and notified Bluetyson.--swfritter 13:51, 4 June 2010 (UTC)

I think your solution is fine considering how the database really doesn't have a way to create variant relationships other than for title and/or author. Thanks. That leaves another Clarkmci hold for a Fixer submission. I'm tempted to hard-reject that one and create another submission for the pub, as it appears to be a UK printing of a title already in the database. Mhhutchins 17:34, 4 June 2010 (UTC)

A crack team of repo men headed by Bluesman is already on the case - see User talk:Fixer :-) Ahasuerus 18:54, 4 June 2010 (UTC)

Hollywood Comics

Just a note that the "Hollywood Comics" books that Fixer has been submitting are bona fide SF novels, mostly translations from the French. The imprint is, um, misleading at best. Ahasuerus 03:03, 2 June 2010 (UTC)

Actually, a lot are collections - previous mods seem to have missed that on several books. And I suspect that if we saw the books themselves, they might well be under the "Black Coat Press" imprint that Hollywood Comics own. Still, good practice for the little French skills I have. We do seem to have some issues about how to credit an "adapter" though - Brian Stableford seems to have been left as co-author at times when he's actually translator and editor of some collections. Surely Help says those roles go in notes for now? BLongley 19:01, 2 June 2010 (UTC)

It sure does: "There is currently no support in the ISFDB for translators, or photographers; this information should just be entered in the notes field." Ahasuerus 22:40, 2 June 2010 (UTC)

Correction to "Y. J. F. Long"

The novel "Les Atlantes du ciel (1954)" is listed under two authors: "Y. J. F. Long" and "Y. F. J. Long". The latter name is correct, as can be seen in a scan of the book cover, available at [4] Chavey 16:43, 2 June 2010 (UTC)

The miscredited record has been deleted. Thanks for bringing this to our attention. Mhhutchins 21:56, 2 June 2010 (UTC)

Fixer: January-June cleaned up

Fixer has cleaned up January-June, so Amazon should be all caught up once the current crop has been processed. They will undoubtedly add more early 2010 books retroactively, but at least we should be able to stay ahead of the wolves. Fixer is working on library catalogs at the moment, but it will be some time before proper submissions can be generated. Ahasuerus 04:37, 5 June 2010 (UTC)

Awards and [The] Stains

This submission would merge away "Stains", leaving "The Stains". The only "Stains" reference is the 1981 British Fantasy Awards win. That site lists it as "The Stains" as well. I was going to accept the merge, but having tried it in a local copy to see if it will work, I notice that the award listing in the resulting title biblio page would still say "1981 - Stains British Fantasy Award, Best Short Fiction (Win)". How does one go about fixing that part? Thanks, --MartyD 10:40, 5 June 2010 (UTC)

You would have to use the Award Editor -- see Al's last comment re: its status. Al started writing Help pages for it when he last appeared in March, but it still needs to be activated/tested. Ahasuerus 19:03, 5 June 2010 (UTC)

Amazon Audio Download oddities

(Posted for mods only, as I think only Fixer submissions are doing this.) Another batch of "Magic Ballerina" submissions looked a bit odd - every one was missing binding, had notes about being unabridged, and when checked at Amazon, the giveaway that they were audio versions was only down to the fact that they listed the "reader". Now they check out at the HarperCollins site under the 13-digit ISBN, for instance here, but Amazon seem to list the download under a separate ASIN like B00383CCNY rather than the ISBN-10 like this. I'm wondering if it's because HarperCollins say they're "Mixed media, B-format paperback plus audio CD" and Amazon don't know how to cope with such yet? BLongley 20:18, 7 June 2010 (UTC)

Anyway, there's the example, and I'd be happy to solve it by nuking all such, but I wondered if anyone else had come across this happening to some important titles? BLongley 20:18, 7 June 2010 (UTC)

Tom Swift Jr. Novels

The Tom Swift, Jr. Novels are uniformly attributed here to Harriet Stratemeyer Adams, writing under the house name of Victor Appleton II. However Fantastic Fiction lists Jim Lawrence as the author of these books. I don't know which attribution is correct (my guess, based on Harriet's Wikipedia page, is that Jim Lawrence is more likely). However, if someone can evaluate this conflict, it might be helpful. Chavey 22:24, 7 June 2010 (UTC)

Actually it is a little more complicated. According to Wikipedia Harriet (Stratemeyer) Adams did the outlines.--swfritter 12:22, 11 June 2010 (UTC)

Fixer - July 2010

July has been submitted. Ahasuerus 16:52, 13 June 2010 (UTC)

December 2009 cleaned up. It's amazing how many books Amazon adds months after the publication date... Ahasuerus 04:06, 22 June 2010 (UTC)

ISFDB editing currently offline

I have run into some issues installing the new pub series software and had to disable editing for a bit. It should be back as soon as I get everything configured. Sorry about the hassle! Ahasuerus 03:46, 27 June 2010 (UTC)

The problem has been fixed and public editing has been re-enabled. Ahasuerus 04:03, 27 June 2010 (UTC)

Fixer - August 2010

Fixer is in the process of submitting August 2010. Since September books often appear as "August 31" in Amazon's catalog and since September is the busiest month in the publishing business, there are likely to be more submissions than usual. Ahasuerus 21:03, 3 July 2010 (UTC)

Possibly too many, too soon. We've still got some of May's Fixer submissions in the queue. Perhaps the "Moderate submission" option means we've got too many Mods self-moderating at the expense of the queue now? :-/ BLongley 00:53, 4 July 2010 (UTC)

I do enjoy a new set of submissions I'd never have thought of looking for myself, but all those "Paranormal Romances" do get me down a bit and I'd often prefer to go back to work on the Stableford collection. And I really should spend more time away from the computer... I've got to get back in practice for Pool-playing, I've got to do the Captaincy for Monday nights now. BLongley 00:53, 4 July 2010 (UTC)

August/September are historically the roughest months, probably because kids go back to school and publishers prepare to attack them -- or at least their parents' checking accounts. I told Fixer to lie low for a bit, we'll need time to digest everything. Also, "Angry Robot" is a UK imprint, but Fixer decided to use US prices since there were no UK prices at Amazon UK. I wonder if the recent reorganization at the publisher may have affected the publishing schedule and some books were postponed/canceled? Ahasuerus 04:25, 4 July 2010 (UTC)

I admit that the "Moderate Submission" option helps in keeping me away from the queue, but I do check in every now and then to look for human-submitted edits. To be honest, looking at Fixer submissions give me a headache. If they were cut and dry, either accept or reject, it would be a lot easier. For the past few months, I've avoided looking at them for this reason, believing my time is better spent on creating records for pubs that I know are A) speculative fiction without question, and B) actually exist. Too many of Fixer's submissions lately have me questioning either A) or B), which has raised my frustration level to the point where I choose to stay away from them entirely. I personally feel someone entering pubs from Locus's Books Received listings represents time better spent, even though they may be a couple of months (or decades) behind. I'd rather be moderating submissions for books that have already been published than those that are forthcoming. Happily there are editors who choose the latter. Mhhutchins 02:40, 5 July 2010 (UTC)

I could not have said it better!!--swfritter 15:54, 5 July 2010 (UTC)

I also would prefer to spend my time on Vector and Paperback Inferno, which I think will provide the nearest UK equivalent to the Locus entries. They're not as detailed, but I usually find a missing UK book or two per Magazine entered. "Happily there are editors who choose the latter" may not be accurate much longer - I'm getting fed up with being one of the few mods that deal with Fixer submissions and want to go back to my own work. Kraang can enjoy such if he wants, but I really feel that the huge numbers of Fixer submissions at times slows down all my other work - the queue should not get so long that I have to wait to see the queue. I'm on broadband and have no problems with a page or two of submissions to review, but when I have to wait 10 seconds for a full list to appear, I'm tempted to go back to self-moderating. BLongley 22:19, 7 July 2010 (UTC)

I'll try to ensure that Fixer's submissions are limited to 2-3 pages at a time. It may separate some books by the same author or other related books, but the alternative seems much worse. Ahasuerus 22:47, 7 July 2010 (UTC)

Another possibility would be to create a "End of Submission List" link that just gave us the end of the queue - so we get to see our own edits, but also some of the other Human entries. BLongley 22:19, 7 July 2010 (UTC)

The A) & B) options keep me busy, lots of borderline stuff and vaporware.Kraang 03:28, 5 July 2010 (UTC)

Fixer can be a frustrating fellow. On the one hand, he is often our only source of information for obscure books and he does a decent job of keeping the database up to date. On the other hand, it can take a long time to figure out whether what he finds is very obscure or simply non-existent. And you don't want to know how much junk he finds that never makes it to the submission queue!

Now, I could ask him to concentrate on library catalogs, which are generally much more reliable than Amazon, but then we would have to give up on the Forthcoming Books section. Decisions, decisions... Ahasuerus 05:27, 5 July 2010 (UTC)

If Fixer did not do the juvenile books I would be a lot happier. Too much of that category is borderline specfict and I spend a fair amount of time thinking about in or out on each title. Something I'd like to be able to to is to feed Fixer a list of ISBNs. DAW Books allocates then sequentially tends to first load the books 9 months out then at 3 to 6 months out the image and price firms up. --Marc Kupper|talk 07:20, 5 July 2010 (UTC)

I would reject about 90% of Fixer's submissions out of hand and I would not have thought that would be acceptable to others. The current crop of YA/New Paranormal writers are not writing in the tradition of Mary Shelley, Jules Verne, H. G. Wells, etc. Stephenie Meyer is proud of the fact that she has never read genre fiction. How can you write genre fiction if you have never read it! Graphic novels would be substantially more acceptable. Their creators are quite obviously aware of the traditions. --swfritter 16:07, 5 July 2010 (UTC)

Well, we all have our preferences, e.g. I am not a fan of paranormal romance, but I have a substantial collection of YA works going back to well before E. Nesbit. However, ISFDB:Policy explicitly rules "non-genre speculative fiction" along with "[f]abulation[s], [m]agic realism and [s]lipstream" in and comics, manga and graphic novels out. Granted, rules can be changed, but any changes that would result in wholesale destruction of thousands of records, which many people spent hundreds of man-hours on, should be approached very very carefully. Ahasuerus 17:23, 5 July 2010 (UTC)

I would not exclude them if somebody thought they were worth the effort of entering manually. Juvenile and YA works need a little bit of retrospective analysis; so in twenty years the pre-humans (aka children) who are reading them now should feel free to enter them into the database if they feel they were a significant part of their spec fiction experience.--swfritter 19:27, 5 July 2010 (UTC)

That would be a very significant change to ISFDB:Policy and one that I would strongly oppose, both on substantive grounds and because it would make Policy much more subjective than it is now. Ahasuerus 21:16, 5 July 2010 (UTC)

How much effort would it be to restrict Fixer's Juvenile additions to those only adding publications to existing titles? I have no real objection to yet more L. Frank Baum Oz publications, or another edition of Alice in Wonderland or Through the Looking Glass. Not so keen on "reworkings" or "adaptations" by other authors, but those could be excluded if they don't match the existing ones. BLongley 21:55, 7 July 2010 (UTC)

That's a very good point - processing an Add Pub submission is much less time consuming than processing a New Pub. I'll add it to the list of Fixer improvements. Ahasuerus 22:58, 7 July 2010 (UTC)

As an aside, there is a distinction to be made between what is "in" based on ISFDB:Policy and what we have the manpower to find and enter given our limited resources. For example, SF pornography is currently "in" even though we have but a small percentage of it on file because no one has been willing to spend much time on this sub-genre so far. And that's fine, of course, but ISFDB:Policy provides a degree of assurance that if you enter something that is described as "in", it will not be deleted by another editor because, e.g. it is not in the tradition of Mary Shelley, Jules Verne, and H. G. Wells. Ahasuerus 21:16, 5 July 2010 (UTC)

The problem that I see with Fixer's submissions at the moment is not that too many books that he is finding are clearly "out" based on Policy -- those can be easily rejected and I usually reject them even before Fixer creates a submission -- but rather that it's not always easy to tell whether they are SF or whether they even exist. The whole idea behind Fixer was to make the process of identifying and entering SF books less time consuming, but the end result is not as clear-cut as I originally hoped it would be.

I can think of a number of things that we could do about it, but they all involve trade-offs. For example, I could ask Fixer to concentrate on library catalogs, which typically have much cleaner data, but then we would have to give up all attempts to keep Forthcoming Books up to date. Alternatively, we could tell Fixer to limit his Amazon imports to a pre-defined list of publishers, which would eliminate (or at least postpone until we have more willing moderators) many vanity/self-published books (many are already auto-suspended), specialty SF romances, etc. The downside will be that we would be missing at least some "under the radar small presses", university presses and other hard-to-find publishers. Finally, I could spend more time processing Fixer submissions, but that would be at the expense of fixing bugs and implementing new features. Decisions, decisions... Ahasuerus 21:16, 5 July 2010 (UTC)

I am thinking that the "low hanging fruit" in this case would be to expand the list of "auto-suspend" publishers so that it would cover even more vanity guys. Ahasuerus 21:16, 5 July 2010 (UTC)

The key point is not whether a pub should be in but whether there are editors willing to make those decisions. My priorities are magazines and free stuff like Project Gutenberg and Librivox. I have also had the good sense to restrict my editing time and increase my reading time.--swfritter 21:55, 5 July 2010 (UTC)

Ah, I see. When you wrote "I would reject about 90% of Fixer's submissions out of hand and I would not have thought that would be acceptable to others", I read it to mean that you thought that other moderators would also reject 90% of the submissions. Now that I have re-read the sentence, I see that you meant the opposite. Sorry, comprehension issues on my end, I may need an English-to-Python dictionary :) Ahasuerus 23:15, 5 July 2010 (UTC)

Perhaps at some point paranormal romances will form a completely separate genre, at which point we may want to split it into a separate database, which will be maintained by a separate team of volunteers (assuming that there are any), but judging by the contents of Publisher:SFBC 2010, I don't think we are anywhere near that point yet. Ahasuerus 17:23, 5 July 2010 (UTC)

Old paranormal (ok by me) vs. buffy-paranormal. I canceled my SFBC membership when they stopped offering Yarbro's St. Germain novels.--swfritter 19:27, 5 July 2010 (UTC)

I would strongly suggest that SF Site be one of our sources for ferreting out titles. --swfritter 16:07, 5 July 2010 (UTC)

Yes, they are one of the better sites out there. They also used to host a copy of ISFDB back when we were having hosting troubles. Ahasuerus 17:23, 5 July 2010 (UTC)

Thanks to Bill for his dedicated effort to make Fixer happy. And thinks to Fixer for trying so hard. --swfritter 16:07, 5 July 2010 (UTC)

He sure tries hard for a two year old! Ahasuerus 17:23, 5 July 2010 (UTC)

One of the huge factors is whether an author is in the database. If Fixer found such a link it would be nice if there were a link to the author's biblio page. The submissions with this data could be cleared much more quickly from the queue.--swfritter 17:02, 5 July 2010 (UTC)

I am not sure I fully understand the suggestion, but if the idea is to link author names in Fixer's (and other) submissions to the author's ISFDB Summary pages, then yes, it's at the top of my list of things to implement. Ahasuerus 21:19, 5 July 2010 (UTC)

Correct. I would come close to automatically approving a submission if the author was in the database and the Amazon listing give even the faintest hint that the work was spec fic.--swfritter 21:55, 5 July 2010 (UTC)

Unfortunately, Amazon doesn't always delete vaporware pubs and Fixer dutifully picks them up. During some runs, over 10% of all submissions are vaporware :-\ Ahasuerus 23:16, 5 July 2010 (UTC)

I think the request remains - on the approval screen, if the author exists in ISFDB, provide a quick link to that author. Maybe also to the Author's home-page, if we have such details. I'm adding such details when I review submissions, so even if I don't care to follow them up, they might help others. BLongley 21:45, 7 July 2010 (UTC)

I'd also suggest a Publisher Link, but in that case not just if the publisher exists in ISFDB, but if the Publisher Web-Page link is present, link to that. Existence or relevance of some entries could not be determined by the Amazon link alone, but sometimes the Author or Publisher pages eventually linked somewhere useful. BLongley 21:45, 7 July 2010 (UTC)

I was working on this very feature yesterday night! It's not as trivial as it may appear to be because the same code is shared by multiple approval screens and has to be approached carefully, but it is certainly doable. Ahasuerus 22:47, 7 July 2010 (UTC)

It would be nice if there were a way to reject a Fixer submission that didn't permanently "ban" it from future Fixer submissions (or correct my misunderstanding if I am mistaken). I know I spend a bunch of time trying to figure out if somewhat incomplete submissions are indeed vaporware. It would be much less time consuming to reject them out of hand for being "incomplete" on the assumption they'll get another chance down the road, if appropriate. I also echo the sentiment about juvenile works -- they're hard to judge; it might be better if Fixer did not submit things marked for low grade levels or "children's" or whatever combination would filter most of them out. --MartyD 00:26, 6 July 2010 (UTC)

(unindent) It may be possible, but it would not be trivial. We could add a new status to submission records, something like "Skip" or "Suspend". Then, when Fixer incorporates the weekly backup into his database, he could check the submission table for any "skipped" records and change the status of the ISBN from "submitted" to "suspended" in his database. Doable, but a bit painful.

Having said that, there are other things that we could do that would be easier to implement.

First, I did a test run for a popular YA author earlier tonight and the quality of Amazon's "library bindings" records was so bad that it made it hard to tell whether they were vaporware or not. It's fairly easy to spot and auto-suspend 80%+ of them, so that should make our life easier.

Second, Amazon uses "browse nodes" to tag books and they can be used by Fixer. Here are some relevant browse nodes and the number of books for each node in Fixer's internal database:

ages 0-2:   2256
ages 0-3:   38
ages 3-4:   4132
ages 3-5:   56
ages 4-8:   14412
ages 5-8:   20071
ages 6-8:   113
ages 9-11:  23376
ages 9-12:  30693
ages 12-16: 33223

I can ask Fixer to auto-suspend "ages 0-8", which should make life easier.

Third, I can expand Fixer's list of "suspend on sight" publishers to include the rest of the known vanity publishers. It's not possible to automate this process because they keep multiplying and because they keep changing their names, but it should help.

Let me give it a shot and see what happens... Ahasuerus 01:18, 6 July 2010 (UTC)

An unfortunate number of titles are still labelled "9-12", but if that's the maximum range of several listed and you take out the ones also labelled up to 8, then maybe I can stop having to review "Magic Puppy", "Magic Kitten", "Magic Bunny" (and there's worse - Sue Bentley also has "Magic Ponies" and "Magic Puffin"!). And nobody should have to sort out Daisy Meadows. BLongley 21:31, 7 July 2010 (UTC)

I am thinking that we can use these browse nodes in combination with the "page count" field to winnow out the riff raff. For example, if a book is under 80 pages and has at least one "ages 0-8" browse nodes, then it gets auto-suspended. Combined with the planned auto-suspension of library editions and additional vanity publishers, it may reduce the volume by 50%+. More importantly, the remaining submissions will require less time to process. Ahasuerus 22:58, 7 July 2010 (UTC)

OK, the first batch of submissions created using new and (hopefully) improved rules is in the queue now. A bit heavy on paranormals in the middle, but that's because it was sorted by ISBN and ended up with a disproportionate number of Zebra paperbacks. Ahasuerus 04:58, 9 July 2010 (UTC)

I did a random sampling (half a dozen or so) of them this morning, and they seemed to be pretty good. The one non-adult book was (a) a true YA book and (b) quite appropriate. No vaporware. Granted, a small sample size, but anecdotally I'd say in submissions past at least half would have been something else.... --MartyD 14:04, 9 July 2010 (UTC)

Yes, not a bad set of submissions. One thing that I've noticed though is that post-approval merges are still quite time-consuming as the check-boxes often default the wrong way - i.e. to the later date, the non-series title and number, and length. So there's often 4 extra clicks required before you can submit a merge. This can be quite time-consuming when you're dealing with Omnibuses, or Collections and Anthologies of previously published material. Can this be improved? BLongley 18:13, 9 July 2010 (UTC)

I don't think the logic that selects the default value is particularly sophisticated at the moment. I am sure it can be improved, but I'll have to review the details before I can estimate how time consuming it will be. Ahasuerus 18:34, 9 July 2010 (UTC)

Angry Robot cancellations

Please note that Angry Robot is in the middle of a UK re-launch and a US-Canada launch, which messed up their schedule. Any Fixer-submitted ISBNs not listed on their 2010-2011 page have been (most likely) canceled. Ahasuerus 03:02, 7 July 2010 (UTC)

That's a pain I just had to go back and redate one to May 2011. Thanks for the warning.Kraang 03:26, 7 July 2010 (UTC)

May 2011? Given the vagaries of the publishing business, you may have more faith in Angry Robot's successful re-launch than some of their employees! :) Ahasuerus 03:33, 7 July 2010 (UTC)

I think I mostly fixed them - unfortunately a lot of the new ISBNs stated on the website are not valid, and they are almost certainly not going to be published with HarperCollins ISBNs as we originally had them. So the 2011 pubs are gone, although some titles remain. It seems all but the Kylie Chan books will appear eventually. (And hopefully the Kylie Chan ones will appear from HarperCollins at some point.) BLongley 21:13, 16 July 2010 (UTC)

Changes to the New Pub/Clone Pub approval screens

There is a new column in the New Pub/Clone Pub approval screens, "Warnings". It will notify you when the publisher and/or the publication series and/or at least one of the authors/artists is not in the database. In addition, any authors, artists, publishers and publication series that are already on file are now linked to their respective pages. (We may want to add similar warnings for irregular page numbers and bindings later.)

Please note that this change affects only the New/Clone Pub approval screens at this time. All other approval screens, which have a column for "before" and another column for "after", are coded differently internally and would have to be modified separately. Ahasuerus 18:29, 11 July 2010 (UTC)

So how does it look? Should we add similar warnings when the binding type is not pb/tp/hc/etc? What about page counts that do not follow the xii+234+3+etc pattern? Prices that start with a number rather than with a currency sign? Catalog IDs that do not start with the pound sign? Ahasuerus 03:45, 12 July 2010 (UTC)

I have yet to see an actual warning. But since the mechanism is there, I'd say drawing attention to anything "odd" would be useful. --MartyD 09:54, 12 July 2010 (UTC)

I've seen one warning so far (only approved a few submissions for new pubs since the new feature was implemented). It pointed out that a cover artist was new to the database. The case where I'm going to be more interested is when a new publisher is being added. This will let me know in most cases when a new editor is using a non-standard form of a publisher's name. Hopefully this will cut down on those "one book" publishers and help consolidate more pubs under a standard publisher name. Thanks for the feature. Mhhutchins 13:34, 12 July 2010 (UTC)

Thanks for the feedback! Just thought of another one -- invalid ISBNs! Ahasuerus 21:46, 12 July 2010 (UTC)

I'm not sure it's added any benefit to the author checks I already do, but it has highlighted some irregular publishers. Good addition. BLongley 23:12, 12 July 2010 (UTC)

How about a warning when the New Publication added to an existing title has an earlier date? BLongley 21:54, 18 July 2010 (UTC)

An excellent idea! Ahasuerus 22:23, 18 July 2010 (UTC)

Done. It was slightly trickier than expected (e.g. the logic ignores 0000-00-00 dates), but hopefully I didn't make any mistakes. Ahasuerus 03:11, 20 July 2010 (UTC)

Fixer and UK submissions

Fixer has finished US books for September and is now processing his UK catch. There appears to be a somewhat higher than usual percentage of irregular prices; some of them look like canceled US books that haven't been deleted from Amazon UK yet. Also, there are a number of Angry Robot books that may have been canceled due to publisher reorganiz(s)ation -- see above. Ahasuerus 03:38, 13 July 2010 (UTC)

There were a few US/UK mix-ups, and one Australian, but not a bad set of submissions overall. BLongley 19:38, 13 July 2010 (UTC)

OK, all August-September books have been submitted. The queue is a bit longer than expected, but I was so close to flushing Fixer's internal queue that I just couldn't stop :-) Now to see what Amazon retroactively added for January-June while we were working on July-September... Ahasuerus 05:03, 14 July 2010 (UTC)

Rerunning Fixer for January and February resulted in only 32 returned records, which were further pared down to 8 new submissions, so not too bad. Ahasuerus 14:14, 14 July 2010 (UTC)

Fixer has re-downloaded March-September. Mostly odds and ends, but also a number of classic reprints and other nifty items. 50% of Fixer's queue has been submitted, the rest will wait until the current crop is taken care of. Ahasuerus 04:22, 17 July 2010 (UTC)

Fixer's January-September cleanup has been completed, what you see in the queue is what's left! Not a particularly interesting batch, but I suppose anything entered after its official publication date is likely to come from marginal publishers. Ahasuerus 00:17, 18 July 2010 (UTC)

New Fixer Algorithm

Now that we are caught up, Fixer is trying a new "weighted" algorithm to identify and submit "notable" works of SF. The results look promising so far, but if you see anything unusual, please post here. Ahasuerus 21:08, 18 July 2010 (UTC)

Hm, the new algorithm is not quite as good as I had hoped, although the results are still better than your average Amazon data. No paranormals and relatively few YAs so far, but a fair amount of borderline "literary fiction" stuff. Oh well, it will probably take a few days to process this batch, then I will have a chat with Fixer. Ahasuerus 03:48, 19 July 2010 (UTC)

The new algorithm has been set aside for now. Fixer is in the process of catching up with Amazon's recent additions for July-September, so we can expect a higher than usual percentage of self-published books and other oddities for another day or two. Once Fixer is done with them, I will ask him to try something new using the table of ISBN correlations that has been built up over the last few months (see above.) Ahasuerus 03:25, 4 August 2010 (UTC)

Help with foreign-language submissions

I need guidance about what to do with some of Benario's submissions that are sitting in the queue. These are French titles by French authors, for which I didn't find an English translation. The ROA seem to exclude these. Yet I see Pierre Pelot with plenty of similar examples. Left to my own devices, I'd be inclined to accept them. But within the bounds of the existing rules, how do I know whether to accept or reject these, and if it's "reject", what is different about them and the set of titles already in the database? Thanks. --MartyD 01:05, 13 August 2010 (UTC)

Well, our current plan is to improve foreign language support and then modify the ROA to open the floodgates, so I see no harm in relaxing the rules ahead of time. The biggest problem is likely to be the language barrier between editors and moderators, e.g. it turns out that Jean-Pierre Andrevon 's Repères dans l'infini is a novelette rather than a novel -- see Noosfere.

P.S. Foreign language support is still on the table, but it will take longer than expected. I plan to finish adding new fields to the Publication table first and then move on to the Title table. Unfortunately, I have had very little time for development over the last 4-5 weeks due to other commitments, but things should improve in October. And if everything goes according to the plan, I should have plenty of time for ISFDB starting early next year. Ahasuerus 02:11, 13 August 2010 (UTC)

I think we could do with some more French-Speaking moderators, and think Hervé might be getting there. BLongley 19:35, 13 August 2010 (UTC)

Yes, he's been doing very good work. --MartyD 10:05, 14 August 2010 (UTC)

Agree, though I haven't seen any multiple-step submissions or title merges from him. Anyone else? --Willem H. 10:53, 14 August 2010 (UTC)

The Robots Are Angry, Very Angry...

Fixer has submitted his Amazon UK findings for October, including a bunch of predictably dubious Angry Robot pubs. This puts us a couple of months ahead of schedule, so Fixer will try a variation on the "new algorithm" next. Ahasuerus 05:54, 13 August 2010 (UTC)

I've rejected the Jeter reprints as those are now put back to 2011, but some of them do seem to be according to the reschedule. I'd advise all moderators to compare with the Angry Robots website in the meantime, Amazon hasn't cleaned up past entries (although I think I have done so here). BLongley

... and yet another cleanup pass, this one for January-June, has been submitted. Ahasuerus 05:31, 15 August 2010 (UTC)

Error occured during Import

I was trying to import from 251593 into 37301, but I mis-copied the source tag (I apparently copied the author's name instead of the tag) and got the following:

http://www.isfdb.org/cgi-bin/edit/clonecontent.cgi

<type 'exceptions.TypeError'> Python 2.5: /usr/bin/python Sat Aug 21 00:30:38 2010

A problem occurred in a Python script. Here is the sequence of function calls leading up to the error, in the order they occurred.

/var/www/cgi-bin/edit/clonecontent.cgi in ()
 170         except:
 171                 publication = SQLGetPubByTag(FromTag)
 172                 clone_from = publication[PUB_PUBID]
 173 
 174         if form.has_key('IncludePages'):

clone_from undefined, publication = 0, PUB_PUBID = 0

<type 'exceptions.TypeError'>: 'int' object is unsubscriptable

A message like "Invalid source tag or record #" would be an improvement, but this isn't the problem I thought it was when I started writing this msg. -DES ^Talk 05:37, 21 August 2010 (UTC)

Yup, looks like a relatively harmless bug. Ahasuerus 19:40, 21 August 2010 (UTC)

Bug 3050469 Bug when importing from invalid tag created at source forge. -DES ^Talk 20:14, 21 August 2010 (UTC)

Cover images from OpenLibrary.org

User:Uzerzero has just submitted a pub update with a cover image hosted at OpenLibrary.org. Based on the statements at http://openlibrary.org/dev/docs/api/covers and http://openlibrary.org/about it looks as if they are basically ok with such use, but they do ask for a link-back and they aren't, I think, one of the sites we automatically handle in that way, nor are they on our whitelist of permitted image sites at ISFDB:Image linking permissions. Should I approve this assuming that the link-back code will be added soon? Should mit be added to the whitelist? -DES ^Talk 07:06, 21 August 2010 (UTC)

Sounds like a good idea. I'll see what I can do tomorrow. Ahasuerus 19:39, 21 August 2010 (UTC)

Thank you. I notice that openlibrary.org is sponsored by the Internet Archive, and that they are interested in bulk uploads. Perhaps we can work with them to some extant. They may also be a good bibliographic source. -DES ^Talk 20:06, 21 August 2010 (UTC)

Done. They would prefer that we link to the publication record, but it wasn't easy to do, so I linked to the main page, which they consider acceptable. Ahasuerus 03:29, 23 August 2010 (UTC)

Thank you. I'll approve the submission i had on hold. -DES ^Talk 06:38, 23 August 2010 (UTC)

Suspect Fixer submissions

I wonder if we should move the suspect Fixer submissions to Fixer's User page and delete them from the queue. I guess all we need to preserve are the ISBNs, which should allows us to go back and re-examine the books later. Ahasuerus 22:04, 22 August 2010 (UTC)

Bug report, author name problem

On the title display for "Poof!", if you click on the author's name, not at the top, but on the clickable link in the publication listing, you get to http://www.isfdb.org/cgi-bin/ea.cgi?John_O%E2%80%99Brien which is invalid and appears to be caused by an incorrect form of apostrophe. The odd thing is that when you follow the link to the Pub record, all the links seem correct and go to valid places. i can't find anywhere to edit to remove the incorrect form. The title record was just created, by me, as part of chapterbook cleanup, and the author name was a copy&paste from the pub record. -DES ^Talk 03:06, 25 August 2010 (UTC)

This looks more like a preschool book that should be deleted.Kraang 03:23, 25 August 2010 (UTC)

The ROA don't say the preschool books are OUT, and we have been adding a lot of them from fixer submissions of late. If there is consensus that such books should be deleted on sight, I'd like that made clear in the RoA. I'll take that to Rules and Standards. But the link bug is not a result fo the age level, and if it happened here it presumably could happen on a different book, too, so lets not delete this one until the developers have a chance to look at it. please? -DES ^Talk 03:39, 25 August 2010 (UTC)

It's a bug. I thought it was already listed on Sourceforge, but apparently not.

As far as preschool books go, the most recent bandaid solution was to ask Fixer not to submit them to avoid queue overload. Ahasuerus 05:53, 25 August 2010 (UTC)

Fixed in r2010-57. Sorry about the development slowdown. It's hard to think at the end of an 80 hour week. Ahasuerus 04:45, 6 September 2010 (UTC)

Adding new author and his bibliography

Dumb newbie mod question, for which I wasn't able to find a clear answer. How can I add a new SF author to ISFDB? His works are listed here: [5] Ruislick0 08:25, 27 August 2010 (UTC)

Hi, and welcome. Not a dumb question, as it is not obvious.... When you add a publication (i.e., using one of the "New Novel", "New Collection", etc. choices from the main ISFDB menu), several other things get added or used automatically: a Title, one or more Authors (these for both the author(s) and any artist(s)), and a Publisher are the primary ones. These are then used to construct the bibliography on the fly. If those other things do not already exist, they are created, and you can then find and edit them once the submission of the publication has been accepted by a moderator. If a particular title already exists, you can either add more publications to it (you'll see "Add Publication to this Title" when looking at the title), or you can "Merge" the title created when you added your new publication with the existing one (easiest way to do this is to use Check for Duplicate Titles when looking at the author's bibliography).

Here, for example, you could add First Ark to Alpha Centauri. If you do not have the book yourself, you could use information from Amazon and its Look Inside, citing those as the source in the Notes. Once that has been approved, you will find an author, "A. Ahad" (we enter names as given on the title page). You could then edit that entry to provide his full name and other pertinent details. Every editing screen has a link to online help to guide you, and you can post questions on the Help desk, too. Also, your submissions are moderated, so a moderator will look them over and give you a hand with anything that's not quite the way it is supposed to be. So go for it! :-) --MartyD 10:45, 27 August 2010 (UTC)

I expanded this answer into Help:How to enter an author into the database for future users with similar questions. -DES ^Talk 14:14, 27 August 2010 (UTC)

Cool. Thanx a lot, I'll get right on it. Ruislick0 17:30, 27 August 2010 (UTC)