Category talk:Articles with invalid ISBNs/Archive2

From Wikipedia, the free encyclopedia

 This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.
Articles with invalid ISBNs/Archive2

Contents

Status

Currenly about 1294 - less than half the recorded high water mark of 2719. Great work. See Dragon's flight Category tracker. Rich Farmbrough, 14:37 15 January 2007 (GMT).

I and J empty. Noticable how many of those left are "foreign" books we would find it hard to find a copy of. Incidentally I have found two "hoax" entries. Rich Farmbrough, 19:28 16 January 2007 (GMT).
Yes, the foreign books do take up a lot of this task. I have been very lucky in locating many. Not surprising, these page owners also drop leading and trailing digits. I think they fear ever using an "X". I am curious as to the "hoax" entries that you located. How about sharing those encounters? Ekotkie 15:47, 17 January 2007 (UTC)
Here and here. Rich Farmbrough, 15:07 18 January 2007 (GMT).

Sherlock moment

"OI Stories" found a photo of the back cover of the book, and read off the ISBN! Rich Farmbrough, 20:06 16 January 2007 (GMT).

Be careful

Just a note, be careful when removing the "Please check ISBN" tag, not to remove to many close brackets at the end, if it's nested in another template! Nearly under 1000! Rich Farmbrough, 23:58 18 January 2007 (GMT).

I have wondered about that. I have followed the word string back, trying to ensure just this point and found that it didn't take when I closed out the page. In many cases the word string never used a bracket at the beginning of the string. It should be no different then a math formula string but I have found it a bit puzzling. Am I not going back far enough? Ekotkie 08:16, 20 January 2007 (UTC)

SmackBot enhancement possibility for someday?

I know we've tried to talk about this before ... an official way of tagging ISBNs that are invalid as published ... For my most recent dialog with someone on the topic, please see this thread. Keesiewonder talk 00:15, 20 January 2007 (UTC)

This is a possibility, though it would just be a template that told SmackBot to ignore the ISBN. In general I think it's fair to still leave these pages in the Category, because they still have invalid ISBNs. The reader should be put on notice that something is the matter with one or more of the ISBNs in the article. This is still literally true, even if the editors prefer not to fix it. So if they are going to be left in the category, then it's an argument for leaving the regular SmackBot template in place. Perhaps instead of making a new template, we could just create a file listing all the invalid ISBNs that editors have requested to be left in place. This would leave open the chance of fixing them or at least re-checking them in the future. EdJohnston 01:26, 20 January 2007 (UTC)
This would be a great feature to have. It is a bit disturbing to find a check digit error that isn't supported by ISBN. I have seen cases where the ISBN is in 4-5 site locations yet errors out on the check digit. Has anyone in botland figured that out? Another plus would be that the bot could be checking Wiki all the time. Wiki could probably be tracked for any new inclusion of ISBN. Ekotkie 08:22, 20 January 2007 (UTC)
In principle, that may be possible, as I may be able to use the toolserver in a while. There is a crude method of keeping SmackBot off a whole article - use {{bots}} or {{nobots}}, at the moment SB will avoid any article with either of these templates if I remember to set the option :-). I have put it in the ISBN article myself. Rich Farmbrough, 22:59 20 January 2007 (GMT).

Mazo de la Roche

(I removed this from my Talk page and put it here as part of the experiment in not breaking up discussions. -EdJ)

Hello Droll! How would you feel about deleting the bad ISBN from this article? I don't want to do anything if you want to study it further. Since OCLC doesn't show an ISBN, I figure they realize it's bad. Also since it's pre-1966 it may not have a real ISBN anyway. National Library of Canada says it has LCCN 60-9741 if you think it would be helpful to add that fact. EdJohnston 02:52, 22 January 2007 (UTC)

I would be happy if you deleted it. I just left the note so that the next editor to check would know that he was not the first to try to find a valid ISBN. I read someones opinion somewhere that the first editor shouldn't do the deletion although I have deleted some on first site if they where to short to even resemble and ISBN after a conscientious search for a valid one. My personal feeling is that no data is better than bad data but that a good faith effort should be make to find valid data. --Droll 03:35, 22 January 2007 (UTC)

Nutrition and pregnancy

Can someone tell me what I'm missing here? Ekotkie 04:58, 22 January 2007 (UTC)

Someone added [[Category:Articles with invalid ISBNs]] at the bottom of the article which made it show up in this category. I removed the category and now it should be OK. I checked the only ISBN on the page. --Droll 08:39, 22 January 2007 (UTC)

Mughatil ibn Bakri

See my suggestion for how to fix this at [1]. EdJohnston 04:59, 22 January 2007 (UTC)

  • Yes, this is fine. Now that the "list" version of ISBNs that need checking is done, I'll be more present here when working on ISBNs. Keesiewonder talk 10:51, 22 January 2007 (UTC)

Listed Invalid ISBNs

Hi, there has been discussion on a number of pages about invalid ISBNs that are published, and whether to show, hide or remove them. Usually in these cases, the MARC records include the invalid ISBN (tagged as invalid), so the ISBN is valid for the purposes of searching for a book in most libraries. Oft OCLCs are being used as replacements for Invalid ISBNs, but to my knowledge MARC records do not list OCLCs, limiting their usefulness. To resolve this, and so that we dont loose too much information in the "Invalid ISBN Cleanup 2007", I would like to propose that where it is obvious, these ISBNs are retained using {{Listed Invalid ISBN}}, and we work out how to deal with these separately. My apologies in advance if this approach has already been rejected. John Vandenberg 08:54, 22 January 2007 (UTC)

I second this. Checksum invalid (but published) ISBNs are how we find books in actual libraries. Cleduc 09:40, 22 January 2007 (UTC)
Sounds like a fine suggestion ... as long as it is not over or prematurely used. Thanks for implementing a solution for all of us! Keesiewonder talk 22:58, 23 January 2007 (UTC)


Note: I've started a discussion on how to present this data more appropriately on Template talk:Listed Invalid ISBN. John Vandenberg 00:51, 24 January 2007 (UTC)

ISBN notation

I noticed using Google that there are over 400 hits for ISBN 13 using this URL. There are fewer hits for ISBN 10 but still over 300. Some of these are alright since an ISBN could start with "10...". I wonder if anyone has discussed this. The magic word does not work with these since the actual use is usually ISBN-13. I realize this is probably not the correct place to discuss this. Could someone point me in the right direction. --Droll 09:37, 22 January 2007 (UTC)

P.S. Check the article Otto Rahn to see what I'm talking about. --Droll 09:44, 22 January 2007 (UTC)

  • Sorry, I'm not sure what your question is. Could you please restate it? I do know that for me, Google is not a good way to identify or confirm an ISBN, regardless of whether it is of length 10 or 13. Keesiewonder talk 23:01, 23 January 2007 (UTC)
  • There are editors who use notation like ISBN-10 0-395-14806-5 and ISBN-13 978-0-395-14806-8. It is possible to find these using Google although there will be false hits. Wikipedia uses a "magic word" (ISBN) to identify ISBNs and generate HTML code. The ISBN-10 and ISBN-13 are not "magic words". I think that's a problem but then I'm a bit obsessive. Who else would try to repair all the invalid ISBNs. --Droll 01:33, 24 January 2007 (UTC)
  • I'm just wondering if anyone thinks its worth the effort to edit articles that currently use ISBN-10 and ISBN-13 notation and convert these to the standard ISBN notation. I mean what sence does it make to have the number in two formats neither of which generate a link. Look at the table in the Otto Rahn article. I think that sort of thing is counterproductive. --Droll 08:22, 24 January 2007 (UTC)
  • Ah! If it were up to me, I'd say yes, let's convert the ISBN-{10,13} style notation over to the standard, link producing ISBN notation. I have less of an opinion on whether we should present both the 10 and 13 ISBN. I guess I prefer just presenting the 10 if both exist. A bot can come through later and change to or add the 13s if need be. The references are listed in a completely non-standard WP way in the Otto Rahn article right now anyway ... IMO, it deserves a clean-up unless there is a special reason why the references should be listed in just the way they are now. Hope that helps, and thanks for your patience with me while I got up to speed with your query! Keesiewonder talk 09:56, 24 January 2007 (UTC)
P.S. I'd be happy to help you convert the (300-700) ISBN-X: style listings over if you'd like. Keesiewonder talk 09:58, 24 January 2007 (UTC)
  • I think we need more of a consensus on this before anything is done. I'm going to continue working on invalid ISBNs for the time being anyway. Lets wait and see what others have to say. --66.228.25.16 08:00, 26 January 2007 (UTC)
    Sorry forgot to sign in.--Droll 12:38, 26 January 2007 (UTC)
SmackBot will correct these, if I remember correctly. Rich Farmbrough, 18:04 27 January 2007 (GMT).

Archiving?

This page is now at 41 Kb. Does anyone object if I create a Talk archive? My idea is to archive old threads by subject. I would put the published invalid ISBNs in their own archive file, for instance. See User_talk:Keesiewonder#Archives for an example of archiving by topic. EdJohnston 16:01, 23 January 2007 (UTC)

  • No objections from ... Actually, while I don't object, maybe you could wait a bit, giving me a chance to work down the notes in this list before they migrate off to their own talk page? Like I'm finding some 'No ISBN Available' notes in the article that I'd like to switch to OCLC if possible. Just my two cents ... I'll follow your lead. Keesiewonder talk 00:25, 24 January 2007 (UTC)
  • Ok, EdJ ... I'm all caught up with what I wanted to do with this page. So, whenever John's ready, I'd say go for it. Keesiewonder talk 00:39, 25 January 2007 (UTC)
I would also prefer a few days to work through this talk page. If we all make sure we start new discussions where possible, we can archive the top half as-is once everyone has finished with the old entries. John Vandenberg 00:55, 24 January 2007 (UTC)
Take a look at Wikipedia_talk:WikiProject_Films and see how their archives are handled. If we do it that way there is no need to wait as access to the archive will be very simple. --Droll 01:39, 25 January 2007 (UTC)
btw, I'm happy for the archiving to go ahead now. John Vandenberg 23:13, 28 January 2007 (UTC)
Due to the popular demand, I did the archiving. Very gracefully done, as you can see from all the fixups of the links. This is not an archive by topic, I just cut off the top part of the file and archived it. Note the elegant archive box at upper right of the page. EdJohnston 04:14, 29 January 2007 (UTC)

Universal IDs for DVDs?

Since Droll and John Vandenberg raised this issue [2] over at User_talk:Jayvdb, I thought I'd try to continue the discussion here. Here is what they already said:

  • A few of the "unable to locate" articles have ISBNs that are supposed to refer to DVDs. I am not sure DVDs have ISBNs. --Droll 01:49, 24 January 2007 (UTC)
  • DVDs can have an ISBN, but usually when accompanied with significant supporting printed material. They also have their own identification system, and we are running into them a lot, so perhaps we need to start arranging how Wikipedia should handle them. John Vandenberg 02:18, 24 January 2007 (UTC)

Hey, John and Droll! This is my cue to point out the section of the ISBN manual [3] that limits usage of ISBNs with 'Videos'. They offer the following as eligible for ISBNs:

  • i. Educational videos specifically aimed at children up to the age of 16.
  • ii. Instructional videos imparting practical knowledge about specific subjects, the content of which is comparable to that available in printed books. For example, a video on how to play golf or how to make wooden toys.

They go on to say that general entertainment videos, including documentaries, are not eligible for ISBNs. Naturally we don't have to follow their rules, but if we want to harmonize our usage with that of other agencies, there could be advantages in doing what is done elsewhere.

So, what are the other players already doing (as opposed to what isbn.org *thinks* they should be doing):

  • Amazon lists DVD movies by their ASIN numbers, and those can be Googled. (They don't use ISBNs for DVDs).
  • Worldcat thinks that the Schindler's List DVD has both an OCLC code and an ISBN. However Amazon does not find anything when you search their holdings for that ISBN, and of course they don't know about OCLCs. And Worldcat does not track ASINs. So it seems there is no universal identifier for DVDs that will be accepted by both bookstores and libraries. Unfortunately. So one option is for WIkipedia to store both an OCLC and an ASIN for each DVD. What do other people think? EdJohnston 03:53, 24 January 2007 (UTC)
See {{Audiovisual works}} for a few related identification scheme's. John Vandenberg 04:23, 24 January 2007 (UTC)

A visit to my local library shows that 8 of 11 videos that I checked have ISBNs. (I looked at a mixture of DVDs and VHS tapes). These ISBNs can normally be looked up in Worldcat. So it appears that people are not obeying the isbn.org rules against ISBNs on entertainment videos! EdJohnston 03:05, 25 January 2007 (UTC)

That is an interesting point. I have an extensive DVD collection. Inspection of my inventory shows that the isbn shows up in support of the EAN bar code. isbn is configured as "X-XXXX-XXXX-X" as the top number. Unlike book bar codes that use the ten digit isbn on the top part of the bar code and the 13 digit isbn as the bottom part of the bar code number, the dvd's appear to have their own lower number. I have not looked at or seen a 2007 released dvd to see if this practice has been continued. I have seen scanners that are available with software that allows the user to "catalog" books, dvd's and cd's. This will remain as something to keep an eye on in the future. 66.242.209.79 15:53, 26 January 2007 (UTC) Interesting, one minute I was logged on and the next I wasn't. Sorry. Ekotkie 15:55, 26 January 2007 (UTC)

Admin needed

The article Tottenham Hotspur F.C. is fully protected and includes 2 invalid ISBNs.

This citation:

  • {{cite book | author=Osvaldo Ardiles | title=Ossie | publisher=Sidgewick & Jackson | year=1983 | id=ISBN 0-287-98872-X{{Please check ISBN|Calculated check digit (4) doesn't match given.}}}}

should read:

  • {{cite book | author=Osvaldo Ardiles | title=Ossie | publisher=Sidgewick & Jackson | year=1983 | id=ISBN 0-283-98872-X}}

and the citation:

  • {{cite book | author=David Bowler | title=Danny Blanchflower: the Biography of a Visionary | publisher=Orion| year=1997| id=ISBN 0-575-06404-4{{Please check ISBN|Calculated check digit (8) doesn't match given.}}}}

should read:

  • {{cite book | author=David Bowler | title=Danny Blanchflower: the Biography of a Visionary | publisher=Orion| year=1997| id=ISBN 0-575-06504-4}}

Thanks. --Droll 08:14, 26 January 2007 (UTC)

Article appeared to be only semi-protected when I stopped by, so I inserted your findings listed above. Keesiewonder talk 09:32, 26 January 2007 (UTC)
Thanks. I guess I wasn't logged on --Droll 12:41, 26 January 2007 (UTC)

14 Digit ISBN?

Please see this which is now in the article on Norwegian military operations abroad. Keesiewonder talk 23:57, 26 January 2007 (UTC)

Yep, that's what I was talking about before. I didn't recall which countries were talking about it or for what reason, but I did know it was European. See what I mean now? Ekotkie 03:10, 27 January 2007 (UTC)
Well, I did faintly remember someone talking about 14 digit ISBNs in Europe before, but, since I didn't immediately find the post, I started this thread. It doesn't surprise me one bit that the length of the number will grow. It has happened with zip codes and telephone numbers too. For me, it is no big deal, really; just a part of life. I'm not complaining that 14-digit ISBNs are on the horizon; I just wanted to confirm that they have already been seen in print. If you want me to re-read your previous posts on this topic, please provide a link or tell me which archive it is in or something. Thanks! Keesiewonder talk 09:58, 27 January 2007 (UTC)
Once again, I stand corrected. I just located this explanation from Google and it appears to be the best one I have seen to date on this subject. http://www.isbn.org/standards/home/isbn/transition.asp One more cog in the seemingly never ending process. I hope Rich will agree that this needs to be include this Wiki page for ISBN. Ekotkie 16:21, 27 January 2007 (UTC)

I have emailed Odin for a correct ISBN. While it is possible that ISBNs will be extended again, or subsumed in an even larger numbering system, it is a bit WP:CRYSTAL. Rich Farmbrough, 18:30 27 January 2007 (GMT).

There is something called EAN/UCC-14 on the page that EdK pointed to. However I don't believe that such a pending standard will affect ISBNs. That 14-digit thingie will put another leading digit in front of the current EAN that could be used to distinguish the ways a given product is batched. (See www.gtin.info). We already have an article EAN.UCC that doesn't mention the length issues, 13 versus 14. That article could probably be enhanced. Also ISBN could be improved to mention EAN/UCC-14. EdJohnston

Reply from Odin Dear sir, This is an unfortunate misprint. Correct ISBN is 978-82-7924-058-7. Yours, XXX XXXX XXXXX The Norwegian Ministry of Defence The Documentation centre Rich Farmbrough, 22:34 29 January 2007 (GMT).

Hidden HTML text -- Please be very careful

FYI, there is a world of difference between placing HTML tags in their persnickety way, and not. I just identified a previously elusive ISBN ... and the visible article grew after my correction. If you want to place html code, it absolutely must start and end with the characters as shown:

<!-- your text goes here; formats nicely -->

For reasons I cannot quite explain, if the HTML is instead like this

<!-- bad format; do not use --text>

Everything looks fine in the article, except that absolutely everything displayed after the above is invisible to the reader. For an example, please see the edit history of FA Norman Borlaug. Kind Regards, Keesiewonder talk 10:26, 27 January 2007 (UTC)

Each of us who has been using hidden text should consider going through our contribution history to see if more cases like this could have happened. It is likely that every such page would still be in the Category, so that should help. EdJohnston 15:29, 27 January 2007 (UTC)
What am I missing here? I just took the above, "bad statement" and dropped it into the ref. section of one of our invalids that are here and saved the page, and reloaded the page to ensure that it took and the statement remained hidden.....Am I being thick headed this morning? I know I need some coffee which I will go make now while someone explains this. Ekotkie 16:37, 27 January 2007 (UTC)
It's hard to explain. Try looking at this link, and scroll down to the bottom to see the last lines of text displayed in the article to the reader. Now try looking at this other link. At least on my computer, the second one has lots of references and external links that the first one has in its code but does not allow to display. Keesiewonder talk 16:43, 27 January 2007 (UTC)
I see says the blind man. It is when I placed the "EdK" up against the ">" in lieu of "EdK-->" (I think this is what he is saying and may be the cause.....sigh. OK Ekotkie 16:59, 27 January 2007 (UTC)

Yeah, that's what she's (me) saying! :-) ... No worries ... Later today I'll see if I can figure out a way to search for these. Keesiewonder talk 17:21, 27 January 2007 (UTC)

She...she....she...so much easier to type he....I don't think it is an issue any longer. The err'ed ones are in Cat Invalid and I just looked through the ones I left and "someone" nicely changed the ones I had left pending. Ekotkie 17:32, 27 January 2007 (UTC)

{{Listed Invalid ISBN}}

I copied the following from my talk page.

Hello Droll. Since it seems this much-discussed template is being put into use, how about discussing minimum standards for using it. I'd claim that either (a) an editor claims to have seen the book with the bad ISBN, or (b) the bad ISBN is found on line (not in a WP-derived source). Otherwise we might just be perpetuating somebody's copying error. And we're claiming to be a reference work! (Though perhaps not quite as reliable as the phone book, at least better than rumors from the neighbors). What do you think? Require some evidence of the ISBN being used (even though wrongly)? EdJohnston 07:06, 28 January 2007 (UTC)

I figured this template was already being used so that is why I started using it but your concern and request are valid I think. I have a couple of comments:

  1. Using this template is better than using "hidden text" because it is easy for anyone to review using "what links here".
  2. My personal criteria so far is that I was able to find the invalid ISBN on at least 2 unrelated google hits that are not wiki hits or wiki clone hits. At least once there was a hit on amazon.com? Several times I found the invalid ISBN on the publishers web page.
  3. This template should only be used after a conscientious effort has been made to find a valid ISBN. I mean every tool available should be used short of searching each library with an online catalog.

A few other thoughts:

  1. I have run across more than a couple of ISBNs that begin 999... In my limited experience there are no valid ISBNs that start that way. Is a practice some publishers use???? Most of the time I was able to find an alternate ISBN.
  2. I saw some discussion about using "hidden text" and the possible consequences. I was thinking about the idea of creating a {{No valid ISBN found}} template. At least if you don't close a template it shows up and probably will be corrected. After google does a fresh crawl of wikipedia this kind of error could be found using google. I'd be willing to go back and change my hidden text to a template if this idea is agreed on. It would give us a record of these cases. It should probably have a field to identify the editor in some way that use the template. I started leaving my user name in "hidden text" comments so that if I go back to and article I can tell if I'm the first or second editor. I feel that invalid ISBNs should not be deleted by the first editor except in the few cases where the number is clearly not is ISBN or and ISSN or a OCLC. I don't think there is any policy about that yet. Ed also has been identifying his comments.
  3. Can you tell I'm a little obsessive. --Droll 10:26, 28 January 2007 (UTC)

Thanks for starting the thread!

  • As I understand the genesis of this template, it is a direct result of the inadequacies of placing information solely in hidden html text.
  • One observation I've had is even when the template is used, if the "ISBN" has less than 10 digits, the text displays (as normal, un-special text), but not as a hyperlink. I do not think this is a problem, but, it did trip me up a bit this morning when I was wondering how something was working.
  • The new Listed Invalid ISBN tag is perfect for works that really do exist with an ISBN, published by the publisher, with an invalid check digit. It allows us to display the publisher's information so the reader can take it to a book seller or library. It also allows us to not constantly go back and check whether the ISBN is valid or not. As I understand it, when this new tag is used, the "invalid" ISBN embedded in the tag is invisible to SmackBot.
  • While I am very glad this new tag exists, I have serious concerns about the general public using it.
  • I am concerned about the tag being set in place prematurely.
1) Those who know my work with ISBNs can attest that most of the ones I have fixed/found are ones that had stumped previous detectives. I enjoy the wikignome activity, but, need to mention details to stress a point. For say 19 out of 20 cases, when someone else says the number cannot be found, I can find it (or its commonly approved alternate). Personally, as I participate in the ISBN clean-up project, I'd rather not have to check articles using this new tag plus those using SmackBot's tag.
2) There's a subtle meaning to the essence the tag is getting at. I believe it is this:
  • If the publisher really did publish a number that does not have a valid check-digit, then using the tag is appropriate. However,
  • if someone just wants the world to think they have an ISBN for something when they do not;
  • or, if sincere but not yet adequate, rigorous checks have been done to identify the proper ISBN (or an approved alternate such as an OCLC);
  • then using the tag is not appropriate. Using the tag inappropriately may cause the articles using {{Listed Invalid ISBN|...}} to very well become the next growing category of things to clean up here ...
  • IMO, We need to be careful when using this tag. Keesiewonder talk 12:37, 28 January 2007 (UTC)
999 It is indeed legitimate for a 10 digit ISBN to start with 9990, 9991... 99954. There is a formatting anomaly in the IIA's list at 99951. If I remember correctly these language areas are very obscure by our standards. Rich Farmbrough, 12:22 28 January 2007 (GMT).
Take a look at isbn.org's list to see the codes that are sometimes called 'language area codes' but isbn.org insists on calling 'group identifiers'. Anything that literally starts with 999 has a five-digit group identifier; they range from 99901 (Bahrain) to 99955 (Srpska). EdJohnston 16:02, 28 January 2007 (UTC)

Random Thoughts

Just my perspective on how I work on ISBNs ...

  • I find it silly to try to stay out of each other's way when fixing ISBNs. This is a wiki environment, after all, where anyone can edit and is encouraged to do so boldly. There are good instructions that pop on the screen for moments when we may get in to edit wars with one another.
  • I don't post here when I complete a certain letter since it seems easy to see what articles remain in the category.
  • I do appreciate the html tags so I can see whether when I start work on a problematic ISBN, I can tell if it is a new check or one that someone else has already fought with.
  • Posting problematic scenarios here seems like a great idea; it'd give all the ISBN editors plus whomever is concerned about the proposed changes to a particular article a centralized place to work matters out. Of course, another good place is the talk page for the article with the ISBN in question.

Keesiewonder talk 11:51, 28 January 2007 (UTC)

Mens Sana Research Foundation

This may be an AfD or even a speedy delete candidate. Please see Talk:Mens Sana Research Foundation. Keesiewonder talk 15:13, 28 January 2007 (UTC)

  • I've removed the ISBNs tagged in error since this really is a journal (with a valid ISSN). I'd leave a note about my AfD idea on the article creator's talk page, but, their last edit was October 13, 2006. Keesiewonder talk 15:29, 28 January 2007 (UTC)
Agree that this should be a candidate for deletion (WP:NOT a directory). I even wondered about Midnight House which looks like a publisher's catalog, but should be keepable because we have articles on many of the listed authors. This is not the case at Mens Sana Research Foundation. EdJohnston 16:19, 28 January 2007 (UTC)
Ok; done; it is the first AfD I've listed ... hope I didn't screw something up ... Keesiewonder talk 16:35, 28 January 2007 (UTC)

Featured Articles should not include any {{Listed Invalid ISBN}} citations

Wikipedia's reference lists should be of high quality, and many of us have been trying to improve them. Here are some questions:

  • What is the degree of checking we should do before listing a book in a WP reference list? (to be sure we are supplying correct information)
  • Are there any situations where it is actually beneficial to the reader to be told about an invalid ISBN for the book?

Here is my proposed descending scale of quality of the reference:

  1. The citation in WP includes a valid ISBN which retrieves proper book details from a major reference provider, such as a library OPAC.
  2. The previous WP editor supplied a valid ISBN for the book, but no online book details can be found anywhere, not even in Worldcat. So we're not sure that the book actually exists. At least, there is nothing that another editor can check, and there's no obvious way to find it in a library.
  3. The book is listed in WP with an invalid ISBN. The bad ISBN can be retrieved from SOME online source. Brief research does not reveal any other ISBN for that book.
  4. The book is listed in WP with an invalid ISBN. Nothing about the book can be found online, neither its ISBN nor anything else.

I believe that a #4 reference should *not* be supplied with {{Listed Invalid ISBN}}. I just don't think it provides enough value to our readers. The invalid ISBN is completely speculative information that is not confirmable by other editors.

I'm not sure about #3. I'd prefer to just drop the ISBN in those cases, but would listen to arguments. In going through WP reference lists, the #2 situation comes up surprisingly often. I would argue that Featured Articles should not include references of lower quality than #1. By introducing the {{Listed Invalid ISBN}} we are adding a template that would have to be *removed* before the article becomes a Featured Article. In my opinion :-). Let me know if you would actually feel comfortable submitting something to the FA process that included such a template. EdJohnston 17:57, 28 January 2007 (UTC)

  • Perhaps this edit of mine was your inspiration; this is a FA. Since there are OCLC's, I do not feel strongly about keeping the ISBN. The two libraries I mention in my edit description do list it, as invalid. Phone rang; more later. Keesiewonder talk 18:16, 28 January 2007 (UTC)
  • Curious....to a degree, has anyone considered the fact that this group is working somewhat in a vacuum? What will keep the originator of the false data just from seeing the change and reverting the information? This could easily come from a lack of understanding of the overall changes to the ISBN process. This is another reason why I asked Rich to get the bot to trigger a scan on any page addition of ISBN. At least at that point a human could step into the loop and explain the problem to the user. We have all seen a multitude of sins being made to create an invalid ISBN. For sure, we will not see a perfect world on this point. Just a thought. Ekotkie 20:51, 28 January 2007 (UTC)
Of course our ISBN changes may be reverted. I barely know of a single case where that has happened, though. For any that may be controversial, dialog with interested parties seems to work well for me. Now that we have the new Listed Invalid ISBN tag, when it is used appropriately, there really should be no problem. Anyone can edit, and is encouraged to do so boldly. Keesiewonder talk 21:48, 28 January 2007 (UTC)
  • Responding to EdJ's list above ...
1) no-brainer for me; certainly list the ISBN;
2) these situations are tough for me; I can go either way; if a) the rest of the article is verifiable, b) there are other resources in the article that are ones we can find, c) multiple proponents of the article attest to the book's existence, d) the ISBN makes sense given its embedded country code, etc., I'm ok with it staying. Have we had an example of this? If so, please forward it.
EdJ>Look at the ISBNs in Mens Sana Research Foundation. At present they are all #2s, all valid ISBNs, but the books are not found in Worldcat. EdJohnston 18:12, 29 January 2007 (UTC)
Yes; how about an example from a situation that is not bound to be deleted post current AfD discussion. Keesiewonder talk 18:19, 29 January 2007 (UTC)
3) essentially a no-brainer for me, and a perfect scenario for the new {{Listed Invalid ISBN|...}} For instance, in the following example, the second OCLC in Harvard's library shows that at least there, the ISBN is labeled as "invalid," yet is still listed in the catalog.
  • Fundamental Surprise: Intelligence in Crisis by Zvi Lanir. ha-Ḳibuts ha-meʼuḥad, Tel-Aviv, 1983. (In Hebrew). ISBN 96502004, OCLC 65842089, OCLC 12420401.
  • Click here and scroll down to see what I mean.
4) I guess the only difference between this and # 2 is that here the check digit test fails and in # 2 it passes? I'd follow the same reasoning as in # 2, except if we cannot find anything and there are no proponents, then it sounds like if the citation remains listed at all, it needs to do so without the ISBN, regardless of whether it is embedded in our new tag or not.
Keesiewonder talk 22:06, 28 January 2007 (UTC)
  • I disagree with this blanket statement of this section header ("Featured Articles should not include any {{Listed Invalid ISBN}} citations"). In what way would this rule help the quality of articles, vis-à-vis how helpful an article would be for a reader to either find a book they want to read, or verify a source? I can easily think of instances where it would make sources more difficult to verify, and books more difficult to find. Aside from a Levitical obsession with "rightness", what useful purpose does this the overall restriction serve? Cleduc 17:58, 29 January 2007 (UTC)
  • Yes, I guess I too disagree with the blanket statement. I like trying to use the tag as appropriately as possible, but, as with most things, this task may be more an art than something that can be put in to a formula. Check out my Yom Kippur War example above. I have not heard or thought of any good reasons for not keeping the invalid as published ISBN in the citation I recently updated. Keesiewonder talk 18:23, 29 January 2007 (UTC)
Hello Keesie and Cleduc. Do you really believe the user would *prefer* to have Author+Title+Invalid ISBN  rather than plain Author+Title? In the Yom Kippur War book, the alleged ISBN was 96502004, only 8 digits. Is this helpful?
I don't believe the user *cares* whether the ISBN is "invalid" as long as it leads to them being able to find the book in a library or bookstore. If it makes that easier, then I think the user does find this helpful. I, as a user of Wikipedia, find this personally quite useful. Cleduc 22:45, 29 January 2007 (UTC)
From WP's point of view, it is extra work to carry along the invalid ISBNs and carefully protect them from the checking-and-fixing process. The Library of Congress tried to keep track of good and bad ISBNs for a while, but seems to have given up. Do you know any major libraries that you think are doing a better job thanks to their patient handling of invalid ISBNs? I know there are some libraries in New Zealand that seem to be tone-deaf to bad ISBNs, but I don't consider this a contribution to the world. There are really three choices for a library: (a) be oblivious (the 'NZ' option), (b) painstakingly track the good and the bad, and (c) delete the bad (apparently Worldcat's choice, and one that I recommend for us). EdJohnston 18:47, 29 January 2007 (UTC)
I see no evidence that Worldcat has deleted checksum-invalid ISBNs, since several can be searched there. BTW, I speak for myself, not WP, but I would like to point out that WP has been very good at applying a lot of (wo)manpower to time-consuming problems. I don't think a blanket solution is necessary or helpful, and an obviously useless 8-digit number is hardly the bedrock on which to rest an argument. Cleduc 19:00, 29 January 2007 (UTC)
  • I think my Yom Kippur War example above is an example of WorldCat not keeping the invalid ISBN around. Click on either OCLC I provided, and all you see is an OCLC in WorldCat. Proceed to one of the libraries where the book is found and you may or may not see the ISBN listed. U of MI lists it. Harvard lists it, and indicates it is invalid.
  • Yes, the ISBN published on the book is helpful, even if its check-digit does not pan out. Especially if all I have is the ISBN for the book I want - sometimes, (often times?) that is all people exchange when suggesting books to one another. Also, not everyone knows how many digits a "good" ISBN has. If I were the author of a book with an "bad" ISBN, I would still like it displayed. It is not too, too much different if the real surname of an author, is, say Spier, but, for who knows what reason, it was published under Speir. It is what it is ...
  • During my ISBN work here, I have run into a small handful of books that have Invalid, Yet Real as Published ISBNs. For a subset of those, the WP editors have specifically requested that the invalid as published ISBN remain in the display. The new tag allows us to accommodate this reasonable request (i.e. an encyclopedia putting forth basic, true information) while also allowing us to behind the scenes distinguish Invalid as Published ISBNs from other ISBN mistakes (typos, fabricated books, ISSNs in disguise, DVDs, etc.). If we don't tag these invalid as published ISBNs, we're basically setting ourselves up to waste our own time or the time of the next crop of ISBN fixers who will time after time tackle the list, only to come to where we are now. The only concern I have about the new tag, if we adopt its usage, is we use it sparingly and appropriately so we don't just create another clean-up area.
  • I don't feel particularly strong about any of this ... Perhaps what is more concerning to me is it is beginning to feel to me like the new tag was implemented prematurely, since, after it has been out and used a bit, we are now discussing whether it is needed. Shouldn't we have had this conversation before the tag was created, not after? Keesiewonder talk 22:58, 29 January 2007 (UTC)
Well, the tag was in part created in response to the summary removal of "invalid" ISBNs from articles. I view the tag as a tool to allow us to continue to work as we try to reach a consensus on how ISBNs should be used in WP. I feel very strongly about the user experience and the utility of the information we provide in WP. Cleduc 02:22, 30 January 2007 (UTC)
  • Ah - so the tag was created because some editors/readers want to see the invalid ISBNs?! Which is one reason why you and I want it to stay, if I understand you correctly. You may also be interested in this. Keesiewonder talk 10:43, 30 January 2007 (UTC)
Exactly. Cleduc 15:29, 30 January 2007 (UTC)

Open Library

Hi, I just ran into http://openlibrary.org, and thought others here may be interested. WP contributors are starting to use this resource with 5 hits. The site is part of Internet Archive and a member of Open Content Alliance [4]. The books appears to be identifiable by (either or one of?) two schemes: number and adhoc names. Any clue what the number "117770257" is? Here is an OCLC 1354145 for the book; havent checked which edition it is. John Vandenberg 23:31, 28 January 2007 (UTC)

  • Wouldn't Google hits be better represented by a search like this? Keesiewonder talk 23:37, 28 January 2007 (UTC)

Vladimir Antonov (scientist)

One tagged ISBN is for this entry:

  • [http://upanishad.swami-center.org/ ''New Upanishad: Structure and Cognition of the Absolute''] (1999) ISBN 5-89455-005-1{{Please check ISBN|Calculated check digit (X) doesn't match given.}}

The web site displays the ISBN cited as invalid ... but I have to admit this feels like a very unofficial, even self published, translation. But, I could use the Listed Invalid ISBN tag, and go with what is on the web site. I also found ISBN 5-89455-0025 or OCLC 51996100, but the latter seems to be a slightly different book ... even to someone who doesn't read Russian ... Keesiewonder talk 00:18, 29 January 2007 (UTC)

This doesn't look like a printed book, it's more like a blog posting. It would be a 'type 3' reference in my proposed system, quite a low grade. I wouldn't suggest using {{Listed Invalid ISBN}} when the reference is of such minimal quality and there is no library which associates that ISBN to a book that they hold. If somebody *referred* to that book in a peer-reviewed source, the situation would be different. Also would be different if the book were in Worldcat. Remember that the WP editor who created the article may have done nothing more than look at the web site, and he or she may have no other evidence that the book exists. When things don't turn up in printed sources, or in libraries, it may be for a reason. EdJohnston 03:31, 29 January 2007 (UTC)
I did a brutal POV edit on this article. It seemed very self promoting to me and make unverifiable claims. I just checked the article's history and a lot of reference information was added by user:Spiritualheart. Spiritual Heart. The Religion of Unity is one of the books listed in the bibliography. It seems like it might be a candidate for deletion. --DRoll 03:42, 29 January 2007 (UTC)
I sympathize with your view. There is a wikilink to an article on the same topic on the Ukrainian Wikipedia, and that version seems to be nominated for deletion. (There's a little garbage can with a question mark on it, and a {{delete}} template at the top). EdJohnston 04:23, 29 January 2007 (UTC)
Thanks for that input. I'm hesitant to go so far as an AfD in the English WP since several works do indeed show up in WorldCat (not all at this link are V.V.A.'s, but many are). Perhaps the ones that are "unpublished" English translations are, for now, the only ones that should go. Keesiewonder talk 12:11, 29 January 2007 (UTC)
P.S. -- I wouldn't describe those edits as brutal POV, though. Read the talk page of something like New Antisemitism or Global Warming for discussions about more brutal POV edits than yours! ;-) Keesiewonder talk 12:17, 29 January 2007 (UTC)

Smackbot chopping

Rich, I saw two more events today where the bot chopped into an ISBN. In one case it moved a correct check digit to the far right and in the other it split the isbn in half. In the later case the isbn was bad, but all the same, it got chopped. Ekotkie 22:10, 30 January 2007 (UTC)

multiple editions for the same work

Do we really need to list multiple editions for the same work? See Mouloud Mammeri for an example. One of the books that remains tagged with an invalid ISBN is one where the article includes 3 ISBNs, 2 of which have been corrected. I suppose we could list all 8 that WorldCat sees, but why do so? How about we list just the most recent, which is probably the one most likely to still be in print? Keesiewonder talk 23:41, 30 January 2007 (UTC)

Sounds like the right thing to do. In rare cases someone will have used page numbers in an article, and if so then using their specific edition might be necessary. In the case of this article there are no page numbers. EdJohnston 23:47, 30 January 2007 (UTC)
I've remembered that often times it is useful to retain the year of publication for the first edition. So, for the article mentioned above, I am listing the original and most recent publication year and the ISBN that goes along with the most recent year, i.e. like this ...
  • L'opium et le bâton: roman (in French). (1965/1992). Paris: Éditions La Découverte. ISBN 2707120863.

OCLC and ASIN have tags for wikilinks

Please remember to use the OCLC and ASIN tags if you cite those catalog numbers. I just fixed two articles that were converted recently that attempted to have the linking work by issuing this OCLC: 25048682 when it should be {{OCLC|25048682}} which yields OCLC 25048682. Keesiewonder talk 01:47, 1 February 2007 (UTC)

ISBN starting 9996 and 9999

Today I ran across an ISBN starting 9996 and another starting 9999. I found one at amazon.com and the other at amazon.co.uk. isbn.com doesn't seem to think they're valid. I marked them as {{Listed Invalid ISBN}}. One was in Shane and Sia Barbi and the other in South Carolina --DRoll 07:09, 1 February 2007 (UTC)

I'd suggest you consider removing these ISBNs. They correspond to no region code or publisher code, and those ISBNs are not found in any libraries that were checked. They seem to be purely made-up numbers, since there is no chance they were officially issued. Some time back we came across made-up ISBNs beginning with 6, used on non-book items. EdJohnston 19:07, 1 February 2007 (UTC)

23 more AfDs per WP:NOT#DIRECTORY?

These seem pretty useless to me ... What do you think?
Spriggan (list of ISBNs)
Unpopular Books
Keesiewonder talk 11:01, 1 February 2007 (UTC)

Both of these technically violate WP:NOT. The first I might cut some slack for, since it is a fan product not a publisher catalog, and these books might not be easy to locate. The second is technically a publisher catalog. An AfD is more likely to succeed with the second, I imagine. EdJohnston 15:27, 2 February 2007 (UTC)
There's also The Cambridge Edition of the Letters and Works of D H Lawrence. Keesiewonder talk 16:17, 2 February 2007 (UTC)
I fixed the ISBNs. But someone who has the patience should rewrite this article with just a couple of links into the Cambridge U.P. web site and remove all the individual books. Cambridge provides all the same information, nicely formatted, on their own site. (Unusual since publishers' sites are typically so unhelpful). EdJohnston 17:25, 2 February 2007 (UTC)
  • I just noticed that Unpopular Books survived an AfD September 20, 2006. So, I guess I'll just opt for fixing ISBNs in both Unpopular Books and Spriggan (list of ISBNs). Thanks for the corrections in the D.H. Lawrence one. Keesiewonder talk 22:02, 2 February 2007 (UTC)

Roy Keane

There's one problematic ISBN on this article's page and so far, neither Droll nor I have been able to find it. I've just left a couple messages on the article's talk page and on the talk page of the user who added the book. Any other thoughts? Keesiewonder talk 11:01, 2 February 2007 (UTC)

I just read through the user's talk page and went through the user's contributions. Taking what I found and the title given for the book it seems reasonable to assume that the book does not exist. I will remove the reference. --DRoll 14:00, 2 February 2007 (UTC)
Excellent; that was kind-of my thought too but I wasn't bold enough to implement it. Thanks for taking care of it! Keesiewonder talk 16:06, 2 February 2007 (UTC)

well done, a whole year it took you ! Unsigned comment at 11:34, February 6, 2007 from Aldini98

Todd Kessler

The books in the bibliography, all with invalid ISBNs, are not even mentioned on Kessler's "official" web site ... ? ... I expect these are all self-published materials. Thoughts? Keesiewonder talk 11:37, 2 February 2007 (UTC)

I spent a good amount of time trying to find reference to these as I am sure you have. I suggest the bibliography be removed. --DRoll 14:19, 2 February 2007 (UTC)
I removed the ISBNs. If you'd like to go so far as to remove the bibliography entirely, be bold! Keesiewonder talk 16:11, 2 February 2007 (UTC)
Done. Did anyone get the impression this whole thing is a spoof? He's not really a motivational speaker, just a satire on them? His website actually sells punk rock records, and he doesn't even perform on them. His alleged books don't exist in any bookstores. Google can't find evidence he has ever been on QVC. An erroneous article in Wikipedia? I'm shocked.. EdJohnston 05:16, 3 February 2007 (UTC)

Well done ladies, its only taken about a year !!!! Aldini98.

FWIW, I associate User:Aldini98 w/Roy Keane, not Todd Kessler. Keesiewonder talk 12:07, 5 February 2007 (UTC)

Risk management

It's curious that the only edit someone made to WP is to insert the book currently in the risk management article that has an invalid ISBN. The author, F. Antonius Alijoyo, certainly exists and has written at least one other well known book (ISBN 9796911264). But,I have not yet been able to find the one listed in the article. Thoughts? Keesiewonder talk 02:11, 3 February 2007 (UTC)

IMHO this one doesn't deserve being kept as an 'official' invalid ISBN. (a) wrong length, (b) not found in a library, (c) not found in a bookstore. Suggest removing the ISBN, since it's not attested in any way. EdJohnston 03:07, 3 February 2007 (UTC)

Being bold and agreeing with EdJohnston I removed the invalid number and left a note on the talk page below the one Keesiewonder had already left. I reaffirmed that the book's existence should be verified. I played with the number using ISBN check but came up with nothing, --DRoll 06:03, 3 February 2007 (UTC)

Butterflies

I left a note at the WikiProject Lepidoptera talk page about the seven butterfly artilces with invalid ISBNs. I'm just going to wait 24 hours and then edit the articles if no one beats me to it. --DRoll 05:42, 3 February 2007 (UTC)

It appears that for every single one of those listings, we already have a valid OCLC. I recommend just removing the invalid ISBNs. Keesiewonder talk 10:11, 3 February 2007 (UTC)
There is a later edition (1999) of J-C Weiss's book with a valid ISBN [5]. If I don't get an objection from the butterfly people, I'll update all seven articles with the newer edition+good ISBN. EdJohnston 21:15, 3 February 2007 (UTC)
It seems to me that there are four distinct books in a series. I could be wrong. I'm just going to remove the ISBNs today.--DRoll 02:30, 4 February 2007 (UTC)
The new ISBN looks good to me. Go for it. --DRoll 03:00, 4 February 2007 (UTC)

physics & physics/wip

FYI ... I have not looked high and low yet for this one, but have left a note on the article's talk page. Keesiewonder talk 20:34, 3 February 2007 (UTC)

issn.org has stopped providing a free ISSN-checker

This Talk page is really for ISBNs not ISSNs, but maybe the collective wisdom here will have an idea on how to solve this new problem.

John Vandenberg, Rich F. and I were looking at ISSNs some time ago and we were using an on-line ISSN-checker at issn.org. The one we were using has just now become a dead link: http://www.issn.org:8080/English/pub/faqs/issn/issnchecking. If you go on issn.org's site they now present you with a login screen before you can check anything, and a free account only lasts thirty days. Some Google searching indicates that the libraries of the world aren't offering an online ISSN-checker anywhere. (Maybe the check digit is patented! (Just kidding..)

I've read somewhere that the ISSN check digit uses the same algorithm as the ISBN-10, though the number is only 8 digits. Another possibility, for those who don't mind running Javascript, would be to add a button on some Wikipedia page (like Booksources) that would verify a submitted number. There is a site that offers some Javascript for the purpose at [6].

I don't think SmackBot is affected by this issue because it uses its own algorithm when populating Category:Articles with invalid ISSNs.

Still another method is to use the old familiar isbn-check.com, and convert the ISSN to a pseudo-ISBN by prepending '00' in front of the ISSN you want to check. (This worked in at least two cases, so it must be right!). If you have any other ideas, please let me know. EdJohnston 22:00, 5 February 2007 (UTC)

Thanks for bringing this up; someone has commented on the ISSN article that it is dead [7]. John Vandenberg 00:26, 6 February 2007 (UTC)

self-organization

What do you think of this? I'm not accustomed to PhD dissertations having ISBNs, but, if you follow the link to ProQuest, an ISBN is listed there that does pass the check digit test. Keesiewonder talk 00:18, 7 February 2007 (UTC)

Seems OK to me. Finding an ISBN at ProQuest is an on-line verification that the ISBN is real, assuming that ProQuest is a recognized custodian of doctoral theses, which I believe is the case. I Googled for 'ProQuest ISBN 0-496' and got 1,170 hits, so maybe ProQuest is assigning ISBNs like a publisher. EdJohnston 02:16, 7 February 2007 (UTC)

The last two: Polikarpov I-180 and Polikarpov I-185

FYI, I have left messages in 3-4 places in hopes of getting some help with the remaining 2 ISBN errors ... [8] & [9]; [10]; [11]. Keesiewonder talk 23:56, 7 February 2007 (UTC)

Discussion is taking place at [12]. The two people I contacted kindly replied ... but a solution is not yet clear. Keesiewonder talk 02:25, 8 February 2007 (UTC)
I looked at the scan, hoping for a photo of the bar code, but no luck. On another front I have changed the template to only list mainspace articles, so these really are the last two! When I recover my other PC I'll do another run, and see what we get... Rich Farmbrough, 10:34 9 February 2007 (GMT).

I've found one online listing that lists the same ISBN, and note that WorldCat lists some other Tankomaster publications with OCLCs, but not this one. I'm leaning toward listing this with the Listed Invalid ISBN tag ... How do others here feel about that? Keesiewonder talk 12:54, 9 February 2007 (UTC)

My preference would be to use Listed Invalid only when a reliable source (such as a library OPAC) connects the specific book with the given ISBN. Otherwise we're just taking the original editor's opinion that the supplied ISBN is correct, and there is no way for any other editors to verify it. Also, in this case there is a scanned version online; that will meet the needs of most readers who care to verify the book details. EdJohnston 16:02, 9 February 2007 (UTC)
So you'd recommend removing the ISBN all together and providing a link to the pdf containing scanned Russian material? That's fine with me ... if I can get the folks who care about the article to agree. Keesiewonder talk 17:56, 9 February 2007 (UTC)
After thinking a bit more about this, I am not comfortable listing the scanned version as a link since I am not able to read Russian and confirm that the material truly is in the public domain. I expect since it is for sale as well, albeit in a monograph form rather than html pages, it is not really supposed to be available for free online. Keesiewonder talk 23:12, 9 February 2007 (UTC)
I've removed the ISBN from these two articles since we have not yet been able to verify it. So, at the moment, there are Zero (0) articles flagged with bad ISBNs. Please help me out if the Polikarpov ISBN returns. Keesiewonder talk 19:11, 10 February 2007 (UTC)

Todd Kessler AfD debate here

User:Droll first ran across this article when he found several wrong-length ISBNs. After further study, it appeared that the books might not exist at all. I removed them here. Finally perceiving all the jokes in the article, I decided it was a hoax, and applied the prod tag. Since someone objected, the prod was removed, and now it's at AfD here. Anyone who has an opinion is welcome to join in the discussion. EdJohnston 17:22, 12 February 2007 (UTC)

Unexpected success: the category has disappeared!

Category:Articles with invalid ISBNs was deleted on 5 March, presumably because it had no entries. Does anyone know if this Talk page will also disappear? I hope that can be prevented. We should be getting new entries in the category when SmackBot begins another run. EdJohnston 03:30, 19 March 2007 (UTC)

Move along. Nothing to see here. :-)
Probably just a mistake; it is quite possible that we neglected to put this category into a Category:Wikipedia_maintenance so that it would be excluded from the logic that fills User:Betacommand/Datadump/To be Deleted. Alternatively, User:Betacommand's logic might need to be tweaked. I have raised a DRV to have it restored at Wikipedia:Deletion review/Log/2007 March 19.
John Vandenberg 04:31, 19 March 2007 (UTC)
I have restored the short cut. Rich Farmbrough, 09:35 19 March 2007 (GMT).
When is another run for invalid ISBNs expected? Keesiewonder talk 10:39, 19 March 2007 (UTC)
I think everyone is waiting with bated breath for the next dump of enwiki for "Articles, templates, image descriptions, and primary meta-pages" which has been "in progress" for an inordinate amount of time. John Vandenberg 11:46, 19 March 2007 (UTC)

Assigning a 978-series ISBN-13 to a book *automatically* assigns an ISBN-10 to the same book

I'm hoping that Rich might respond to this comment. My proposal is to strike out the following phrase (in the text part of CAT:INVALID) that's been used to describe the relationship between ISBN-10 and ISBN-13. The following is the full section, and I've bolded the clause I'd like to remove. I'll give reasons below:

Many books published between 2004 and 2006 have also been (explicitly) assigned ISBN-13s (they all have one based on the 10 digit ISBN, or 9 digit SBN). Until books with the 979 prefix start to appear, however, it will be possible to uniquely refer to any book either by its ISBN-10 (although there is no guarantee the book will have been issued with a 10 digit ISBN) or its (equivalent) ISBN-13 and both numbers will continue to be reserved for use by that book by the local ISBN agency. Beginning with the 979 prefix, only the ISBN-13 will work or be meaningful for the given book since there will no longer be an equivalent ISBN-10.

There is no way a publisher (at present) can issue an ISBN-13 without also making a 'hole' in the ISBN-10 range for the numerically-converted value of the ISBN-13. So, in my view, it is technically *impossible* for a publisher to release a book with an ISBN-13 beginning with 978 without also making it possible to refer to the book using the converted ISBN-10. This fact is demonstrated by the way that Amazon looks up ISBN-13s. What they do is they first 'mechanically' convert it to an ISBN-10, and then they look up the resulting ISBN-10 in their system. (There is no way the corresponding ISBN-10 could ever be assigned to a different book).

So, in a sense, it doesn't matter if the publisher EXPLICITLY releases the book with an ISBN-10, it will automatically have one thanks to the conversion rule. I don't see this written down anywhere at isbn.org but it seems to mathematically follow from their other rules. Striking out the bolded phrase in the above passage will make the situation more clear to our readers.

Though this might appear to be a subtlety, it could affect the urgency with which people have to stop referring to books using ISBN-10s. I'll let you decide if you agree :-). EdJohnston 16:30, 20 March 2007 (UTC)

Eliminate ASINs?

Links to Amazon have been taking a lot of abuse over at WP:AN. (It's fun to see the collective horsepower of WP's administrators being harnessed to look up obscure book references). There's a very thorough discussion of ASINs at User:Shimgray/ASIN that captures a discussion from 2005. I haven't read this yet, but it looks useful. EdJohnston 17:53, 20 March 2007 (UTC)

The recent WP:AN discussion was following by a thread of about fourteen mailing-list messages starting with [13], in which John Vandenberg put in a good word for the ISBN-fixing effort. Andrew Gray spoke in favor of what I would call 'naked references' in which you only have the basic author-title information, without trying to amplify it with an ASIN and perhaps with no catalog number at all. (Gray identified himself as a librarian). He also made some criticisms of Worldcat that I didn't fully understand.
I think ISBN-fixing, at its best, was resolving errors in book details, as well as just adding correct ISBNs. If ASINs were forbidden, it would not be a heavy blow to the encyclopedia, but it suggests some misunderstanding of the benefits of online-checkable book details. If anyone can think of a better way of stating the benefits of what we did, it would be helpful. I know we eliminated some completely bogus references. The next step would actually be to check references with *valid* ISBNs, because many of them are presumably bogus as well. However, that's not a quick job :-). EdJohnston 02:38, 25 March 2007 (UTC)

Library Thing is doing a 'wholinkshere' via Wikipedia ISBNs

Some of you might have heard of http://www.librarything.com, a site that allows people to register their books, and compare them with the holdings of other members. The following URL, which may or may not work for anyone else, is now listing at the bottom of the page a list of all the Wikipedia articles that refer to the ISBN of the selected book. They say that the listing is updated monthly. I wonder if they use a WP database dump, or if they crawl through our pages, or if they search Wikipedia using Google. EdJohnston 18:50, 3 April 2007 (UTC)