User talk:Rich Farmbrough

Note: I will answer on your talk page (and usually copy here), and look for your responses here. If you see my answer here and it's not on your talk page, I'm either not happy with it, or I forgot to copy it over. R.F.
ISBN

---

If you (or someone else) have (has) corrected an ISBN in an article I labelled using this account as having an invalid ISBN, and I have re-labelled it using this account before 14:34, 27 August 2006 (UTC), it is an error on my part. Please accept my apologies. Any other ISBN problems (even if I have already been told of a similar one), please leave me a note, and I will fix as required. R.F.

FAQ

---

Please feel free to read my FAQ. R.F.

FULL ARQUIVE

---

Alternatively browse my Talk Archive Index. R.F.

Note to self

1930206151. Rich Farmbrough 11:42 31 August 2006 (GMT).

13 digit ISBN

I looked at the following invalid isbn:

• J.K. Beatty, C.C. Petersen, A. Chaikin, ed. (1999). The New Solar System (4th ed.). Cambridge press. ISBN 9780521645874.{{cite book}}: CS1 maint: multiple names: editors list (link)

It seemed completly invalid at first, until I realised. Its using the 13 digit ISBN number. Unfortunatly, most libraries doesn't appear to track 13 digit isbn numbers yet. Don't know if you're accounting for this yet or not. Kevin_b_er 20:45, 1 September 2006 (UTC)

Hi, Kevin, I'm working on it. It's one of the reasons I'm doing the project in the first place (see Template:Auto isbn). Thanks for the info. Rich Farmbrough, 20:52 1 September 2006 (GMT).

ISBN 9780521645874 is not an ISBN is an EAN article code for the ISBN which is possibly 052164587X. (note the difference final character which is check digit on a different scheme than the full EAN code). The EAN code is what is used for the bar coding used. :: Kevinalewis : ^{(Talk Page)}/_(Desk) 12:57, 4 September 2006 (UTC)

Just when you thought you knew something the world shifts from under your feet - thanks for the update. Nice use of shifting terminology EAN loses it's "European" meaning and ISBN-13 is shrunk to ISBN; leaving us all to be confused. :: Kevinalewis : ^{(Talk Page)}/_(Desk) 13:36, 4 September 2006 (UTC)

On August 19 you fixed a broken (13 digit, written with spaces) ISBN in Margaret Mahy to one written with hyphens and 10 digits. However, you did not alter the last digit accordingly, so the link was broken. The trailing -0 has now been changed to -7. Please always verify that the resulting ISBN link works and actually points to the right book. ISBNs with 13 digits aren't broken per se, but having white space among the digits is wrong. When changing between 10 digits and 13 digits, the prefix is almost always 978 (except when it's 979), but the last digit, the checksum, also changes. --LA2 20:25, 7 September 2006 (UTC)

Sorry, that was stupid of me. Rich Farmbrough, 20:58 7 September 2006 (GMT).

A Certain Woman: publisher error on ISBN?

From going down your list of books with invalid ISBNs, I came across one that looks like a publisher's error: [1] 1. Check digit doesn't match so it's invalid, 2. Book not in Library of Congress, 3. Not in Amazon, 4. Copy for sale on Ebay for $39.00, found by Google for the given ISBN, 5. Worldcat shows it at Harvard College Library where they have the entry shown below (with the bad ISBN). 6. Google shows another copy at a library in Holland, also with the bad ISBN.

Author : Arishima, Takeo, 1878-1923.

Title : A certain woman / Arishima Takeo ; translated, and with an introd. by Kenneth Strong.

Published : Tokyo : University of Tokyo Press ; [Forest Grove, Or. : distributed by ISBS], c1978.

Location : Harvard-Yenching W 5933 04 Holdings Availability

Location : Widener WID-LC PL801.R5 A813 x Holdings Availability

Description : 382 p. ; 24 cm.

Series : UNESCO collection of representative works : Japanese series

Notes : Translation of Aru onna.

Notes : Includes bibliographical references.

ISBN : 0680082377 :

I don't believe I can update the article to fix this, without just plain removing the ISBN. Or, I could add a note that ISBN is invalid due to a possible error by the publisher, but it can still be used for some web searches. Do you have a recommendation? The benefit to our readers from checking ISBNs is mostly when there is a GOOD ISBN somewhere in the world and we have a BAD one. If there is NO good ISBN, we should put in a suitable flag. Maybe a template?

I checked with the British Library, so yes, I think just a note at this point. There was another book like this, but i can't find the discussion. Rich Farmbrough, 13:08 2 September 2006 (GMT).

#Published ISBN is invalid here. Rich Farmbrough, 13:32 2 September 2006 (GMT).

I prefer to join the English page

Some users gave me advices how to edit my page and how I can introduce myself but there is still the same problem. I click to main page and then I'm trying to open the "edit page" but there still appears this f....ng index. I open this index and then there is an ampty page, I recognise that there, beside thet Start menu, appears a link which says "download", I click download and then another index(2) opens and so on.

I've solved my problem

I've already solved my problem by my own, but thanks for giving me advices

The phrase 'Cancelled ISBN' already appears in Wikipedia

See the Ralph_Ginzburg article.

From the article:

100 Years of Lynchings edited by Ralph Ginzburg (New York: Lancer Books, 1962; Baltimore, MD: Black Classic Press, 1988, cancelled ISBN 933121180

The situation is murky, however. There could be some politics involved in the 'cancelling' of this ISBN, due to the various obscenity trials of Ginzburg. For instance, the number given in their records (above) is only 9 digits, where we expect 10 for an ISBN! If you put a zero in front of it, you get a valid ISBN, which however does not return anything when you search for it in LOC. So this is not a good example. It's just a case (confusing at best) where they use the phrase 'Cancelled ISBN'.

The pages containing publisher errors where 'Cancelled ISBN' could be needed are (so far):

• 2-10-4 Both the old and new ISBNs for the Jack Farrell book are invalid. Marked as 'No ISBN available'. See my note on Talk:2-10-4.
• A Certain Woman, Now marked as 'No ISBN available'. History comment says 'Please supply corrected ISBN if found.'
• I'saka language, Now marked as 'No ISBN available'. Left link to online article by same authors.
• Ahmad al-Alawi Now marked as 'No ISBN available'. Comment left in Talk page.
• Ajahn Sobin S. Namto Now marked as 'No ISBN available'. Comment left in Talk page.

From Wikipedia's ISBN page I found this document, [2], which asserts: "Most national bibliographies and especially the Books in Print correct ISBN mistakes. The systems then provide cross references ("incorrect ISBN -> correct ISBN")."

It would be good to see an example of that, anywhere! EdJohnston 17:58, 2 September 2006 (UTC)

Since writing the above I was able to get access to www.booksinprint.com. I found that it will not even ACCEPT an invalid ISBN in the search field. So much for the elegant (but nonexistent) solution hypothesized by the IETF document above. (I believe that Books In Print [Bowker] IS the American custodian of the ISBN system). How about we just DELETE from the articles all the ISBNs that are published invalid, of course after explaining properly on each Talk page why that's a reasonable thing to do? Otherwise these cancelled items (of which there might be 200 or so in all of Wikipedia) will need to be processed by hand after every robot visit. EdJohnston 00:37, 5 September 2006 (UTC)

Clarification on R.R. Bowker and ISBNs

From http://www.isbn.org/standards/home/index.asp:

"As the U.S. ISBN Agency, R R Bowker is the exclusive US source of publisher prefixes and accompanying ranges of ISBN numbers for eligible publishers. It provides information and advice on the uses of the ISBN System to publishers and the book trade and promotes the use of the Bookland EAN bar code format. In addition to their ISBN prefixes, publishers also register their titles for inclusion in the Bowker Books In Print databases."

So Bowker could police the invalid ISBNs if they chose to. At present there is no evidence that they try to correct mistakes or keep records on the ISBNs that are published invalid. You just can't search Books in Print for the invalid ones (though you can search Amazon and sometimes find the book you want). EdJohnston 03:02, 5 September 2006 (UTC)

Chew Valley edits

Thanks for your edits on Chew Valley while I was away. — Rod ^talk 07:39, 3 September 2006 (UTC)

you wrote: "Could you look at the "invalid ISBN's"., also for [[Chew Magma], Chew Lake and Chew, at some point?" I presume you mean Chew Magna & Chew Valley Lake but Chew is not anything to do with me or the valley - was there another one?— Rod ^talk 20:31, 3 September 2006 (UTC)

I've now fixed these ISBNs (with one exception I couldn't find which I've removed). I thought ISBNs were unique and fixed but obviously not.— Rod ^talk 08:22, 4 September 2006 (UTC)

ISBN hyphenation

I'm not convinced that hyphenation adds value; in fact, it may even be a liability in web searches. Also, correct hyphenation depends on the number. For example,

99921-58-10-7

9971-5-0210-0

960-87363-3-1

80-86119-13-0

4-8457-0667-9

0-14-015098-6

0-446-60098-9

0-8044-2957-X

0-85386-070-X

0-901690-54-6

0-9500000-0-0

are all correct. Worse still, the proper hyphenation is not determined by a centralized international authority, and requires consulting ever-changing tables like this one. There is a regular pattern for 10-digit numbers beginning with "0", but even books published in English need not be in that group. Is your bot prepared to cope? --KSmrq^T 19:52, 3 September 2006 (UTC)

Yes, it is up to date with the full hyphenation table. Rich Farmbrough, 19:58 3 September 2006 (GMT).

Where did you find a full international hypenation table? --KSmrq^T 20:20, 3 September 2006 (UTC)

The table was found on the ISBN Agency's web site here. Rich Farmbrough, 20:28 3 September 2006 (GMT).

After the world converts to 13-digit ISBNs, won't it be hard to visually scan them (or copy them with pencil and paper) if they are unhyphenated? A better system would be fixed grouping of digits, as in the social security number. But that's out of our control. Perhaps WP could offer a script to do the hyphenation. Or add isbn.org/convertpub.asp to the Special:Booksources page so editors could use it routinely. EdJohnston 20:38, 3 September 2006 (UTC)

Last point is excellent. Rich Farmbrough, 20:42 3 September 2006 (GMT).

My reg-exes are published under the WP:AWB pages. Rich Farmbrough, 14:17 4 September 2006 (GMT).

Could you be more specific? I didn't see regular expressions anywhere, but I also don't know where to look. --KSmrq^T 05:02, 6 September 2006 (UTC)

Wikipedia:AutoWikiBrowser/Settings - they are auto generated from the table, and hence are a little long winded. Rich Farmbrough, 20:47 7 September 2006 (GMT).

Internet radio article

Hi Rich. I re-arranged the external links on the Internet radio article. One section, I made, "sample internet radio stations". It seems to be an advert area for odd-bit stations. I am not sure it should be there at all. Should it be deleted ? Thoughts ? Bests. --- (Bob) Wikiklrsc 13:52, 4 September 2006 (UTC)

Pygmies & Dwarves

FYI, re this edit, <str> doesn't strike text, <s> does. Guettarda 12:51, 5 September 2006 (UTC)

You're gonna love this one!

Hi

My neck of the woods is the Westcountry in the UK. I was "wandering" around articles there when I spotted something that I was not sure about. Moretonhampstead - at the end of the page it states "This page contains information orginally provided by Moretonhampstead Development Trust" and I wondered about copyvio. Looking back I saw that it originally said "Written by the Moretonhampstead Development Trust" and that your edit [3] was the one that changed it. I guess I'm not really worried (and not experienced enough to knwo better!) but is this ok? Sorry to bug you but having found it I thought I ought to ask. Regards --Nigel ^(Talk) 16:35, 5 September 2006 (UTC)

Makes perfect sense to me and all part of the learning curve. Thanks for the speedy answer - regards --Nigel ^(Talk) 18:18, 5 September 2006 (UTC)

Proposal to mark published-invalid ISBNs as 'No ISBN available'

Yet another invalid-as-published ISBN turned up in a book by Johan Cartigny in the Ahmad al-Alawi article. In this case I took the number out and marked the Cartigny book as 'No ISBN available', but left commentary as to what was happening. (See Talk:Ahmad al-Alawi as well). I left the invalid template in hidden text as a souvenir of the problem. Upon saving, the page took itself out of the invalid ISBN category, as expected. I hope this saves some work long term, because if you leave no mark on the page, a future editor could put back the bad ISBN, and if you leave the page in the invalid Category, the next ISBN-fixer will try again to fix it.

If this approach is not one you would adopt, please let me know, since I am working my way down through the Category to fix the invalid ones. As I go, I am only intending to suppress from the visible page the ones that were invalid-as-published).

Bowker's decision to refuse to search for the invalid ones has some logic to it, because the alternative is so much fuss and bother. Essentially Bowker is forcing the publisher of the invalid number to re-issue the book properly if they want to sell any copies through Books in Print. EdJohnston 19:02, 5 September 2006 (UTC)

Signpost updated for September 5th.

The Wikipedia Signpost

Volume 2, Issue 36	5 September 2006	About the Signpost

Everyking desysopped	Explicit images spark debate
Report from the Italian Wikipedia	The English Wikipedia reaches 1,000 administrators
Voting begins in Board elections	Wikipedia in the news
News and notes	Features and admins
Bugs, Repairs, and International Operational News	The Report on Lengthy Litigation

You are receiving this message because you have signed up for the Signpost spamlist. If you wish to stop receiving these messages, simply remove your name from the list. Ralbot 06:39, 6 September 2006 (UTC)

SmackBot is rejecting *valid* ISBN-13s

From American conservatism:

+ *Bruce Frohnen et al eds. American Conservatism: An Encyclopedia (2006) ISBN 9781932236446 Template:Invalid isbn, the most detailed reference book

But don't feel bad, Amazon rejects it as well! Amazon finds the book after you use isbn.org/converterpub.asp to make an ISBN-10 from the ISBN-13. EdJohnston 03:31, 7 September 2006 (UTC)

WP:RFBOT

Task IX has been approved for immediate use, see Wikipedia:Bots/Requests for approval/SmackBot for details. — xaosflux ^Talk 02:08, 8 September 2006 (UTC)

Jan 1, 2007 and 13-digit ISBN

Regarding the Jan 1, 2007 deadline, it's incorrect to assume that as of Jan 1 all ISBNs will be issued as 13-digit. From the [ISBN.org website http://www.isbn.org/standards/home/isbn/transition.asp], this quote:

The new 979 prefix will not be utilized until all 978 pre-fixes have been assigned. However, 13 digit ISBNs that begin with 979 can never be converted to 10 digit ISBNs. They have no 10 digit counterpart.

So as of Jan. 1 we should be prepared to start getting 13-digit ISBNs, but we'll still get mostly 10-digit ISBNs - especially since the ISBNs are entered for the most part by "real users" and not folks in the book industry.

Also, it's questionable whether we should be converting 10-digit ISBNs to 13-digit ISBNs for linking. That will probably work for vendor maintained databases, such as WorldCat, or for major catalogs, such as LoC or British Library. But I seriously doubt that most libraries are going to scan their DBs to reindex their 10-digit ISBNs as 13-digit. It will probably require a software upgrade on a library's part for their OPAC to understand the incoming 13-digit '978' as a valid 10-digit.

In addition, if the book has a 10-digit on the back and that's what the user put in their reference, that seems like the correct ISBN to display in the reference. 10-digits should still work for linking for the forseeable future, otherwise OPAC vendors would break almost every ISBN link into their OPAC.

I'm going to put this comment on the Auto isbn template page as well.

Thanks. RainbowCrane | Talk 15:21, 8 September 2006 (UTC)

Hi Rainbow, the "sunrise" for 13 digit ISBNs was 1 January 2005. As to the support of 979s, I hope and believe this present exercise (with ISBNs, not Wikipedia) will be long out of the way before we need worry about them. I'm sure that most systems will continue ot support 10 digit ISBNS for a long time, but they are not required to. In terms of supporting 13 digit ISBNs all significant systems will need to support them natively, legacy data can be dealt with in two ways, either by checking the converted number as well as the input number (which is a trivial piece of programming) or by running a conversion routine on the data, which is almost as trivial, the hard part, potentially, being the change to allow longer numbers, which will have been needed anyway. Regardless we need to keep thinking about the best process, as I say "auto isbn" can be used in a number of different ways. Rich Farmbrough, 16:23 8 September 2006 (GMT).

While all OPAC vendors will probably be ready, my point was that many, many libraries won't upgrade for a while. I work in the library tech industry, and my company (who shall remain nameless) typically has to support legacy software releases for 6 months or a year, at minimum, because libraries have to coordinate upgrades with their school year, fiscal year, etc. So it's probably not going to work to switch to 13-digit for linking a book that has a 10-digit ISBN printed on the back - the links will break. We should trust our editors to enter the ISBN that's actually shown on the book, and use that ISBN for linking. If there's a 13-digit on the book and someone has cataloged it then it should be valid for linking, and we're fine, but we shouldn't convert all 10s to 13s.

Thanks RainbowCrane | Talk 16:46, 8 September 2006 (UTC)

Easter egg link in your FAQ

Under the Dates section, point 4, you link to Easter egg, but contextually it should probably go to Easter egg (media). Just a thought. --Bdoserror 16:15, 8 September 2006 (UTC)

Cheers. Rich Farmbrough, 16:23 8 September 2006 (GMT).

Deletion of quotes from the Poverty article

Hi Rich. Any comments on this deletion and discussion on the Poverty article ? User_talk:Ultramarine#Quotations_you_deleted_from_the_Poverty_article. Thanks and Bests. --- (Bob) Wikiklrsc 18:34, 8 September 2006 (UTC) (talk)

Hi Rich. Thanks for your kind response. If I understand what you said, I would agree that quotations, for example, could be included in Wikipedia, as well as Wikiquote, etc. There is no easy linkage between WP and WQ. So, can one put pertinent quotes in articles in WP without having been told to put them in WQ or for fear they will get expurgated from WP ? --- (Bob) Wikiklrsc 13:26, 9 September 2006 (UTC) (User talk:Wikiklrsc)

Ohio State ISBN

Thanks for the info. I just made the change, and it works fine.--Sam Harmon 21:37, 8 September 2006 (UTC)

ISBN puzzle

Your smackbot placed invalid ISBN tags on several ISBNs in the Thomas Edison article. Apparently, from what I can make out in the ISBN article, we have to change the ISBN printed in older books to longer ones and remove the dashes. You page has FAQ but I don't see simple basic instructions for updating these arcane digit strings to some modern standard. Perhaps you could be so kind as to add a paragraph to the ISBN article telling explicitly and in simple terms what we should do to update ISBN numbers. I have never understood the point of them, nor have I ever heard of anyone making use of them. Most people go to a library or Amazon,com and find a book by author or title. It seem like it would make as much sense to add the Library of Congress catalog card number. But I really hate to see something I painstakingly typed into an article labelled as INVALID as if I had been careless typing (Typos happen all to often in fact). So when I checked some of the ISBNs which were invalid against what was printed in the book, I thought maybe the Bot had slipped a cog. Thanks Edison 22:14, 8 September 2006 (UTC)

This actually seems a bit more complicated. You've quoted amazon as having 0703304682, which it does for a book called "Edison a Biography", however the book in question is apparently just titled "Edison" which Amazon has as 0070330468 [4], these of course may indeed be the same book but it is certainly confusing. --pgk 22:37, 8 September 2006 (UTC)

Thanks for the info on SBNs, as I said elsewhere I'm certainly not an expert, just responding to a {{helpme}} request, and my little knowledge of stuffing a zero on the front appeared to work... --pgk 22:45, 8 September 2006 (UTC)

Pgk/Edison, the 0-07-033046-8 number that pgk found is probably the correct one. Reasons:

1. The layout is consistent with the book
2. (Most importantly) 0-07- is McGraw Hill's publisher ID
3. The checksum is correct

I suspect that somewhere someone added a number on the end (as I did), and the data got "borged" - after all the purpose of ISBN searches is to enable people to find a book, so it makes sense to allow as many possible variants.

Good detective work pgk! Rich Farmbrough, 08:39 9 September 2006 (GMT).

Woojgie's ISBNs

I took the ISBN Nrs from the actual books on my shelf, I don't know how your automated computer routine came up with a different ISBN.--Woogie10w 22:49, 8 September 2006 (UTC)

Hi Woogie, I'm afraid I will need a little more information. Which article would be a start. Rich Farmbrough, 08:39 9 September 2006 (GMT).

Woogie10w is refering to the World War II casualties article. Shanes 08:31, 9 September 2006 (UTC)

Thanks for adding the zero, anyway Amazon also drops it for Dower's book ISBN Nr.--Woogie10w 16:47, 9 September 2006 (UTC)

Ok, both needed a leading zero. for example. Maybe SBN -> ISBN or just the 0 got left off for some reason. Rich Farmbrough, 09:25 9 September 2006 (GMT).

invalid ISBN

Your bot inserted an invalid ISBN notice to [Quartal and quintal harmony]. See: [5] and [6]

But the ISBN is quite valid: Amazon link, and DTV.

Is your bot equipped to verify ISBNs of books written in other languages? - Rainwarrior 00:46, 9 September 2006 (UTC)

• In this case, the bot was correct and caught an ISBN which was too short. I corrected the hyphenated ISBN that Rainwarrior mentioned from ISBN 3-4230140-6 to ISBN 3-423-30140-6 -- TruthbringerToronto (Talk | contribs) 06:17, 9 September 2006 (UTC)

Ah! I missed one of the numbers. Okay, that makes me feel better about the bot. Thanks! (I should probably go fix the original German page it was taken from too...) - Rainwarrior 06:46, 9 September 2006 (UTC)

Another invalid ISBN

Smackbot flagged ISBN 186448408X (Leichhardt: On the margins of the city, by Peter Reynolds) as being invalid on a number of articles, for example Ballast Point, New South Wales. It's an Australian book published by Allen & Unwin in 1997 and the ISBN was taken out of the front cover. The state library finds it OK here. Am I missing something? amitch 08:46, 9 September 2006 (UTC)

Erm where to start...

1. SB spotted the 9 digit incorrect number and labelled it with invalid
2. You changed the number by adding an X on using AWB (presumably having got the X from the above library?), but changed it in the invalid tag as well.
3. SB changed the name of the tag to something considered politer.

So no problem, you could simply remove the tag. I have done so for Ballast Point, and hyphenated the ISBN 1-86448-408-X. Regards, Rich Farmbrough, 09:41 9 September 2006 (GMT).

Stop!

Stop deleting my work, it takes time you know! from homersimpson742