Goodreads Librarians Group discussion
note: This topic has been closed to new comments.
Archived
>
Announcement: Updated Policy on ISBNs and ASINs
Clare wrote: "Thanks. I like to be able to find the edition I have read and place it on my shelf. The cover usually helps, but not all covers are shown. "That's why we used to create ACEs—at reader request—so different covers were shown.
Content entered about a book or author on Amazon can often be mis-spelled for older works."
This policy doesn't address misspellings in an edition's description. If you're referencing edit fails in a book's title or author name, something rarely found with professionally published books, those issues have historically been addressed on a case-by-case basis by Goodreads Support.
Published books, like spoken words, can't be unpublished.
Msg 189 — Chiquita wrote: "Hi,I updated the ISBN for the currrent book on sale. Are you able to check a nd make sure everything is correct?."
Please see Rivka's Msg 175.
ETA: Updated message numbers to reflect renumbering caused by deleted posts.
Msg 199 — Ian wrote: "I've followed the instructions - I think meticulously - but the old cover still appears in my dashboard at https://www.goodreads.com/book/show/3.... Are you able to update my o..."Please see Rivka's Msg 175.
ETA: Updated message numbers to reflect renumbering caused by deleted posts.
Derek wrote: "I would tend to think that's the best reason for this policy change. ISBNs are not supposed to be reused for different books. So, if we have a new book that provably has this ISBN, it should mean that the previous book was entered with an incorrect ISBN. Now, I'm sure that isn't always the case, but it's not our (or Goodreads) job to deal with the fact that somebody misused an ISBN. "The vast majority of duplicate ISBNs are those reused by the publisher. Those entered in error by someone on Goodreads are dealt with on an individual basis. As is appropriate. Not tossing out the baby with the bath water, as it were.
To me it sounds as if the current owners of Goodreads wants what appears on their website to be the same as what appears on Goodreads.Not to throw a wrench in the works regarding ISBN's, I have 2 different books, with different titles, by different authors, that have the same ISBN and ISBN13 as other books. ie when I do a search using the ISBN on either book, the results display books other than what is in front of me. I am aware that this is 'supposedly' impossible, but there it is.
Miffy wrote: "If ACEs are an issue, to me it doesn’t make sense to move the ISBN. Why not make it that librarians can add multiple covers to a single ISBN and all the ACEs are visible to all users? The ISBN belo..."BINGO!!!!!!
David wrote: "To me it sounds as if the current owners of Goodreads wants what appears on their website to be the same as what appears on Goodreads."You probably meant one of those 'Goodreads' to read 'Amazon' :)
And that will never be possible. Goodreads is/aspires to be a database of all books ever published. Books currently in print are just a fraction of the total number of books ever published, and books offered by Amazon are even fewer than that.
David wrote: "TNot to throw a wrench in the works regarding ISBN's, I have 2 different books, with different titles, by different authors, that have the same ISBN and ISBN13 as other books. ie when I do a search using the ISBN on either book, the results display books other than what is in front of me. I am aware that this is 'supposedly' impossible, but there it is. "That is what this new policy should address, but doesn't. Yes, your search can take you to whichever book has the ISBN. But the description should refer with a link to other editions that use the same ISBN. Librarians should be required to provide this link. The new policy fails completely in this regard.
To me, it doesn't really matter which edition has the ISBN, and it shouldn't really matter to anyone else as long as the edition one wants to add can easily be found.
Elizabeth (Alaska) wrote: "To me, it doesn't really matter which edition has the ISBN, and it shouldn't really matter to anyone else as long as the edition one wants to add can easily be found. ."Well, it matters to me.
Dobby0390 wrote: "My conscience won't let me deliberately make it more difficult for a reader to find a specific edition."How is this policy change doing that? It will now be more difficult to find the first edition, less difficult to find the most recent, and just as difficult as always to find any other ACE. I think it's a poor policy decision (because it accomplishes nothing), but it is not any worse a policy than the original.
Dobby0390 wrote: "The vast majority of duplicate ISBNs are those reused by the publisher."
Well, that's my point. They're evil bad publishers for doing so, but it's not our (either the GR librarians or GR itself) job to do anything about that. All we can do is put ISBNs we know are valid on editions that really have that ISBN—whichever edition we choose to use.
Derek wrote: "I think it's a poor policy decision (because it accomplishes nothing), but it is not any worse a policy than the original."It is worse, because it involves much more work for the librarian adding the ACE. Suddenly the librarian now has to start researching which edition is the latest and so gets the ISBN, whereas it used to be that the edition already in the system kept the ISBN.
It is not always that clear-cut which is the later edition. Not every ACE is a book by an SPA wishing to update their covers. And since librarians are expected to work according to the new policy, they risk losing their status if they get it wrong.
And moving ISBNs between two completely different books that happen to be published with the same ISBN is just complete bonkers. Who is to say the latest book is the correct one? Reusing ISBNs usually means that the earlier book is the correct one.
MrsJoseph wrote: "Elizabeth (Alaska) wrote: "To me, it doesn't really matter which edition has the ISBN, and it shouldn't really matter to anyone else as long as the edition one wants to add can easily be found. ."..."
That is because you want to treat GR as your personal book catalog (based on your comments) and disregard that you are not the only user here.
As long as books are properly entered it doesn't matter what edition the ISBN is on in the end.
Cal wrote: "As long as books are properly entered it doesn't matter what edition the ISBN is on in the end."It does matter to librarians, because they are burdened with much more work now. I take it you haven't added many ACEs.
Cal wrote: "That is because you want to treat GR as your personal book catalog (based on your comments) and disregard that you are not the only user here.As long as books are properly entered it doesn't matter what edition the ISBN is on in the end. "
You didn't read what I wrote because you have said exactly what I said and yet argue with me.
Elizabeth (Alaska) wrote: "Cal wrote: "That is because you want to treat GR as your personal book catalog (based on your comments) and disregard that you are not the only user here.As long as books are properly entered it ..."
I wasn't responding to your post.
Cal wrote: "That is because you want to treat GR as your personal book catalog (based on your comments) and disregard that you are not the only user here.As long as books are properly entered it doesn't matter what edition the ISBN is on in the end."
Only to you and your argument. And it is my personal book catalog. That's the why and the how I became a member. I'm not here for your BS.
Cal wrote: "I wasn't responding to your post. ."Sorry, I misread your reply quote.
lethe wrote: "It does matter to librarians, because they are burdened with much more work now. I take it you haven't added many ACEs. "No, I think Cal's post is an argument *against* the new policy and agreeing with the majority posting here. As it doesn't matter which edition has the ISBN, then GR should not have bothered to change the policy. What matters is finding your edition, so that the policy change should have required cross-linking editions with the same ISBNs.
Mod
Elizabeth (Alaska) wrote: "lethe wrote: "It does matter to librarians, because they are burdened with much more work now. I take it you haven't added many ACEs. "No, I think Cal's post is an argument *against* the new policy..."
That is not what I gleaned from his earlier comments: https://www.goodreads.com/topic/show/... (and onwards).
Generally speaking this change doesn't do anything useful for the average user. While I don't agree that it is quite the time sink that some people do, I don't deny that it creates additional work for Librarians.Instead of this GR should have just provided better tools to link books together so it wouldn't matter where the ISBN/ASIN was originally placed.
What this ends up being is a zero cost effort by GR to fix something that wasn't really broken instead of putting that Amazon money to work and improving the site in a meaningful way.
We can agree on that, Cal :)---
To return to my previous example of Turtle Diary:
The first Picador edition was published 1977:
https://www.goodreads.com/book/show/1...
The film tie-in edition was published 1985 (very likely, because that is when the film was released), but has a pub date of 1977 and a [7th printing] in the colophon (I know because I own this edition):
https://www.goodreads.com/book/show/3...
When I click on the edition with the ISBN, I'm taken to the Amazon page with yet another cover (not on Goodreads) and a pub date of 1977:
https://www.amazon.com/gp/product/033... (the 'Look Inside' links to the NYRB edition)
Worldcat has a pub date listed of 1991, but no cover:
https://www.worldcat.org/title/turtle...
Can we really be sure that 1991 is the right pub date for the green cover on Amazon? 1977 is definitely wrong, because I know the 1977 edition on GR has the correct cover. Maybe the green cover was published yet another year.
Also, the 1977 edition was added by 77 people, whereas the 1985 edition was added by only 7 people. The 1991 edition is not even on Goodreads.
Can you really oblige people (librarians) adding their own books to add an edition they are not familiar with because that is supposedly the latest edition with that same ISBN? Or should they go to the trouble of moving the ISBN to the 1985 edition, crosslinking with the 1977 edition, only to find someone else later adds their own 1991 edition and the ISBNs have to be moved yet again?
Merry wrote: "Regarding ACEs, allowing the addition of multiple covers to a single book record would be ideal and would reduce the number of separate records in the database. "As mentioned upthread, sometimes the editions have different page counts, even different formats, and they always have different publication dates (and that doesn't even address the issue that sometimes they are not even the same title/author). Combining different editions into a single edition would not only not be ideal, it would not be even marginally acceptable.
Merry wrote: "I must admit I'm a bit confused about one point in Rivka's original post -- * Members will see the newest available edition of a book when searching the Goodreads database." when compared to the information about primary editions in the Librarian Manual that I linked to above."Rivka meant when searching by ISBN (which is the best way of finding the book in hand, provided the ISBN was not reused).
When searching by title, the default edition will be found.
I don't mind this idea, it's just that I will have to update my "ACE" text! And it's not going to undo all that ACE work in the short-term, because it's going to be an "update as you go" situation. I'm guessing much of the old ACE work won't be affected, because there's too many books in the database to really be able to do much about it.But, for newer books, especially with this hype of regularly updating covers, I can imagine that having the ISBN/ASIN on the latest edition would be more ideal. However, I do agree with these two comments:
Emy wrote: "May I suggest that rather than requesting reuse of an ISBN, we add a cross-linking field. In a similar way to the cross referencing in Genres/Shelves, where we can say that B is also A.
Regardless, I think this aspect is possibly a discussion best taken to Feedback."
Lizz wrote: "As a developer, I propose that Goodreads implement a search function that shows all editions with the same ISBN/ASIN, so that the readers can choose among them. The database can keep them separate by item or entry ID. I do not believe that it should be one way or the other. You can accommodate both sides of the argument."
Lizz's idea would be an ideal one, but an easier way of cross-referencing different editions with the same ISBN/ASIN is definitely needed. We can do that manually, via notes etc., but it would be far better if it was included in the database.
With regards to ebooks, unless you purchased via somewhere like Smashwords then sent the book manually to your Kindle or ereader, the latest cover is what you'll always have. Paperbacks, however, need diligent cross-referencing work, so a perfectionistic reader can always find their precise edition.
With regards to paperbacks with the same publication date but different edition number, I think it needs to be possible to take the edition into account (but some books even state "revised edition [date]", or even have a separate date listed somewhere on the page).
But, anyway, I really hope that in future there will be some more automatic system linking, so we can add the ISBN onto the edition we're adding, and the database will automatically link to the rest with the latest edition as the "default" (obviously with some failsafes in place for duplicate ISBNs).
Andrew wrote: "I think you should have made ISBNs non-unique in the database before making this change and any further changes that have anything to do with ISBNs."^ Exactly.
After reading >200 comments, I understand, that this new policy is for ACE only. (And not as I first understood it, for new editions of a book in general.) This is a relief.But still I'm worried in not finding my specific edition of a book anymore.
Taking the example from the manual, if I search for ISBN 0671536753, I only get one work (Nora Roberts: https://www.goodreads.com/book/show/3...), but not the other one in the search result (Catriona, https://www.goodreads.com/book/show/3...), as the description is not a searchable field.
So even if the different (cover) editions are linked to each other in the description field, there's no way of finding them without having to click through several entries. :-(
Merry wrote: "... Surely all that's needed is for the properties of the ISBN/ASIN fields to be changed to allow duplicates?... So the database would be filled with same-cover, same-ISBN duplicates because there would be no safeguard in place to circumvent that. What a nightmare.
Regarding ACEs, allowing the addition of multiple covers to a single book record would be ideal and would reduce the number of separate records in the database...
And how would one flip through the various cover images, and how would one select the cover image that matches the book-in-hand?
"...Then functionality that detects the latest cover based on the date it was added to display that cover on the book page could be added as well..."
It's not at all unusual for me to add ACE editions that were published prior to the edition on Goodreads that holds the ISBN. That would be the latest cover added, but it wouldn't be the most recent cover.
...Primary (default) editions are currently system-set ...
This policy change isn't about default editions, so not sure where you're going with this.
... It would also be interesting to learn if Amazon has any plans to integrate their database with the Goodreads database, especially in any ways that may be beneficial for users of both websites; just for example, when a book is published on Amazon, it is automatically added to Goodreads and perhaps editing or updates only needing to be done in one place.
That system is in place and has been in place for quite some time. The Amazon bot is one of the bots that we've been discussing.
Msg 225 — Merry wrote: "...I hope this hasn't led you down the wrong path, but I'm sure someone will pipe up if it has! ..."So that specific instructions don't get lost in this discussion thread, please start a new thread in Book Issues as Rivka has requested.
Urgh, I have been a member of LT since 2005 (lifetime subscription) but I moved to GR because their interface is clunky and incredibly it's still the same. But like many others this development is quite disturbing, instead of about reading books GR seems to be turning into a place to push Amazon products down our throat.
Barbara wrote: "So even if the different (cover) editions are linked to each other in the description field, there's no way of finding them without having to click through several entries. :-("That has always been the case though with ACEs. It is not introduced with the new policy.
lethe wrote: "We can agree on that, Cal :)---
To return to my previous example of Turtle Diary:
The first Picador edition was published 1977:
https://www.goodreads.com/book/show/1...
The fil..."
That is a fun example.
The green cover actually looks like a regional ACE. The problem would probably be related to that being from a third party seller and not actually from Amazon.
I have no idea who, if anyone, verifies how accurate the information from a third party seller is. Amazon itself isn't always that accurate.
In message 231 Barbara wrote: "But still I'm worried in not finding my specific edition of a book anymore.Taking the example from the manual, if I search for ISBN 0671536753, I only get one work (Nora Roberts: https://www.goodreads.com/book/show/3...), but not the other one in the search result (Catriona, https://www.goodreads.com/book/show/3...), as the description is not a searchable field.
So even if the different (cover) editions are linked to each other in the description field, there's no way of finding them without having to click through several entries. :-(."
This change in policy does not change that aspect. Librarians have always been able to update the description of the the edition records to explain that the ISBN has been used for more than one book and to link to the records for the other books. Librarians can still do that. Librarians have never been required to update the description. This policy does not change that.
It used to be that once a particular record had an ISBN and/or cover image that we were stuck with that forever, even if it was wrong.
The policy change is that librarians are now given the option to, but, are not required to, move the ISBN from one record to another, as long as you are moving the ISBN to a record with a newer publication date.
Barbara, in the example you used:
The Law Is a Lady was published in July 1984 with ISBN 0671536753. GR at present has the ISBN on this record.
Catriona was published in June 1997 and also uses ISBN 0671536753.
In the past we were stuck with Goodreads searches for ISBN 0671536753 taking us to the record for "The Law Is a Lady". As it's now within policy we are now allowed to move the ISBN to the record for "Catriona".
An Abebooks search for ISBN 0671536753 finds 24 records for "Catriona" and two records for "Law is a Lady".
A WorldCat search for ISBN 0671536753 finds a record for "Catriona" and another for "The Law is a Lady". Ironically, WorldCat seems to only allow one cover image per ISBN and is showing the cover for "Catriona" on the record for "The Law is a Lady".
Marc wrote: "The policy change is that librarians are now given the option to, but, are not required to, move the ISBN from one record to another, as long as you are moving the ISBN to a record with a newer publication date."We are required to. See rivka's original post:
"Going forward, new editions added to Goodreads should adhere to this policy. As you make librarian edits to book records already on Goodreads, you can update the ISBN/ASINs as needed."
We are, however, not required to crosslink.
Marc wrote: "The policy change is that librarians are now given the option to, but, are not required to, move the ISBN from one record to another, as long as you are moving the ISBN to a record with a newer publication date."This is an error of interpretation, I believe. The new policy is that now we are required to move the ISBN to the edition that most recently uses the ISBN.
Marc wrote: "In the past we were stuck with Goodreads searches for ISBN 0671536753 taking us to the record for "The Law Is a Lady". As it's now within policy we are now allowed to move the ISBN to the record for "Catriona"..."Someone has changed the entry since my previous posting, so the librarian's note is visible at one glance (and not at the end of the description anymore).
This is still an misleading example in the GR manual, as the ISBN is NOT used for the newer edition. No wonder I'm confused.
Barbara wrote: "Someone has changed the entry since my previous posting, so the librarian's note is visible at one glance (and not at the end of the description anymore). This is still an misleading example in the GR manual, as the ISBN is NOT used for the newer edition. No wonder I'm confused. "
Staff seems confused by the new policy.
lethe wrote: "Marc wrote: "The policy change is that librarians are now given the option to, but, are not required to, move the ISBN from one record to another, as long as you are moving the ISBN to a record..."While Rivka used the word "should" I interpreted the later use of "can" and "as needed" in the explanation as as saying that we "may" and not "must" be moving ISBNs.
In other words, I do not believe there's an expectation that librarians are now required to be attempting to discover the publication date of each cover, checking for newer editions and moving the ISBN to the most recent printing if one is discovered. Instead, it's that in the course of working with a set of records if we discover there is a newer printing or publication with the same ISBN that we can, but are not required to, move the ISBN.
Also, I had thought the old policy was that the ISBN had to remain on whatever edition record someone used it on. It could have been the record for the first, middle, or most recent printing of a title and/or could have publication of a work or title. Rivka says the old policy was "Requiring that ISBN/ASINs remain with their oldest edition ..." It's not clear if the use of the word "oldest" meant oldest by publication date or oldest by length of time a record has been on Goodreads.
Marc wrote: "While Rivka used the word "should" I interpreted the later use of "can" and "as needed" in the explanation as as saying that we "may" and not "must" be moving ISBNs. "I'm afraid your interpretation is wrong.
She said we do not have to seek out existing ACEs to change them, but when we come across them while doing other edits, we are expected to update them.
And certainly when adding new ACEs we do not have the option not to move the ISBNs (or ASINs).
The previous policy was that the ISBN should remain with the edition first added to GR.
Professional catalogers and interlibrary loan librarians use large databases (e.g., WorldCat) when looking for records. Sometimes that means a specific "edition." In most cases, you will get ALL the records that have been assigned that ISBN when originally cataloged. So, why can't all records show the ISBN that they actually hold and let the user narrow their search themselves. This would align with what users of other public sources, like worldcat.org, must do.
Dobby0390 wrote: "Merry wrote: "... Surely all that's needed is for the properties of the ISBN/ASIN fields to be changed to allow duplicates?... So the database would be filled with same-cover, same-ISBN duplicates because there would be no safeguard in place to circumvent that. What a nightmare."
Valid point, but I think there are several ways to prevent this from happening: a) not allowing automatic import of data using an ISBN already in the databse, b) not allowing users to add entries with an ISBN already in the database (pretty much just rolling with how it currently works), c) only enable librarians to add reused ISBNs and in case you don't have any faith whatsoever in librarians doing their job properly and checking whether that ACE is already in the system d) only allow super librarians to add reused ISBNs.
From what I understand, a)'s already in place (I think rivka mentioned it in this thread somewhere), b)'s just how it works right now, c) might just be added like any other editing ability we already have and d) since super librarians already have special editing abilities probably could be added to their special powers.
Kathy wrote: "Professional catalogers and interlibrary loan librarians use large databases (e.g., WorldCat) when looking for records. Sometimes that means a specific "edition." In most cases, you will get ALL th..."Because GR was set up to treat the ISBN as a unique identifier, which it is not. They would need to completely overhaul the database so that more than one record could have the same ISBN
Ellie wrote: "Dobby0390 wrote: "Merry wrote: "... Surely all that's needed is for the properties of the ISBN/ASIN fields to be changed to allow duplicates?... So the database would be filled with same-cover, s..."
I very much agree with Ellie’s idea.
Another possibility, that granted would take more programming, would be for users to create a "draft" of a duplicate ISBN record to be checked and approved by librarians or super librarians. Basically hold the record as pending until it can be checked and then either added to the database or merged with the existing record. This would be manual add only, not auto import.
Kathy wrote: "In most cases, you will get ALL the records that have been assigned that ISBN when originally cataloged. ."I certainly don't have the same experience with WorldCat, but when I search by ISBN, I get one result. Sometimes that one result has additional ISBNs listed in the record. Several of you have mentioned WorldCat, but I don't see it as being useful for this policy.
Elizabeth (Alaska) wrote: "I certainly don't have the same experience with WorldCat, but when I search by ISBN, I get one result."The ISBN on Goodreads always links to one Worldcat record, but when I put the Turtle Diary ISBN in the search box in Worldcat, I found 3 records.
lethe wrote: "The ISBN on Goodreads always links to one Worldcat record, but when I put the Turtle Diary ISBN in the search box in Worldcat, I found 3 records. "Ah, well that's a poor search. Thanks, had not seen this before because the search from the GR page is by ISBN, not by title. Sort of like here, except here at least you get format and probably cover and more complete details.
EDIT: As I said upthread, one WorldCat record can have multiple ISBNs. I don't see how this would be an improvement, if I understand that some are suggesting GR should work like WorldCat.
...Or just let all editions have ISBN?
I read this new policy in total incredulity, especially after I researched ASIN (Henceforward to be pronounced "a sin", or better, "asinine"). If, as seems to be inferred, this datum is to co-habit with ISBN, this is a total violation of good data management. Having worked much of my engineering career in untangling data from within old data systems, I learned that principle #1 is "Never lose control of the data!" Number 2 is "Never allow two types of data to combine in one data field, requiring an interpretation key to disentangle them."
I, further, own two books of same ISBN with different titles. Are they to be treated as alternate cover editions, since the content remains unchanged? But who can be sure which is the original edition except the publisher?
As a librarian, I shall boycott this policy and refuse to comply. Amazon can go chase a goose.
Most of Russian e-books published by Litres.ru (they are leading e-book publisher in Russia) are share ISBNs with print edition of same book (from different publisher). Print edition usually first one to be published.
Their support says that non-unique ISBNs are ok.
Does this policy mean that if I add new e-book from Litres AND print one arleady exists, ISBN should be moved to e-book edition?
This topic has been frozen by the moderator. No new comments can be posted.
Goodreads Librarians Group
Books mentioned in this topic
Срыв (other topics)The Law Is a Lady (other topics)
Catriona (other topics)
Faith of our Families: Everton FC, an Oral History (other topics)
White Wolf Of Avalon: Werewolf Knight (other topics)
More...
I won't be doing ACE requests any more either, Michael. I completely agree, it's just not worth it in terms of time and effort (and frustration). My conscience won't let me deliberately make it more difficult for a reader to find a specific edition. It's a shame that Goodreads implemented an illogical, faulty policy aimed at satisfying the the lowest common denominator rather than maintaining orderly records.