Publication vs. printing

[John]

For many of the books I’ve been editing/adding recently I’ve found an existing copy with the same ISBN or the same cover or both, but a very different publication date. The one on file may say published in 2008 while the one in my hand says 1964. My best guess is that the difference would be a difference between the printing date and the publication date. Would I be correct in changing that, or is it a valid distinction?

[Michael]

I would add another edition, leave the ISBN fields blank, but record them in the Librarian Notes and, if the editions states what printing it is, record that in the Editions field.

Please don't change the publication date on the original edition as it might be one on my shelf, and then that would be wrong ;-)

[Paula]

per rivka - In general, I would say that all printings of a given edition are one edition, with two exceptions: if the ISBN changes (very rare, I think); or if the cover changes (in which case an alternate cover edition would be created). Therefore, the publication date of that edition should be from the first printing.

Check these threads for further guidance:

http://www.goodreads.com/topic/show/3...

http://www.goodreads.com/topic/show/3...

[Michael]

Well, I don't agree on that. If I have a seventh printing of a book with a publication date 5 years after the first printing, I want the edition on my GR shelf to match the edition on my physical shelf.

I disagree that this is irrelevant.

[Jan]

I'm with Michael, because often with translations, corrections are made between print runs. I think that printings could be noted down in the Editions field however. If we had more of them, we could simply list the print runs and dates there.

[John]

It's a question of definition, which is why I brought it up. You wrote: ... a book with a publication date 5 years after the first printing ...; I would have said that was a printing date five years after publication. Again, in my personal understanding, publication is a one-time thing--creating a book where you had nothing--whereas printing just means the crates in the warehouse were empty. Maybe with a new printing we'll throw on a new cover, but usually not. So I guess I'm saying that if there's no difference between the third printing and the fourth, say, except that printing date, I'd definitely be against making them separate Goodreads books, since it would then be much more difficult in the proliferation of editions to find the copy online that corresponded to the one in my hand. And if that's the case, I think the "publication date" field should be labeled "printing date" instead.

Again, all my definitions, my opinions, my perceptions, with absolutely no intention of saying anyone else's definitions, opinions, or perceptions are wrong; just that a policy decision is necessary. And I really apologize if I've run on too long.

[Paula]

Current policy is to use the date the edition was published NOT printing dates.

[Elizabeth (Alaska)]

Paula wrote: "Current policy is to use the date the edition was published NOT printing dates."

I find this to be inconsistent with current practice. Why do we have an "originally published" date and a published date?

[vicki_girl]

Each edition will have it's own published date. The "originally published" date is the publication date of the first edition.

[Marilyn]

http://www.goodreads.com/work/edition...
I sorted editions of Heinlein's "Assignment Into Eternity" by date and noticed the multiple listings with same dates and different ISBN's.
http://www.goodreads.com/book/show/11...
http://www.goodreads.com/book/show/10...
http://www.goodreads.com/book/show/10...
http://www.goodreads.com/book/show/10...
I counted 5 Penguin editions. The only difference is they list different ISBN's.
None of these books should have an ISBN as they were printed in 1954. In a similar case I tried to delete a book in a duplicate of 3, but each listed 51 reviews. I didn't know how to merge and couldn't delete. What is best way to get/make necessary corrections? Thanks in advance for your help.

[vicki_girl]

Marilyn wrote: "http://www.goodreads.com/work/edition...
I sorted editions of Heinlein's "Assignment Into Eternity" by date and noticed the multiple listings with same dates and different ISBN's."

Each ISBN is a separate edition in the GR database, always. Booksellers will often assign different ISBN to books being released at the same time, e.g. the Canadian edition might get a different ISBN, though in appearance, text, cover, etc. it is identical to the US edition.

Of the editions you linked, three of them are no longer there, so I can't help with those.

Many times, the first printing of an edition did not have an ISBN, as they were printed prior to that system being in place. Then, after we had ISBNs, the ISBN was added to later printings of that edition*.

Sometimes, an edition is listed with an ISBN, and an ~earlish date, because the only date printed in the book is the original copyright date. There is no known "publication date" for that edition*.

*Both of these issues crop up quite a bit with Heinlein's novels. I have been very, very slowly working through these to get the data straightened out. Most of the editions of his work printed before 1980 are a complete mess. The site isfdb.org is the best place to get info on these. Also this thread has more info on sorting out Heinlein editions.

[Marilyn]

vicki_girl wrote: "Marilyn wrote: "http://www.goodreads.com/work/edition...
I sorted editions of Heinlein's "Assignment Into Eternity" by date and noticed the multiple listings with same dates and different IS..."

ISBN's started at the same time as the first barcode testing very late 50's established in the early 60's. I am fortunate that my career involved working with extensive databases and barcoding of addresses and schemes while working for the USPS. I wasn't left behind as hardware and software improved.

I appreciate the Url you gave me for the best look-up site. However, when I tried to list that as my reference in saving editions it showed up as an invalid url. I'll try again.

I managed to fix the othe "bad" listings. They were all Italian language copies. I combined and deleted identical editions, keeping the most popular Italian edition.

Today I am adding all the old Heinlein editions for which I have access. I never know what I will do from day to day.

[vicki_girl]

I was able to use isfdb.org as a source without any problems. If you run into issues again post a link to the particular edition, and I can see if it'll work for me.

As long as the identical Italian editions did not have unique ISBNs, I think you're okay. GR does not remove editions with valid ISBNs.

[Kyle]

I find the following Librarian Manual section on publication date confusing.

"This [publication date] field is for noting the date the edition associated with the listed ISBN was published. Date published can most usually be found on the copyright page of a book. That page will tell you when the work was copyrighted - and if the book is a first edition, the copyright date will be the same as the date published. If it is a later edition, the date will be different. Some books, not all, will list dates for subsequent publications somewhere on the copyright page. The last date listed is what should be used to fill in the date published. If you know the book is a reprinting/reissue published later than the copyright date, but no other dates are given you can A) if the book already has a date published supplied by Ingram, leave it as is; B) research the publication date for the edition you have using Worldcat, Library of Congress, the author's website or numerous other resources; or C) use the copyright date as the date published" (http://www.goodreads.com/help/show/27..., emphases mine).

As you can see, the manual uses the terms "edition," "publication," "reprinting," and "reissue" in rather synonymous ways. Do clearer definitions or policies regarding these issues exist elsewhere? (I already read the thread Paula referenced earlier but didn't find much help there.)

[RB]

I think that the first sentance says it clearly. The rest of the test is simply trying to clarifying (albeit in a confusing way)

"...noting the date the edition associated with the listed ISBN was published."

I understand it like this:

1) You have the unique ISBN number of your book in hand.
2) You have the year of printing of your physical copy of the ISBN in hand.
3) Your copy could be a) a reprint (with the same or different cover as the original one), b) a first edition, or c) first print.
4) You log the data that you have: ISBN and most recent date of print that you can see.

If your book is a different "edition", as in a different version of the book where changes/additions have been made to the text, then the ISBN nr would be different as it's it is in essence a different book.

A different cover would not constitue a different book and ISBN nr, which is why you don't log the ISBN, but instead log it elsewhere and have the books combined so it's listed as a different edition of the same book (but not to confuse this with a say 3rd or 12th edition).

I know it's confusing but I hope that this helps (at least if I'm correct in my assumptions).

[Kyle]

Rita, if I'm understanding you correctly, your argument then is that all editions (I'm using this term to mean a notable change has occurred in the text) and all reprints (I'm using this term to mean another printing of an existing edition, whether the cover has changed or not) should be entered into Goodreads as different editions (I'm using this term here in the way Goodreads uses it in the sense of "combining editions") of the same book. Am I right that this is what you're saying? If I am, you're presenting a different view than Paula (in this post) and rivka (in this post).

I should also note that many of the books I work with are pre-ISBN, so I am looking for a policy that goes beyond using ISBNs.

[vicki_girl]

Reprints of the same ISBN do not get added as distinct editions unless there is a change to the cover, or other significant data (like the publisher). The published date listed for a given edition should be the first printing of that edition.

For pre-ISBN books, it is similar. Reprints do not get added as distinct editions unless there is a change to the cover, or other significant data (like the publisher). The published date listed for a given edition should be the first printing of that edition. Keep in mind however, that changes to the cover can be subtle. For example, a cover may use the same artwork, but different type for the title.

It took me several minutes to realize the following had different covers. ( I added the second as an alternate cover edition when I realized the title was different.)

http://www.goodreads.com/book/show/77...
http://www.goodreads.com/book/show/13...

[Kyle]

vicki_girl, that all makes sense to me. Thank you very much. Just to clarify, this means that the date of a reprint with no changes (of an edition already existing in Goodreads) is not noted anywhere in a Goodreads listing, correct?

If I am correct about that, would it makes sense to suggest a new area on book pages for "reprinting of this edition" dates?

Does anyone here know if it would it be possible for the Librarian Manual to be edited to clarify these points (assuming this position explained by vicki_girl is the "official" Goodreads policy)?

[RB]

Well it's hard to explain, it's obviously open to interpretation.

Basically I'm on the same page as Michael in msg #2.

Reprints of the same book (same ISBN, text, same or different cover, same or different publisher, different year) should be logged without the ISBN (though it's referenced to elsewhere in the book data) and then combined - if you don't do this they you change someone else's book data which by all intents and purposes can be correct. Rivka basically said the same (unless I have completely misunderstood her): all printings of a given edition are one edition - I interpret this as "one edition = combined edition". Following this logic allows you also to combine, for instance, translations.

What I find more tricky, though, is how to handle same ISBN's (maybe with even same covers but different print years) but different page counts. Personally I would treat those as in the "same ISBN but different cover" scenario (log them separately, omit using the ISBN with is referred to elsewhere and then combined).

[Kyle]

Rita, I understood Rivka differently. When she says "all printings of a given edition are one edition," I interpret "one edition" as "one edition of a book." You equate "one edition" with "combined edition," while I would simply call a "combined edition" a "book."

Following the policy vicki_girl outlined, translations, to use your example, would still be allowed to exist as combined "editions" (using the term as Goodreads does) of the same book because a translation is a "change" to a book. I think that different page counts (to use your other example) could also be allowed to exist as combined "editions" (using the term as Goodreads does) of the same book because a change in page numbers is a change to a book. However, this policy from vicki_girl would not allow, reprints with no changes to exist as separate "editions" (using the term as Goodreads does) of the same book.

Having said all of this, there is a very minor change that has occurred to a book in any reprinting: the line on the copyright page (or elsewhere) about the reprinting number and date. So, if the policy state by vicki_girl were to work, the policy would need to specify what constitutes a "change" to a book.

[vicki_girl]

Rita wrote: "Rivka basically said the same (unless I have completely misunderstood her): all printings of a given edition are one edition - I interpret this as "one edition = combined edition"."

I cannot speak for Rivka's intent. However, based on reading many of her comments I have always interpreted this as "one edition = one record/listing in the GR database". Editions are then combined into a single work. For example, Jane Austen only has one work titled "Pride and Prejudice" that has many editions by many publishers, as well as many translations. Reprints of those editions, with no differences, other than a new printing date are not listed in the GR database. As John stated in msg #6, a new printing just means there were empty crates in the warehouse so they printed more. Sometimes corrections are made between printings, but they are rare, especially after the second or third printing, and do not generally constitute significant changes.

One of the reasons that later printings are not added is simply volume. As an example, this edition of How the Grinch Stole Christmas had its first printing in 1957. It has been reprinted every year since then, sometimes multiple times. It has 60+ printings of just that ISBN.

In addition, many people have trouble identifying the ISBN of their book and logging the correct edition on GR. Identifying the right printing would be even worse.

[RB]

vicki_girl wrote: "...
One of the reasons that later printings are not added is simply volume. As an example, this edition of How the Grinch Stole Christmas had its first printing in 1957. It has been reprinted every year since then, sometimes multiple times. It has 60+ printings of just that ISBN...."

Very good point you bring up. It cleared it more up for me, I think.

[vicki_girl]

Looks like Kyle & I cross posted.

I always used a rule of thumb that a different edition has to have one of the following:

1. A significant change to the text of the book itself (e.g., "revised & expanded edition"). A change on the copyright page wouldn't count.
2. A change to the book's data, e.g. title, cover, ISBN, author/pseudonym, pages, publisher, etc.

Also, as an aside, if the only difference is page count, then one of the page counts is probably wrong. It's not impossible that two identical editions have different page counts, but it's unlikely unless one is an abridged or expanded edition. The publishers and the booksellers (and ingram from whom we import data) often use page counts that are approximations based on the number of "sheets" to be used (1 sheet = 8 pages for most books).

[rivka, Former Moderator]

Kyle wrote: "If I am correct about that, would it makes sense to suggest a new area on book pages for "reprinting of this edition" dates?"

You can suggest it in the Feedback group, but I rate the likelihood of it being added as very low. Too few GR users care. Plus there's the question of how many dates it should hold. I have books in their 30th printing less than 10 years after initial publication. Would that mean trying to find out what those 30 dates are? Assuming at least one was each year and adding each year?

It's just not realistic to try to start tracking that info for EVERY EDITION of every book. And if the info wouldn't be tracked for 98% of our books, it's not a useful field.


Kyle wrote: "Does anyone here know if it would it be possible for the Librarian Manual to be edited to clarify these points (assuming this position explained by vicki_girl is the "official" Goodreads policy)?"

We have a folder for suggestions about changes to the Manual. Feel free to start a thread, and include your suggested changes in language there.

[rivka, Former Moderator]

vicki_girl wrote: "However, based on reading many of her comments I have always interpreted this as "one edition = one record/listing in the GR database"."

Indeed.

[Kyle]

rivka wrote: "We have a folder for suggestions about changes to the Manual. Feel free to start a thread, and include your suggested changes in language there."

Will do (eventually). Thanks.

[Andy]

Most of the books I read don't have more than a couple editions, so this hasn't been much of a problem until now. I've been collecting and reading Agatha Christie books for the last couple months, and there are TONS of different editions to choose from.

The question I have sounds kinda dumb, but it's got ME totally confused. How does one determine the publication date?

According to The Mystery of the Blue Train it was first published 1927 with this particular edition published October 2nd 1987. One copy I have, that has a matching cover says it is copyright 1928, copyright renewed 1955 and the cover art is copyright 1987. Lower on the page, it says the first Pocket Books printing was March of 1940, and the next line says 50 49 48 47. It doesn't say publication date anywhere. :-(( :-((

One I have with a different cover says basically the same thing, copyright 1928, renewed 1955, cover art copyright 1985 and cover design copyright 1985. First Pocket Books printing March 1940, and the next line 50 49 48 47 46.

I also have two different copies of Death Comes as the End. This was supposedly first published 1945 with this edition published March 2nd 1984. Both copyright pages are the same, copyright 1944, renewed 1972, cover art copyright 1984, cover design copyright 1984. First Pocket Books printing was April 1959, and the next line says 35 34 33 32. Oops! One was printed in USA, and one in Canada, but the copyright holders for the different covers are the same. :-s

help? please?

[rivka, Former Moderator]

You probably cannot tell for sure. But the latest date (which sounds like cover copyright in all of those examples) is probably a good guess.