Audio Books?

I notice that many audio editions are automatically combined, sometimes regardless of whether they are abridged or not.

Is there a consensus, e.g. Unabridged (combine), abridged (don't combine)

or just list them all separately?

I'm not sure if there really is a consensus. But I would say: separate abridged, and combine unabridged (unless it's more an audio play, in which case I feel it still has to be separated as an adaptation).

In practice, since I'm largely unfamiliar with audio books, I tend to leave them where I found them either way - assuming that the person(s) who combined or separated them did know.

The consensus is to combine unabridged, and separate abridged, regardless of medium. (i.e. abridged print versions of books should also be separated from the main work). If you are entering audiobooks, please indicate whether they are abridged or not in the title.

Kind of what I thought except it's not always easy to tell from the title if they are abridged or not - I have only listed one in my collection so far but I will take on board Lorax's comment and include 'Unabridged' in the title.

I don't put anything in the title about its being an audiobook anymore, unless it's abridged, in which case I also indicate that. (I got tired of the lack of automatic combining, the automatic uncombining, and the uncombining by people.)

I can see it both ways, but—all things being equal—I'd combine abridged and unabridged. I think they're socially the same, or mostly so, and combining is about that, not the content per se. If that argument doesn't persuade, there's the simple problem of how to keep track of this. Most audiobook editions do not have different titles, so they're virtually impossible to pick out of a list of author/title editions on the combine page. It's too much to ask for people to Google every ISBN to see if it's an audiobook, and if so, try to suss out if it's abridged.

T

Yes, I think in practice we won't be able to keep uo the separation. But I'd combine abridged and unabridged. I think they're socially the same? You've got to be kidding. Are you saying an abridged audio version of a classic is socially the same as a full printed edition?

6>

Unabridged audiobooks get combined with the actual books, though, and I certainly hope you aren't arguing that an abridged book should be combined with the real thing! They aren't even remotely the same socially or content-wise -- at some point, content has to be considered, or we should just lump everything into a big fuzzy lump of Book.

I'm not asking "people" to google every ISBN. I'm asking that the efforts of the combiners who take the time to try to clean things up be respected, and that an hour-long audio abridgement of Moby Dick not be combined with the actual novel, when it's known to be abridged. Because they aren't the same thing.

>Are you saying an abridged audio version of a classic is socially the same as a full printed edition?

Do the cocktail party test. "Have you read Life of Pi?" "Oh, yes. It was great." Maybe someone is lying, though.

Shall we start decombining when someone admits to skimming?

If they read the Reader's Digest Condensed Version, I wouldn't think they read anything remotely the same. To me, it's like the picture book version. The person is familiar with the story, perhaps, but not the work. To me, those are more like saying "no I didn't read Harry Potter, but I did see the movie." Maybe other abridgements, including audiobook ones, aren't quite so bad though. I've never listened to an abridged audiobook though, so I can't personally say how I'd consider it though.

That being said, abridgements could be considered to some degree like different editions are. We combine the 1st and 2nd edition of a book, and those don't have the same content. By the 10th edition of a textbook, the content may not even resemble the original book, but it's generally still combined. Combining isn't a strict content comparison.

Do the cocktail party test. "Have you read Life of Pi?" "Oh, yes. It was great." Maybe someone is lying, though.

I admit I never fully understood the lemonade stand test. But now imagine "Have you read The Iliad?" "Oh yes, Brad Pitt was great."

And yes, people may be lying - but that can't be solved. We have no way of knowing if any user has ever read any of the books they catalogue. Deal!

But does that mean we should combine everything that is even remotely similar? Walt Disney's Lion King with Shakespeare's Hamlet?.. In that case let's get it over with and combine all of LT into one big happy work. Gosh, they all have print and/or pictures - they must essentially be the same.

>>I'd combine abridged and unabridged. I think they're socially the same

No, they are not. It is not the same if you read the novel or the abridged version (in 50 pages for example)... If we go combining like this, then let's just combine all poetry books from one author in the same work - socially you had read the poetry of him so why bother...

Combining abridged books by mistake is not a problem for me, or even having abridged ones combined in the big ones when it is not caught... but having a rule saying that abridged book is the same as an unabridged might be ok for someone that does not like reading but is just wrong. And I thought that LT is for people that enjoy reading.

Sadly, I suspect that for most people and for many users of this site, the abridged and unabridged versions of a book are socially the same. I do not know where the line is drawn for most people but I think that it is somewhere between abridged audiobook and movie adaptation. Much of the need for this level of analysis would evaporate if LT supported the identification and linking of editions.

The differentiation between levels of reading (close reading, partial reading, skimming, reading the blurb, and so forth) is a very interesting problem of self-reporting. I commend Bayard's musing on this in How to talk about books you haven't read. He also proposes an interesting classification system of books.

Where does the idea that we should have read the books in our libraries come from? I've got lots of books that I haven't read; some are TBR, some MTBRIIGNBTD*, others UETBR**, yet others WOEDIEBTB?***

With all of them, however, having bothered to enter them on LibraryThing, I enter them as accurately as I can, try to put the right cover picture on, etc. My tidy mind prefers to have my books combined with others with more or less the same content, although it's no big deal if they are not.

I did think that this site had some aspirations to literacy, so rather than the cocktail-party test, how about the library/bookshop test? E.g. "Have you got a copy of Great Expectations?" "Yes, here it is. If you want a graphic version or an audio book they are on another shelf."

*Maybe TBR If I've Got Nothing Better To Do
**Unlikely Ever TBR
***Why On Earth Did I Ever Buy This Book?

Do the cocktail party test. "Have you read Life of Pi?" "Oh, yes. It was great."

"Oh, what did you think of the part where..."

"Huh? I don't remember that."

"Well, then what about...."

"Sorry, I don't remember that either. It probably wasn't in the abridged version I read."

Sorry, Tim, this is your site and all, but on this issue you're just wrong.

Though I think combining full and abridged versions of any work violates the part / whole rule, this isn't just about being dogmatic - at least not for me.

Would somebody who owned / read a popular abridged version of Gulliver's Travels - rarely all four voyages, so that scenario lorax' proposes would be a pretty sure thing, but anyhow - be more interested in funny fairytale like adventure stories, or in 18th century literature / satire? I think the former.

>15 lorax:, 16

Well said. Thank you for expressing my arguments more clearly than I would have done.

>14 abbottthomas: Errr... I think that "the idea that we should have read the books in our libraries" comes from you. It certainly was not what I was implying. Perhaps I should have included "not even examined" in my partial list of states (and, yes, I do have books in that category; I call that collection Unread unowned).

Sorry, Tim, this is your site and all, but on this issue you're just wrong.

I may be. But it's comforting to know that I've never been wrong before. :)

What if there were a way to add metadata to editions—original edition, trivial part/whole (eg., volume 1, volume 2), and abridged. Would that satisfy people?

19 > Put a saddle on that pony!

#19 Sounds like a good direction to me.

(I wrote this before seeing #15 and later posts.)

What, exactly, constitutes passing the cocktail party test?

- "So you like the Young Wizards and Star Trek. Ever read Spock's World?"
- "I listened to it."
- "I loved the chapters about Vulcan history, especially the one about Surak."
- "Huh?"

- "'Twas brillig, and the slithy toves
Did gyre and gimble in the wabe;
All mimsy were the borogoves,
And the mome raths outgrabe."
- "Nimm dich in acht vorm Brabbelback,
mein Sohn! Er beißt, wenn er dich packt.
Reiß aus, reiß aus vorm Sabbelschnack,
vorm Jubjub, der dich zwickt und zwackt!"
- "What? I'm sorry, I don't understand much German..."

- "What about Frankenstein?"
- "One version. The early one."
- "There are multiple versions?"
- "Two. Most people know the later one."
- "Are they much different?"
- "I dunno. I just read the early one because it happened to be handy. I'd like to find out though. Do you remember it well?"

- "Moby Dick, now, that's a great story."
- "Oh, yes! I was totally crazy about that when I was a twelve. I made my own harpoon and everything."
- "Really?"
- "I never read the actual book, mind you. Kind of intimidates me. But I read the Classics Illustrated comic over and over."
- "Oh, right. Well, you probably missed a lot of encyclopedic stuff about whales that way. Not that a lot of people don't skip that anyway."

- "As You Like It?"
- "Yeah, I read that one. Several times."
- "But you didn't see it? That's half the fun, you know!"

13: I do not know where the line is drawn for most people but I think that it is somewhere between abridged audiobook and movie adaptation.

Some abridged audiobooks are actually full-fledged dramatisations, so perhaps those are the line.

Look, these issues are always unanswerable. The distance between editions isn't a boolean. It isn't even a scalar, it's an array, or maybe can't be expressed in math at all.

Take a look at Gibbon sometime. http://www.librarything.com/author/gibbonedward It's a total mess. Most people do not have complete editions. There are all sorts of big, thick and impressive editions that aren't actually complete. So almost any combination is to some degree a lie. And if it's not a lie, it's a fact nobody knows, which is no better than a lie. LT combines title/author/ISBN strings, not "editions" (which have no existence in the data). You just can't tell what's a complete Gibbon and what's not. Often an LT edition will include examples of both.

So, we start at aporia, and let's admit it. Our rules are therefore best-efforts at best results. And in this case, although I have a complete Gibbon, I think I share something important with people who have a lesser set, and not just because I haven't read all of Gibbon either.

Gibbon looks like he needs something like a meta-work.

23 >

just because I haven't read all of Gibbon either

If you've read more than half, are the volumes gibbous?

23>

Take a look at Gibbon sometime. http://www.librarything.com/author/gibbo... It's a total mess.

And it's something that Combiners work very hard on to make less messy. Imperfections don't mean it's not worth doing.

Look, we agree that it's messy. I just disagree that the right answer is to say "Everything vaguely related -- the original, translations, a heavily abridged audiobook, the Cliffs Notes, the Books-a-Minute summary is the same." A richer 'relatedness', where we could say "This work is part of that work" -- whether it's Part One or an abridged edition -- would go a long way toward solving this. I have to say I would far prefer this -- creating a way to indicate that works are related, though not the same -- to your suggestion in #19 of lumping everything together and flagging partial versions in the metadata. No amount of metadata or wiggle-room in "sameness" will make The Fellowship of the Ring be the same thing as Return of the King, but your original proposal (to combine both of these with Lord of the Rings, and thus with each other, and have metadata tags for "Part One", "Part Three", and "the whole thing") would suggest exactly that.

> 26: ditto (including the part about it being messy)

26,27> Me too.

Once again, lorax illustrates my position beautifully. And, once again, I agree with Tim's pragmatism. Metadata tagging of the elemental level of an edition would help combiner/seperators, particularly if the set was bounded. Ultimately, however, I would prefer this to be addressed by the technically more complex/demanding contains/contained.

Once again, lorax illustrates my position beautifully. And, once again, I agree with Tim's pragmatism. Metadata tagging of the elemental level of an edition would help combiner/separators, particularly if the set was bounded. Ultimately, however, I would prefer this to be addressed by the technically more complex/demanding contains/contained.

ETF tyop

I seem to have put the cat amongst the pigeons here and I'm sure that more than one dedicated combiner may have had an aneurysm just reading this thread only to find all their hard work and attention to detail so casually dismissed as socially irrelevant.

Sticking to Audio books

If the book and the audio contain every word they are essentially the same thing (regardless of whether the end user chooses to skim or fails to listen) - if they are abridged they are not - to quote a popular advert on UK TV - 'Simples'

On the wider theme -

The ability to enter and edit editions is both the glory and the downfall of the site when it comes data integrity...the entry form hardly dictates a rigorous standard and combining is therefore by definition a Sisyphean task...and experience testifies to the dubious quality of some of the electronic data sent out by publishers to spawn in the vast databases wielded by Nielsen and the like.

But I hardly think the 'cocktail party' test withstands even a cursory viability check. I would suggest that the 'social' model proposed would not only combine cliffs notes and bluffers guides with their original text (as has already been commented on) but by this standard there would only ever be one work on speaking each language. All phrase books and language courses in any format by any publisher might 'socially' be considered the same ...well it's all about talking English / French/ German after all?...all books about String Theory, all books about Container Gardening, the logic can expand or contract in either direction.

There may be no easy answer to Gibbon but that doesn't mean we should choose to add to a problem rather than avoid repeating it or solving it where an obvious answer beckons.

Few would opt to make the mess worse by lumping everything together based on the lowest common denominator.

And I post comment #31 with all due respect to 'He Who Will Be Heard' (otherwise known as 'The God Of Library Thing' the right honourable Mr Tim Spalding .

and I particularly enjoyed the phrase 'You just can't tell what's a complete Gibbon and what's not' - which when taken out of context is a real show stopper.

PS. I've never been to a cocktail party and I now know that I never want to.

Can I have my 'Hug of Death' now...

>No amount of metadata or wiggle-room in "sameness" will make The Fellowship of the Ring be the same thing as Return of the King, but your original proposal (to combine both of these with Lord of the Rings, and thus with each other, and have metadata tags for "Part One", "Part Three", and "the whole thing") would suggest exactly that.

Wait, I didn't propose that at all. They have traditionally been separate works, albeit closely linked. They have different titles. They have coherent story arcs which, while part of a larger arc, is also distinct. Nobody would want them to be combined into a single work.

No, I mean that if a coherent work is sometimes one volume, sometimes two and sometimes three, and that these are just printing decisions, and decisions about how much leather you is allocated to a binding, that shouldn't result in four works (all, 1, 2, 3)--especially provided we have some way of indicating partial-ness at the edition level.

For example, leaving aside the issue of dead languages, I'd like to get to a place where we can combine all the editions of Herodotus, whether they're one volume, two volume or ten volume. The Histories are a single coherent work. Splitting them into volumes is an arbitrary marketing decision. Indeed, the "books" its split into (numbered, and sometimes named after muses) aren't even original to the author, but some Hellenistic editorial decision. And it's even further absurd for people who share "Herodotus, v2" to share with almost nobody but other people who have editions composed of at least two volumes (but not necessarily split into the same number).

The issue of works split into volumes is definitely a problem, but one I thought would be eventually solved with a contained/contains relationship between works not by saying that the volume 1's are the same work as the volume 2's. I'm not opposed to doing it this other way, but if you do make them all one work I presume it will have the smarts to not tell someone with both volumes of something that he has duplicates?

But Lord of the Rings was written as one book by the author. It was split into 3 books by the publisher for a marketing decision (too large and costly as a single book).

I have never understood the whole cocktail party thing. Why would you want to discuss a book with someone who has read only an approximation of the book ? At the core what do people who think thats OK have in common with those who don't ? What would be the purpose of this meeting or chatting ? Empty noise ?

I think editions and metadata are desperately needed. And the non-fiction edition example is spot on, Edition 1 and Edition 10 are related but not the same.

I have never understood the whole cocktail party thing. Why would you want to discuss a book with someone who has read only an approximation of the book ?

Those are different ideas. Do you understand the cocktail party thing—that you might want to talk to someone who's read the book?

Not if they have only read Cliff's notes, or some abridgment, only seen the movie/TV series, even audio books are iffy for me.

Yes, people who read books tend to absorb less than when they listen to it at a spoken pace.

Actually people learn differently and some are visual and some are audio, and some need both.

My problem with Audio books of any stripe is that the talk will invariable end up about the reader, and the voice, and the accent/non-accent, and did the person sound like you imagined, and did they try different voices for different characters, and wasn't this scene boring/thrilling ....

Sort of how almost anything about King Tut is really about Carter and his discovery. Total waste of time.

I don't see them as different issues. Its like you talking abut X and the other person talking about Y.

Right. And some are French and some are German. The French ones keep noticing the food in books, and the Germans have no sense of humor. Let's separate works according to who reads them...

The core issue here is: Are we talking about books, which are a discussable thing, or are we talking about objects, which mostly aren't. Do you want to relate to people around the objects you own, or the content they contain? If the latter, it doesn't matter whether you read something with your eyes or your ears, whether you read it on paper, on a Kindle or in Braile.

Now, if you accept that idea, while audio books are clearly combinable, there's no question abridged versions—audio or not—are a problem. At one end you have trivial abridgements--minor differences between editions and etc. At the other you have one sentence Twitter versions of classics. Put more realistically, at one end you have Moby Dick and Gibbon, and at the other end you have Readers' Digest.

All things being equal, I'd favor putting abridged audio in with regular audio—which is to say in with printed books and Kindle editions too. But I can it both ways. What tips it for me, however, is the great difficulty of deciding or knowing the facts in many circumstances, and the many editions that will contain both abridged and unabridged versions.

Argh, I've been trying to clean up Gibbon, very gradually over the last few weeks. I agree it's a mess, but it needs to be cleaned up, and in the process that means separating, recombining, a lot of work.

Some kind of additional data indicating abridged, audiobook, whatever at the edition level (even just listed after each one on the editions page) would be wonderful for helping make sense of combining/separating work. I agree that it's nice to know if you share Gibbon with someone, even if you have the Modern Library abridged version and they have the 6 volume set, but I don't think the way to do that is to ignore abridged vs. complete. It makes sense to be able to separate out vol. 1 from vol 2, from vol 1-6, as separate works. You don't want your copy of volume one automatically combined with volume 4, etc. (And, if someone has that 6 volume edition, they might well want to know who else shares that complete edition, separate from the 1000s of people who have various abridged and partial editions.) The better solution would be some way of linking works and/or a part/whole relationship.

My problem with Audio books of any stripe is that the talk will invariable end up about the reader, and the voice, and the accent/non-accent, and did the person sound like you imagined, and did they try different voices for different characters, and wasn't this scene boring/thrilling ....

Blind person: "My problem with Print books of any stripe is that the talk will invariably end up being about the cover design and illustrations, and did the person look like you imagined, and did they use different typefaces for different characters/times/states of mind, and wasn't this poem interestingly printed..."

"Invariably" is a big word to use.

Audio books are certainly no longer only aimed at blind people. But other than that...

I'd favor putting abridged audio in with regular audio—which is to say in with printed books and Kindle editions too

Problem is: why put abridged audio in with the full versions but keep abridged print separate? And where do we find the divide between "normal" and "trivial" abridgements? It's putting the whole part/whole thing on the slippery slope.

I agree with you that abridged versions are a problem - as are adaptations by the way. And yes, I dislike the fact that my full multi volume sets of Vasari's Lives don't get somehow connected to users who have either entered these volumes separately or have a popular abridgment. But I think only a more sophisticated relations feature could solve this.

Running into problems until then, imho, is however no reason to combine stuff that really isn't the same.

p.s.: touchstone doesn't refer to the full edition I'm talking about - which is sort of what we're talking about I guess.

43: True, but I don't think I implied anything to the contrary. It's just that a blind person would be most likely to experience a similar annoyance with the few print readers who will derail a conversation this way. (You're not going to talk about the cover design when discussing the book with a blind person, or about character voices when talking with someone who didn't read the audio version - except if the person is curious about these things.)

Deleted. Oops (blush)

>46 vaneska:

Got multiple tabs open? :)

45> (You're not going to talk about the cover design when discussing the book with a blind person, or about character voices when talking with someone who didn't read the audio version - except if the person is curious about these things.)

Of course they do, Its all they talk about in my experience. A few words about the actual book and then its off to the reader and the listening experience.

Just like those who watch movie/TV versions only talk about the casting and the actors.

Invariably is the right word to use.

> 40 Do you want to relate to people around the objects you own, or the content they contain?

Exactly and that is why the content is the most important issue. If its not the same content you end up discussing the object.

>48 FicusFan: Did you forget to open your irony tag? :)

33>

I don't understand what you're talking about, I guess.

As it stands, entries that say "Complete Herodotus" are (or should be) combined, regardless of whether they say "Two-volume set" or "Ten-volume set". But entries that say "Herodotus, Volume One" are not combined with the Complete Herodotus. You're proposing that we combine those with the complete thing -- but that we don't do exactly the same thing in the case of the Lord of the Rings? (That's a single coherent work too, that Tolkien considered a single book -- splitting it into three was a publishing decision. If anything the seven-volume set you see occasionally (six books plus one for the appendices) is closer to Tolkien's intention than the usual three-volume set.) Why does one author in this case get the "the part is equal to the whole" but another doesn't? How do you determine whether to say "Lump everything" or not?

36>

The cocktail-party test can be used to demonstrate either side of any combining discussion, since you get to construct the conversation in your head. You might "want to talk to" someone who's read the brand-new sequel to a book you've read, too, but that doesn't mean the sequel and the original are the same thing.

>50 lorax:

Let's look at the consequences.

If Tolkien were combined, there'd be one work page for all of them. Differences in the tag clouds, reviews of one or another, ratings, Common Knowledge, etc. would all be squished. Whatever Tolkien had in mind, the division is so basic to most people's understanding that, I think, people would freak.

What happens if we combine a two volume set of Herodotus into Herodotus. Well, a editions that nobody knew what they contained (does v. 1 have books 1-4, 1-5, 1-2 or were the "books" ignored alltogether?) is combined into the larger work, and people who've read Herodotus are connected to people who've read Herodotus. If you leave them uncombined, the central fact about your copy of Herodotus isn't what you think of it, but whether it was bound as one, two or three volumes. And since most people read it in a one-volume edition—a big clunking un-beddable brick—you're basically alone.

How do you determine? I'd say that you determine by thinking about what most people consider to be the work? At a cocktail party you'd say "Have you read Herodotus?" You wouldn't say, "Have you read Herodotus, volume 1, which in my edition is, I think, books 1-4."

Now, an exception would be that Herodotus is sometimes published in special single-book forms. Some of the books do have a coherence. Book 2 is Egypt, for example, and six is Scythia. So there are editions with lots and lots of commentary, maps etc. that take just the edition as the topic.

Again, we're not making decisions about things that always have unambiguously clear answers. There's no answer to "what should get combined" that would satisfy both the person who uses books for kindling and the rare book dealer. We're trying to get at a rough approximation of what most people consider to be a work, specifically in a social context.

49: >48 FicusFan: Did you forget to open your irony tag? :)

Sounds more like sample bias than irony to me. Must be the wrong cocktail parties. (Said the recluse.)

Am I reading this wrong or is the proposal to actually combine based on different rules based on who the author is? Because your explanation for Herodotus may be applied quite ok for Tolkien if you have someone that does not read Fantasy - for him it will be as ok to combine all LOTRs (volumes or full ones) as it is to do it for Herodotus... Not to mention that if we go based on numbers and then more copes are added, someone will have to separate them when it goes over 10 copies (or whatever the threshold is).

Or let's get all the novels in any series together - technically if you had read a book from a series, you most likely have what to talk about with someone that reads the same series, just different books.

If it is a different text, it should not be combined... unless if a middle level is introduced (edition) in which case I do not mind getting in one work somewhat different editions for connection purposes... But in the system now - it just makes no sense... and sooner or later someone will start screaming why their 10 different books are reported as duplicates when they do not have anything besides the author in common.

>52 timspalding:
I can't picture someone at a cocktail party asking "Have you read Fellowship of the Ring?" either.

>52 timspalding:

to me The Lord of the Rings is the work (maybe because I had a big-fat-book-without-appendices first), and "The Return of the King" is just a part of it, and I find it hard to think of why people would buy just one of the volumes (maybe you found one in a second-hand-shop, and are waiting until you find a copy of another part?)

But, if you did only have one part of Herodotus, but the parts were combined with the whole, then you would be reported as sharing a work with people who only had a different part of it, which is not at all ideal. I suppose at least that is not likely to be true with abridgements of books (... well books such as "selected poems" could have that zero-overlap, I suppose)

(on trying to touchstone The Lord of the Rings, it offered me first "The Two Towers", which is not the same thing at all, and the the one I picked turned out to be a "photonovel" whatever that is... so I gave up AAARgh!)

What happens if we combine a two volume set of Herodotus into Herodotus. Well, a editions that nobody knew what they contained (does v. 1 have books 1-4, 1-5, 1-2 or were the "books" ignored alltogether?) is combined into the larger work, and people who've read Herodotus are connected to people who've read Herodotus. If you leave them uncombined, the central fact about your copy of Herodotus isn't what you think of it, but whether it was bound as one, two or three volumes. And since most people read it in a one-volume edition—a big clunking un-beddable brick—you're basically alone.

I'm not following this logic. The thing is, there are other ways to figure out that connection - i.e. people who've read Herotodus. Perhaps the database could be made to find ways of displaying shared authors, or ways of establishing part/whole or "related" connections between books. What you're suggesting would collapse single partial volumes together with abridgments together with complete and unabridged (whether multiple volumes or one very big book) together. On that logic, why not combine all Shakespeare plays into the "complete works of Shakespeare" (since I have that book).

Another case: I have 3 volumes out of a six-volume set of Gibbons. These 3 are each cataloged separately. On the logic you're proposing, each of these three would be combined into the overall Gibbons work. That would affect data in my and everyone else's catalog, and basically make the database no longer distinguish between those three volumes. They are not duplicates, but they would be counted as such.

I actually think there's a pretty clear consensus here among combiners, though Tim doesn't agree with it.

Tim doesn't agree with it because he's trying to shoehorn some exceptions into the system. What we need is a change to the system, relationships and/or contained/contains.

Maybe what you/we need is a new or redifined set of concepts, with a database structure that accommodated it, such that:

work = the whole enchilada, the title, the concept. (edited to add: I see this as potentially a concept a bit more abstract & overarching than the current one - anyone who has a partial version or full version could be shown as linked to this work somehow, but with ways of clearly seeing which version they have and who they share their particular version with)
part = a partial version of a work, which could be linked to the work somehow but not seen as identical to it
edition = the particular version of physical book, ebook, audiobook that you have, which might actually be a book with several "works" in it, or several "parts" of works, or even just a single part of a work. This could be linked to the various works related to it by some kind of part/whole or "relation" data.

(As I write this, I'm a little confused myself on the distinction between part and edition, so maybe we just need a more robust way to distinguish between editions that are shared and works that are shared.)

I'm just brainstorming and throwing this out there. Not wedded to it, very open to criticism. However, I do think that some kind of intermediate step and way of indicating relationships between books and works would help a lot.

Also, on the Herototus case, I don't think anyone has a problem with combining a 4 volume version of Herodotus with a 1 volume version, provided they're both complete versions (and cataloged as single entries). There is a pretty big difference between a 4 volume version and a 100 page abridgement, though. And, if someone wants to catalog their 4 volumes separately, they shouldn't be automatically combined together as if those entries were identical. But combining volume 1 only of a 4 volume set OTOH, with a 100 page abridgment of the whole thing OTO, is where it gets even more problematic, since those are clearly not the same thing in any meaningful way. Added: someone who bothers to catalog their volumes separately presumably does so for a reason, and doesn't intend the database to treat them as identical.

57>

On that logic, why not combine all Shakespeare plays into the "complete works of Shakespeare"

Don't even say that. As a combiner I live in fear of the day -- which will come -- when someone does that. And without any form of separation than book by book, it will take weeks to undo what they did in seconds.

I actually think there's a pretty clear consensus here among combiners, though Tim doesn't agree with it.

Agreed.

61 - I'm surprised it hasn't happened already!

Data can be nice, but there's also such a thing as quality of data.

Should Mister Magoo's A Christmas Carol be (re)combined with A Christmas Carol? I don't even understand why that user attributed his entry to Dickens in the first place. If this should be the quality of connections data provided by LT I'll start ignoring connections.

We have had several questions from people recently about 'Why does LT say I have two copies of the same book, when they are volumes 1 and two of such and such, and not the same at all.'

People make a decision as to whether they want such works catalogued as a whole or as single volumes. What we do as combiners needs to respect those who split their Herodotus into volumes when they enter it.

(Even if such works are really hard to clean up because of it.)

64 Agreed.
(Some of those cases, and some of the mess, however, arises when people enter from sources that don't distinguish the volumes via title or ISBN. This can happen with amazon sometimes, but also with library records, which often don't catalog separate volumes. So even if you search on the ISBN for vol. 2, it associates that ISBN - along with all the other volume-specific ones - with a single multivolume record, and so it only imports one ISBN, usually for the first volume.)

added: It can also happen when people edit their publication data, title, or ISBN so that these no longer correspond. Things that are not the same can get automatically combined because they have the same ISBN in the database, even though they're two different works. If everyone were fastidious about checking their data, there would be no problem, but since people aren't - and it's not realistic that everyone would or should be - it's always going to make tricky work for separating and combining.

I'm as big of a "lumper" as you are likely to find in the Combiners! group and yet I find myself echoing the others who are dismayed at learning Tim's opinion on this question.

>33 timspalding: No, I mean that if a coherent work is sometimes one volume, sometimes two and sometimes three, and that these are just printing decisions, and decisions about how much leather you is allocated to a binding, that shouldn't result in four works (all, 1, 2, 3)--especially provided we have some way of indicating partial-ness at the edition level.

Until we have a "contained-in/contains/related-to" function I think that the all/1/2/3 model that we have been using is the most functional thing we can do with the current set up. Any complete set (regardless of # of volumes) is combined with any other complete set (or single huge volume).

When I actually care about the connections I enter both in LT - an omnibus work and individual works as "inclusions". Is anyone clamoring that they can't figure out who else on the site also likes to read Shakespeare?

My opinion - abridged versions (audio or print) should be separated from "main" work when this is obvious (by ISBN or by having "abridged" in the title) - ambiguous entries go with the main/all work. Personally I would lump all the "abridged" together...but I know that others would split even further if given their druthers.

To be considered "abridged" I think a work has to be missing a significant percentage of the content of the main work. (For instance, cut vs uncut Stranger in a Strange Land doesn't make the grade for me - as I've stated before - I won't combine them because I know others feel strongly in the opposite direction...but for me it is a distinction without import.)

I'm not following this logic. The thing is, there are other ways to figure out that connection - i.e. people who've read Herotodus.

I'm sorry. I'm using "Herodotus" in the technical sense of The Histories. Herodotus only wrote one work. It's preserved. Like saying "Thucydides" you can use the author's name as the work's name. You can talk about sections (eg., the Scythian logos), but the book divisions are conventional, and volume 1 of a two-volume set is a meaningless division.

This can happen with amazon sometimes, but also with library records, which often don't catalog separate volumes.

They generally don't catalog separate volumes. A four-volume set is a single record in professional library cataloging. You'll even see this applied when the book wasn't released together. For example, the LC has a record for John Meier's A Marginal Jew, a three-book work that took ten years to complete.

*nods* But you wouldn't stick all of Plutarch's Lives together either.

So hurry up and add some type of "related/abridged/expanded" relationships. :)

67 - I got that part of it. It's the other part that doesn't make sense to me (or to others who've posted here, if I'm reading them correctly). Volume 1 of a 2 volume set is not a meaningless division for people who are cataloging their books, and who have separately entered volume 1 and volume 2 -- or simply volume 1, while they save up the money for volume 2. Combining them makes them into duplicates for LT purposes, which they are not. It may be somewhat meaningless at cocktail parties and for social connections in that sense, but not for maintaining "library-quality catalogs" - unless you want to implement a sub-level for works to clearly indicate which volumes of a multi-volume work someone has (like RL libraries do in some cases). But until that kind of precision is possible, or some other kind of part/whole relationship, I want my 3 out of the 6 volumes of Gibbons listed as separate books and works in my catalog.

Okay, but whether or not the system alerts you about duplicates is a single feature, not a logical consequence of the intellectual structure at stake here. We can change how that part works—I'm not sure work duplicates are ever very useful, frankly.

>70 timspalding:
The Work duplicates feature is useful for alerting you that you've added the same book twice. I would hate to see it disappear from the add books page.

As for the rest of this discussion, my feelings are mixed, so I'll just wait for the dust to settle.

It's not just about being alerted about duplicates: it's that everything in the system considers them the same book once they're combined....

It seems like what just about everyone is saying is it's the intellectual structure that needs to be changed, with some kind of part/whole relationship or whatever you want to call it integrated in, and then the features changed around that.

Okay, it looks like I'm going to lose on the idea of combining lonely and arbitrary volume 1&2s of works into the work itself. At least for now.

So let's get back to the audiobook issue. I think there's simply no reason to keep audiobooks separate. If we're going to do that, we should track down all the Braille editions, and then we might as well remove translation too.

So, I'm left with the question of abridged audiobooks. My argument is that, on balance, I think they're the same work—more stuff lines up similar than different for me. But, as I said, I can see it both ways and I'd go with the flow if it weren't for the second matter—that there's no way to tell whether something is abridged from LT, and it's often hard to tell from the thing itself.

Agreed, it's sometimes difficult to tell if something is abridged, but actually, many of the combiners and separators do spend time looking into isbns, publishing details, etc. when possible, to separate out those that can be clearly determined to be abridged (whether audiobooks or print books). As someone else said above, cases that can't be determined tend to be left with the complete work. I don't see why abridged audio books would be treated any differently from abridged print books, and in what seems to be the majority view here, abridgments should be separated out / not be combined, when it's possible to tell the difference.

I think everyone is more or less in agreement that (complete) audio, Braille, etc. editions are to be combined with the (complete) print work. (The problem, while we're on the question of different formats, is DVDs. They usually get entered with just the title and jumbled with all the title-only editions of the book, and there's no way to separate them as editions from anything else: no isbn, etc.)

Also agreed, but speaking more to the point of adaptations rather than abridgments, where another person has been given credit for a work, (on the title page and in library catalogues), but bad data persists in listing the author of the original work as author, well it's hard. It would help combiners if there was more than one type of disambiguation notice, one place to put an author or author role another for edition information, perhaps a third for authors sharing a same name etc.

BTW The Diary of Anne Frank by Anne Frank is still being mixed up with the The Diary of Anne Frank , the play, by Frances Goodrich. See isbn 082221718X. Where the LT system doesn't work, combiners are quite willing to step in and make the separations. Please Tim, all we ask is your your moral support.

64, 65. It's older books with loooong titles that are the main problems in this.

70) Duplicate works is very useful. Sometimes it points out a book mistakenly reentered. Sometimes repurchased (My son buys a lot of second hand science fiction.) Sometimes it helps chase down combination mistakes. Sometimes it's just good to have the reminder that we have two copies of a book.

What MarthaJeanne just said. I occasionally scan my duplicates quickly and it is a useful way to pick up miscombinations, like recently when I found some infernal idiot had combined all of a particular author's books into one.

In general, count my voice in the anti-Tim chorus here ;)

v

It's not really anti-Tim. It's just that now and again, he proves that he is human, and not perfect. I'd be the last person to be anti-someone just because they aren't perfect.

I'm British which means that things like that are rarely meant to be taken literally ;)

v

I have multiple language translations of some of my books. At the moment, they show up as duplicates. I like having that reminder of the multi-language part of my collection. Whether as "duplicates" or a translation association between works, I want that fact about my collection noted.

I was working on something else and came across Anna Sewall's page:

http://www.librarything.com/author/sewellanna

I think this is a case of separating run amuck.

I understand the desire to designate abridged vs unabridged but - Holy Moses!

Personally I find this page practically useless - imagine what a mess it would be if the poor girl had written more than one book!

OK Tim - it pains me to say it, but this example has sold me - we need a way to distinguish "editions" under the "works" designation so we can associate all of these together without designating them a completely different "work". I'm much more interested, personally, in a "contained in" feature but am now convinced that more people are interested in keeping their edition(s) differentiated than are interested in cataloging contents of anthologies/collections/omnibuses.

This looks like a nice job to move some of these to the right author pages, for example. I will see what I can do to get started. As for the Black Beauty video recordings I could care less and don't feel they belong on the book page.