1jasbro
I’m unsure whether we ever developed a consensus on unabridged audible.com works, https://www.librarything.com/search.php?search=%3DHYPERLINK%28%22https%3A%2F%2Fa...
On the one hand, it seems they should be combined with their corresponding work; on the other hand, properly combining them as such yields hundreds, if not thousands of combination proposals with all other works a available in audible.com editions.
So, do we combine or separate them? Thanks for your thoughts …
On the one hand, it seems they should be combined with their corresponding work; on the other hand, properly combining them as such yields hundreds, if not thousands of combination proposals with all other works a available in audible.com editions.
So, do we combine or separate them? Thanks for your thoughts …
2wester
It seems to me the cocktail party test says to combine them.
If one person has read a book and the other has listened to it, they still know the same book. It's not like one has read the book and the other has seen the movie.
But there might be complications here I have not thought of.
If one person has read a book and the other has listened to it, they still know the same book. It's not like one has read the book and the other has seen the movie.
But there might be complications here I have not thought of.
3r.orrison
>2 wester: In general, yes.
But the question is about the *things* listed in the search results in OP's message. Hundreds of often zero-copy editions where the title is of the form "=HYPERLINK("https://audible.com/pd/B00354ZSS2?ipRedirectOverride=true&overrideBaseCountry=true";"Neverwhere")" which all suggest each other as possible combinations. Look at the suggested combinations at https://www.librarything.com/work/30472746/editions
Someone has been putting canonical titles on them, which pretty much guarantees that someone else will do the obvious combination at some point.
I really really wish that some things could be permanently deleted from LibraryThing's database.
I also really wish that the default after adding a book would be to go to the Edit Book page for the user's entry, so that people who add these things would see what they've done. Many of these *things* aren't zero-copy, and someone actually has them in their catalog looking like that.
Not that it matters, but that looks like a link from an Excel spreadsheet cell, I suspect that someone did an import from a spreadsheet and LibraryThing took the formula from the cell instead of the value. Probably an easy thing to check for while importing, but I doubt that will happen.
But the question is about the *things* listed in the search results in OP's message. Hundreds of often zero-copy editions where the title is of the form "=HYPERLINK("https://audible.com/pd/B00354ZSS2?ipRedirectOverride=true&overrideBaseCountry=true";"Neverwhere")" which all suggest each other as possible combinations. Look at the suggested combinations at https://www.librarything.com/work/30472746/editions
Someone has been putting canonical titles on them, which pretty much guarantees that someone else will do the obvious combination at some point.
I really really wish that some things could be permanently deleted from LibraryThing's database.
I also really wish that the default after adding a book would be to go to the Edit Book page for the user's entry, so that people who add these things would see what they've done. Many of these *things* aren't zero-copy, and someone actually has them in their catalog looking like that.
Not that it matters, but that looks like a link from an Excel spreadsheet cell, I suspect that someone did an import from a spreadsheet and LibraryThing took the formula from the cell instead of the value. Probably an easy thing to check for while importing, but I doubt that will happen.
4r.orrison
To answer the question in the original post:
I combine them. Personally I think the bogus list of suggested combinations is less important than the pollution of search results by having them separate.
I combine them. Personally I think the bogus list of suggested combinations is less important than the pollution of search results by having them separate.
5Nicole_VanK
Personally I agree with >4 r.orrison: . But yes, the complications are "most unfortunate".
6waltzmn
>5 Nicole_VanK:
I too agree, both with >4 r.orrison: and >5 Nicole_VanK:. :-)
I like that "cocktail party test," FWIW.
I too agree, both with >4 r.orrison: and >5 Nicole_VanK:. :-)
I like that "cocktail party test," FWIW.
7jasbro
A couple of further questions:
Has anybody yet seen an abridged audible.com edition? I haven’t (yet), but personally, I’ve only only done a small, irregular sampling.
From a coding perspective, is it possible/practical to omit combining suggestions based on “=HYPERLINK("https://audible.com/pd/…” alone?
Has anybody yet seen an abridged audible.com edition? I haven’t (yet), but personally, I’ve only only done a small, irregular sampling.
From a coding perspective, is it possible/practical to omit combining suggestions based on “=HYPERLINK("https://audible.com/pd/…” alone?
8Nevov
Ditto I combine when I meet them, considering them the drunk at the cocktail party who is rambling... barely understandable, but you can tell what book they're talking about... just.
The small discussions previous mentioned, were in the fix this book #85 topic, at posts: 15-16 (last August), 45-55 (October), then 122-129 (January).
>7 jasbro: Yes, I've definitely seen one where an abridged already existed in 'correct' form with multiple copies, so I combined the 'hyperlink' version onto it (I always checked the Audible site for what the hyperlink represented). I can't recall if I dealt with any that had no copies via a canonical title, or if I just skipped them entirely.
To the autosuggestions pollution: in general it's a bit incongruous there is not a way to do the equivalent 'never combine' that we can do on an author page, as a way to suppress bad suggestions.
The small discussions previous mentioned, were in the fix this book #85 topic, at posts: 15-16 (last August), 45-55 (October), then 122-129 (January).
>7 jasbro: Yes, I've definitely seen one where an abridged already existed in 'correct' form with multiple copies, so I combined the 'hyperlink' version onto it (I always checked the Audible site for what the hyperlink represented). I can't recall if I dealt with any that had no copies via a canonical title, or if I just skipped them entirely.
To the autosuggestions pollution: in general it's a bit incongruous there is not a way to do the equivalent 'never combine' that we can do on an author page, as a way to suppress bad suggestions.
9Taliesien
>7 jasbro: You can select for Abridged/Unabridged on Audible's search menu. Here is the Abridged search results page -> https://www.audible.com/search?crid=1FJTBPLURXI2U&feature_nine_browse-bin=18...
As for the bad titles, wouldn't the first step be to contact the user (Mellow.Mel) and tell them their import messed up hundreds of titles and could they either edit them, or if they aren't even going to use LT (which appears the case as it was an import hit and run) could they delete them? I'd at least make that attempt before wasting my time combining garbage edition records into the work. If it became clear they had no intention of using LT I would definitely not waste my time combining since the uncombined titles don't "pollute" (meaning they aren't going to supplant the valid work in a search) the search results to begin with. Better to leave them separate IMO but I've got no skin in the game as none of them overlap with my catalog, so I personally don't care. Ask yourself this, is it better to corrupt good data with bad if the bad isn't hurting anything by being off by itself? I say leave the trash outside, not in the kitchen. :D
ETA: my rule of thumb for combining since I joined LT is, would it make things better or worse? If a stray edition has garbage data that I can't fix at edition level (bad title, wrong isbn etc) I choose to not combine. If it's something I can fix at the work level, I fix it then combine. In this case canonical title is a band-aid, not a true fix. Bad data is also IME, almost always tied to inactive users and as such I feel no need to "help" correct an inactive user's catalog when they can't be bothered themselves.
As for the bad titles, wouldn't the first step be to contact the user (Mellow.Mel) and tell them their import messed up hundreds of titles and could they either edit them, or if they aren't even going to use LT (which appears the case as it was an import hit and run) could they delete them? I'd at least make that attempt before wasting my time combining garbage edition records into the work. If it became clear they had no intention of using LT I would definitely not waste my time combining since the uncombined titles don't "pollute" (meaning they aren't going to supplant the valid work in a search) the search results to begin with. Better to leave them separate IMO but I've got no skin in the game as none of them overlap with my catalog, so I personally don't care. Ask yourself this, is it better to corrupt good data with bad if the bad isn't hurting anything by being off by itself? I say leave the trash outside, not in the kitchen. :D
ETA: my rule of thumb for combining since I joined LT is, would it make things better or worse? If a stray edition has garbage data that I can't fix at edition level (bad title, wrong isbn etc) I choose to not combine. If it's something I can fix at the work level, I fix it then combine. In this case canonical title is a band-aid, not a true fix. Bad data is also IME, almost always tied to inactive users and as such I feel no need to "help" correct an inactive user's catalog when they can't be bothered themselves.
10MarthaJeanne
The problem, of course, is that deleted titles are not really fully deleted, but stick around as ghost editions.
11r.orrison
>8 Nevov: it's a bit incongruous there is not a way to do the equivalent 'never combine' that we can do on an author page, as a way to suppress bad suggestions.
https://www.librarything.com/topic/99884
>9 Taliesien: Bad data is also IME, almost always tied to inactive users and as such I feel no need to "help" correct an inactive user's catalog when they can't be bothered themselves.
You can't correct what's in someone's catalog, but combining stray editions helps other users - it reduces clutter on search results and author pages, making it easier to find what someone is actually looking for, and if anyone in the future adds the same edition the autocombiner will be more likely to do the right thing.
https://www.librarything.com/topic/99884
>9 Taliesien: Bad data is also IME, almost always tied to inactive users and as such I feel no need to "help" correct an inactive user's catalog when they can't be bothered themselves.
You can't correct what's in someone's catalog, but combining stray editions helps other users - it reduces clutter on search results and author pages, making it easier to find what someone is actually looking for, and if anyone in the future adds the same edition the autocombiner will be more likely to do the right thing.
12Nevov
Yes, the original owner altering the title after the fact, or deleting the book, doesn't make any difference to combining, the edition just now becomes a 0-copy in the site (still shows up in search results etc).
Something worth saying about these ones is they aren't involving mismatch between title/author/ISBN or any ambiguity, the title could be called extreme non-standard but the product code should be reliable (a few cases I saw the item was no longer present on Audible's site, but a websearch on the code found it). Though if there was ambiguity I'd certainly leave it uncombined. It sort of reminds me of handling editions that are in non-Latin alphabet that I don't read, and have to either leave things as-is trusting what's been done before, leave it for someone else, or make a large effort to work it out myself.
>11 r.orrison:
:-)
Something worth saying about these ones is they aren't involving mismatch between title/author/ISBN or any ambiguity, the title could be called extreme non-standard but the product code should be reliable (a few cases I saw the item was no longer present on Audible's site, but a websearch on the code found it). Though if there was ambiguity I'd certainly leave it uncombined. It sort of reminds me of handling editions that are in non-Latin alphabet that I don't read, and have to either leave things as-is trusting what's been done before, leave it for someone else, or make a large effort to work it out myself.
>11 r.orrison:
:-)
13cjbanning
In my experience, abridged titles on audible.com (and I'm assuming also other audiobook sites) tend to be older audiobooks, hearkening from the days when one had to purchase them on physical media. Newer audiobooks tend to be unabridged.
14AnnieMod
>13 cjbanning: There are also the "new books in old series/from old authors" cases - these occasionally get both an unabridged one (modern days and so on) and an abridged one (matching the rest of the series/author work narrations).
15cjbanning
>14 AnnieMod: Another case I've occasionally seen is where the abridged audiobook is read by a celebrity, and the unabridged audiobook is read by someone less famous.
