author division problem

Go to the author Cox and attempt to divide the works. You will notice that only the numbers 1-14 are available. 8 and 13 are already used. The rest, if used, alias the work to another author. There is no way that I can see to assign the remaining unknown works to their correct authors.

I'd say that the disambiguation notice is next to useless now. The only 4 works left attributed to Cox are shown, 2 of them are aliased and 2 are unassigned. This has probably been cleaned up with Other Authors. I'd suggest if you know who wrote any of those works that you use the Other Author feature to assign an author and put them on the correct page. As for the Disambig notice, I would delete all the irrelevant authors since there are none of their works still left on that page, and you won't need any numbers after 14.

Does using the Other Author feature in this way really seem like a good idea? I had the impression from Tim's posts on the subject, that that was not a good use but I may be wrong. In any case, it seems at least bad design and probably a bug that it fails in the way that it does.

Oh and yes I know who some of the others are, otherwise why would I post?

I figured you knew some of the others, that's why you were working on the page.

I'm not sure where Tim gave the impression that this was an inappropriate use of Other Authors but if you point me in the right direction I'll have a look. It seems like a good use of the feature to me if you have better information about the work. To the person who originally entered it, the book will still show in their library with Cox as the author, but you will be able to combine it correctly as far as everyone else's works are concerned.

You can undo the aliasing on the numbers that are no longer needed.

There's nothing there to undo anymore, unless you want to use the other authors to update the two divisions still left, all of the other 12 are empty.

>7 Collectorator: "This is a fine example of everything that is wrong with author aliasing and using other authors.

Would you mind elaborating on this statement please? You seem to feel very strongly on this and I don't quite understand how you think this is being misused.

I agree that simple combination is the quickest and easiest solution, but do you know for a fact that that was possible for all the works on this page? Are not some other tricks sometimes necessary?

Also reporeted here: http://www.librarything.com/topic/194822 as a problem for http://www.librarything.com/author/paperback

Also a problem on a page like this: http://www.librarything.com/author/magazinetheeditorsof

What's the bug here now?

>14 timspalding: What's the bug here now?

Go to http://www.librarything.com/author/magazinetheeditorsof. Note that Divisions 1 - 67 are spoken for. If you want to assign one of the new works to yet a different magazine in Division 68, you can't, because the dropdown in the assignment page only goes up to 24.

I don't see that Divisions 1-64 are spoken for. Books have only been assigned through number 16. The others are words in the text blob that is the disambiguation notice--authors who don't exist on LT, or, if they do, have names that put them not on the author page.

This about it another way. LT members add 1,000 authors to the text file, but they're not actually used. How is LibraryThing going to know this?

The fix is simple—if something needs to be assigned to split 92, make it split 17, or whatever and reorder the artificially inflated list of disambiguations.

You have to wonder why all the disambiguations are there in the first place, apparently someone has distinguished all those authors on the page at some point.

Maybe if the disambiguation notices were tied to the individual splits then if the splits are rearranged it won't mess it up.

Good idea. It does make it harder though to figure out who's who in the zoo.

I think users think of those notices as being more than just a blob of text. And indeed maybe we should make it "data," but the idea is that you can put ANYTHING there, and often what needs to be there is general, not merely a list of personages.

True, but it also makes sense to list the people and who they are. Why not have a general disambiguation notice and a field for each split.

>16 timspalding:
The fix is simple—if something needs to be assigned to split 92, make it split 17, or whatever and reorder the artificially inflated list of disambiguations.

You seem to be unaware that works remain tied to their divisions even if they temporarily move off the page, so if any of the works previously assigned return to that page, you have the potential for a contaminated division when you reuse numbers. Not to mention, the CK left in the divisions that potentially needs to be changed.

Reusing split 17 might make for some raised eyebrows. Currently that division's page is titled "The Editors of Southern Living Magazine (17) : The Editors of Penthouse Magazine", so whoever reused that division might be surprised/confused why their Christian Parenting magazines or whatever were labeled as by the editors of a porn rag. (Yes, a recalculation would fix it, but how many users would know to do that?)

Don't try to change the author split. Combine your copy with the existing work.

Yes, it's a workaround, but needs doing anyway.

I believe much of this comes from the fact that Roy Thomas is a common name so a lot of people have this name. Perhaps if you had the artist's middle initial, you'd not have to worry about assigning it.

Otherwise, Tim will need to dig into this.
It would also happen when new books appear on John Smith:
https://www.librarything.com/author/smithjohn

(Also, if you still have the problem, you want to reopen the bug, which I just did, not set it to Needs Discussion. Needs Discussion means it's on the back burner to be ignored.)

>28 PlaidStallion:

the works are all assigned to author Roy Thomas (unknown), and I can't change to 1

Really odd.

I tried to re-assign them, and when I hit "save", a blank page came up. When I checked the author page, it was clear that the re-assignment didn't save.

>30 lilithcat: Should we reopen this bug then?
https://www.librarything.com/topic/302493

>31 gilroy:

Yep. I'll do that.

Okay, closing this one because the other was reopened.