The "42 Quixotes" Bug

TalkBug Collectors

Join LibraryThing to post.

The "42 Quixotes" Bug

This topic is currently marked as "dormant"—the last message is more than 90 days old. You can revive it by posting a reply.

1SilentInAWay
Edited: Jul 6, 2007, 4:36 am

When a member who owns multiple copies of a work compares their library to another member who also owns multiple copies of that same work, then the results are incorrectly multiplied when determining shared works (this is almost certainly the result of LT looping through your books, rather than your works when comparing with the other library).

As a result, if you view the books you have in common with another user, you see their books repeated for each book you have of the same work. I like viewing other users' covers, but not repeatedly.

For example, I own seven different one-volume copies of Don Quixote (different translations). When I view the books I have in common with antimuzak (who has six Quixotes), 42 Quixotes are displayed.

2SilentInAWay
Edited: Aug 10, 2007, 11:40 pm

I'm going to have to rename this bug the "80 Quijotes" bug!!

I bought another translation of Don Quixote and I stumbled across the library of ManuelaPK, which contains 10 copies of the Quijote!! When I view the books that we have in common, sure enough, the catalog view is filled with 80 copies of the book. Click here to see our shared books--to really see the problem, you should sort the view by title and view 100 covers at a time. Skip past the 24 Aeneids (which are incorrectly replicated as well) and you'll soon come to the first 8 Quijotes. At the bottom of the first page (assuming you're viewing 100 at a time) you'll find another 15 and then, continuing onto the next page, you'll come to the last 57. This is ridiculous.

Unfortunately, the duplication isn't the worst part of it. I've just verified that some books do not display in the shared catalog view when your (own) catalog contains multiple versions (e.g., translations) of the same work. The catalog display for shared books appears to cut off when it has displayed the proper number of shared books (which, correctly, is larger than the number of shared works listed on the profile page). However, because this cutoff does not take into consideration the incorrect replication of displayed works due to multiple copies in your own catalog, not all of the other members' books appear in the list. This is not good.

I'm going to create another thread for this problem with a less whimsical topic header. Here it is

3oregonobsessionz
Aug 11, 2007, 1:02 am

Interesting. This is probably related to something I have seen in my own catalog. Mine is private, so I can't illustrate what is happening.

I have a number of titles where I am the only user with that title, but I have 2 different editions and I have combined them. In catalog view, each title appears twice (once for each edition), with 2 users each time (both users being me).

I don't know if this is affecting the total count for my catalog. I suppose it is affecting obscurity statistics.

4infiniteletters
Aug 11, 2007, 2:43 pm

Sounds like a bad rendition of a one->many database relationship to me.

5antqueen
Aug 11, 2007, 4:48 pm

I've noticed the multiples in others' catalogues too and would like to see it fixed. Fortunately, I don't have 10 copies of anything :)

#3, The main count on your profile page is the number of items you've listed, but you can see the number of distinct works in the top left of the 'fun statistics' page.

6SilentInAWay
Edited: Aug 11, 2007, 10:24 pm

I don't have a problem with seeing all the copies I have of a work when I'm viewing my own catalog. Nor do I have a problem seeing all the copies that someone else has when I'm viewing their catalog. I could even put up with all the "false multiples" (silly as they are) if it weren't for the fact that they prevent actual shared books from being displayed.

7SilentInAWay
Sep 14, 2007, 5:33 pm

This has been fixed. Thanks guys!