1.5 million books and 2 million tags
LibraryThing has hit 1.5 million books cataloged and 2 million tags, all since August 29.
What used to be a straight line of 9,000/day has become a bending curve—gulp! Even so the site pretty-much stayed up today. The database isn't out of the woods yet, but it's in better shape. You'll see some stutters, but downtime should be minimal and I'll be getting a second server for some of the "thinking."
Some other updates:
What used to be a straight line of 9,000/day has become a bending curve—gulp! Even so the site pretty-much stayed up today. The database isn't out of the woods yet, but it's in better shape. You'll see some stutters, but downtime should be minimal and I'll be getting a second server for some of the "thinking."
Some other updates:
- Profile pages now show ten random books from the user's library. People were doing this with widgets, but that presented some security problems and also looked bad with the new design. I will probably add the ability to customize what people see (most recently-read books, and so forth).
- Acquisition, started and finished dates have been added. This feature hasn't been extended everywhere—eg., you can't use it in a widget. That will come soon. You can enter dates in North American or European format, but the catalog currently only shows them in the righ—um, North American—way.
- I added a statistic to the Zeigeist page listing the "50 'completist' authors." These are authors that, if you have one of their books, you have a lot. Think people who collect every single Agatha Christie, although the actual authors may surprise you. To qualify the author needs to be read by at least 100 users. Come to think about it, I could do a statistic for whose library is the most "completist." I am quite sure that a science fiction fan would win.
posted by Tim at 6:07 AM
23 Comments:
The completist feature is really interesting, but perhaps it would be more informative if you divide the average number of books by the total number of books from that author. So, the ranking would be a number between 0 and 1, or you could multiply it by 100 and it would be a percentage instead. That way, authors like David Eddings (for instance) who have written >20 books, or Agatha Christie, will not outweigh folks like J.R.R. Martin who have only a few. You may need to limit the completist ranking then to authors that have at least say 5 books or something like that.
For date formats, I highly recommend the ISO format of YYYY-MM-DD, as it seldom causes confusion between days and months. As an added convenience, it's the same as MySQL's date format, so you don't need to massage the output.
Cool changes, especially the various dates.
Completists? If it's just for tracking reading, the mystery fans will be heavyweight competitors. For ownership, well, it's up to the eccentrics no matter the topic.
I don't know if it's a result of recent changes or older and I merely haven't noticed, but I seem to have a lot of records where the 'Summary' display is broken. I've edited and re-edited 'The Time Traveler's Wife', for instance, but it keeps coming up blank. Ordering by Summary I see 15 records where this is happening and I can't find any link between them. Ideas? Particularly on how I can get the edit to stay?
I second skud's comment on dates in ISO-format.
This might sound pedantic or nitpicky but I'm not from Europe nor from North America. I believe the date format you've called "European" actually covers most of the world. In fact I'm traveling in one of the 3 countries of North America now and here we don't use the "North American" date format but the "European" one.
That's the end of my rant but I really hope you can improve the real international issues such as accents and authors' names in non-Latin alphabets. Keep up the great work!
I had a friend talk about this this ages ago, but with the 'completist' author data, would it be possible to eventually have 'add all books in a series'? Perhaps with check-boxes, to save from deleting books later.
Otherwise, perhaps a 'add all books by this author with at least 100 readers'. This should cover the most common books by them (though perhaps the number should be lowered, maybe 50?)
You're not measuring 'completeness', you're measuring 'addictiveness': discovering which authors are instantly habit-forming. You will get interestingly different (and possibly more meaningful) results if you changed the initial threshold from 1 to 2.
(Note that since many "graphic novels" are actually chapters in an ongoing serial, their authors get a statistical boost.)
'Completeness' might be, for example, the number of people who have a large percentage of the author's output; but this is impossible to measure because of the variant-edition problem.
I find this kind of data mining fascinating. My own Eroica Project site does the same kind of exploration, using performances of Beethoven's "Eroica" symphony as a database.
Keep up the good work!
I have a quick question: you posted in your blog a list of terms you could use with searches, like "+" and "-". Is that list posted anywhere else on the site? Could there be a "help" tag, or something, with a list of what the search terms are? Or perhaps it could go on the "about" page. I can't find it in the blog now.
Also, could one of the search terms be "and not"? So, for example, if I want a list of ONLY my Robert Heinlein books (but not my Robert Bolt or Mary Heinlein books), can there be some way to look for "Robert Heinlein and not not-Heinlein and not not-Robert"? Or if there were some way to put the terms together, say by putting quotes around them and so looking for only "Robert Heinlein" and that's all you would get, that would work too.
Sorry to harp on this. Perhaps I just don't understand how the search function works.
I hate to be the bearer of bad news, but the "search" function from "add a book" does not seem to be working properly. A search for a legitimate ISBN will yield a line which indicates that there is one Amazon result, but no way to access the result, and a line stating that no books were found.
I think the plus and minus stuff only applies to tag searching, but if Tim were to implement the same functionality in book searches you could easily do +heinlein -mary -bolt.
I don't think it's possible to do an accurate book search yet.
Excuse me if this shows up a second time.
Give me an example of something that doesn't work in addbooks. It's working for me. Give me a ISBN?
Tim
Tim, where would you prefer us to leave bug reports?
I've thought about posting my bug under the latest blog post, but often the bug is unrelated to it, and I hate taking a thread off topic.
Sometimes there's an earlier blog post that may be relevant, but I'm not sure if you go back and read the older ones. (I've done it that way before, but I've never known if you actually saw the message.)
Then of course, it's also possible to:
- leave a comment on your Profile page
- e-mail you
- add a new post to the LibraryThing Google Group.
Which option is best for you?
I like the completist list but there are of course problems.
Say someone has 18 shakspeare plays and a book of sonnets. 19 books, about half his writings. I have two copys of complete works (a 'show' copy leather bound 1907, and a 'reading' copy) 2 books everything he wrote twice!
What might be nice with the acquisition data would be simple check boxes: Owned, Read. I say this because there are books in my library I bought and read more than 5 years ago, but for the life of me, I don't know when nor do I still have the receipt. (That tax season is long gone. :>) But being able to just say "I own it." and "I've read it." would benefit me as much as a date might.
Another user asked about my duplicates, so I went looking, and it turns out around 400+ of the books in my 'collection' are duplicates, which is not true in real life.
Unfortunately, that's an import error on the part of LibraryThing. I tried to import my DeliciousLibrary books via CSV, and evidently it double-recognized a huge number of books. What's worse is that there's also a huge number of books that it didn't import, because they're too old to have ISBN's.
It's *really* frustrating, and I don't have the time to go through and hand-delete all the duplicates right now, much less hand-re-enter all the information for my older books, not to mention figuring out which books are and aren't in the two libraries.
One thing most of the good desktop software products have is import tools from each other. :(
For the deletion, a 'checkbox' next to all the books, and 'Delete all checked', or similar functionality, would help a lot. As would being able to see 100+ on the screen at a time. (Of course not duplicating books on load might have helped also.)
Not every book has an ISBN (specifically older science fiction), and since I've entered the data for those books once, I really shouldn't have to again. Importers, importers, importers.
But the duplication is really the worst part. I can't even clean it up easily, because I *do* have duplicates of some books (hardcover+softcover usually), so I have to actually think about each deletion.
And I can't select the 4 extra copies of 'A Million Open Doors', and delete them all at once, nor can I see more than 20 items at a time in a mode that lets me delete...
I imported my books during the period of time when the site was yo-yo-ing, and unfortunately only now have been able to see what ended up in my list. My Delicious Library collection contains 1750+ books, the import feature said it found 1600 books, and I ended up with over 2000 books in my collection, which I presume means it duplicated some 400 of my books in some way.
'Completist' is a cute feature, as are the 'social' features, but I think the site seriously needs better mass booklist management tools, and a few real importers (not just ISBN scanners) from the more popular desktop packages. It's not as fun a feature, but for mass input of books (i.e. the people who are going to need more than 200 books in their collection), it's a necessity to get it right.
Sorry to post this on the blog, but I'm not sure where you want bug reports to go, and I thought I might not be the only person running into a bit of frustration with the importer and book management tools.
Greetings,
I stand, sit, or meow corrected.
I found the 'advanced editing' mode (which only apparently allows tagging, but would be the PERFECT place for 'Delete all selected', for instance). Since it gave a full view list, without paging, I was able to mark all the books that got duplicated (that I could tell with a quick skim) with the tag 'delete', and then display normal mode searching for that tag. (Some 327 of them, after I'd already killed off a bunch via the normal interface before giving it up as a bad job.)
Use Firefox's ability to middle-click to open-in-a-new-tab without focusing to that new tab, middle click down the line of 'X's, don't bother going to the tabs, right-click the main tab, choose 'Close other tabs', reload the tab to get the next 20, and continue.
Lather, rinse, repeat 16 more times.
Could the completist feature (in whatever final form) also be shown on the author's page?
Just thought you might be interested to note a nice writeup/review of LibraryThing at http://www.powells.com/blog/?p=735.
Is the "similar libraries" page gone forever? I'd be really sad if it was, cause I liked the "sort by obscurity" section. A lot of my most obscure books are my most treasured ones, and it was nice to be able to find other users having them...
Similar libraries: No, you can still get it at http://www.librarything.com/profile_similars.php . I need to put an icon for it on the profile page.
great to hear, thanks :)
The completist feature is interesting, if only for the simple fact that many of the listed authors seem (for lack of a better word) obscure. Is anyone else amused that Aleister Crowley is fairly high on the completist list?
Post a Comment
<< Home