Hacker News new | ask | show | jobs
by anyaya1 28 days ago
Does Anna's Archive use a completely different "source repository" from LibGen?
2 comments

AA compiles from everywhere; LibGen and Z-Lib served as the major sources of books. This has unfortunately led to search results for a particular book containing multiple versions of that book, and it is not readily clear which one is the highest-quality version. A real library would have librarian staff who carefully curate everything, but in the pirate world this isn’t realistic so it just gets all thrown together.

LibGen is now more or less a dead project. The servers of the original version were reportedly seized a couple of years ago already, and other sites under the LibGen name were notorious for piggybacking the original collection and just plastering it with ads. If one wants to upload stuff, better now to upload it to Z-Lib (not a perfect site, but still) and it will then get picked up by AA in a few months.

Surely this is realistic now (or soon) in the form of LLM curation. A few auto-librarians reading everything, looking at different versions side by side, making choices, etc.
LOL, not realistic at all. The differences between book versions lie in more than the raw text. Moreover, an archival project would be loathe to favour or disfavour a version unless an actual human made the call.
Donate your tokens then.

LLMs are many things, but one thing they definitely are not is cheap/free to run at scale.

annas archive is practically a compilation from all sources possible (including libgen afaik)