27 comments

[ 3.0 ms ] story [ 72.1 ms ] thread
OHMIGOD, It's full of stars!! There are tons of great books here. The download rate is a slow (of course). Reddit guys are working on putting the stuff on torrent somewhere.
Any word on licenses/copyright?
Some of these books are already freely available. The others are illegal copies, as far as I can see. I have no idea how they got their hands on so many texts. The number is quite mind boggling. This won't last long, though.
Yeah. It's worth browsing, and if you see something you would download, google for copyright info.

almost 38,000 books, it is incredible.

I can't log in. Is there another way to get the books?
I'm waiting for a torrent to be made available.
Books by language:

English 26352

Russian 10124

German 1950

French 685

Spanish 31

Ukrainian 8

Polish 5

Latin 3

Bulgarian 2

Hebrew 2

Portuguese 1

It would be great to get some idea of the categories, titles and authors...I'm intrigued as I can't view anything in work.
This site rocks. Some pirated books well worth reading (buy the book if you can afford it):

Witten, Moffat, Bell. Managing gigabytes: compressing and indexing documents and images (How to build a search engine)

Aho A.V., Lam M.S., Sethi R., Ullman J.D. Compilers: Principles, techniques, and tools (Dragon book)

Anyone else have any suggestions?

Cormen T.H., Leiserson C.E., Rivest R.L., Stein C. Introduction to algorithms
The Feynmann Lectures on Computation. It's a scan of a library book but has come out really well.

A History of Algorithms too.. looks amazing, lots of diagrams, and shows how algorithms were derived in the past.

I'd never heard of the .djvu (Déja-vu) format before trawling through that site. I grabbed a few books to test, got a DjVu viewer (MacDjView) and was seriously impressed. Small file size yet you get the "as original" experience. No horrible plain text or HTML conversions, just renderings that look like the original book (if a little cruftier after scanning).
It's not that hard to download ridiculous amount of ebooks from torrents. The hard stuff is to throw out all the ...for Dummies, ... in 24 hours, ...all the irrelevant thesis and articles, all the duplicates (and always throwing out the least convenient format, lowest quality) and to properly categorize or tag them.

But I have to say that this seems to be structured quite nicely. The thing is that for a collection like that you would really need domain experts in each field to look for the important books (even if they are in low quality). Here for example a quick glance at Bioinformatics tells me that Durbin's Biological sequence analysis and Xiang's Essential Bioinformatics is missing even though it can be found on torrents. They are relevant much more than the rest of the books

There's got to be a way to organize a distributed mirroring effort. Each person has a file list in a flat text file in a shared Dropbox folder, maybe?
Commonly when directories like this are shared on Reddit, people are encouraged to use the Coral cache to get a distributed caching effect. You just add .nyud.net:8080 after the hostname.

http://lib.homelinux.org.nyud.net:8080/_djvu/_catalog/index_...

I'm not sure the Coral cache will work with a username/password. Right now, it looks like multiple people are trying to do separate comprehensive crawls to do a complete torrent, without any coordination.
Use this link for a search interface on a faster site with essentially the same contents. Just try any subject/author keywords: http://gen.lib.rus.ec

The menus there link to contents, torrents and other related stuff. E.g. the torrents are at http://free-books.dontexist.com/repository_torrent/.

This library (of which the link of reddit is one of the mirrors) has been maintained by Russian volunteers, mostly mathematicians and CS people, for a number of years. It used to be informally known as kolkhoz (Russian for 'collective farm').

Seriously? The link provided on reddit is some server some guy found that happened to have credentials listed on bugmenot. Nearly every book on the list is of dubious copyright status (e.g., do they have permission to redistribute it?). Essentially, it's a warez site.
It feels different from the regular warez, though.
Then a library is a warez building?
Just to play devil's advocate (and examine it rationally rather than legally), what exactly is the problem here? Libraries allow multiple people to read the same book. I'm sure you've loaned books to friends. Is it that it's OK for more than one person to read a purchased book, as long as it doesn't happen simultaneously? What if two people read a library book at the same time? Is that stealing? What if this site implemented some DRM that only allowed a book to be read by one person at a time?

This site isn't doing something wrong, merely something illegal.

Thanks, this link is great.

I enjoy copying content that others worked hard to prepare, against their express permission, and then not compensating them, so this is great.

THANKS!

Seriously? We're discussing warez sites on HN now?

Half you people would throw a tantrum if some site published 0days or shared logins for your webapps. But it's OK if it's just someone else's books?

That's not a valid comparison. Approaching everything under the sun with the same copyright approach is untenable, even the greedy music industry and others (e.g., see Hulu) are no longer totally following this approach. A lot of these companies have found that the strong "I won't give you a single bit of my digital goods if you don't pay" idea, is unpractical and not good for business. In fact, new web apps do distribute shared logins, on reddit or elsewhere, to get traction, e.g. see this story from just a couple of days ago http://www.reddit.com/r/programming/comments/ckcbr/mizage_a_....

That point aside, your comparison is still invalid. Money earned from a webapp goes directly to the developer. However, authors, especially of academic books, which is the kind that is mostly on this site, make very little money from each book sold, just ask a professor who wrote such a book. The books are mostly for academic prestige. Publishers, who still live in a world of their own, try to sell these books for unbelievable amounts, generally $50-$75 a pop. My company buys me (almost) any book I want, but when I was a student in a different country I suffered a lot because of this; we used to Xerox books for almost all our courses.

This is not an endorsement to steal any intellectual work. Just a thought that things are not so simple.

P.S. For an additional exercise to see why the simplistic approach fails, look into the intellectual battle between poor countries like Brazil and US pharma companies on AIDS IP. Would you say it's stealing in that case, too? Now, work backwards.