Does anyone know of an ebook sourcing search engine?

[Krystian Galaj]

Let's say I found an ebook somewhere on the Internet. All I have is the filename, the book inside, in some specific format. Or possibly a zip, with a handful of html files inside, which form a single book. I'd like to know if that ebook is an illegally created one, or possibly if it's a legal offering, for example that some site offers it on promotion so I could go there and get it from its real source.

I was thinking there should be a search engine where I could go, upload the book or, as it's usually done, point some web form to it so it can read the file, calculate crc32 or md5 sum from it, and look up those sums in its database, and tell me what is the book's source. Often, when I think of such things, and then look for them, they turn out to be existing sites with big databases already. This time, I could find nothing. Does anybody know of such a site?

[mr ploppy]

MD5, etc wouldn't really work on a pirate ebook because it would change when the DRM restrictions were removed.

[Krystian Galaj]

Originally Posted by mr ploppy

MD5, etc wouldn't really work on a pirate ebook because it would change when the DRM restrictions were removed.

Many, if not most, pirate ebooks aren't made by removing DRM restrictions, but by scanning paper books.

Also, in that case the database would just have two entries for that book, one with DRM, the other one after it was removed - or wouldn't include DRMed books at all (I don't know exactly how DRMed books work, it's possible that the resulting DRM-ed file from the same book would be different for every user).

Edit: though I remember I read here once about books being offered for free, but with DRM (so requiring to have an account on the site, and often also working credit card, even if there would be no charge for the book).

[speakingtohe]

can't you just search for the title/author?

[Krystian Galaj]

Originally Posted by speakingtohe

can't you just search for the title/author?

No, because well-known book would likely return such a number of results that wouldn't be able to go through them all to find any meaningful promotion links, and you wouldn't be able to say if it's available anywhere for free or legally just by going through Google results.

[Richey79]

How fantastic would it be to have an equivalent of ISBN that followed each book on the WWW, even if there were OCR errors or if they were PDFs?

A hypothetical way for the system to work might be how Windows Media Player finds album art for music, whatever its provenance, or those pieces of software that allow you to hum a few bars of a tune and tell you its name and the composer.

This is an idea that hadn't occurred to me, and a really interesting basis for a thread.

[BenG]

I seem to remember years ago on the Usenet e-book groups someone wrote software that could tell if two files were actually the same file or from a different source. I'm not sure if he kept a database or not.

[Falcao]

Originally Posted by Krystian Galaj

Edit: though I remember I read here once about books being offered for free, but with DRM (so requiring to have an account on the site, and often also working credit card, even if there would be no charge for the book).

Indeed. I downloaded some free ebooks from Amazon and most have DRM.

[Krystian Galaj]

Originally Posted by BenG

I seem to remember years ago on the Usenet e-book groups someone wrote software that could tell if two files were actually the same file or from a different source. I'm not sure if he kept a database or not.

There are many programs that can compare files by content to see if they're identical. If you mean comparing with some kind of database, I'd be happy to find that program and learn if the database was systematically expanded with new ebooks showing up around the Net.

[Ken Maltby]

Perhaps it is the cynic in me, but I would think the most obvious players that whould try
and compile and keep such a data base would be the legal staff of the DRM proponents.

Luck;
Ken

[Krystian Galaj]

Originally Posted by Ken Maltby

Perhaps it is the cynic in me, but I would think the most obvious players that whould try
and compile and keep such a data base would be the legal staff of the DRM proponents.

Luck;
Ken

They would have to be very clean about it though, and it's hard to compile such a database without potentially downloading some illegal books. Unless their database was limited to known legal/officially sold books only, in which case it wouldn't be of much use for removing pirated versions.

Even if they did have all illegal books in such database, I'm not sure it would benefit them. If they spot some book on Rapidshare etc., going by a link posted on the Net, they probably can't even ask Rapidshare to check its database for all copies of the ebook on various user accounts and remove them all, because some of those may be legally made backup copies by the original creator of the ebook. (I assume that Rapidshare saves its space, by calculating various checksums internally, and if they have identical 700MB video file on 100 different accounts, they store it only once on their servers).