The problem with the upvote/downvote mechanism for things like books is that popular books tend to get voted higher than better, less popular books. If only 5 people have heard of a book, but they all think it's amazing, it should probably be around equal or higher than a book that a hundred people read and all thought was okay.

Come on we have an entire community of math and computer science wizards here, I think heuristics like that can be quite easily written. Especially using a tab for "Popular Books" and another "Highly Rated"

Of course. It's not a hard problem to solve, but it is one to consider.

You mean something like this: http://www.hackernewsers.com/books.html ? ;-)

I was just waiting for a link like that, I was sure someone had already done this

Edit: I do still think an up/down vote system to build a Top Hacker News suggested books, could be really cool to see

I agree. A first step is to search HN for "book review"[1] and see what the review says and how many upvote it has. I know it's not exactly what you ask for but it's a start.

[1] https://encrypted.google.com/search?q=site%3Anews.ycombinato...

http://www.hn-books.com/ appears to be something of an attempt to capture the book fancies of the HN crowd.