How do websites like Shelfari and Librarything collect book information?

Question

Do they manually add each book with its cover and author or do they use a scraper?

Answer 1

There is a pretty good write up on Google answers. It's from 2006, but I don't think the ISBN database market landscape has changed much since then.

I'd start by looking at Bowker

Answer 2

LibraryThing uses Amazon's API and volunteer libraries' Z39.50 connection to get book data. Each book is added as a LT member adds it to their catalog (which makes it basically manual).

The third way they get book data is from users manually entering it (also manually adding covers). Otherwise covers come from Amazon.