Hacking the library

I'm working on a Python script to look up basic information about books using a Library of Congress call number. The purpose of this script is to make my life easier. What I'm wondering is if there is a simple database out there my script could rely on to turn the LC numbers into book data. The actual Library of Congress Online Catalog might work, but it seems very slow. In brief, what I need is a database that has the following characteristics:

• It has to be relatively easy for my script to access. Straightforward HTTP requests, hopefully. The data has to arrive in a format that is easy to work with.
  
• It has to be reliable. I'm going to be using it a lot, so it has to be online most of the time.
• It has to be fast. As mentioned, the L of C's online catalog is probably too slow for my needs.

Oh yeah, and it has to be free, too. But there's got to be something out there, right? I was thinking maybe I should just use a college library's server to do my dirty work. BGSU's server is kind of clunky and the output it generates will be difficult for my script to process properly. At the end of the day, I may end up using it, but it would be nice to find something better.

I realize this post has nothing to do with politics or the law. Or philosophy. But I DO have other interests!