I am obsessed with having correct metadata in my eBook library and am constantly trying to get the most complete, correct information for my books. Sometimes but not very often I will find a book that cannot be found from the other sources for whatever reason. I can almost always get the missing information and the ISBN by looking it up on
http://www.goodreads.com.
Since this site is such a good source for information, would it be possible to add it as a metadata source? I haven't looked in to the python scripts too closely yet but I'm assuming you just use some form of xml based scraping to get your information.