Well, for what its worth, I did manage to find a package that would convert pdf files to text, which makes indexing possible, yet it would be much better (and cooler) to be able to suck the text out of the pdf's with our favorite language.

Maybe I'll get ambitious and figure out the structure and do it by hand, but I'm not sure I'm ready (perl-wise) to tackle that kind of package. Laziness, hubris ...