document similarity - automatic bookmark sorter

i've had this idea in the back of my head, to write something that would
take the bookmarks i've collected over past months, fetch their pages,
compare for similarity, and group accordingly.

i wonder first if anyone has already seen this utility

i wonder second if there is a good similarity-tester(*) to use.  if
so, it should be pretty easy to script up something workable.

* - based on word distributions?

On Wed, 22 Sep 2004 07:52:41 -0700, socks wrote:

> i wonder second if there is a good similarity-tester(*) to use.  if
> so, it should be pretty easy to script up something workable.
>
> * - based on word distributions?

no replies, so maybe there is no easily available unix/c/perl/whatever
library or utility.

fwiw, i happened to be reading the GMail blurbs yesterday and thought over
their "Search, don't sort" slogan in this context.  maybe what i want is
just a browser that offers an easy search of past bookmarks.  ideally the
bookmark menu would be a rolling history - the 20 things most recently
marked would be directly accessible - the rest would be content-cached
for easy searching.

it would be more like "tag this page as interesting", and "search
interesting pages for ..."

that way i could surf all day for info on new tvs, but when i went back to
look for info on that samsung, i could search my interesting/bookmarked
pages for that word.  (the browsers i have can only search history for
title or location.)