levenshtein distance and html content comparison

dear l33t readers,

i'm writing a (currently simple) ruby script that posts random values to html form variables, and checking the results. each value may give different results. so what i want to do is calculate the difference between the results so that i can group them together. to do this, i perform a post with defined values and set this as a baseline to be used as a comparison for the next post (the result of the first post is sort of like a 'signature' for the web page). and then i start posting data, and calculate the difference with the initial baseline. this can be done by levenshtien distance. unfortunately, my test scripts runs extremely slow*, and i have no idea on how to speed up the operation (one way is to use curl and use my ruby script to calculate the distance).

i can continue with this, but i'm wondering if there are better alternatives. so got one?

* it runs slow simply because i'm comparing tons of strings... maybe i should just strip the html tag...