I agree that legislating the use of proprietary filtering products that do not work and often 
further the agenda of their parent corporations is wrong.  I also believe that there is much 
content on the net that needs to be screened out for children.  As such, I wonder, can a bayesian 
filtering algorithm similar to what has been proposed for screening out spam be applied to the 
problem of inappropriate content and protecting children from it?   Most browsers already keep a 
cache of content we have viewed.  How difficult would it be to add a mechanism allowing the user to 
drop bad content into one cache folder for filtering out bad content?  It would be fairly safe to 
assume that content not labelled bad was at least acceptable if not good.  Could organizations that 
police adult sites then publish lists of words and their weights from the corpus of sites they have 
on blacklists for use by parents and other concerned entities?  How difficult would it be to 
utilize this in a pro!
xy?  It would seem as though there is a a great need for filtering software that actually works and 
can be flexible enough to keep kids from inappropriate content, or workers away from games without 
rendering a large part of the internet inaccessible.  Just some thoughts sparked by the recent 
discussion of filtering content.  One other note, could such a tool also be used to increase the 
relevance of internet searches?  Any thoughts, ideas, links or dialogue are greatly appreciated.
Thanks.
Rob