[cdonnellytx]

Looks like a clear case of the Scunthorpe problem:

http://hashpix.herokuapp.com/s/scunthorpe (thanks @DjangoReinhardt!)

http://en.wikipedia.org/wiki/Scunthorpe_problem

(EDIT: still can't format comments correctly)

[DjangoReinhardt]

Heh, good catch! :P

I've been lazy about writing a good filter to weed out the nasty ones. I considered quite a few approaches - ranging from an NLTK-based algo to basic brute-force list-search with word-stemming. Then I realized, it would take me a lot more time to develop that, than the rest of the project.

So, I chose the easy way out. Turn out it saved me quite a bit of effort, too. :)