Hi! I'm the maintainer of mrjob. I really enjoyed your blog post (it's now linked from the mrjob readme). I just wanted to fix a couple of things to make sure your example continues to work in the future and is easy to get running.
Stuff I did:
Various improvements to docs and mrjob 0.4 compatibility
Link to blog post from README
Also fix up vectorSimilarities.py
Thanks irskep I will still improve the code with more examples! Stay tunned :)
Thanks for your collaboration.