NOTES 10/14/08:
*  email genic folks about effect of generation size and how good genic is about predicting number of clusters
*  evaluate distance metrics
*  get hamlet clustering working with ourmine
*  text corpus vs source code vs uci
*  research text clustering in news articles
*  lsi unix telcordia
*  lucene


NOTES 10/21/08:
*  lsi to create an index, allows similarity between documents, play with weight.  
*  committee clustering for next week
*