update link in doc

Readme.md

@@ -1,6 +1,6 @@
 ## Prototyping FTI for CouchDB databases in CouchDB
 
-**NOTE: This project is now incorporated into [Bitstore](http://github.com/bdionne/bitstore/tree/bitcask/src/search/ "Bitstore"), and will be continued there**
+**NOTE: This project is now incorporated into [Bitstore](http://github.com/bdionne/bitstore/tree/master/src/search/ "Bitstore"), and will be continued there**
 
 Chapter 20 of Joe Armstrong's <a href="http://www.pragprog.com/titles/jaerlang/programming-erlang">Erlang book</a> provides a nice example of the use of processes to do full text indexing with map/reduce. The essential idea is to spawn a process for each document to index and let the reduce function populate the inverted index as it collects the results of the map phase. I recently heard mention of <a href="http://dukesoferl.blogspot.com/2009/07/osmos.html">osmos</a> in a talk from the NoSQL east conference and it struck me as the ideal data structure for storing an inverted index, particularly since it supports user-defined merging. So when one encounters the word Neoplasm in multiple docs one can just write the key/value to the store and let a defined merging function sort things out.