J - 2003-05-02

How can I pinpoint a specific string's position in an html page so that if the string changes, I can find out what it is from that position.  For example, I parse http://sourceforge.net index page and find the position of [Project of the Month] (for April 2003 JBoss).  Let's say that the position is (3, 4, 5); I'm just making this up.  Unless SourceForge changes the format of their index page, I can get [Project of the Month] next month from the position (3, 4, 5).  How can I do that?  htmlparser appears to provide enough tools to do something like this, but I don't know how to make the best out of it (yet).  Any help would be appreciated.