Going through my #evolutionarily logs it is becoming very apparent to me that I need a good text mining application. Something that will work like the Bank of English but that allows me to use just my own text selections. As I piece together more and more log files with the help of current community members, the task of extracting trends in the way people discuss community is appearing increasingly daunting...

All I really want to do with this is compile lines that have the word 'community' in them for statistical analysis. But just going through it once, the shallow impression I have is that for #evolutionarily, the IRC channel of a clan initially devoted to playing Natural Selection, community is constantly under negotiation, and frequently the subject of many fights. Community is extremely contested in the data I have, which spans the year from May 19 2003 to Saturday November 1 2003.

#evo is looking more interesting than #ns for discussions of community since it's more tightly-knit, and can occupy a critical insider/outside stance when it comes to discussing 'the community'

Another thing I've noticed is that 'the community' is extremely important - it's often used as a yardstick to measure how good/bad a mod is (it has a active or supportive/lame community), or how good a forum/chat member is (he contributes a lot to the community /or/ i've been in this community for X years).

There's also lots of nostalgia - apparently my logs date back only to after the release of Natural Selection, and many users warmly remember the pre-release days, 'before all the noobs came'

I also need to find a way to establish some demographic information. I believe it should be in here, spread out over the years, because #evo became quite a personal community. Just need to develop some methods for searching them out.

And let's not even talk about the IRB. Though it appears I need my faculty advisor to do it for me? this needs more looking-into