OK, OK, so I already tweeted this, but it’s interesting. Hillary Mason set up intelligent email autoresponders to deal with repetitive email enquiries and politely nag people for replies. It’s good stuff, and something I’ve never really got around to doing myself, for a couple of reasons (besides the obvious); I use Gmail, and to be honest, no two emails I send are the same.

(It’s an interesting overlap with Project India, by the way… I don’t know why I’m so excited about this year’s Group Projects. Either because they’re kind of real, or because I miss academia. Or both?)

The downsides of using Gmail haven’t really affected me personally, but thinking about it, I would like to be able to actually access my raw email to set up better, NLP-based filters. I have a lot of email filters, and a lot of labels, and a system that just about works (thanks to superstars and multiple inboxes). But, you know, it could be better, and despite IMAP access it doesn’t quite flow; if I wanted to process mail, I’d have to access it all on a random box, and then what? I can’t apply Gmail labels or superstars or mark as read, can I? I guess what I really want is to operate on both the protocol/content and the interface itself, and that’s asking a wee bit too much. Oh well. Time to hack on a communication platform that I can play with…

(credit to Craig, who linked me the video, and will complain if I don’t acknowledge his genius.)