Dynamic Language Symposium

I enjoyed DLS this year. It had a mix of good and bad. I think the morning was better. The middle section was focused on papers that were doing something type related with dynamic languages. This was my biggest frustration with the sessions. I'm coming to believe that type theory is a pariah.

The other talks were good though. I already blogged on the APL one. The O-meta stuff was cool. It was interesting to listen to the Ambience guys talk about multi-methods. And once again we ended up at the question of "what happens when you have different paths with increased specifity." As usual... there was no real answer. This remains the achilles heel of the otherwise very powerful concept. Not a limiting one by the way. Just the thing you have to look out for and be aware of.

There was an interesting presentation about Scheme packages, and how they try to make scheme packages loadable into multiple dialects. I had distinct recollections of when I noticed that often this "interface" code can be as large as the library itself. SUnit being somewhat of an example.