I had a productive conversation with Didier Rémy this morning about various type inference topics; he had given a very nice talk on the latest MLF research yesterday. He also seemed quite interested in OCamlTeX. I read through some papers by Pottier on solving subtyping constraints in preparation for beginning work on InforML.
I'd like to think that I should have the higher-rank extension to AspectML finished soon, and then be done with the initial InforML prototype in a couple months. Really, how soon I graduate is proportional to how quickly I can get it running. It is unfortunate that I have a very clear idea of most of the algorithms and things that need to be done, but that there will be a considerable amount of grubby hacking and plumbing to realize them.