Caution: dissertating in progress

I haven't really been saying nearly enough about my research lately. However, for the most part it has involved hacking away on AspectML to slowly remake it as InforML. The new frontend is kind of finished. I still need to fix the parser so it generates output that satisfies the Barendregt convention. Aaron Turon deserves considerable credit for all the help and hacking he's been doing to get ml-antlr (an LL(k) parser generator for SML) and ml-ulex working for me and the peculiarities of my implementation style.

The past week or so I've been focusing on extending my current »specification« with descriptions of the algorithms for solving label constraints, kind and type subsumption, and type joins and meets. This has been particularly hard on my brain. But I think I'm closing into something that should be sound and close to complete, at least without having explicitly done the proofs. But my committee decreed that I should not prove anything from my proposal on out, so it will have to do. I'm hoping maybe that tomorrow I'll have settled on a essentially final design and will be able to get back to implementation.

The sort of vague plan is to aim for an ICFP submission (which I just learned today will be doing double-blind reviewing this year) with the plan to defend in time for an August graduation date. It could happen.

I spent some time of Friday putting my current dissertation draft into the correct format for the Ruler Lady. I am hoping to provide an alternate format for pleasant reading; I did some pricing on Lulu and definitely think I'll print some nice copies for a few people when I'm done.

Leave a Comment