I’ve just posted a draft of my thesis proposal, which, barring disaster, I’ll be presenting to my research committee next Friday. To spill the beans, here’s my thesis statement:
Lattice-based data structures are a general and practical foundation for deterministic and quasi-deterministic parallel and distributed programming.
Most of the material in the proposal already appears, in one form or another, in our existing papers on LVars. Section 4 of the proposal, though, is new and describes how I plan to investigate the relationship between LVars and CRDTs. On that note, I’ve just updated my post from earlier this month to reflect my recent thinking on threshold reads for CRDTs.
On another note, for most of the next week I’m volunteering as a resident at Hacker School. This is my second residency with them, but my first time in their recently acquired permanent space, a pleasant, high-ceilinged, bean-bag-chair-strewn second-floor office in lower Manhattan. (In fact, as far as I know, this is the first time they’ve had a permanent space.) I’ve already gotten to talk with a few amazing people, and I look forward to meeting many more when school is back in session on Monday!