I'd have one more talk... actually, if people are interested, and we can meet in Palo Alto, and it won't be Thanksgivings Eve (say, in Google we can schedule it for any other day), and Mike agrees, we could cover this topic again.
There's one thing in this Rosetta that totally puzzles me: why? Curry-Howard isomorphisms with two more components... or is it? Or is there something simpler, like just monads, whereby all 4 are (platonic) projections?
Thanks,
-Vlad