PSSL review

[pozorvlak]

My review of the 83rd Peripatetic Seminar on Sheaves and Logic, in rec.juggling HLGCB style:

High: Francis Borceux's seminar on synthetic differential geometry (constructing infinitesimals by replacing the topos of sets with something more fancy) or Tom's four-line definition of Lebesgue integration. Both seriously cool. And the dinner at Balbir (

Comments

[michiexile]

I have, at one point in the past, managed to grok derived categories. It is a while ago, and afaicr, the basic deal is that you want to take a chain category and localize it in the morphisms that are quasi-isomorphisms; so that in the derived category objects and their resolutions end up being "the same thing".

I did get this told to me from a bunch of people doing almost only homological algebra; so my understanding probably is rather skewed by this.

[pozorvlak]

That was one of the things the guy said, but it's good to have independent corroboration. The other use of them is apparently to measure the extent to which left-exact functors fail to be right-exact (and dually). G-d knows how that's done, though...