Skip to main content
8 events
when toggle format what by license comment
Jul 3, 2019 at 10:46 comment added David Corwin Looking back at this, the main problem I had with the D-G proof was that I did not understand (co)simplicial objects very well then.
Apr 10, 2018 at 6:48 answer added მამუკა ჯიბლაძე timeline score: 3
Jan 20, 2013 at 0:40 answer added Justin Curry timeline score: 15
Jul 26, 2012 at 1:03 comment added David Corwin Thanks, I had seen that proof, but I found it to be unnecessarily non-elementary.
Jul 25, 2012 at 19:09 comment added AFK You can also rewrite the proof quite easily if you remember that (1) Goncharov's objective is to compute $H_n(P^n_{x,y}X,\partial P^n_{x,y} X)$ ($P^n_{x,y}X=\{x\}\times X^n \times \{y\}$ the cosimplicial path space) (2) In the formalism of 4 operations $$H_n(X,Z;A) = {}^\tau H^{−n}(a_{X!} j_{∗} j_{∗} a^{!}_X A)$$ with $\tau$ is the t-structure, $a_X:X\to pt$ the structural morphism, $j:X\setminus Z \hookrightarrow Z$ is the open immersion. (3) The proof in Deligne-Goncharov is Poincaré dual in the sense that they compute $$ H^n(X \mod Z;A) = {}^{\tau} H^n( a_{X } j! j^! a_X^{} A ) $$
Jul 25, 2012 at 19:02 comment added AFK The proof was rewritten in the framework of sheaf cohomology in Deligne and Goncharov's "Groupe fondamentaux motiviques de Tate mixtes". This is french but should still be easier to understand.
Jun 21, 2012 at 13:50 comment added S. Carnahan Wild guess for (4): Given a subvariety, you may push the constant cosheaf on that subvariety forward along the inclusion map, and take the derived functors.
Jun 19, 2012 at 4:07 history asked David Corwin CC BY-SA 3.0