I remember talking about the game, and trying to codify the rules, but I don't think I ever saw it played. I think moves were specifying a group at a vertex, or specifying a morphism along an edge between two adjacent specified vertices.You could score a point and capture a square and take another turn if you could prove that the square commuted. Don't remember whether the rows and columns were required to be exact, or just that the composition of two adjacent vertical or horizontal arrows had to be the 0 map. Andy On Wed, Dec 3, 2014 at 12:25 PM, James Propp <jamespropp@gmail.com> wrote:

Does anyone besides me remember this game (Andy Latto, maybe)? I never played it, but when I was an undergrad math major at Harvard around 1980 I think I saw some grad students play it.

If I recall correctly, a legal move involved adding a new object and a new morphism to an ever-growing commuting diagram, subject to rules about exactness of sequences etc.

Jim _______________________________________________ math-fun mailing list math-fun@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun

-- Andy.Latto@pobox.com