Want to take part in these discussions? Sign in if you have an account, or apply for one below
Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.
I have separated Eilenberg-Watts theorem from abelian category and added the references and MR links. One of the queries from the abelian category is moved here with backpointer there. I cleaned up some typoi.
The following discussion is about to which extent abelian categories are a general context for homological algebra.
Zoran: I strongly disagree with the first sentence, particularly with THE (it is THE general context for linear algebra and homological algebra). MacLane was (according to Janelidze) looking whole life for what is the general context for homological algebra, and the current answer of expert are semi-abelian categories of Borceux and Janelidze, and homological categories…Linear algebra as well makes sense in many other contexts. This “idea’ is to me very misleading. MacLane in 1950 was lead by the idea to axiomatize the categories which behave like abelian groups. Grothendieck wanted to unify on the obsrervation that the categories of abelian sheaves and categories of R-modules have the same setup for homological algebra as in Tohoku.
There is much linear algebra you can do with cokernels, for example, as well as much linear algebra which you can not do if you are not over a field for example. So, saying that abelian categories are distinguished is only among categories which have closest properties to abelian sheaves and R-modules, not among principles for homlogical algebra and linear algebra that uniquely (although the strong motivation was ever there).
Mike: I changed it to “a” general context; is that satisfactory? Once we have pages about those other notions, there can be links from here to there.
Toby: I've made the phrasing even weaker. Abelian categories are pretty cool, but (if you don't already have the examples that make it so useful) the definition is a fairly arbitrary place to draw the line.
Tim : I note that sometimes we (collectively) take parts of a discussion and turn it into part of an entry, because of that I would like to note two points here. The first is that the accepted first definition of semi-abelian category is in the Janelidze, Marki, and Tholen (JPAA, but we have a link on the semi-abelian entry.)
The other point is that Tim Van der Linden’s thesis does a lot of stuff that could be useful. It is available online http://arxiv.org/abs/math/0607100
1 to 1 of 1