Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

In his paper "Categories and cohomology theories" Graeme Segal considers the category of finite length chain complexes of finite dimensional vector spaces: Let $n = (n_i)_{i \in \mathbb{Z}}$ be a sequence of positive integers almost all zero. Then he claims that the space $K_n$ of chain complexes $E$ with $E^i = \mathbb{R}^{n_i}$ is a real algebraic variety. I am no algebraic geometer, but I suspect this to be similar to flag varieties somehow. So my first question is:

What is the topology on $K_n$?

As morphisms in the above mentioned category he takes chain homotopy equivalences between these complexes and claims that these also form a topological space. So:

What is the topology on Mor($E,F$) for two chain complexes $E$ and $F$ as described above?

share|improve this question
    
$K_n$ is the subset of $\oplus_{i}{\rm Hom}_{\bf R}(E_i,E_{i+1})= \oplus_i M^{n_{i+1}\times n_{i}}({\bf R})$ (where $M^{n_{i+1}\times n_{i}}({\bf R})$ are the $n_{i+1}\times n_{i}$-matrices with real coefficients) consisting of the elements $(d_i)$ such that $d_{i+1}\circ d_i=0$. So this suggests that the topology should be induced by the natural topology of the real vector space $\oplus_i M^{n_{i+1}\times n_{i}}({\bf R})$. –  Damian Rössler Sep 28 '11 at 7:45

1 Answer 1

up vote 7 down vote accepted

The space $K_n$ sits inside the space of sequences of linear maps $$L_n = \Pi_i Hom(E^i,E^{i+1}).$$ This is just a space of sequences of matrices, so it is a real vector space of dimension $\sum_i (n_i \cdot n_{i+1})$. We give it the usual euclidean topology for real vector spaces.

The subspace $K_n$ consists of those sequences of linear maps $\{f_i\}$ which form a chain complex - i.e., $f_{i+1} \circ f_i = 0$. This condition is polynomial in the entries in the matrices, so $K_n$ is a real algebraic affine subvariety inside $L_n$. We give $K_n$ the subspace topology.

The topology on the space of morphism can be described similarly by embedding the morphism set $Mor(E,F)$ into the space of sequences of linear maps $\Pi_i Hom(E^i,F^i)$, which is again a real vector space. The condition of being a morphism of chain complexes is real algebraic so the morphism space is again a real algebraic affine variety.

share|improve this answer
    
Thanks! That was much easier than I thought. –  Ulrich Pennig Sep 28 '11 at 8:21

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.