text/ncat.tex
changeset 796 d30537de52c7
parent 795 4d66ffe8dc85
child 797 40729de8e067
equal deleted inserted replaced
795:4d66ffe8dc85 796:d30537de52c7
   717 This map is natural with respect to the action of homeomorphisms and with respect to restrictions.
   717 This map is natural with respect to the action of homeomorphisms and with respect to restrictions.
   718 %\item Product morphisms. \nn{Hmm... not sure what to say here. maybe we need sets with structure after all.}
   718 %\item Product morphisms. \nn{Hmm... not sure what to say here. maybe we need sets with structure after all.}
   719 \end{itemize}
   719 \end{itemize}
   720 \end{axiom}
   720 \end{axiom}
   721 
   721 
   722 
   722 \medskip
   723 
   723 
   724 
   724 When the enriching category $\cS$ is chain complexes or topological spaces,
   725 \nn{a-inf starts by enriching over inf-cat; maybe simple version first, then peter's version (attrib to peter)}
   725 or more generally an appropriate sort of $\infty$-category,
   726 
   726 we can modify the extended isotopy axiom \ref{axiom:extended-isotopies}
   727 \nn{blarg}
   727 to require that families of homeomorphisms act
   728 
   728 and obtain an $A_\infty$ $n$-category.
   729 \nn{$k=n$ injectivity for a-inf (necessary?)}
   729 
   730 or if $k=n$ and we are in the $A_\infty$ case, 
   730 We believe that abstract definitions should be guided by diverse collections
   731 
   731 of concrete examples, and a lack of diversity in our present collection of examples of $A_\infty$ $n$-categories
   732 
   732 makes us reluctant to commit to an all-encompassing general definition.
   733 \nn{resume revising here}
   733 Instead, we will give a relatively narrow definition which covers the examples we consider in this paper.
   734 
   734 After stating it, we will briefly discuss ways in which it can be made more general.
   735 
   735 
   736 \smallskip
   736 Assume that our $n$-morphisms are enriched over chain complexes.
   737 
       
   738 For $A_\infty$ $n$-categories, we replace
       
   739 isotopy invariance with the requirement that families of homeomorphisms act.
       
   740 For the moment, assume that our $n$-morphisms are enriched over chain complexes.
       
   741 Let $\Homeo_\bd(X)$ denote homeomorphisms of $X$ which fix $\bd X$ and
   737 Let $\Homeo_\bd(X)$ denote homeomorphisms of $X$ which fix $\bd X$ and
   742 $C_*(\Homeo_\bd(X))$ denote the singular chains on this space.
   738 $C_*(\Homeo_\bd(X))$ denote the singular chains on this space.
   743 
   739 
   744 
   740 \nn{need to loosen for bbc reasons}
   745 
   741 
   746 %\addtocounter{axiom}{-1}
       
   747 \begin{axiom}[\textup{\textbf{[$A_\infty$ version]}} Families of homeomorphisms act in dimension $n$.]
   742 \begin{axiom}[\textup{\textbf{[$A_\infty$ version]}} Families of homeomorphisms act in dimension $n$.]
   748 \label{axiom:families}
   743 \label{axiom:families}
   749 For each $n$-ball $X$ and each $c\in \cl{\cC}(\bd X)$ we have a map of chain complexes
   744 For each $n$-ball $X$ and each $c\in \cl{\cC}(\bd X)$ we have a map of chain complexes
   750 \[
   745 \[
   751 	C_*(\Homeo_\bd(X))\ot \cC(X; c) \to \cC(X; c) .
   746 	C_*(\Homeo_\bd(X))\ot \cC(X; c) \to \cC(X; c) .
   752 \]
   747 \]
   753 These action maps are required to be associative up to homotopy,
   748 These action maps are required to be associative up to coherent homotopy,
   754 %\nn{iterated homotopy?}
       
   755 and also compatible with composition (gluing) in the sense that
   749 and also compatible with composition (gluing) in the sense that
   756 a diagram like the one in Theorem \ref{thm:CH} commutes.
   750 a diagram like the one in Theorem \ref{thm:CH} commutes.
   757 %\nn{repeat diagram here?}
   751 %\nn{repeat diagram here?}
   758 %\nn{restate this with $\Homeo(X\to X')$?  what about boundary fixing property?}
   752 %\nn{restate this with $\Homeo(X\to X')$?  what about boundary fixing property?}
   759 On $C_0(\Homeo_\bd(X))\ot \cC(X; c)$ the action should coincide 
   753 On $C_0(\Homeo_\bd(X))\ot \cC(X; c)$ the action should coincide 
   766 and we can replace the class of all intervals $J$ with intervals contained in $\r$.
   760 and we can replace the class of all intervals $J$ with intervals contained in $\r$.
   767 Having chains on the space of collar maps act gives rise to coherence maps involving
   761 Having chains on the space of collar maps act gives rise to coherence maps involving
   768 weak identities.
   762 weak identities.
   769 We will not pursue this in detail here.
   763 We will not pursue this in detail here.
   770 
   764 
   771 A potential variant on the above axiom would be to drop the ``up to homotopy" and require a strictly associative action. (In fact, the alternative construction of the blob complex described in \S \ref{ss:alt-def} gives $n$-categories as in Example \ref{ex:blob-complexes-of-balls} which satisfy this stronger axiom; since that construction is only homotopy equivalent to the usual one, only the weaker axiom carries across.)
   765 One potential variant of the above axiom would be to drop the ``up to homotopy" and require a strictly associative action. 
       
   766 (In fact, the alternative construction of the blob complex described in \S \ref{ss:alt-def} 
       
   767 gives $n$-categories as in Example \ref{ex:blob-complexes-of-balls} which satisfy this stronger axiom; 
       
   768 since that construction is only homotopy equivalent to the usual one, only the weaker axiom carries across.)
   772 
   769 
   773 Note that if we take homology of chain complexes, we turn an $A_\infty$ $n$-category
   770 Note that if we take homology of chain complexes, we turn an $A_\infty$ $n$-category
   774 into a ordinary $n$-category (enriched over graded groups).
   771 into a ordinary $n$-category (enriched over graded groups).
   775 In a different direction, if we enrich over topological spaces instead of chain complexes,
   772 In a different direction, if we enrich over topological spaces instead of chain complexes,
   776 we get a space version of an $A_\infty$ $n$-category, with $\Homeo_\bd(X)$ acting 
   773 we get a space version of an $A_\infty$ $n$-category, with $\Homeo_\bd(X)$ acting 
   777 instead of  $C_*(\Homeo_\bd(X))$.
   774 instead of  $C_*(\Homeo_\bd(X))$.
   778 Taking singular chains converts such a space type $A_\infty$ $n$-category into a chain complex
   775 Taking singular chains converts such a space type $A_\infty$ $n$-category into a chain complex
   779 type $A_\infty$ $n$-category.
   776 type $A_\infty$ $n$-category.
       
   777 
       
   778 One possibility for generalizing the above axiom to encompass a wider variety of examples goes as follows.
       
   779 (Credit for these ideas goes to Peter Teichner, but of course the blame for any flaws goes to us.)
       
   780 Let $\cS$ be an $A_\infty$ 1-category.
       
   781 (We assume some prior notion of $A_\infty$ 1-category.)
       
   782 Note that the category $\cB\cB\cC$ of balls with boundary conditions, defined above, is enriched in the category
       
   783 of topological spaces, and hence can also be regarded as an $A_\infty$ 1-category.
       
   784 \nn{...}
       
   785 
       
   786 
   780 
   787 
   781 \medskip
   788 \medskip
   782 
   789 
   783 We define a $j$ times monoidal $n$-category to be an $(n{+}j)$-category $\cC$ where
   790 We define a $j$ times monoidal $n$-category to be an $(n{+}j)$-category $\cC$ where
   784 $\cC(X)$ is a trivial 1-element set if $X$ is a $k$-ball with $k<j$.
   791 $\cC(X)$ is a trivial 1-element set if $X$ is a $k$-ball with $k<j$.