4
$\begingroup$

While studying a completely unrelated problem, I have proved something on the following line: given a diagram of simplicial sets $X: C \to \textrm{sSet}$, some deformations of the geometric realization $|X_c|$ fit into a diagram indexed by $C$ if relations are allowed to hold only up to homotopy coherence (in other words one get a functor from $N(C)$ to the $\infty$-category of spaces).

I do have a relevant application coming from my original problem, but I wanted to see if other nice examples can be given in different contexts. sSet-valued functors seem to be usually called simplicial functors, and the latter are apparently common in derived geometry to give higher analogs of classical concepts.

Let us for example consider a simplicial presheaf over a manifold (or some more general context if that is beneficial, e.g. a Grothendieck site). Is the openwise geometric realization of this presheaf worth to be studied? As an example of what I have in mind, it could be that the simplicial homology of the total sections admit an approximation in terms of simplicial homology over smaller open sets. The latter may be simplified by looking at the geometric realization, as the simplicial set could be a large model of a simpler object, up to homotopy. However, deforming the spaces openwise does not guarantee that the presheaf constraints are still satisfied, and one may have to move to the world of homotopy coherent diagrams.

My impression is that the geometry of these simplicial sets are not studied, and that simplicial sets are regarded as a categorical machinery to extend ordinary notions to higher analogs. However, since I am an outsider, it's worth asking. Thanks!

EDIT: my original example is quite simple but occurs naturally in several contexts. If one considers a cosimplicial diagram with values in simplicial sets, there is a (co)homological spectral sequence converging to the (co)homology of the homotopy totalization of the diagram (I think it is called Bousfeld-Kan). If one is ultimately interested in the geometry of such homotopy totalization, it could be convenient to replace the given simplicial sets with homotopy equivalent simpler spaces.

$\endgroup$

1 Answer 1

9
$\begingroup$

If I understand your question correctly, you may find some interesting results of a similar nature in the paper by Vogt (Homotopy limits and colimits, Math. Z., 134, (1973), 11 – 52.) which was followed up by Jean-Marc Cordier and myself in some papers (Vogt’s theorem on categories of homotopy coherent diagrams, Math. Proc. Cambridge Philos. Soc., 100, (1986), 65 – 90) and (Maps between homotopy coherent diagrams,Top. and its Appls., 28, (1988), 255 – 275) in particular. The results there are particular cases of results proved later by Joyal and by Lurie in a much more general context, but in the specific case that you mention in the question, i.e. functors with values in the category of simplicial sets, it is perhaps easier to unpick the geometric interpretation of the results and thus their application to cases of interest to yourself.

My own feeling is that the simplicial set valued functors often involve combinatorial structures that would be obscured if you pass to the geometric realisation. In the case of presheaves on spaces, you seem to be heading off into the realm of stacks, gerbes, etc. and for the geometric interpretation of these, a good point of entry is Larry Breen's notes (Notes on 1-and 2-gerbes, in J. Baez and J. May, eds., Towards Higher Categories, volume 152 of The IMA Volumes in Mathematics and its Applications, 193–235, 2009).

As to how one regards simplicial sets or more generally simplicial objects in some nice category, I have found the simplicial approcah an extremely useful means of encoding combinatorial relationships (e.g. in homotopy coherence). If one gets used to working with the face and degeneracy relations and with, e.g. products of simplicial sets via shuffles and similar devices, then one can obtain explicit formulae for relationships, whereas if one uses the corresponding spatial machinery one has to work quite hard to get out as much detail.

There are more recent expositions of all these ideas e.g. in Riehl and Verity's book (Elements of ∞-Category Theory, Cambridge Studies in Advanced Mathematics, Cambridge University Press), in Jacob Lurie's online Kerodon, and, of course, quite a few other sources.

I hope this helps.

$\endgroup$
8
  • $\begingroup$ Thanks. Two examples that Breen's notes made me think about are the classifying space and the nerve of a category. These are maybe different from the initial spirit I had in mind, since the domain is large category and not a geometric object (e.g. poset of open sets), but it can still be a nice source of examples. In the first case the geometric realization is relevant as e.g. the cohomology is interesting. As far as I understand, you seem to confirm that the combinatorics of these 'simplicial resolutions' is more interesting than its geometry, and so the realization has limited interest. $\endgroup$ Commented Nov 12 at 20:24
  • $\begingroup$ On a side note, I see a possible source of misunderstanding in my question, as there are two "higher" involved. First is replacing objects with simplicial resolutions that want to capture higher data. Second is replacing diagrams with homotopy coherent diagrams. I am interested in examples of strict diagrams of simplicial objects, so there is only the first "higher" involved. These are the 'input' of the lemma I came up with. The output of has also the second higher involved, but it's not quite relevant. Sorry if that was misleading. $\endgroup$ Commented Nov 12 at 20:30
  • $\begingroup$ In reply to your first Comment, I am not 100% what you mean by geometry when applied to geometric realisations. There is a combinatorial geometry which can come from a resolution. I enjoyed reading Loday's Homotopical Syzygies, in Une d ́egustation topologique: Homotopy theory in the Swiss Alps, volume 265 of Contemporary Mathematics, 99 – 127, which looks at both the idea of resolution and of the geometric structures involved,also M. Kapranov and M. Saito, 1999, Hidden Stasheff polytopes in algebraic K-theory and in the space of Morse functions, in , volume 227 of Contemporary Mathematics.. $\endgroup$
    – Tim Porter
    Commented Nov 13 at 6:24
  • $\begingroup$ For your second comment, any homotopy coherent diagram can be rectified to give an equivalent commutative diagram. That is more-or-less wht Vogt's theorem states. Note that Cordier's definition of homotopy coherent diagram involves a simplicial resolution of the domain of the functor defining the diagram. Playing around with that resolution is great fun and is quite insightful for what is going on. $\endgroup$
    – Tim Porter
    Commented Nov 13 at 6:28
  • $\begingroup$ Maybe I am wrong, but I think of simplicial sets being used for two different purposes: algebraic and geometrical. Algebraically they can be used e.g. as a non-commutative version of chain complexes, or to provide resolutions in higher context. The nerve of a category provides a nice algebraic example. Geometrical simplicial sets, instead, are mainly employed to give a combinatorial basis to the (co)homology of their geometric realizations. [cont'd] $\endgroup$ Commented Nov 19 at 3:41

You must log in to answer this question.