Notes on 2-groupoids, 2-groups and crossed modules (Q858573)

The starting point for this paper is the relationship between homotopy 2-types of topological spaces and 2-groupoids: given a topological space \(X\) one can associate to it a 2-groupoid that encodes the homotopy 2-type of \(X\). Going further, one can ask whether this correspondence can be extended so that homotopy classes of maps between homotopy 2-types can be described as morphisms between the corresponding 2-groupoids. The answer to this question is no: in general there are too few 2-groupoid morphisms to produce all homotopy classes. The author writes: ``There are two standard ways to go about this problem. One is to enlarge the set of morphisms by taking into account weak morphisms between 2-groupoids. The other is to endow the category of 2-groupoids with a model structure and use a cofibrant replacement procedure to define derived mapping spaces. There seems to be a common understanding among experts how these things should be done and agreement that both approaches yield essentially the same results. But to my knowledge, there are not many, if any, places in the literature where one can look for precise statements, let alone proofs. This article is an attempt to fill this gap. The goal is to present various models for the mapping space between two given 2-groupoids (or weak 2-groupoids) and to verify that such models are naturally homotopy equivalent and that they provide algebraic models for the mapping space between the corresponding 2-types.''