On the structure of Hom quandles (Q2001439)

Quandles are algebraic structures used extensively to construct invariants of knots. The study of general quandles can be difficult, thus the need of restricting oneself to some special classes of quandles. The Hom-quandles is an example of a sub-class that is worth studying. This class of quandles was previously investigated in [\textit{A. S. Crans} and \textit{S. Nelson}, J. Knot Theory Ramifications 23, No. 2, Article ID 1450010, 18 p. (2014; Zbl 1295.57007)] in which it was shown that if $(Y, *)$ is a medial quandle, that is \[(x*y)*(z*v)=(x*z)*(y*v),\quad \forall x,y,z,v \in Y,\] then the set of quandle homomorphisms $\mathrm{Hom}(X,Y)$ from any quandle $X$ to the quandle $Y$ has a natural quandle structure. The authors of [loc. cit.] used this Hom quandle to enhance the counting invariant of knots, that is the cardinality of $\mathrm{Hom}(Q(K), \mathrm{Hom}(X,Y))$, where $K$ is a knot and $Q(K)$ is its fundamental quandle. \par The paper under review uses a structural theorem of [\textit{P. Jedlička} et al., J. Algebra 443, 300--334 (2015; Zbl 1326.57026)] to show that it is enough to consider the source quandle $X$ being medial. This means that if both $X$ and $Y$ are medial then the structure of $\mathrm{Hom}(X,Y)$ is completely understood using the result of [loc. cit.]. The main theorem of the article under review is Theorem 3.14 which relates the Hom quandle to group homomorphisms between the components of the source and target. As a consequence, if the target quandle $Y$ is $2$-reductive, the authors can count the number of elements in $\mathrm{Hom}(X,Y)$ and give a characterization of $\mathrm{Hom}(X,Y)$.