Title: Discussion of "A Gibbs sampler for a class of random convex polytopes"

Abstract: An exciting new algorithmic breakthrough has been advanced for how to carry out inferences in a Dempster-Shafer (DS) formulation of a categorical data generating model. The developed sampling mechanism, which draws on theory for directed graphs, is a clever and remarkable achievement, as this has been an open problem for many decades. In this discussion, I comment on important contributions, central questions, and prevailing matters of the article.