Reproduction: How High is 'High'? Rethinking the Roles of Dimensionality in TDA and Manifold Learning