In this talk I will discuss joint work with Bernhard Leeb and Misha Kapovich. I will explain why the three linear algebra problems discussed by Fulton in his Bulletin article (BAMS 37) all lead to the same set of inequalities - we do not discuss the fourth problem of decomposing tensor products. I will then discuss the possible generalizations to other semisimple algebraic groups G split over Q. At the moment we have only partial results for the p-adic case for general G. This is the generalization of the problem of specifying the invariant factors of a product in advance. In terms of geometry it involves constructing n-gon linkages with given “side-lengths” and given vertex type in Bruhat-Tits buildings.