Local uncertainty relations serving as measures of entanglement in a bipartite two-level system

We comment on the recent suggestion to use a family of local uncertainty relations as a standard way of quantifying entanglement in two-qubit systems. Some statements made on the applicability of the proposed "measures" are overly optimistic. We exemplify how these specific "measures" fall short, and present a minor modification of the general theory which uses the same experimentally gathered information, but in a slightly different, better way.