On Two Different Kinds of Computational Indeterminacy