@Himeshi My plan is that I will try to apply the floating-point semantics of Ruemmer and Wahl. Please feel free to provide feedback on this. This should be done after #8 is resoved.
We are to use our own precision error expressions for general numerical precision analysis not restricted to floating-points only, so this issue is irrelevant.
@Himeshi My plan is that I will try to apply the floating-point semantics of Ruemmer and Wahl. Please feel free to provide feedback on this. This should be done after #8 is resoved.