Open elexunix opened 3 years ago
Yes, of course, they are tautologies
But it prevents proving&testing propositional lemmas... As random structs are no longer considered to be suitable for substitution. Seems like one would always have to construct silly wffs and use them to check lemmas of this kind. Maybe there is a better solution?
It would be better for the axioms from namespace propositional (hypothesis_addition, conditional_modus_ponens, ...) to check whether their arguments are well-formed, wouldn't it?
For example, do we permit substituting variable<conjunction<a, b>> and b into the axiom x -> y -> x? Or it is better to generate an illformed type or compile-error at this moment?