Re: Declarative constraints in practical terms

vc wrote:
> It's a fascinating statement in the light of more than 40 years of
> denotational semantics history for imperative languages.

Yes, and it's taken that amount of time to achieve rather limited results. I admit that the word "useful" is a value judgement, but are you really disputing the difference between "*easy* to reason about" and "*possible* to reason about (to a very limited extent and then only after a great expenditure of effort)"?

> On the other
> hand, all the 'pure' languages can have operational semantics
> (Mercury/Prolog ->WAM).

They all have to run on something, but how that happens is not (cannot be) mandated by a pure language.

• Ralph