Re: Formal semantics for SRFI 226
Wolfgang Corcoran-Mathe 14 Nov 2021 16:03 UTC
On 2021-11-13 20:46 +0100, Marc Nieper-Wißkirchen wrote:
> I am sorry for this rather negative answer, but the problem with writing
> down formal semantics is that they first have to be complete, secondly,
> they have to be correct, and thirdly, their completeness and correctness
> have to be verified. For anything less, they would lose their raison
> d'être.
As I wrote, it's a big request, so I appreciate your position.
I'm also glad to hear that you want the semantics to be verified; I
think this is a very important step. (Indeed, it's one reason that
I wish R7RS had used the PLT-Redex'ed R6RS semantics instead of the
old R5RS semantics, which were never formally verified and allegedly
have some inconsistencies.)
So it's something to work toward. Thanks very much for your response.
--
Wolfgang Corcoran-Mathe <xxxxxx@sigwinch.xyz>
"I once received a letter from an eminent logician … saying that
she was a solipsist, and was surprised that there were not others."
--Bertrand Russell