Re: type safety

Subject: Re: type safety
From: Macneil Shonle (Macneil.Shonle@Sun.COM)
Date: Mon Feb 04 2002 - 11:05:08 EST

Hi Matthias,

Then where does the compiler-correctness fit in to all of this? Would that mean
a statically typed program still won't ever "go wrong"? Or, to phrase my
question another way, how can a statically typed program ever signal a type


> No no no. Typed programs go wrong when they signal a type error.
> The type soundness theorem is nothing more than a semantics coherence
> theorem that says there should be no other behavior.
> If the VM/compiler doesn't live up to that, it is wrong. It's its
> specification.
> -- Matthias

This archive was generated by hypermail 2b28 : Mon Feb 04 2002 - 11:05:13 EST