Re: type safety
On Mon, 10 Jan 2000, Joe Wells wrote:
> type safety: safety (for one program or for an entire programming
> language) enforced by requiring well typedness in some type
> * Some people might define safety to also require the absence of
> uncatchable exceptions. However, as long as these uncatchable
> exceptions generate well defined behavior (e.g., halting), I think
> it is better to consider them safe.
Joe, I infer from these statements, together with the fact
that we can't tell at compile time whether an array index is in
bounds, that any language which does not check array bounds
dynamically cannot be type-safe. Is my inference correct? If
not, how are such languages handled in your view?