Date: Mon, 13 Oct 2025 16:51:44 -0500
On Mon, Oct 13, 2025 at 3:32 PM John Spicer via SG15 <sg15_at_[hidden]>
wrote:
> Ryan has a paper (which will hopefully be in a mailing soon), where he
> says:
>
> [image: Screenshot 2025-10-13 at 4.27.49 PM.png]
>
> The problem with contracts, as I describe in the mail quoted here and in
> P3835, is that in a large number of cases the programmer does not know
> whether or not the contract check will “stop the program from continuing”.
>
> For any function whose definition is in a header file, the no one knows
> what will happen when the function is executed unless every single object
> file (including libraries) in the whole program is compiled with an
> enforcing semantic.
>
Did you mean ".. whole program is compiled with a __consistent__ semantic"?
wrote:
> Ryan has a paper (which will hopefully be in a mailing soon), where he
> says:
>
> [image: Screenshot 2025-10-13 at 4.27.49 PM.png]
>
> The problem with contracts, as I describe in the mail quoted here and in
> P3835, is that in a large number of cases the programmer does not know
> whether or not the contract check will “stop the program from continuing”.
>
> For any function whose definition is in a header file, the no one knows
> what will happen when the function is executed unless every single object
> file (including libraries) in the whole program is compiled with an
> enforcing semantic.
>
Did you mean ".. whole program is compiled with a __consistent__ semantic"?
-- -- René Ferdinand Rivera Morell -- Don't Assume Anything -- No Supongas Nada -- Robot Dreams - http://robot-dreams.net
Received on 2025-10-13 21:52:02
