Date: Thu, 9 May 2024 14:00:01 +0100
On Thu, May 9, 2024 at 12:29 PM Sebastian Wittmeier wrote:
>
> What do you gain by standardizing pointer size? Some exotic
> architectures would not be C++ compliant any more for current
> standard versions.
Inclusion of minorities is a good thing (both people and compilers).
All of the pedestrian crossings in my town have dimples in the
concrete so that blind people know where to cross, and I'm in favour
of these dimples even though they bug me when I'm out rollerblading.
But I'd want the dimples gone if there were zero blind people in
existence. As far as I know, there isn't any C++23 compiler that has
sizeof(char*) bigger than any other data pointer. If a minority exits
then I'd like to continue to accommodate the minority, but if there's
no minority then there's no point in limiting the majority. There are
a few benefits to making all data pointers the same.
>
> What do you gain by standardizing pointer size? Some exotic
> architectures would not be C++ compliant any more for current
> standard versions.
Inclusion of minorities is a good thing (both people and compilers).
All of the pedestrian crossings in my town have dimples in the
concrete so that blind people know where to cross, and I'm in favour
of these dimples even though they bug me when I'm out rollerblading.
But I'd want the dimples gone if there were zero blind people in
existence. As far as I know, there isn't any C++23 compiler that has
sizeof(char*) bigger than any other data pointer. If a minority exits
then I'd like to continue to accommodate the minority, but if there's
no minority then there's no point in limiting the majority. There are
a few benefits to making all data pointers the same.
Received on 2024-05-09 13:00:13