revisiting the "What am I running on?" question

[eryk sun]

On 2/20/19, Thomas Jollans <tjol at tjol.eu> wrote:
>
> I'm fairly sure "win32" was used on W9x as well. In any case it *was*
> correct at the time, as early versions of Python also ran on DOS and
> Windows 3.1. "windows" would not have been suitable.

DOS Python would be a separate platform. But if there was really a
Windows 3.x version of Python, I wouldn't have a problem calling it a
"Windows" platform. We could differentiate implementations via os.name
("win3", "win4", and "winnt" -- not "nt"). On a technical note, the NT
subsystem name is "Windows", not "Win32", so there's that.

> But yeah, we're stuck with it. There are obviously good reasons that it
> wasn't changed when we moved to amd64, but it is annoying.

The win-amd64 ABI is significantly different, but at the API level
there isn't a drastic difference between 32-bit and 64-bit Windows, so
there's no cognitive burden with perpetuating the Win32 name. The
official API name was actually changed to "Windows API" or WINAPI (or
WinAPI). But it would require a massive effort to change the culture.
There's no pressing need to expend that much time and energy over a
name.