On 2020-May-17 16:08:26 -0400, Clem Cole <clemc@ccc.com> wrote:
>On Sun, May 17, 2020 at 12:38 PM Paul Winalski <paul.winalski@gmail.com>
>wrote:
>
>> Well, the function in question is called getchar().  And although
>> these days "byte" is synonymous with "8 bits", historically it meant
>> "the number of bits needed to store a single character".

8-bit bytes, 32/64-bit "words" and 2's complement arithmetic have been
"standard" for so long that I suspect there are a significant number of
computing professionals who have never considered that there is any
alternative.

>Yep, I think that is the real crux of the issue.  If you grew up with
>systems that used a 5, 6, or even a 7-bit byte; you have an appreciation of
>the difference.

I've used a 36-bit system that supported 6 or 9-bit bytes.  IBM Stretch even
supported programmable character sizes.

>DEC was still sort of transitioning from word-oriented hardware (a lesson,
>Paul, you and I lived through being forgotten a few years later with
>Alpha);

The Alpha was byte addressed, it just didn't support byte operations on
memory (at least originally).  That's different to word-oriented machines
that only supported word addresses.  Supporting byte-wide writes at
arbitrary addresses adds a chunk of complexity to the CPU/cache interface
and most RISC architectures only supported word load/store operations.

-- 
Peter Jeremy