Re: normalization of pointers...

In article <ff0s53dcohboqf049p7kea8i37rb71hf21@xxxxxxx>,
Mark McIntyre <markmcintyre@xxxxxxxxxxx> wrote:
On Wed, 30 May 2007 21:31:11 +0100, in comp.lang.c , "Malcolm McLean"
<regniztar@xxxxxxxxxxxxxx> wrote:

Most people program PCs most of the time,

Datum: I do servers, workstations, appliances, and occasionally my
Palm Pilot[1]. You could make a case for calling the workstation a PC,
but not any of the others, and workstations take up less than half of
my time-spent-programming and, in any given month, probably less than
at least one of the others. Not exactly "most of the time".


I beg to differ. Most people I know programme functionality not
hardware.

The functionality I do is recognizeably, though perhaps not tightly,
bound to the type of hardware (at least at the general-class-of level,
though usually not at the architecture level) it will be running on.


And I betcha the number of salaried embedded programmers still
outweighs wintelers.

Datum: At my day job, we ship a system running on Wintel that interfaces
with the outside world as a network-connected appliance (i.e. much the
same way some embedded systems would, and not at all like a PC would),
and everybody involved (including our manager) dreams of eventually moving
it to something that looks and feels more like an embedded system inside
the box as well. (We'd still need enough memory and CPU cycles that the
*real* embedded programmers would laugh and say "You call that embedded?",
but definitely not something any sensible person would call a PC.)


dave

[1] And I've written some code that could, and probably eventually
will, run on all of them, with no changes except which compiler I
feed it to.

--
Dave Vandervies dj3vande@xxxxxxxxxxxxxxxxxxx
I presume the difference is that Laurier's off-campus environment includes the
University of Waterloo, whereas the best we can offer by way of off-campus
environment is, well, Laurier. . . . --Chris Redmond in uw.general
.