> I merely found the concept offered, that a processor architecture
> isn't a business decision, to be, shall we say, quaint.

It can be, but isn't always.

> From the people who conceive it, thru the people who implement it,
> thru the people who code for it, to the people who use it, processor
> architecture is an amalgam of ideas and compromises driven by
> business purposes, ultimately to serve business purposes.

This is true to the extent - and *only* to the extent - that business
ends are what drive those various steps.

Consider the recent thread talking about doing a VAX-11/780 in an FPGA
(or at least, something vaguely along those lines; I don't know all the
precise technical terms).

That was not conceived for business purposes.  If it's implemented, it
won't be for business purposes (at least not if it's done per sketch).
The people who code for it - heck, most of the coding I do is done on
my own time at home, and is by no stretch of the imagination done for
business purposes.  And my use of my machines, well, if you stretch a
point, three of my machines can be considered to be used for business
purposes, the three involved in my receiving email - but that's really
stretching a point, and even then, it accounts for only three machines
and two architectures, of the 14 machines and 7 architectures I have
set up, never mind the dozens of machines that would need work before
they'd boot.

No, I make no claims to being typical of anything.  Just pointing out
that not everything is driven - or even really influenced - by business
purposes.

(For the curious: the seven architectures are SPARC, 68k, MIPS, VAX,
PowerPC, Alpha, x86.)

/~\ The ASCII				der Mouse
\ / Ribbon Campaign
 X  Against HTML	       mouse@rodents.montreal.qc.ca
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B