23 Sep
2001
23 Sep
'01
7:57 a.m.
To describe this application as 'microcode' is
indefensibly wrong.
Bob et al.
Any blanket statement like this is generally false and this one is not
excluded. (For you language weenies, parse that one :-))
One would be more clear if they wrote:
"If you describe this application as 'microcode' then I won't understand
what you are talking about."
That of course is provably correct :-)
Computer geeks have been abusing the English language since the the moment
they co-opted the term "computer." If you want to get particularly anal
there are only two types of implementation, those that use state machines
and those that use combinatorial logic.
The Feynman lectures on Computation are an excellent discussion of this.
On this mailing list, we're interested in _communication_ more than a
particular set of definitions thus if someone says "Microcode a PDP-11 on a
Pentium" and someone finds the statement confusing because their definition
of the word micro-code doesn't allow for considering assembly code
microcode, and yet at the same time they recognize what the author was
trying to say, then lets just leave it at that. Now if the reader doesn't
understand what the author is trying to say then asking the question
regarding the definition of microcode that the author is using is both a
valid question and helpful.
A lot of great stuff has come out of this conversation so far, this
includes various microcode design references, further illumination on the
architecture of the PERQ and other machines. Lately however this thread is
showing dangerous signs of becoming chaff.
Sorry for intruding,
--Chuck