On Tue, 10 Jul 2007 15:48:21 -0600, "Bob Myers"
wrote:


"Jeff Liebermann" wrote in message
.. .
In college, the senior electrical engineering class was polarized into
two camps, analog and ditital, largely by their choice of senior
projects. I made the mistake of designing a project that straddled
both camps (Secode Selector using RTL and DCL). Life was hell. The
debate came to a grinding halt when someone noticed that DNA sequences
were digital. So, if you dig deep enough into an analog world, you
eventually hit a digital bottom. I'll call it a win for whichever
side pays better this week.

I've tended to take a slightly different approach, which I hinted
at in my earlier response. It generally gets me some odd looks and
a "no, that can't be right" sort of reply, but I find it IS a helpful way
to look at things - at the very least, a different perspective that can
give you some new insights into how all this stuff "really" works.

That would be nice, but that's not the way engineering works these
days. At some point, most designers end up being either analog or
digital. Except in systems design, it's a rare engineer that can
function well in both camps. The result is usually microprocessor
acting as a marginal replacement for an op amp, or an analog circuit
that can't work in the real world because the tolerances and error
accumulation far exceed what could be done with digital. If you only
have a hammer, everything looks like a nail.

As I'd said, I tend to think of the "real world" as just that - it is
neither
"analog" nor "digital." From this perspective, those two terms simply
point to different means of encoding information for communication or
storage. I find that, all too often (again, at least from this
perspective),
we tend to use the words "analog" and "digital" when what we really
mean to say are things like "linear," "continuous," "discrete," "quantized,"
and so forth.

I just hate to agree with anyone, but you're correct. Analog/Digital
have become so vague that more specific terminology is required.
Still, the terms will not go away and must be dealt with as they
appear. Try searching Google for "analog engineering" and "digital
engineering" with the quotes. It's going to take a while for all
those hits to go away and be replaced by something more specific.

Fundamentally, I tend to see "analog" as simply meaning
" a system whereby information about a given parameter is encoded
by causing some other parameter (voltage, for instance) to vary in an
analogous manner." It doesn't necessarily mean "linear" or even
"continuous." Similarly, "digital" winds up with an even simpler
definition - "information encoded in the form of digits (numerical
values."

Agreed.

I've never found a situation where I couldn't use these words
with those interpretations.

Now you've done it. I'll be spending most of the day dreaming up
situations where the type of information encoding is ambiguous.
Offhand, quantum mechanics doesn't it either world, but then it
doesn't really fit any sane world, so that's not a good example.

And like I said, it IS often helpful - for one
thing, you wind up with a much better feeling for the real advantages
and disadvantages of "digital" and "analog" systems. (And you also
wind up not worrying about certain sillinesses, like whether power
systems are "analog" or "digital" - since the world no longer has to
be divided up exclusively as one or the other.)

Agreed.

Some people can't seem to wrap their minds around such things, but
then, I'm not really going to worry about that.

There's always a way to misinterpret something, no matter how clearly
it is stated. Besides, I like my illusions, even if they're wrong.

--
# Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060
# 831-336-2558
# http://802.11junk.com
# http://www.LearnByDestroying.com AE6KS