SemiOT: Mourning for Classic Computing

From: Chuck McManis <cmcmanis_at_mcmanis.com>
Date: Fri Aug 17 21:45:40 2001

At 06:28 PM 8/17/01 -0700, Sellam wrote:
>This is only the case if you are taught this way (methodical), or do not
>develop skills
>beyond this method of programming. The only reason we all (or most of us
>at least) drive cars in a civilized manner is because we are taught to do
>so. If it was up to us to drive the way we wanted, imagine the choas (and
>fun) on the roadway.

Sounds like Italy (the driving anyway!) I do recall that when I learned to
program in Fortran the school had a you send a deck of cards in that they
would run and then you would get your results back. It did make you stop
and think. Still I have to agree that teachers have a lot more influence
than language environment.

> > Needless to say, I consider 'modify at random until it works' to be a
> > very Bad Thing!
>
>One has to start somewhere! This is the best way to learn. Or did you
>just pick up a soldering iron one day and simply know what to do? I don't
>think so. Trial and error is the best learning tool.

My feeling is that trial and error develops an internal understanding
faster than methodical analysis, however good designs do require
methodology. In part because trial and error requires that you hit all
possible errors, which as the system gets larger becomes less and less likely.

>Tony Wrote:
> > I've seen hardware 'designers' do something similar with FPGAs (and
> > other technologies that are easy to modify). Things like 'maybe it'll
> > work if I change this AND gate to an OR gate' or 'I'll try inverting
> > that clock signal'. Or 'Maybe I need one more state in that counter'.
> > No real idea as to what they should be doing, and why.

You should meet Mike Kahn. He used an HP1600A and bascially started poking
around inside a Nintendo to make it do mysterious things. It made for a
very interesting VCF exhibit even if maybe only 15% of the people seeing it
understood what was going on. I have to admit that I have, on occasion just
programmed and FPGA to see if it would work rather than loading the
simulator, generating a stimulus file, and methodically simulating the part
before programming.

>Granted, if you want to become an engineer, or in the very least a
>"professional", you have to evolve beyond trial and error. But as I said,
>you have to start at the bottom and work your way up. I would never start
>a student off in a rigorous, structured environment. What a perfect way
>to turn them off.

This is very true, let someone get a feel for just how wide open the
choices are when programming and then introduce them to the most efficient
way to get the results they want.

--Chuck
Received on Fri Aug 17 2001 - 21:45:40 BST