Thoughtful Programming
- To: <MISC>
- Subject: Thoughtful Programming
- From: "Myron Plichota" <myron.plichota@xxxxxxxxxxxx>
- Date: Thu, 25 May 2000 11:39:04 -0400
I check into Jeff Fox's website every once and a while and recently noticed
the essay "Thoughtful Programming" at
http://www.ultratechnology.com/forth.htm
that presents the case for Machine Forth well enough to (finally) penetrate
my own habits and biases as a long-time Forth(79!) programmer. It is
extremely well written and organized and hits the nail on the head in
explaining the technical issues as well as the political collision between
standard industry practice and lean computing. It is also a manual for
writing good Machine Forth. It gave me a good kick in the head and I am now
experimenting with an emulation of the F21 instruction set and stack
machine model and many of the other ideas presented.
Not everyone will want to accept the challenge of the extreme level of
software engineering involved in getting the best possible performance and
memory footprint, but I intend to persue it for my own satisfaction. My
main problem is that there are large gaps in my education re: integer
numerical algorithms that obviate the need for floating-point.
I propose setting up a clearing house for such arcana. I can contribute a
few of the insights and algorithms that I am familiar with such as the
CORDIC sin/cos algorithm and some DSP and miscellaneous stuff. I would also
like to see a bibliography of any already existing reference material that
anyone may be aware of. (For example has anyone seen or owned a copy of
"The Kitt Peak Primer" and does it contain such material?) This would
definitely assist novice Machine Forth programmers concerned with such
things to hit the ground running and also benefit anyone using non-stack
integer processors. Collected into a printed edition, it could be a
fund-raiser for the F21 initiative. I would gladly spend US$100 for such a
compendium ("The 2's-complement Cookbook"?). What is the readership's
reaction to this idea? What language(s) or psuedo-language(s) should be
used as the publication standard?
Myron Plichota