trolfs@vax1.tcd.ie (Thomas Rolfs) (12/15/89)
ForthNet interaction to comp.lang.forth:
PORTED FROM xCFB's=>
------
Date: 12-07-89 (16:07) Number: 1429 (Echo)
To: GARY-S Refer#: 1425
From: STEVE PALINCSAR Read: NO
Subj: BASICS OF THE FORTH LANGU Status: PUBLC MESSAGE
I'd like to echo J. Wavrik's sentiments about the steepness or lack
thereof of the forth learning curve. To me, a "steep learning curve"
implies immense initial difficulties. I think the initial learning
curve with forth can be quite shallow indeed. After all, you only need
to learn one or two things to start fooling around with forth, learning
RPN, etc. You can try any concept by itself, in a small, simple word
that takes stack arguments (which are *quite* visible if you use .S) and
actually watch what it does. You can even run it with a debug word that
will let you trace its action. The scope of your problem is quite
limited, and you can play with it until you understand thoroughly what
it is doing. That's a far cry from steepness to me. As I understand
"steep," it's typified by a MS Windows or OS/2 Pres. Mgr. program that
requires about a thousand lines of cryptic preamblejust to print "hello
world" on the screen. I.e., you need to know a hell of a lt just to
get moving.
This incremental learning in small bits is what I understand by the idea
of "shallow" rather than steep. And it's true whether you have a simple
FIG type implementation, or something sophisticated as can be, like
HS/Forth, with its segmened dictionary andseveral hundred word lng
word list.