Re: Re: objects and forth
"John R. Strohm" <[email protected]> Sun, 8 Feb 2009 12:15:16 -0600
| Newsgroups | gmane.comp.lang.forth.colorforth |
|---|---|
| Message-ID | <58A442AF4310402CAA31D813A5C4840A@HAL> |
From: <[email protected]> To: <[email protected]> Sent: Sunday, February 08, 2009 11:57 AM Subject: Re: [colorforth] Re: objects and forth > > On Sun, 8 Feb 2009 00:28:24 -0600, "John R. Strohm" <[email protected]> > said: >> The question was whether learning C, Smalltalk, or LISP before attempting >> to >> learn FORTH would help or hurt. >> >> Smalltalk is, as near as I can tell without spending a LOT of time >> learning it >> in detail, inextricably tangled up in twisted weave of classes. None of >> C, >> LISP, or FORTH have that particular feature. All can be extended with >> classes. >> C has two main ways, Objective C and C++. LISP does it with CLOS. >> FORTH, you >> roll your own, but look at CREATE-DOES> for a starting point. >> >> It is not clear to me that requiring the student to master many >> intricacies of >> object-oriented design and programming is all that likely to help him >> learn the >> mindset of a language that explicitly does not include object >> orientation. >> >> On The Gripping Hand, as Niven & Pournelle fans say, the RIGHT way to >> learn >> FORTH is to grab a copy of Leo Brodie's wonderful book "Starting FORTH", >> and >> work through it. Follow that up with "Thinking FORTH", by the same >> author. > > I like the two mentioned books by Brodie, but _so much has changed_ > since they were written that I'm not sure if they should still be the > recommended introduction to Forth. (Bottom-posting to keep Aleksej from getting his knickers in a twist) While I grant you that "Starting FORTH" used polyFORTH-16 for the exercises and examples, and polyFORTH-16 is long gone, I remain convinced that the concepts behind FORTH at that time are still valid. Of course, I'm also the guy who got into a mild argument with Elizabeth Rather about how much of FORTH's productivity gains came from the highly-interactive nature of the implementation, and how much of it came from the block systems that forced ruthless factoring of definitions until every individual definition would fit in no more than 16 lines of at most 64 characters.