From: Conrad Schneiker/Austin/Contr/IBM <schneik@...>
Date: 2000-11-09T05:44:01+09:00
Subject: [ruby-talk:6179] Re: Mapping text to method calls

Stephen White writes:

# On Wed, 8 Nov 2000 hal9000@hypermetrics.com wrote:
# 
# > Just a few thoughts.
# 
# That was a cool wish-list. We want a lot of things in common!
# 
# Here's my wish-list, 

<snip, snip, snip, snip, snip, snip, snip, snip, snip, ....>  :-)

It sort of seems like what you are (implicitly) asking for is (to mention 
one of several possibilities) an IDE development language, a long list of 
(dynamically extensible) behavioral specifications in terms of such a 
language, and something of an expert system to weave everything together 
into a coherent system. 

What sort of overall model would others suggest as a general framework for 
dealing with some reasonable subset of the Hal and Stephen lists (plus 
related suggestions)? And how would you generally employ Ruby for 
implementing that model?

(Side note to Hugh: I forgot to mention that when I said "klunky" in 
connection with Tk, I didn't mean that as a synonym for "impractical" or 
"bad" per se; I just ment "not as nice". I probably should have said 
"somewhat klunky". Sorry for the needless confusion. Along these lines, I 
still think that we could still perhaps do a tolerably serviceable Ruby 
class browser and help system in Ruby/Tk that could serve as an adjunct or 
"virtual extension" to gvim and xemacs, with some suitable additional 
macros.)

Conrad Schneiker
(This note is unofficial and subject to improvement without notice.)