🚀 go-pugleaf

Matthew Dillon (dillon@overload.Berkeley.CA.US) wrote:
: In article <Bz5x32.80A@hpuerca.atl.hp.com> sailer@hpuerca.atl.hp.com (Lee Sailer) writes:
: >jake@skatter.usask.ca wrote:
: >: From article <1fg7kqINNi29@darkstar.UCSC.EDU>, by isbell@cats.ucsc.edu (Art Isbell):
: >: >
: >: >
: >: > What Edit needs is emacs' CM-b and CM-f functionality to conveniently jump to
: >: > the start and end of a bracketed clause so that those starting brackets can be
: >: > easily inserted.
: >: > --
: >: >
: >: > Art Isbell 						     Cubic Solutions
: >:  What Edit needs is to be able to interpret elisp.
: >:
: >: jake
: >:
: >: --
: >
: >What NeXTstep needs is an independent USER AGENT object that sits outside
: >of any and all programs, accepts commands in some suitable language, and
: >executes them.  If the language happened to be elisp, then that set of
: >capabilities could be extracted from emacs.  Emacs could still use it to
: >work exacctly the way it works now.  Edit could use it.  NeXTMail could
: >use it.  All god's chillun could use it.
: >
: >   USER AGENT	USER AGENT  USER AGENT	USER AGENT  USER AGENT	USER AGENT
: 
:     REXX REXX REXX REXX REXX REXX ...
: 
Matt is quite correct that the Amiga implementation of REXX would be
a good model of what I called a USER AGENT.  I must point out that Amiga
REXX can do things that IBM Mainframe REXX cannot dream of.

However, I want NeXTstep to go one step further.  The main problem with
REXX in 1992 is that it offers yet-another-syntax to learn.  Instead
of REXX, I'd prefer an end-user oriented visual programming language
that incorporates Unix shell functionality, REXX functionality, elisp
functionality, and NeXTstep idioms in a general way.

NO, I do not mean a language that allows REXX, elisp, shell syntax, etc.,
just that if it can be done it is clear how to do it in USER AGENT.


-- 
Lee Sailer

					- Let's leave my employer
					  out of this, OK?