Subject: rmas and code ...
From: "Leonard H. Tower Jr." <tower@ai.mit.edu>
Date: Sun, 14 Dec 1997 05:55:16 -0500 (EST)

   Date: Fri, 12 Dec 1997 16:40:20 -0800
   From: Joseph Arceneaux <jla@arceneaux.com>

   A related issue here is the GUI.  rms has been consistently against
   devoting any effort to GUIs, and yet such effort would have advanced
   the cause of free software a long way.

rms is blind to graphics and their good use.  It's a sad consequent of
his ability to process text at a rate that would blind most of us.
He's not opposed to others doing it.  Part of the problem MAY BE that
                                                          ^^^^^^
he's realized at some level that he doesn't know how to make the right
decisions here, so he leaves others to deal with it - this is mostly
conjecture on my part mixed with having worked very closely with him
for a dozen years - Richard doesn't make much of his inner self
available to others.

IMHO, I agree that it has hurt the free software cause.

   E.g., I sent rms changes to gud.el which, amoung other things, replace
   the stupid ascii arrow pointer with a highlighted line which is much
   easier to see.  This is only about 10-20 lines of code, but he never
   installed it.

Bad example of the point you are trying to support.

Sigh.  "replace" fails.  Not all terminals do highlighting, so the
code has to provide the ASCII arrow pointer for those that do not,
while using your highlighted line for those that do.  If your code did
this, cleanly, in a way that would work in all cases, he would have
installed it.

This is a classic example of people seeing a behaviour, and
misinterpreting why it happened.

rms had made the point repeatedly, that GNU Emacs will continue to
support terminals as well as GUIs.  I can understand why he could get
tired of providing it several times a week...

best -len