Aucbvax.4234 fa.editor-p utzoo!decvax!ucbvax!editor-people Mon Oct 5 15:31:56 1981 forwarded note #3 >From sklower Mon Oct 5 15:31:32 1981 Mail-from: ARPANET site MIT-MULTICS rcvd at 2-Oct-81 0733-PDT Date: 2 October 1981 1012-edt From: Bernard S. Greenberg Subject: Re: emacs and unix To: CSVAX.wilensky at BERKELEY Cc: CSVAX.emacs at BERKELEY, CSVAX.wnj at BERKELEY, arpavax.fabry at BERKELEY, editor-people at SU-SCORE At this stage of retrospect, I think that the general approach I took (in Multics Emacs) of lumping all the functionality of which you speak into one program was technically wrong but morally defensible. It seems pretty clear now that there ought be window managers, video systems, terminal-support paradigms, input-line editors, etc, outside of the editor, and the Editor ought use a maximal subset of them. This is what the Lisp Machine does, and it operates very smoothly. Much of Multics Emacs' vast (too vast?) functionality was due to political and management problems at Honeywell. It was a lot easier to rebuild a better world bottom up inside Emacs than to convince distant, uninterested troglodytic legions that the product should be redone bottom up. It was extremely important to jam people's eyes open to see what kind of things could be done, and Emacs did this well; the price that was paid is Emacs' competition with Multics as a user environment as opposed to being a part of it. Now Multics people are seriously developing generalized video and window support; this could not have happened had there not been Emacs first. Multics Emacs turned into an attempt to build a Multics Lisp Machine. As such, it is grossly successful. As an editor it is grossly successful. As an integrated part of Multics, though, I guess it is a flop; precisely because it calls every conceivable package of functionality in Multics, in order to have the "Emacs Mail System" (from which I am sending this, of course) to compete with the "Multics [extended] Mail System" and so on, it has created divisive and counterproductive duplication of function all over the Multics user ring. It took Multics people many years to see the difference between terminal support, input line editing (!!!!!!!!$%&%#!!!), managed video, windows, and Emacs. That's not the way it ought to be. Hopefully, the advent of couple-of-kilobuck office automation systems with sophisticated screen management paradigms will help people be able to evaluate these different levels of facility and their needs for same in a more precise manner. Of course, as soon as I type ^X^C, I will STILL be back to counting #'s ..... nihil novum sub sole... ----------------------------------------------------------------- gopher://quux.org/ conversion by John Goerzen of http://communication.ucsd.edu/A-News/ This Usenet Oldnews Archive article may be copied and distributed freely, provided: 1. There is no money collected for the text(s) of the articles. 2. The following notice remains appended to each copy: The Usenet Oldnews Archive: Compilation Copyright (C) 1981, 1996 Bruce Jones, Henry Spencer, David Wiseman.