Aucbvax.6462 fa.works utcsrgv!utzoo!decvax!ucbvax!works Fri Mar 12 11:45:34 1982 Unix, IBM, humankind, Smalltalk, animation, parallelism >From RYLAND@Sri-Kl Thu Mar 11 21:16:30 1982 Please, let's not crank up the old Unix religious debate on WorkS. See back issues of nearly every other digest on the net for similar debates. I think it's clear that the critical resource in today's computing world is manpower, not machinery. The result? Systems which encourage and support software movement among a variety of hosts are bound to be popular; Unix and CP/m fill that bill reasonably well (no comparison betwixt the two intended). Since this list is supposed to be a collective soul-search for the ultimate workstation (whatever that is), what might conceivably serve as a basis for the next wave, after Unix? I propose Smalltalk (ST80 for short). ST80 sits just about on the edge of what today's hardware (read: 68K equivalents and follow-ons) can support "reasonably". I've seen ST80 in use, and know that I'd much prefer that world to anything else I've seen (including Lisp machines, which have perhaps the most complete, if also the most arcanely complex programming environment today). I think the XEROX Systems Concepts Group (or whatever they call themselves now) also knows it (why else would they have devoted so many years to ST?) It's a shame that they're having such a hard time releasing ST to the world, but I hear they're on the verge of such a release after a long battle to get licensing going. (I hate to speculate like this, but I can't say all that I hear.) ST80 has exactly the properties proposed recently: an extensible system with an "animated" feel. It doesn't use parallelism in any large way, but I'm convinced that we don't understand how to use massive parallelism in any but the crudest sense (e.g., weather prediction). My ideal workstation, perhaps buildable by 2000, would place me in a virtual world (3d and all) where the hallmark was malleability of the objects and actors in the universe. This is an old idea, but makes complete sense to me. And no one is doing anything seriously along these lines. Boils down to the same old problem: no one knows how to use massive amounts of computing power. The old paradigms just don't make sense anymore if you have billions of hardware actors (in the Hewitt sense) following quadrillions of scripts. ------- ----------------------------------------------------------------- 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.