Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 Fluke 1/4/84; site fluke.UUCP Path: utzoo!linus!decvax!tektronix!uw-beaver!microsoft!fluke!kurt From: kurt@fluke.UUCP (Kurt Guntheroth) Newsgroups: net.micro Subject: Re: "Smalltalk Coming to Micros!" Message-ID: <1163@vax2.fluke.UUCP> Date: Tue, 24-Jul-84 11:58:26 EDT Article-I.D.: vax2.1163 Posted: Tue Jul 24 11:58:26 1984 Date-Received: Sat, 28-Jul-84 06:47:20 EDT References: <191@sri-arpa.UUCP> Organization: John Fluke Mfg. Co., Everett, WA Lines: 23 The thing about the smalltalk image from xerox is that it is not a very imaginative implementation. It is kind of like Wirth's P4 Pascal implementation; simpleminded, but it works and is well tested. It turns out that by tuning the implementation you can get substantial speed improvement out of smalltalk. See the last POPL procedings for 2 articles on doing just that. As has been pointed out before, what kills smalltalk on really small machines is that the whole smalltalk system strains the very limits of a machine with a 64K segmented architecture. Xerox now seems to consider such a machine too small to run smalltalk, although early implementations ran on such a machine. The smalltalk image is also quite large and access to it in a reasonable time requires a hard disk. These factors tell you that while you may soon see smalltalk on a 'personal computer', it is not going to be your garden variety home system. -- Kurt Guntheroth John Fluke Mfg. Co., Inc. {uw-beaver,decvax!microsof,ucbvax!lbl-csam,allegra,ssc-vax}!fluke!kurt