Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!samsung!uunet!olivea!apple!heksterb From: heksterb@Apple.COM (Ben Hekster) Newsgroups: comp.sys.mac.system Subject: Re: True Multitasking Summary: So what is it, anyway? Message-ID: <48110@apple.Apple.COM> Date: 15 Jan 91 19:14:10 GMT References: <48107@apple.Apple.COM> <1991Jan15.183312.27926@maverick.ksu.ksu.edu> Organization: Apple Computer Inc., Cupertino, CA Lines: 75 In a previous posting, I said: >> Perhaps Mr. Kirksey will be so kind as to refer me to some computer science >> literature where the term `true multitasking' is defined. jxf@altair.cis.ksu.edu (Jerry Frain) responds: > I believe that Mr. Kirksey is referring to what many of us label > "preemptive multiprogramming." Well, if that is what he meant, why didn't he just *say* it? The computer science field is extensive enough that we don't need to invent new terms to describe concepts for which perfectly good ones that have been in use for decades exist already, just to satisfy the egos of certain individuals..! `True' multitasking is a vague term which says nothing, and was obviously coined for its implication that other kinds of multitasking (depending on what you feel `true multitasking' means) must necessarily be `false'. Preemptive/nonpreemtive *says* something. > Yes, yes, preemptive multiprogramming has been around for some 25 years > now, but Apple still hasn't figured out how to do it on the Mac (except > under A/UX -- but AT&T wrote the part of the OS that does the scheduling). Presumably it would not be impossible to implement some sort of preemptive multitasking scheme for the Finder, but frankly I don't see the burning necessity for it. The Macintosh, under MultiFinder, already *has* multitasking. Without necessarily wanting to get into The Great Multitasking Debate again, let me just point out that non-preemptive multitasking also has its benefits--in the case of the Macintosh, the application that the *user* sends events to gets control of the processor. At the moment, I have MPW (compiling), MacBrowse and (obviously) Finder in the background, and I have no wish for them to preempt *me*. Having MPW suspend while I edit this message is fine. I am not debating the point that preemptive multitasking has advantages also--which is the better may be a matter of personal preference. As it is, the Macintosh's form of multitasking does satisfy my requirements, and there I really see little need to resort to multitasking-mudslinging. I also said: (uh oh) >> As it is, the term >> seems only to serve the purpose of consoling insecure Amiga owners. to which Jerry Frain responded: > And give Apple employees a way to divert attention away from the real > question, which is, > > "Why hasn't Apple implemented a preemptive multiprogrammming OS?" That's funny, I always assumed that `true multitasking' was invented to obscure `preemptive'. > For me, the answer to this question is not important. If you have looked > at the Q&A list for System 7.0, the issue of preemptive multiprogramming > was barely addressed. It seems that Apple is now researching ways to > possibly provide this feature. I agree that the question has little importance. Let me point out, though, that I am not a real Apple employee, but a student intern with his own personal opinions. > I would much, much rather see protected for the OS and all of the applications > than preemptive multiprogramming. Again, I agree, although even the benefits of protected-mode execution are also debatable. >> Followup to: comp.sys.amiga.has.true.multitasking.na.naah.na.naah.naaah > A very mature response. Glad you could help. No problem, dude.