Path: utzoo!news-server.csri.toronto.edu!rutgers!cs.utexas.edu!usc!ucla-cs!ucla-seas!turing!plinio From: plinio@turing.seas.ucla.edu (Plinio Barbeito) Newsgroups: comp.sys.atari.st Subject: Re: (none) Was: Multitasking Message-ID: <2183@lee.SEAS.UCLA.EDU> Date: 13 Mar 91 00:13:35 GMT References: <74457@bu.edu.bu.edu> <1010002@hpmwmat.HP.COM> Sender: news@SEAS.UCLA.EDU Organization: SEASnet, University of California, Los Angeles Lines: 82 In article <1010002@hpmwmat.HP.COM> mikep@hpmwtd.HP.COM (Mike Powell) writes: > > > There seems to be at least a couple of definitions that can be applied > to multitasking... Definitions? Don't you mean perceptions? > First, there is the technical implementation of multitasking[...] > Second, there is the USER level definition. This is the most important > of the two definitions for a number of reasons. Users far outnumber > technical 'experts'. If fact, the technical 'experts' are a > subset of this user group. Also, the advantages of a true multitasking > environment are independant of the technical methods used to > implement it. Users are just as interested in a fast, stable system as experts are. The performance of a multitasking system varies very widely with the different kinds of possible implementations. > The user definition of a true multitasking environment would be > that the user is allowed to run multiple tasks of his/her choice > at the same time (provided memory exists to run them). On multitasking systems, you are constantly running software that is not of the user's choice (it may be significant, or it may be negligible, but it is there) -- the O/S. How much time is spent doing this decides just how much of the user's control of his own machine is being usurped. > A good multitasking system shows little or no signs of slowdown > when multiple tasks are run. The degree to which this is true > is one good measure of how well a multitasking system it is. There are ways to make a system more responsive, but this is implementation-related. Sounds as if you are arguing against what you said above, that it didn't matter what were the "technical methods used to implement it". > When xxigx users say they have a true multitasking system, it is ^^^^^(my x's) (net.police hat on) Remember this is an st group you are posting to, not .advocacy or comp.os.misc. In the future, try to help avert flame wars by either using the appropriate groups, or not mentioning the flame war topic specifically. Moreover, your willingness not to mention it will be seen as a measure of your true interest in the original, constructive discussion, and not a selfish interest in trying to prove which computer you perceive as better. (net.police hat off) > certainly true... They can run multiple tasks of their choice > at the same time with little or no signs of slowdown in most > all cases. How the Amiga does this is not really important to them. > But the increase in productivity and enhanced utility of the > system due to the true multitasking environment (user definition) is > quite important to them. I would let each individual user decide for himself whether he likes a particular implementation. In trying to limit our discussion to non-technical issues, it seems as if you would just as soon rather keep people uninformed about their alternatives (if they have any). Far worse would be for the 'experts' to draw their own conclusions and impose them on the users. > So when someone has a USER question about multitasking, be sure > to use the user definition.... anything else is simply a diversion. It's not a diversion, inquiring minds want to know why certain multitasking systems are so slow (for example). I think ST users are interested in knowing the caveats and issues "behind the scenes". They want to know how much of their money and time is being wasted (or saved) by the software they use. > -Mike- plini b -- ----- ---- --- -- ------ ---- --- -- - - - plinio@seas.ucla.edu I don't think therefore I'm not: People don't exist while watching TV