Path: utzoo!mnetor!uunet!husc6!cmcl2!nrl-cmf!ames!pioneer!eugene From: eugene@pioneer.arpa (Eugene N. Miya) Newsgroups: comp.arch Subject: Re: Universal OS & universal language Message-ID: <8331@ames.arpa> Date: 5 May 88 16:50:01 GMT References: <1596@pt.cs.cmu.edu> <1575@hubcap.UUCP> Sender: usenet@ames.arpa Reply-To: eugene@pioneer.UUCP (Eugene N. Miya) Followup-To: comp.os.research,comp.lang.misc Distribution: na Organization: NASA Ames Research Center, Moffett Field, Calif. Lines: 68 I think we are beginning to stray from architectures, I will follow this up to os.research and lang.misc In article <1575@hubcap.UUCP> fpst@hubcap.UUCP (Steve Stevenson) writes: > . . . summary: languages are personal >Compilers developed to meet the needs of early programmers >with different needs. The issues in the '60s was fast compilation and >optimal runtimes. I'll add: to minimize storge for expensive memories >It is time to reevaluate that objective function. I personally think >it stinks. This is my comment as it pertains to architecture: we must not forget that we built and use computers to accomplish specific tasks: first to compute ballistic trajectories, then simulate thermonuclear processes, model the weather, fly spacecraft, and so on [fun now for instance]. I can't easily justify the first two personally, but I have worked on weather, image processing, and space problems. It's kind of a silly joke to hear about the 24 hour weather forecast which takes 27 hours to run. But weather is a more serious matter outside California. ;-) A different physicist I am working with (in his 60s, in fact he can remember the comment about the most machine ever needed [live]) made a really neat comment when the people at Purdue mapped a cold virus. He said, "That's neat, that's what computers are for!" And he is kind of right. I wonder if Turing and von Neumann would turn in their graves if they saw video games and word processors as the chief use of computing (Don't post, it's retorical, think about it for yourself)? This contrasts with the islands he vaporized doing experimental physics ;-). My point is we have to keep some balance and perspective why we have made computers. I believe Cray (the person does this). Now if you architects can only get some quiet time to build better machines. Enough rambling (sorry). >The question - now that some folks want to have the >fate of the world in the hands of programmers - is getting things >to work right every time. This is why Ada(tm) was developed, see the Rationale document. (not an endorsement of this language). >As to the universal OS - it is a "language" too, at least in the formal sense. Re: the topic of problem specific language is always bandied about: i.e. editing as a language, programming languages, graphics as a language, etc. I was given a reminder in a different discussion. Check out: %A Ivan E. Sutherland %T Computer Graphics - Ten Unsolved Problems %J Datamation %V ?? %N ?? %D May 1966 %P 22-27 %X (Surprise!) When Datamation was a respectable technical publication. A very good article. See problem 3 on "coupling." This latter (reminder) given to me by Frank Crow (PARC). Neat old classic paper (youngins' like me) need reminders like this. Another gross generalization from --eugene miya, NASA Ames Research Center, eugene@ames-aurora.ARPA resident cynic soon to be aurora.arc.nasa.gov at the Rock of Ages Home for Retired Hackers: "Mailers?! HA!", "If my mail does not reach you, please accept my apology." {uunet,hplabs,hao,ihnp4,decwrl,allegra,tektronix}!ames!aurora!eugene "Send mail, avoid follow-ups. If enough, I'll summarize."