colonel@sunybcs.UUCP (Col. G. L. Sicherman) (11/05/85)
["You saved my life, Captain Buffalo! Have a CIGAR!"] > With a few notable > exceptions (such as hash-coding), it seems that almost all the "good > stuff" in computing evolved in theoretically-oriented research environments > (often academic), i.e. out of the efforts of people involved in doing > computer science. > Ada! (and what ever happened to Jovial by the way?). > and intellectual insights of DoD's great achievments in this area: Cobol > exactly they failed, Colonel, [END OF ARTICLE] Looks like this was written on a sinking ship.... Well, yes, Cobol! A perfect illustration. When a new idea works well in practice, like WHILE-loops, theory pronounces it a good idea, and explains why. If theory cannot explain why, as with Cobol, it is reduced to sneering at it (I name no names (but his initials are E.D.!)). And if a theoretical idea works badly in practice, no explanation is wanted; nobody wants to hear about it any more. I'm not talking now about the theoretical research that compares disk-scheduling algorithms or ease of using text editors; this is obviously valuable. It's the theory of programming I mean. And just because an idea comes out of a "theoretically-oriented research environment" (university) doesn't make it a product of theory. The people who do computer science are practitioners--they have to be. A researcher who strives to be purely theoretical is often just an obstacle to progress. At the least, he's trying to maintain the old distinction between theory and practice in a field where they're inseparable. -- Col. G. L. Sicherman UU: ...{rocksvax|decvax}!sunybcs!colonel CS: colonel@buffalo-cs BI: csdsicher@sunyabva
tedrick@ernie.BERKELEY.EDU (Tom Tedrick) (11/06/85)
******************************************************************** Warning: Flame ahead (still time to hit 'n' ...) >[ ... criticism of theory is the issue ... ] >It's the theory of programming I mean. And >just because an idea comes out of a "theoretically-oriented research >environment" (university) doesn't make it a product of theory. The >people who do computer science are practitioners--they have to be. >A researcher who strives to be purely theoretical is often just an >obstacle to progress. At the least, he's trying to maintain the old >distinction between theory and practice in a field where they're >inseparable. Ah ... Now I am beginning to understand. Although the theory of algorithms, theory of computation, etc. seems very beautiful to me, I have to agree that the "theory of how people should program" is a field infested with intellectual lightweights who have generated lots of nonsense to burden the rest of us with, so as to justify their existence. I think programming is mainly a matter of practice and common sense with a reasonable dose of intelligence thrown in, and theory has contributed very little of practical value to the problem of how to teach programming. I guess eliminating "go to" was a success though. The worst thing is when students with little experience have been indoctrinated into some absurd style of programming, then treat the paradigm they are familar with as some kind of holy revelation, requiring them to exterminate anyone who believes in a different style. I have had first hand experience of persecution at the hands of some of these small minded idiots who were working as lab assistants for me. I am trying to develop a "theory of theories" (purely for my own enlightenment) which explains why these "theoretical parasites" exist. I think it has to do with funding being available to subsidize research in the field. End Flame. ***************************************************************** -Tom tedrick@ucbernie.arpa