Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!pasteur!ames!ucsd!sdcsvax!trantor.harris-atd.com!melmac!chuck
From: chuck@melmac.harris-atd.com (Chuck Musciano)
Newsgroups: comp.arch
Subject: Re: Complex Instructions
Message-ID: <2037@trantor.harris-atd.com>
Date: 9 May 89 17:50:55 GMT
References: <57252@yale-celray.yale.UUCP> <4101@tolerant.UUCP> <134@dg.dg.com> <504@daitc.daitc.mil> <1277@l.cc.purdue.edu> <2249@pembina.UUCP> <2024@trantor.harris-atd.com> <2223@wpi.wpi.edu>
Sender: news@trantor.harris-atd.com
Reply-To: chuck@trantor.harris-atd.com (Chuck Musciano)
Organization: Advanced Technology Dept., Harris Corp., Melbourne, Fl.
Lines: 68

In article <2223@wpi.wpi.edu> lfoard@wpi.wpi.edu (Lawrence C Foard) writes:
>Come on who defines qualified professional??? Of course it will depend on how
>many expensive slips of paper you have (degrees). History is full of great
>idea's that where at first rejected by the "qualified professionals". If the
>car maker's only explaination is "it isn't done that way" there is no reason
>why you should except that for an answer. There may be a hundred reasons why
>SL might be a bad idea but you didn't name any you just stated that you where
>a qualified professional and that pions arn't allowed to question you! This
>isn't the dark ages where the pope was infallible, people no matter how well
>trained or professional can still make mistakes and many times it takes an
>outsider to see them. If you had brought the idea of Risc upto the "qualified
>professionals" in the VAX development team, I wonder what the answer would have
>been?

     "Qualified professional" is a term defined by other qualified 
professionals.  If they came from real schools, those "expensive slips of
paper" represent large amounts of knowledge.  When you have the slips of
paper, you are a graduate, nothing more.  After you go into the world and do
something with the knowledge, you begin to attain the rank of professional.

     In my example, the car maker doesn't necessarily say "it isn't done
that way", he will probably say "it can't be done that way".  I don't have
enough knowledge to dispute him.  If he is a responsible professional, he'll
mull over my suggestions, and perhaps modify his thinking.  If I'm an
interested third party, I'll educate myself to better understand the problem.

     I didn't say anything about SL.  I can't, since it doesn't exist.  The
point I was trying to make is that before one begins taking aim at langauge
designers, one should learn a lot about language design.  Without sufficient
background, you simply cannot comment.  I can't count the times people have 
come to me with some new language feature, not realizing the impact on the
rest of the language.  Lots of things that people think are simple represent
massive work, based upon the way a particular compiler has been built.  Other
features are completely incompatible with the language philosophy and 
aesthetics.  Still others are not extensible to include existing language
features.  My job as a professional is to sift the wheat from the chaff,
and do the things that are right.

     Anyone can feel free to question my work.  The weight I give their
questions is commensurate with their professional background.  I try to
give an acceptable answer to anyone who asks me things, couched in a way
that matches their knowledge.  When my Mom asks how computers work, I
don't launch into a discussion of the impact of cache loading on pipeline
stalls.  When a coworker asks why I chose a particular algorithm in some
piece of code, I'll go into great detail, as needed.

     My wife is an industrial engineer.  I don't presume to tell her how to
design automated factories.  She doesn't tell me how to build compilers.
I might offer some input when she is designing assembly method sheets, since
the psychology of instruction writing is somewhat similar to the psychology
of user interface design.  In that realm, we are qualified peers.

     Certainly, professionals make mistakes.  I've made my share.  Almost
all of them have been pointed out by other, equally qualified professionals.
This field is so complex, the issues so detailed, it is difficult for laymen
to understand the subtle shadings particular issues have.

     While the merits of the Vax architecture design have been debated in this
group ad infinitum, I would still hazard the guess that the Vax team would
have correctly rejected RISC had it been presented to them.  They were filling
a particular market niche that RISC, at that time, did not meet.  That isn't
to say that they would not have given it a lot of thought, and perhaps begun
to think of their next machine which might use RISC technology.

Chuck Musciano			ARPA  : chuck@trantor.harris-atd.com
Harris Corporation 		Usenet: ...!uunet!x102a!trantor!chuck
PO Box 37, MS 3A/1912		AT&T  : (407) 727-6131
Melbourne, FL 32902		FAX   : (407) 727-{5118,5227,4004}