Xref: utzoo comp.misc:11481 comp.lang.misc:6634
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sdd.hp.com!spool.mu.edu!news.cs.indiana.edu!news.nd.edu!mentor.cc.purdue.edu!pop.stat.purdue.edu!hrubin
From: hrubin@pop.stat.purdue.edu (Herman Rubin)
Newsgroups: comp.misc,comp.lang.misc
Subject: Re: standard extensions
Message-ID: <6100@mentor.cc.purdue.edu>
Date: 17 Feb 91 18:48:43 GMT
References: <14814@lanl.gov> <1991Feb15.192653.9846@rice.edu> <27BDC456.20D6@tct.uucp>
Sender: news@mentor.cc.purdue.edu
Followup-To: comp.misc
Lines: 43

In article <27BDC456.20D6@tct.uucp>, chip@tct.uucp (Chip Salzenberg) writes:
> [ Followups to comp.misc ]
> 
> According to hrubin@pop.stat.purdue.edu (Herman Rubin):
> >Montgomery and Silverman are number theorists, I am a professor
> >of Statistics and Mathematics.  We are expected to do other
> >things.
> 
> I'm a programmer, but that doesn't mean I've got all the time I want
> to do projects for J. Random Person.  I'm expected to do other things,
> too.
> 
> >I believe that such a macro translator could be produced ...
> 
> A spec, Herman.  Surely you can squeeze a few hours our of your busy
> schedule to produce a spec.  Or is your desired language a mystery
> even to you?

I am not saying that every programmer should drop what he is doing to
produce the appropriate tools.  But the problems faced are problems which
can be handled, and even anticipated, by including flexible tools.  They
are not going to be even treated by moving in the direction of more and
more rigid structures, hardware, software, and OS.

In the next few weeks I will produce something which I believe will 
indicate what should be done.  It will definitely not be in the direction
which programmers and language desingners are moving now; the great 
introduction of variables by Diophantus made it possible to communicate
algebraically without confusion.  Unfortunately, we seem to refuse to 
consider this simple idea, and instead concentrate on manipulations.

The same is true in the current approach to computing.  Instead of setting
up tools so that the person expressing ideas in a notational language can
have this automatically, as far as is possible, translated into what the
sub-imbeciles known as computers can understand, we state that this processor
can only use these idea.  How efficiently a computer can handle things depends
both on the hardware and software, and I suggest that the first step should be
to make a preliminary translation of what the user wants to write, in the user's
notation, into "computerese."  THEN worry about how to do things efficiently.
--
Herman Rubin, Dept. of Statistics, Purdue Univ., West Lafayette IN47907-1399
Phone: (317)494-6054
hrubin@l.cc.purdue.edu (Internet, bitnet)   {purdue,pur-ee}!l.cc!hrubin(UUCP)