Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!pt.cs.cmu.edu!sei!firth
From: firth@sei.cmu.edu (Robert Firth)
Newsgroups: comp.arch
Subject: Re: Compiler Costs
Message-ID: <7861@fy.sei.cmu.edu>
Date: 13 Jul 90 16:41:15 GMT
References: <1797@apctrc.UUCP> <373@e2big.mko.dec.com> <2319@l.cc.purdue.edu>
Reply-To: firth@sei.cmu.edu (Robert Firth)
Organization: Software Engineering Institute, Pittsburgh, PA
Lines: 23

In article <2319@l.cc.purdue.edu> cik@l.cc.purdue.edu (Herman Rubin) writes:

>To produce good code, it is necessary to
>understand the cost considerations, and I find it difficult to believe that
>this can be done at the limited level of HLLs.

What cost considerations, pray?  The costs that usually matter are
the costs of designing, producing, testing, debugging and
maintaining the code, since these things are very intensive
of human labour.  Most people who build software are very
sensitive to these costs, which is the reason then use high
level languages.

A few people, however, are completely insensitive to most of the
costs that really matter, and focus on the cost of executing the
code on a machine.  These people code in assembler, and they can
be an expensive luxury as their tight, efficient programs
repeatedly fall behind schedule, fail on simple test cases, and
eat up unbelievable amounts of maintenance time.

I find it very sad that so many of these discussions take the
standpoint that it is the purpose of the human to serve the
machine.