From: Jeff Dalton
Subject: Re: C vs LisP yet again (long but thoughtful)
Date:
Message-ID: <CwJL6v.4sE@cogsci.ed.ac.uk>
In article <······················@groucho.cse.psu.edu> ········@groucho.cse.psu.edu (Scott Schwartz) writes:
>·····@rheged.dircon.co.uk (Simon Brooke) writes:
> This is because I value my time: programmers are expensive, silicon
> is cheap. Or, to put it another way, people are more important than
> things.
>
>That sounds suspiciously like you are saying, "I don't care how slow and
>bloated this application is, it was easy to write." A better philosophy
>is, "other people are important and numerous, so code as efficiently as
>possible in order that they can make the most effective use of their
>things."
Why is your extreme better than the other? Surely a balance is needed.
>Your analysis of C and Unix misses this important factor. In the Unix
>community, Perl, and interpreted language, is now very commonly used for
>systems programming. It's expressive (your point) and efficient (my
>point), and so people often use it instead of C, even though it means
>loading a large (360K) runtime system.
Perl is not "as efficiently as possible".
360K is not very large, especially compared to typical Common Lisps.
-- jeff
····@aiai.ed.ac.uk (Jeff Dalton) writes:
Surely a balance is needed.
Agreed.
Perl is not "as efficiently as possible".
Very often it is. And Perl users are sensible; they look for the
balance you mention above, and discover something that meets their
needs. The point is that perl is a counterexample to the assertion that
C users are wedded to the idea of arduous low level bit twiddling.
360K is not very large, especially compared to typical Common Lisps.
Yup.