Newsgroups: comp.lang.dylan
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornell!travelers.mail.cornell.edu!news.kei.com!news.mathworks.com!newshost.marcam.com!uunet!sytex!smcl
From: smcl@sytex.com (Scott McLoughlin)
Subject: Re: But Can It Do...
Message-ID: <cH2a2c3w165w@sytex.com>
Sender: bbs@sytex.com
Organization: Sytex Access Ltd.
References: <gat-010395095412@milo.jpl.nasa.gov>
Date: Wed, 1 Mar 1995 21:56:47 GMT
Lines: 31

gat@robotics.jpl.nasa.gov (Erann Gat) writes:

> > "The measure of how good a programming language is comes down to how good
> > a 'shoot-em-up' game you can produce with it."
> 
> By that metric, assembler is the last word in programming languages.
> 

Howdy,

There is a good point in the "shoot-em-up game metric". Another
way you might say this is "The measure of how good a programming
language is comes down to how well you can access host system
services".

I've been wondering. Lisp has WRITE-CHAR and WRITE-BYTE defined
as standard functions.  Why not WRITE-PIXEL or COMMIT-TRANSACTION?
I am not saying that Lisp _should_ have "primops" for every 
possible i/o function.  But why consider "stream i/o" somehow
priveleged or fundamental. From a language design point of
view this is purely historical accident and a mistake IMHO.

A dyanamic language definition that relegated typically
essential system services to libraries would soon become
a dynamic language with decent facillities for inline
assembler, a decent FFI, etc.

=============================================
Scott McLoughlin
Conscious Computing
=============================================
