Newsgroups: rec.arts.int-fiction
Path: nntp.gmd.de!news.ruhr-uni-bochum.de!news-koe1.dfn.de!main.Germany.EU.net!Germany.EU.net!EU.net!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!cam-news-hub1.bbnplanet.com!uunet!in2.uu.net!192.35.48.11!hearst.acc.Virginia.EDU!murdoch!not-for-mail
From: nr@mamba.cs.Virginia.EDU (Norman Ramsey)
Subject: Re: Questionable Languages & Compilers
X-Nntp-Posting-Host: mamba-fo.cs.virginia.edu
Message-ID: <59uhn6$c0@mamba.cs.Virginia.EDU>
Sender: usenet@murdoch.acc.Virginia.EDU
Organization: University of Virginia
References: <32b72f97@beachyhd.demon.co.uk> <1996Dec19.165314.307@wcc.govt.nz> <59cimp$2hf@life.ai.mit.edu>
Date: Thu, 26 Dec 1996 18:55:34 GMT
Lines: 37

In article <59cimp$2hf@life.ai.mit.edu>, David Baggett <dmb@ai.mit.edu> wrote:
>[it should be Ok to criticize tools as well as games]

Agreed.  Let's have at it :-)

>I've spent a great deal of time reading about [programming-language
>design and implementation] over the past few
>years, and it's quite clear to me that there are serious problems with
>*all* the existing IF development systems, and that none of them deserves
>anything like a "professional-quality compiler" label.
>
>One very basic requirement for a "professional quality compiler" is that it
>perform nontrivial optimizations like strength reduction; do any of these
>compilers do that?

I think you're assuming that the goals for an IF language or compiler
should be the same as the goals for a compiler for a general-purpose
programming language, e.g., the generated code should run fast.
What evidence have you got in favor of this position?  

My pet peeve at the moment is that existing tools seem to offer little
in the way of static type checking.  A language like Inform leaves a
lot of checking for run time.  For example, according to the language,
every property is meaningful on every object.  This rule offers
tremendous flexibility, but it also leads to heavy overloading of
properties like `general'.  On the other hand, the obvious way of using
a fully static type system like that for C++, Java, or Modula-3 would
make Inform programming significantly more awkward than it is today.
Reconciling these competing goals would make a good master's project :-)

It's easy to write long lists of desiderata for these languages and
tools.  What's on other people's lists?

N
-- 
Norman Ramsey
http://www.cs.virginia.edu/~nr
