Newsgroups: rec.arts.int-fiction
Path: nntp.gmd.de!newsserver.jvnc.net!raffles.technet.sg!pith.uoregon.edu!cliffs.rs.itd.umich.edu!howland.erols.net!news-peer.gsl.net!news.gsl.net!ix.netcom.com!netcom.com!erkyrath
From: erkyrath@netcom.com (Andrew Plotkin)
Subject: Re: Questionable Languages & Compilers
Message-ID: <erkyrathE2svyK.Hur@netcom.com>
Organization: NETCOM On-line Communication Services (408 261-4700 guest)
X-Newsreader: TIN [version 1.2 PL1]
References: <32b72f97@beachyhd.demon.co.uk> <1996Dec19.165314.307@wcc.govt.nz> <59cimp$2hf@life.ai.mit.edu> <erkyrathE2q30q.48I@netcom.com> <59fmnj$krf@life.ai.mit.edu>
Date: Sun, 22 Dec 1996 05:44:44 GMT
Lines: 66
Sender: erkyrath@netcom3.netcom.com

David Baggett (dmb@lf.ai.mit.edu) wrote:
> In article <erkyrathE2q30q.48I@netcom.com>,
> Andrew Plotkin <erkyrath@netcom.com> wrote:

> >This becomes a sore point, because there *is* a difference between Graham 
> >and his spare time (on one finger), and MetroWerks (on the rest of the 
> >platoon's appendages).

> Is it also a sore point that we're writing IF games in our spare time, so
> that when someone compares our works to "real" literature, it's not fair?
> I say that's garbage.

Writing *is* different. Everyone who writes does it in their spare time 
(except a small minority, just about all of whom have been doing it for a 
while) and there are no programming textbooks to study. No backlog of 
technical info which applies to the field. There's just books (of which I 
read as many as anyone, give or take) and stuff like writers' workshops 
and composition classes, which are known to be optional.

Writing is also a singleton sport, which was my point about Metrowerks. For 
most kinds of programs, you *can* do twice as much by hiring twice as 
many programmers. (Up to obvious limits.) My Matrowerks C compiler has 
buttloads of nice features -- source-level debugger, syntax-based 
coloring in the editor, pop-up lists of functions, lots of libraries, 
kept up to date with Mac system software, etc, etc. One person couldn't 
do that, not at that rate. Kernighan and Ritchie didn't do it.

Whereas a book is written by one person staring at enough blank paper for 
enough time.

> And if it's a sore point, why claims like this:

>     "But Inform, though it is certainly an amateur piece of work,
>     is a quite high-performance compiler by comparison with many,
>     perhaps even most, professional works."

Well, I didn't say that. Now that you've brought up all these issues like 
optimization, I'm wavering on it (mostly because I don't feel like 
re-reading the Inform tech manual chapters on optimization, so I am, to 
coin a phrase, ill-informed.)

> I didn't say that lack of optimization was a major problem.  I said it was
> an indication of a non-"professional" compiler.

Well, but I agree with you on that. :-) I'm saying I don't expect as much 
from Graham as I do from my commercial compiler/IDE.

>  If you want an example of
> a major problem, how about this: why are we writing text-based interactive
> fiction in variants of a low-level systems programming language?  This is
> *not* a systems programming task; in fact it's at nearly the exact opposite
> side of the spectrum.  (This doesn't just apply to Inform, of course.)

Oop, different thread. The choice of design language features will be 
dealt with next post. I *don't* think it's fair to criticize Graham (or 
Mike Roberts) for their choice of tasks to tackle. Especially since, at 
least for *me*, their tackles have been very successful, and enabled me 
to do exactly the tasks I want to do. I'm a classic (ok, really ANSI) C 
recidivist.

--Z

-- 

"And Aholibamah bare Jeush, and Jaalam, and Korah: these were the
borogoves..."
