Newsgroups: comp.lang.dylan
From: cyber_surfer@wildcard.demon.co.uk (Cyber Surfer)
Path: cantaloupe.srv.cs.cmu.edu!nntp.club.cc.cmu.edu!hudson.lm.com!news.pop.psu.edu!news.cac.psu.edu!howland.reston.ans.net!pipex!demon!wildcard.demon.co.uk!cyber_surfer
Subject: Re: Will Apple use Dylan for development?
References: <9501041714.AA13315@newton.apple.com>
Organization: The Wildcard Killer Butterfly Breeding Ground
Reply-To: cyber_surfer@wildcard.demon.co.uk
X-Newsreader: Demon Internet Simple News v1.27
Lines: 45
X-Posting-Host: wildcard.demon.co.uk
Date: Sat, 7 Jan 1995 14:14:06 +0000
Message-ID: <789488046snz@wildcard.demon.co.uk>
Sender: usenet@demon.co.uk

In article <9501041714.AA13315@newton.apple.com>
           KenD@apple.com "Ken Dickey" writes:

> doing so gives a strategic advantage.  I.e. if someone can deliver an
> application in 1/2 to 1/3 of the time it takes to deliver the same app in,
> e.g., C++ and can evolve features faster as well, then why should that
> someone want to tell the C++ based developers about this?  On the other
> hand, if that someone does disclose that a dynamic language was used, the
> competition can start beating the drum about what a bad thing GC is and all
> that misinformation biz.  It pays in this case to gain market share by
> keeping ones mouth shut.

Exactly. I'm writing a Lisp to C compiler right now, and so far,
the code it generates looks to me as good as I could write by hand.
Of course, I write C code in "cliches", so this is easy. Also, I
like using lists a lot!

Naturally, if this makes me more productive, while losing little
performance, then I can just keep quiet about the language used.
I don't yet know what effect the GC will have on the performance,
but using the compiler could certainly make the I/O a lot faster,
by the use for directly translating the "Lisp" I/O function calls
into calls to the standard C I/O functions. The app I'm testing
this on will write a lot of literal strings to an output file
(this is a real app, by the way, not a contrived benchmark), so
I expect it do as well on I/O as a handcoded C version.

> Perhaps you will be different.

Ask me again in a year, when I have a handful of utilities and
apps written with my compiler, and a lot of experience with them
on real data. Even if the performance isn't as great as I could,
in theory, get from handcoded C versions of the same apps, I don't
always have the time to work that way. This is why my answer to
any question that starts, "Why do you use...", will be, "coz I can".
If I couldn't do it, then I wouldn't be able to do it. There are
a lot of tools I _don't_ write in C, but if I _do_ write them in
Lisp, it's coz I can.

I'm hoping that Dylan will be another language I can use to write
my tools with. ;-)
-- 
CommUnity: http://www.demon.co.uk/community/index.html
CommUnity: ftp://ftp.demon.co.uk/pub/archives/community
Me: http://cyber.sfgate.com/examiner/people/surfer.html
