Newsgroups: comp.lang.dylan
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!newsstand.cit.cornell.edu!news.kei.com!news.mathworks.com!gatech!swrinde!howland.reston.ans.net!ix.netcom.com!netcom.com!netcom9.netcom.com!haahr
From: haahr@netcom.com (Paul Haahr)
Subject: Re: implementing Dylan
In-Reply-To: erik@lexmark.com's message of Tue, 30 Jan 1996 19:07:08 GMT
To: erik@lexmark.com (Erik Ackerman)
Message-ID: <HAAHR.96Jan31100932@netcom9.netcom.com>
Sender: haahr@netcom9.netcom.com
Organization: NETCOM On-line services
References: <tigrfhur-2801961507360001@asd05-04.dial.xs4all.nl>
	<erik-2901960739040001@c21macpm8100.pfv.prtdev.lexmark.com>
	<822990198snz@wildcard.demon.co.uk>
	<erik-3001961407080001@c21macpm8100.pfv.prtdev.lexmark.com>
Date: Wed, 31 Jan 1996 10:09:32 GMT
Lines: 12

Erik Ackerman <erik@lexmark.com> wrote:

> Actually, I just got done doing a bit of poking through my Technology
> Release (esp. the DRM) and realized that Dylan could do a heck of a
> job of bootstraping itself given the BNF grammar and LALR(1) grammar
> in the back.  Lex and Yacc just LOVE these.  So use lex and yacc to
> get a base Dylan going on any platform, and then use Dylan itself to
> do a solid implementation.  If no-one beats me to this, I'm gonna give
> it a shot when my BeBox arrives (any time now).  What do you think?

I think implementing Dylan is a lot harder than parsing it, and I think
parsing it is a lot harder than just yacc, given the macro system.
