From owner-freebsd-hackers Sun Apr 2 12:07:25 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA08442 for hackers-outgoing; Sun, 2 Apr 1995 12:07:25 -0700 Received: from snoopy.mv.com (snoopy.mv.com [199.125.64.182]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id MAA08434 for ; Sun, 2 Apr 1995 12:07:22 -0700 Received: (from pw@localhost) by snoopy.mv.com (8.6.9/8.6.6) id NAA04498; Sun, 2 Apr 1995 13:38:04 -0400 Date: Sun, 2 Apr 1995 13:38:04 -0400 From: "Paul F. Werkowski" Message-Id: <199504021738.NAA04498@snoopy.mv.com> To: hackers@FreeBSD.org In-reply-to: <199504021500.KAA03948@bonkers.taronga.com> (message from Peter da Silva on Sun, 2 Apr 1995 10:00:18 -0500 (CDT)) Subject: Re: CLISP clarification, Was: New Snapshot...Good and Bad.... Sender: hackers-owner@FreeBSD.org Precedence: bulk >>>>> "Peter" == Peter da Silva writes: >> Hmm, that beats GCL which starts life at 2.4 MB. Seems like it >> used to be smaller than that. Anyhow bear in mind that Lisp >> grows like a wart once it has to actually do anything. Peter> That's not inherent in Lisp. I was doing useful stuff in Peter> Lisp 1.5 on a PDP-11. Probably not recently, I'll bet :) I perhaps should have said that the Lisp process grows quickly as code and data are added. Peter> I can't imagine what GCL could possibly have in it to Peter> require that sort of resource utilization. I've seen some Peter> of the syntactic sugar people have added to lisp, but Peter> surely all that stuff can be considered optional? Actually GCL (formerly KCL/AKCL) is a pretty sparse implementation of CLtL1 with its kernel and many runtime functions in C. A large part of the code is the compiler which does the lisp to C conversion (with optimizations). Common Lisp the Language 2nd Edition (CLtL2) documents the additions (not really options) that are included in the ANSI Common Lisp spec - or at least what the ANSI committe was thinking in 1989/90. The first big chunk is CLOS, the object system, which has features C++ can only dream about and is one of the main reasons I prefer to work in Lisp. I can pretty quickly prototype a class structure in CLOS and test out concepts before (yuk) grinding the result into C++. comp.lang.lisp frequently erupts into discussions of the relative merits of C vs Lisp and there are folks who would like to see CL broken into a core + optional subsets just to get the memory use down - but then it wouldn't be CL. It's probably all moot anyhow as former Lisp'ers seem to be looking at Dylan for future work. But still, a language that is now around 40 years old and still represents state-of-the-art may just be around a lot longer. But this all digresses from BSD kernel hacking. Waiting for cheap gigabit RAMs and tera-ops CPUs.. Paul