Newsgroups: comp.lang.smalltalk
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!yeshua.marcam.com!usc!cs.utexas.edu!utnut!nott!cunews!tina.mrco.carleton.ca!knight
From: knight@mrco.carleton.ca (Alan Knight)
Subject: Re: Self (Was: Re: Smalltalk in CS 1)
Message-ID: <knight.783753083@tina.mrco.carleton.ca>
Sender: news@cunews.carleton.ca (News Administrator)
Reply-To: knight@mrco.carleton.ca (Alan Knight)
Organization: Carleton University
References: <9410211433.AA13245@sci.brooklyn.cuny.edu> <tms-251094103822@stambaugh.tiac.net> <38jk2v$ia6@crchh497.bnr.ca> <tms-261094133117@stambaugh.tiac.net> <mmkCyD4xF.Ktx@netcom.com> <patrick_d_logan.160.000F8CCF@ccm.jf.intel.com> <1994Nov1.191010.26637@xilinx.com>
Date: Wed, 2 Nov 1994 05:11:23 GMT
Lines: 46

In <1994Nov1.191010.26637@xilinx.com> bretts@fred.xilinx.com (Brett Stutz) writes:

>In article <patrick_d_logan.160.000F8CCF@ccm.jf.intel.com>,
>patrick_d_logan@ccm.jf.intel.com (Patrick D. Logan) writes:

>|> Is anyone else who attended OOPSLA '94 as tired as I am of seeing
>|> Self put up 
>|> on an almighty pedestal?

Yes, I'm a little tired of it too. Neat stuff, but overhyped. Perhaps
that's inevitable in this industry.

>I'd also like a relatively unbiased account of why Self is (perceived
>as) faster than
>Smalltalk.

It's perceived as faster because it is very significantly faster on at
least some benchmarks. It is faster because they've put an enormous
amount of effort into making it fast, and it's one of their primary
design goals.

IMHO, Self is not primarily an experiment in language design, but an
experiment in optimizing compilers for dynamic languages. The earlier
papers described it as an optimizing compiler for alangage even harder
to optimize than Smalltalk. It's a fairly straightforward
prototype-based language with some amazing compiler optimizations. On
the other hand, it runs on one platform with what I understand are
pretty large space requirements, which conflicts with PPS's goals of
complete portability and at least semi-reasonable space usage. They're
also still working on the tradeoff betwen optimizing compilation and
development environment responsibeness (there was a paper on that at
OOPSLA too).

I'm sure we will see some of the ideas of Self's compiler feeding back
into dynamic language compilers, but probably not quite as
aggressively, or as quickly as we, the users, would like. (I know I
want something that runs significantly faster than C on low-level
benchmarks, compiles instantaneously, is fully dynamic, and delivers a
full development environment in 640K. I need it by next week, and it
can't have runtime fees :-).

-- 
 Alan Knight                | The Object People
 knight@acm.org             | Smalltalk and OO Training and Consulting
 alan_knight@mindlink.bc.ca | 509-885 Meadowlands Dr.
 +1 613 225 8812            | Ottawa, Canada, K2C 3N2
