Newsgroups: comp.lang.c++,comp.lang.smalltalk,comp.object
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!uunet!world!edwards
From: edwards@world.std.com (Jonathan Edwards)
Subject: Re: c++, smalltalk, and real-time
Message-ID: <D75oFM.MoB@world.std.com>
Organization: IntraNet, Inc.
References: <3mjk4m$ijf@news.Belgium.EU.net> <patrick_d_logan.2.00163EC9@ccm.jf.intel.com> <D74s1F.CL4@world.std.com> <ethridge-1604951748280001@dallas-1-6.i-link.net>
Date: Mon, 17 Apr 1995 01:35:46 GMT
Lines: 23
Xref: glinda.oz.cs.cmu.edu comp.lang.c++:123456 comp.lang.smalltalk:22821 comp.object:29662

In article <ethridge-1604951748280001@dallas-1-6.i-link.net>,
Allen B. Ethridge <ethridge@i-link.net> wrote:
>Since Smalltalk systems come with a "virtual" operating system this makes it
>difficult for Smalltalk systems to make use of the multithreading and
>multitasking capabilities of the "native" OS.  C++, being just a language and
>not a system, provides no mutlithreading or multiprocessing capabilities.
>This gives C++ an advantage over Smalltalk?

But C++ does not hinder support for multithreading and multiprocessing,
and all the major commercial implementations are in fact doing so if they
haven't already. Microsoft, Borland, Sun and IBM already support 
multithreading in C++. MS even has builtin support for thread-local storage.

So yes, it is an advantage.

Of course it is far outweighed in my opinion by the massive disadvantages of
C++ for most of my needs. So I am using Smalltalk where I can, but I will have
to code the kernel of my system in C/C++ in order to do multithreading. 
-- 
Jonathan Edwards				edwards@intranet.com
IntraNet, Inc					617-527-7020
One Gateway Center				FAX: 617-527-6779
Newton, MA 02158
