Path: utzoo!attcan!uunet!husc6!bloom-beacon!mit-eddie!uw-beaver!fluke!mce
From: mce@tc.fluke.COM (Brian McElhinney)
Newsgroups: comp.sys.mac
Subject: Re: 2 megs and LSC kiss of death
Message-ID: <4139@fluke.COM>
Date: 17 Jun 88 16:57:58 GMT
References: <16044@uunet.UU.NET>
Sender: news@tc.fluke.COM
Organization: John Fluke Mfg. Co., Inc., Everett, WA
Lines: 26

In article <16044@uunet.UU.NET> mo@uunet.UU.NET (Mike O'Dell) writes:
> THINK's PASCAL system has a source-code debugger in 1 megabyte.
> How come their C system can't manage that? 

Wirth designed Pascal to teach students about compilers, and thus designed it
to make it easy to write the compiler.  That's why you have bogus requirements
about no semi-colons before ENDs and the like.  I have heard that C compilers
(just the compiler) are roughly twice the complexity Pascal compilers, and
that C++ compilers are another two times more complex.  Anyone at THINK care
to comment?

> I predict that the 2 meg requirement will significantly impact
> THINK's market share, particularly if BORLAND introduces their
> source-code debugger and manages to keep it within 1 meg.

I have never seen Turbo C, but I suspect it does not have the speed nor user
interface of LSC.  You win some, you lose some.  I doubt that LSC is in
trouble because RAM happens to be expensive at the moment.

I predict that market shares wills stay about the same until virtual memory
becomes standard for developers (two years?).  At that time MPW will start to
pull ahead.


Brian McElhinney
mce@tc.fluke.com