Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!usc!apple!shebs
From: shebs@Apple.COM (Stan Shebs)
Newsgroups: comp.sys.mac.programmer
Subject: Re: Don't use MPW GCC if you respect RMS's wishes
Message-ID: <11493@goofy.Apple.COM>
Date: 11 Dec 90 19:39:46 GMT
References: <333@unf7.UUCP>
Organization: Apple Computer Inc., Cupertino, CA
Lines: 67

Just a couple reactions:

In article <333@unf7.UUCP> shite@unf7.UUCP (Stephen Hite) writes:
>
>   Yes, I  ftp'ed MPW GCC from apple.com (funny, I checked today and it's
>gone)

A screwup on my part - the files have been restored.

>[...] I've decided to shelve MPW GCC out of respect for RMS's wishes. [...]
>
>   I decided not to support RMS by getting rid of my SE/30, though. [...]

So basically you're supporting RMS as long as it's not too inconvenient.
There are people who really need MPW GCC to do their work, for whom there's
more at stake than convenience.

>   I can understand a programmer porting GCC to the Mac with the good 
>intention of wanting to create a free standalone development tool 
>(i.e. free headers, libs, etc).   However, it makes no sense for Apple 
>developers to do it and include a "readme" that discourages Apple developers 
>to use it on their own projects.

Oops, that file wasn't supposed to be distributed.  It shouldn't be too
surprising tho - RMS is enough of a loose cannon and the law is vague enough
that I would be most uncomfortable betting the whole company on his goodwill.

Compilers are used for a lot of things besides compiling product code.  I
did MPW GCC as part of our ongoing research; no other compiler that I know
of can touch GCC for quality, portability, and retargeting support.

>They (Apple) include another "readme" 
>that freely admits that GCC is a "better" C compiler than  MPW C.  About 
>the only thing MPW C "won" was the compilation speed test.

You're misinterpreting the intent of the statistics.  I was supplying some
information that potential users could study to decide whether it was worth
their while to try MPW GCC.  In fact, for the kind of work I do, compilation
speed is more important than code quality, so MPW C or THINK C is better for
me.  (Actually, I really prefer to hack Lisp!)

>  The Apple developers involved with the GCC port have proven that they
>are great systems programmers / integrators.  Unfortunately, they have
>also demonstrated that they do not realize the implications (internal
>and external) of what they do.

*I'm* the "Apple developers" you're referring to and I can assure you that
I've spent a great deal of time considering all the implications - I've been
hacking on GCC for a year and a half now, and have used GNU software for
many years.

Personally, I'm sympathetic to the ideals of the GNU project and have produced
much "free" software (to my wife's horror, BTW).  On the other hand, I don't
think RMS is Gandhi, and I don't think that the boycott was a particularly
clever idea (Apple's support isn't important to GNU, and GNU's support isn't
important to Apple, so leverage is minimal).  In fact, demonstrating the
value of free software is much more effective in the giant anarchy that is
Apple Computer.  Compare strategies from the executives' point of view - "We're
supposed to drop a lawsuit that's protecting our company because a bunch of
competitor-supported Unixoids aren't buying Macs?" vs "We're more profitable
and doing better research because we're using and distributing free software?"

'Nuff said.  You're free to make your own decision.

						Stan Shebs
						Apple ATG System Software
						shebs@apple.com