Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!think!mit-eddie!zrm
From: zrm@mit-eddie.MIT.EDU (Zigurd R. Mednieks)
Newsgroups: net.micro.mac
Subject: Re: Easy of programming, Mac, Amiga
Message-ID: <3379@mit-eddie.MIT.EDU>
Date: Sat, 4-Oct-86 13:31:35 EDT
Article-I.D.: mit-eddi.3379
Posted: Sat Oct  4 13:31:35 1986
Date-Received: Tue, 7-Oct-86 19:24:00 EDT
References: <8609152222.AA23166@cory.Berkeley.EDU> <13032@amdcad.UUCP>
Reply-To: zrm@mit-eddie.UUCP (Zigurd R. Mednieks)
Organization: MIT, EE/CS Computer Facilities, Cambridge, MA
Lines: 63

There are a few key points the discussion of Mac programming vs. Amiga
programming has missed. Most important, since the discussion hinges on
the Byte article by an implementor of True Basic, is that his main
gripe is rooted in the job he had to do, not the Mac. He had to port a
system designed to be "portable" to an environment that loathes
"ported" software. Of course it would be unpleasant! True Basic's
goal of making all True Basic implementations the same is in direct
conflict with Apple's goal of making all Mac applications behave the
same. The Amiga makes far fewer demands on its applications to look
alike.

As for the Mac being difficult to program, I can say from experience
that it is loads easier to bang out a quick program for a Unix machine
than a Mac. Since the Amiga is sort of like a Unix system with a
window system layered on top, it would be about as easy to program as
a Unix system. Unless you need advanced features, you can ignore them.
Makes sense.

But what do we mean by "easy?" What do we mean by "program?" If we
mean something that strips the high bit off every third character in a
file, the Amiga wins. If we mean writing something like DBase Mac or
More or LightspeedC, I think the Mac wins hands down. It is much
EASIER to produce software that is very impressive on the Mac. And
that it what sells.

But enough talk! What did I actually DO about the fact that the Mac is
hard to program? Two things: First, I wrote a book called "C
Programming Techniques for the Macintosh" with Terry M. Schilke.
Howard W. Sams & Co., Inc.  publishes it. Second, I started a
consulting firm for those folks without the time to read the book. The
book is what I wish I had when I, a real Unix Wizard, was helplessly
cussing and flailing through my first Mac program. There are many
other good books. Knaster's book should really be called "How to
REALLY Write Macintosh Software" since it points out all the pitfalls
that snag the first-time effort at bigger applications. His book
should take a big load off Apple's developer support people.

And what do I actually feel about programming, now that I  have done a
lot of it on several machines? I don't hack Unix much anymore, unless
somebody pulls a big wallet on me. Mac programming is just more fun,
more intellectually stimulating, and it produces more gratifying results.

Lastly, Apple is doing something about programmer productivity on the
Macintosh. MacApp will make the Mac not only the most featureful
machine around, but it will also make it the fastest to program for
medium to large applications. And it will infuriate people porting
software to the Mac, because it will make Mac programming still more
"different." 

I think Apple has stumbled upon something unique in the computer
industry with the Macintosh. It has produced a machine with a
higher-quality "feel" to it. Programming a Mac is like working with
really good tools, or playing a really good instrument. You have the
confidence that the people who wrote the underlying software are at
least your intellectual peers and, in many cases, have something to
teach you about computer systems. There are few other machines that
seem to win this kind of respect from programmers, and most other
micros don't even come close.

Well now that I have made everybody without a Mac or some avant garde
Lisp or Smalltalk workstation feel inadequate, flame off.

-Zigurd