In <uNqdnahp1_BPUK3ZRVn-sg@comcast.com> Michael J. Mahon wrote:
> PZ wrote:
>>
>> Supposedly Heineman had AppleWorks GS 2.0 - 98% completed but bailed
>> after contract issues with Quality Computers.
>
> So the stories about the source being completely unmanageable were
> just spin?
I don't think so, not if the following were even only half true. This is
part of a message from Jerry Kindall of Quality Computers announcing the
official death of AppleWorks GS. It was posted to comp.sys.apple2 on 23
June 1994.
"When we took over AppleWorks and AppleWorks GS, we assumed that
updating
AppleWorks GS was going to be a fairly straightforward task.
Unfortunately,
we were wrong. The source code for AWGS is 7 megabytes in size. Those
who
have seen it have called it the most poorly organized and documented
source
code they've ever laid eyes on. Claris was unable to provide us with
any
form of documentation for the source code, nor were they able to provide
us
with their official bug list.
The source code was designed to build under an old version of MPW
(Macintosh Programmer's Workshop). I have it on good authority that
even
the old MPW wasn't actually capable of compiling the source code as it
was
provided to us; it seems likely that it was compiled in pieces and then
patched together by hand. We did not receive any documentation on this
process -- in fact, there probably never was any.
To give you an idea of how bad the AWGS source was, consider that it
took
Jim Merritt, who we originally contracted to lead the project, four
months
just to get the source code Claris sent us to produce an executable
version
of AppleWorks GS 1.1. Even then, the program was not 100% byte-for-byte
identical with the shipping version because of the hand-patching which
was
used in the original version.
Jim Merritt, as you may know, is no slouch. He's the one who, working
at
Apple, coordinated the development of the IIGS System 5 Finder, among
other
things. The original plan was for Merritt to divide the program among
several programmers (of his choosing) and have them work on the project
independently. Merritt would be responsible for coordinating things and
making sure all the program segments could be combined into a fully
functional program.
This turned out to be an impossibility, because the source code simply
was
not arranged in any coherent fashion. I've been told that there are
sections of the AppleWorks GS source code which exist mainly because
nobody
knows exactly what they do -- Claris was afraid that removing them would
cause the program to stop working!
At around the same time another programmer decided he wanted a shot at
it.
This programmer was Bill Heineman, author of Harmonie, Out Of This World,
and other programs, and renowned around the IIGS world as an extremely
competent programmer. As the weeks went by, we realized that it was
beyond
his abilities, as well.
A third team, led by Steve Disbrow, also had a look at AWGS. After
spending
a few weeks with the source code, Disbrow recommended to us that we
scrap
it and rewrite the entire AWGS application from scratch. This is, for
obvious reasons, an economic impossibility. So, after almost ten months
of
effort by the top programmers in the IIGS world, we have concluded that
the
task of creating a major upgrade is a near-impossible one."
--
Roger Johnstone, Invercargill, New Zealand
http://roger.geek.nz/
________________________________________________________________________
No Silicon Heaven? Preposterous! Where would all the calculators go?
Kryten, from the Red Dwarf episode "The Last Day"