In which PythonCard Forks...

[Richmond Mathewson]

Alex Tweedly wrote:

"Yeah, it's unfortunate (IMHO) that he has forked the project without 
ever offering any enhancements to the original project. If he has the 
time and energy to improve the project, he could easily have done so 
within the existing project - but he has not (AFAICT) ever posted any 
code or specific suggestions to the Pythoncard-developers list."

and

"It should be interesting. While I wish him luck, I do wish he wasn't 
including inadequately tested code (mine!)  in his initial release."

Comment on first quote:

I suppose this is the downside of Open Source initiatives; while it
can encourage creativity that is unfettered by commercial considerations,
it is also open to all sorts of abuse, and leaves room for an awful lot
of hairy-chested egoism to go unpunished. 

Comment on second quote:

This seems to be a universal problem; having experienced it both in the
commercial sector and the free one. In both cases it looks rather like
why Communism doesn't work; not many people can fill their bellies on
high-sounding ideas - there have to be stronger motivating factors to
get involved in Beta-testing; whether financial or otherwise.

Certainly the profit motive (however 'dirty' the profit motive may be to
 ideological purists) is a very important factor: and it cuts both ways. 
I worked on a commercial project where the financial backer refused 
Beta-testing because he wanted to maximise his profits. Of course when 
a few glitches became apparent in the product AFTER it had been released
 the programming team were blamed, even though they had urged the backer 
to pay people to Beta-test the product. Needless to say, the backer
 ultimately lost out as end-users stopped buying the product as word 
of its flaws spread rapidly.

Certainly, all I know about good business practise (whether 
commercial, open source, or just Plain Richmond programming for the 
few computers in his micro-school) involves several stages of Beta-testing
in the SDLC (my GANTT charts seem to consist of nothing else!!!!).

I am a lucky person as the children I teach have, over the 2-3 years I have been teaching them, become ruthless, evil-minded software testers,
who do not feel bound by the usual painful social conventions, and are
more than happy to tell me WHERE and WHY something 'stinks' in any of
my programs. I foster this by having a box of small Oxford EFL Learner's
Dictionaries which are much sought after by the kids: they find a 
problem in one of my programs, they get a shiny, new dictionary: pure
profit motive - and it works! They are, also, becoming increasingly 
sophisticated at assessing GUIs - and, "just for fun" I took the 
advanced kids through the Nielsen stuff:

http://www.useit.com/

sincerely, Richmond Mathewson.

____________________________________________________________

A Thorn in the flesh is better than a failed Systems Development Life Cycle.
____________________________________________________________