Re. PythonCard - was Re: Event-driven GUIs ...

[Kevin Altis]

I just had to chime in. I was a hardcore HyperCard (HC) developer at one
time and built a variety of add-ons and stand-alone applications, including
Organizer + (a PIM), my first commercial app. Anyway, Python is the closest
I've come to the fun I had creating and exploring in HyperCard, but of
course, Python "out-of-the-box" is much more capable at doing real-world
tasks and interfacing to other components and applications. As others have
said, what is missing is the tight coupling of a GUI environment. All of the
current GUI solutions feel like work, compared to whipping up something in
Visual Basic, Delphi, etc. Given that many people think MS has basically
broken whatever simplicity VB had with the transition to VB .NET, there is
actually a huge opportunity to woo VB developers over to Python if a
commercial-quality Python GUI/IDE environment was available.

I'm guessing that at some point, O'Reilly will probably catch a clue and
write a "Python for VB programmers" book, but they haven't done a "Python
for Perl programmers" book yet, so I may be wrong.

Getting back to HyperCard. Remember that HC was closed and proprietary and
the HC team tooks its sweet time in adding features. Color support was
always a hack job and even though 2.0 added printing, menus and other UI
elements that looked like what users expected on the Mac, it was difficult
to impossible to actually create an HC app that had the Mac look and feel
and operated the way a Mac user would expect. In Python terms, it is like
trying to make a first-class app for Windows or the Mac using tk, you just
can't do it, your app will always look like a tk app. The good news is that
wxPython (and maybe other UI toolkits or DHTML/DOM) looks like it will fit
the bill. HC was also difficult and error-prone to extend, Python is easy to
extend.

So, I'm in favor of a PythonCard environment as well as a set of modules
that would ship with the standard distribution, to simplify running anything
built in PythonCard. If enough people are interested, we could even form a
SIG. It would probably benefit everyone to take a look at projects such as
Sash http://sash.alphaworks.ibm.com/ and Squeak http://www.squeak.org/ to
better define what people want and don't want. If you wanted to run
arbitrary PythonCard code from the net, then you probably want to run the
Python interpreter in some form of restricted mode, and/or require digital
signatures to be associated with projects to reduce the risk of running
viruses, trojans, etc. Right now, Python by default is wide open.

ka

"Spicklemire, Jerry" <Jerry.Spicklemire at IFLYATA.COM> wrote in message
news:mailman.992975010.18077.python-list at python.org...
> Hi Roman, Neil, et. al.,
>
> Neil suggested:
>
> "However, the market, which was given several HyperCard like products to
> choose from, has decided against this form of development environment."
>
> to which Roman replied:
>
> "The market ways are unpredictable, because decisions are made not on the
> basis of technical merits but with some strange reasoning behind."
>
>
> Just my $.02 about "the market" where Open Source and FREE software
> is concerned. There is actually something of a parallel universe
> phenomenon at work, an inevitable side effect of the Web and it's
> predecessors. Python is a great example of a tool that is virtually
> invisible to "the market", but which has been thriving nonetheless.
> Think Linux, prior to about 1997.
>
> As many Java and C++ refugees frequenting these lists can attest,
> "the market" no longer has the last word. We're witnessing the
> birth of a "paramarket". Python has risen on its merit entirely
> independent of "conventional wisdom", "market forces", "public
> relations", "strategic imperative", and other counterproductive
> distractions.
>
> So, as for Python Card- Go For It! If HyperCard faded due to
> "extenuating circumstances", perhaps a comeback is in order.
>
> Later,
> Jerry S.
>
>
>
>
>