wxruby-dev
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: RE: [Wxruby-dev] Cable ties


From: Curt Hibbs
Subject: RE: RE: [Wxruby-dev] Cable ties
Date: Mon, 17 Feb 2003 22:05:14 -0800

Bob Calco wrote:
>
> I vote for Cable - I have a design plan for creating the Ruby
> emitter, which I will attempt to make public this week so we can
> collaborate on it.
>
> Sorry for my long silence; my wife has been in the hospital on
> bedrest (32 weeks pregnant) for the last four weeks, and will be
> in another 4-8 weeks, and I've been doing a lot of Mr. Mom lately.
>
> - Bob Calco

Bob, on a pure development design basis I would side with you and push for
the Cable route. But the pragmatic side of me wants to be sure we actually
get results. Neither your nor I have had the recent time or bandwidth to
push forward. It doesn't really sound like that is going to be changing for
you any time soon and, given a choice, I would rather be working on my
growing FreeRIDE wishlist (which includes the i18n support the wxRuby would
give us).

And here we have Peter who has actually taken the time to do an assessment,
who was the energy, interest, and (at least some) time to spend on it -- not
to mention the fact that we actually have a working version of wxRuby using
SWIG. Because of this I really have to support Peter in his desire to push
forward with SWIG.

I see a couple possibilities for you...

1) You could continue to work on a Cable path for wxRuby (while Peter works
on the SWIG path). If it turns out that you are able to make significant
progress and the results look compelling, then I think it is a slam-dunk
that we would switch to the Cable path.

2) You could decide, instead, to shelve the Cable path (at least
temporarily) and assist with the SWIG effort. One thing that I know would be
of tremendous value is a test harness that would allow us to create a suite
of platform-independent unit tests for wxRuby. This would really help us in
porting and testing on multiple platforms. I know that GUI unit testing is
particularly difficult, but I also know you have some special expertise in
test harnesses.

As for me, I could act as the first real-world customer by implementing the
plugin that uses wxRuby as FreeRIDE's GUI renderer.

I can see good reasons for you to choose either path, and I will respect
your choice. But until there is a compelling reason to the contrary, I will
support Peter in pushing forward with the SWIG path.

Curt


PS to Richard Lin (and others),

Do you still have the interest, desire, and time to spend on wxRuby? If so,
this would be the time to join in.






reply via email to

[Prev in Thread] Current Thread [Next in Thread]