> In fact there is nothing stopping anyone from putting together
> a port which is based on wxWindow (or any other x-platform
I don't know why this discussion turned like this, really. It's absolutely
clear that nobody in their mind would start completely rewrite of a project
more than a year since it beginning. Writing a wx port of AbiWord is IMHO
absolutely not interesting - especially because it already works (I think)
on all platforms supported by wxWin.
It would have only made sense from the very beginning because (at the price
of having a bit prettier font dialog) it would allow saving all the time
AbiWord developers put into developing the GUI parts of the program.
BTW, I really think this time was not negligible - it seems like there a
lot of people working on AbiWord and, building on one GUI framework they
would probably achieve much more in the time passed since the beginning of
the project.
Of course, I'm saying this because, in the back of mind, I also think that
wxWin could have profit from it too (I don't care much about nicety of font
dialogs - this can be changed in minutes, but a wxBe port would be fantastic),
but it doesn't change at all my deep feeling that AbiWord lost too (and
probably more) from this decision.
I appreciate Larry Wall a lot, but let's not turn his words into something
completely different from what they mean. The Perl motto means that the
language is so flexible that it allows to do anything in the way most
appropriate for the program(mer) - it never meant that each program(mer)
should do everything in each of the possible ways! IMHO, the example at
hand goes against the most important virtue of programmer (quote of the
same author) - lazinessl; because instead of using others work you preferred to
redo it all yourself.
More seriously, duplication of efforts is bad. I understand perfectly well
that AbiSource has means to duplicate our effort (count the people who work
on wxWindows), but I still don't understand why this solution had to be
preferred (rather than help us improve wxWindows - and helping other OpenSource
projects which could use it too). But it's none of my business I guess...
Sorry for a (probably) confused argumentation (or lack thereof), but I have a
feeling of arguing for something so obvious (it's better to reuse code rather
than duplicate it) that I don't even know how to do it. Last point: I obviously
don't agree with Robert G. Werner when he says that this diversity is a force of
OpenSource movement. IMNSHO, it's its worst weakness. If it were not for the
lack of coordination between different projects there would be nothing else
except OpenSource programs since a long time ago.
> I would say take up the challenge and try to do a port of
> AbiWord using the wxWindows library ..
If I really had so much free time I would rather write a BeOS (or QNX) port of
wxWindows :-/ BTW, although it doesn't seem that wxWin is very popular among
abiword developers I'd still like to use this opportunity for a small ad
campaign: if there are people knowing BeOS who would like to create a wxWin
port to this platform, we would be (more than) glad to help them to do it.
IMHO, such port would be very interesting not only for wxWin, but mainly to
Be users who could profit from the programs which already run under Windows/Unix
(and even Windows only - porting from MFC to wxWin is much easier than
directly porting from MFC to BeOS API).
Best regards and good luck with AbiWord development!
VZ