Re: Gnustep + mac + windows?)

[Fred Kiefer]

Richard Frith-Macdonald wrote:
>> You don't have to try and make people switch to unix or GNustep.
>> Andreas said "Windows developers could be interested in GNustep,
>> if they joined it would help". Nobody's saying you have to try
>> and convert them. Just that if GNustep had a windows GUI, more
>> people would join, which would be good for GNustep.
>
>
> I think this is true ... but Fred has done the major work of implementing
> the windows gui to the point where less experienced people ought to
> be able to take development forward, now is probably the time for
> people who want to develop on windows to have the vision to work on
> it.  It's unreasonable to expect core GNUstep developers who work
> entirely on GNU/linuxto put a lot of their time into windows
> development (though I think we've all offered to help and advise).

Now that Richard has brought up my name, I cannot stay out of this
discussion, as I would have prefered.

Yes, I did write the Windows backend, but although we had requests for
this every other month, I did not write it for somebody else, I did
write it for me.
During my stay in the UK I only had a loaptop along and somehow the
modem refused to work under Linux, whereas it worked great for Windows.
This way all the CVS code ended up in some Windows directory and after a
short time I did give up on copying it over (I had given up on the modem 
driver before this). To be able to continue to work on GNUstep I had to 
write a Windows backend, no way out. So I did sit down for three long 
weekends and wrote the code for this.

As usual in free software project not the secret master plan did bring 
us a Windows backend, but pure chance and personal need. OK, there was 
one more reason to do it, I have to admit. Some months before this, 
there were some discussions on how a Windows backend should be 
structured (We used to have this once a year). And I just wanted to 
finish those off by showing that Adams backend structure would be the 
ideal framework for this task. So yes, I also did it show that GNUstep 
is just great. Now if anybody else would want to show how easy it is to 
do a GNUstep backend for some strange operating system or some obscure 
graphics framework, this would be a good thing to have. Even though I 
probably would never use it myself.
And the same is true for any other feature currently missing in GNUstep, 
if you need it, implement it. And if you cannot do so, ask for it, but 
only if you really need it. As Richard noted despite all the requests 
for it there hasn't been much feedback on the Windows backend.

Finally, as I am now back on my old Linux computer I am no longer 
interested in the Windows backend myself. If somebody will need it, 
there will be further progress on it, otherwise there wont.