Re: The Path of GNUstep (Was: Re: Gnustep + mac + windows? Possible?)

[Tim Harrison]

Pete French wrote:

> Themes are pointless window dressing. No point in it looking good if it
> doesnt run the code. Cross platform abilities is our best Unique Selling Point
> really. Not only is it a fantastic ddevelopment environment but you
> can write once and then deploy on UNIX / Linux / OSX / Windows.

The problem is, if the underlying code is not functional, then it's 
broken on all platforms.  Fix the underlying code, *then* worry about 
getting it running on non-Unix platforms.  It seems that the primary 
platform GNUstep runs on (or is intended to run on) is Unix, so maybe 
getting everything working on Linux and *BSD should be priority, and 
then moving GNUstep to other platforms.

I just worry about effort being expended on the wrong priorities.  Of 
course, I don't know what the priorities are.  I can just state what I 
think.

What about a feature freeze, and a massive bug week (or month) to hammer 
down any pesky bugs, remove any nasty workarounds, and basically clean 
up what's already in CVS?  Chris and Richard have been going back and 
forth on [auto]gsdoc.  Maybe it's time to rip out the deprecated code, 
clean up a bit, and force those who need to upgrade to do so?  At 0.8.0, 
it seems to me it should be time to worry about what's going to remain 
in the source tree for a 1.0 release, not about who's still using some 
old workaround from 0.5.0.

Maybe a little housecleaning would do wonders.  List out all the major 
things that need to be fixed (prioritise the FIXMEs), and just put 
everything into bringing stuff up to snuff.

As I say, it's just an opinion, and from my point of view, it would be 
massively beneficial.


--

Tim Harrison
tim@linuxstep.org
http://www.linuxstep.org/