Re: Project leader candidacy

Kristian G. Kvilekval wrote:
> I've been commiting Ed's patches, just more slowly and after a bit of
> review and testing.   You can also commit his changes  you if you think
> they are the right ones.   The whole scale interface changes probably
> should be agreed to by at least one other developer (even inactive
> ones), the locking code will go in when someone has had chance to check
> it thoroughly and it does not conflict with other goals (win32).
> 
> That said, the interfaces need to be frozen for a short period. The
> win32 code will probably be re-introduced shortly.   The new build
> system should allow us to use mingw32 for cross compilations (minimal
> testing).


that's fine. But i develop on my local branch whether or not my stuff is 
committed to cvs.  Not accepting changes because they may break win32 as 
you re-introduce it (a mistake in my opinion anyway) will not save the 
win32 port from being broken.  Eventually the changes will be committed 
and unless win32 developers can actually keep up with the unix 
developers, the port is going to fail again for the same reasons it 
failed before.  As it stands, My todo list doesn't have interface 
changes for some time.
I don't think lesher's efforts are fruitless but i do think he needs 
more active win32 developers to help handle the two fronts he's got to 
face. You cant port to a new build system without being able to test the 
building of the files, you cant build the files without updating them so 
that they're compatible up to date with at least the bugfixes.  This 
means incredibly slow progress with just one person so if zinf cvs 
development is to be frozen someone better be helping lesher develop the 
win32 port who has the power to freeze the code base.




-------------------------------------------------------
This SF.net email is sponsored by:
The Definitive IT and Networking Event. Be There!
NetWorld+Interop Las Vegas 2003 -- Register today!
http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en