Re: Cocoa/Windows parallel dvlpmt
Re: Cocoa/Windows parallel dvlpmt
- Subject: Re: Cocoa/Windows parallel dvlpmt
- From: Dave Thorup <email@hidden>
- Date: Wed, 4 Feb 2004 17:45:45 -0500
I think Florent makes some great points here, and here's my opinion...
On Feb 4, 2004, at 11:02 AM, Florent Pillet wrote:
Now there are a few things about GS that need to be very clear:
- NeXT having been absorbed by Apple (though some will say that it
happened the other way round :-)), the new baseline OpenStep APIs are
Cocoa. Any attempt to ignore this fact will lead to Cocoa developers
turning away from GNUStep. Let me repeat that: OpenStep is history.
Cocoa is the new specification. Hence, GS should at least provide the
equivalence to the N-1 version of Cococa (ie currently offer the set
of APIs that were in Jaguar would be deemed acceptable by developers).
The importance of this cannot be overstated. GS really needs to make
it a #1 priority to maintain 100% compatibility with the new OpenStep
which is Cocoa. If this goal does not become one of the top priorities
of the GNUStep project, then the majority of Cocoa developers will
never look into it seriously, nor will they contribute to it. For
example, what if the WINE project decided that they wanted to implement
some of Microsoft's Windows APIs differently, or not at all. Sure,
their implementation could be "better," but if it were different and
didn't behave the same way then it would lose its appeal.
- I think that the look and feel issue needn't to be overlooked. As
Alex mentioned it, people must overcome their own personal taste and
realize that in any given platform, the only acceptable way of
displaying the UI is by respecting the platform look and feel.
This is also a hugely important priority. In order to appeal to Cocoa
developers, GNUStep will need to make their AppKit implementation use
native Windows controls or at least controls that look exactly like
native controls. If this is not a priority of the project then Cocoa
developers won't be interested in using or contributing to the GS
AppKit.
In reading the GNUStep mission I see that maintaining Cocoa
compatibility is there, but does not look like a huge priority. That
may be fine for the GNUStep project as it is now, but in order to
attract more developers the mission should be changed to strive for
Cocoa compatibility. I think Florent's suggestion of N-1 version
compatibility is acceptable (N-2 wouldn't be too bad). Remember this
is marketing. In order to attract more Cocoa developers, the goal of
100% (or 99%) Cocoa compatibility needs to exist.
_____________________________
Dave Thorup
Software Engineer
email@hidden
http://www.kuwan.net
Defaults Manager - The premier editor for Mac OS X's User Defaults /
Preferences database.
_______________________________________________
cocoa-dev mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/cocoa-dev
Do not post admin requests to the list. They will be ignored.