Re: Minutes from July 16 Release Wranglers call

On Fri, Jul 16, 2004 at 01:36:15PM -0700, Torrey Lyons wrote:
> Some comments on Mac OS X support:
> 
> 1. If the next release is going to be a major release and not a 
> stability X11R6.7.1 release, I was planning to pull all the recent 
> XDarwin work over from the XFree86 tree. Any objections to merging 
> this into HEAD? (I see from a comment by Roland Mainz that Mac OS X 
> does not currently work.)

We discussed several features on today's release wrangler's call, and I
think this would be a good one to discuss as well.  The feature deadline
is two weeks from today, so there is still time to discuss and bring
this code into the tree.

> 2. Adding XDarwin server side support for many of the new extensions 
> contemplated for this release will not be trivial. The release 
> schedule is very aggressive since there is not yet a clear consensus 
> on exactly what is going into it.

Yes, this is a very aggressive schedule, but it is not required for
XDarwin to implement support for all of the new extensions.  At this
point, the extensions under consideration are XFixes, Damage, Composite
and XEVIE.  On today's call, the Composite extension was discussed,
which appears to be the main point of contention.  The resolution was
two fold:

1. Investigate the Composite code to determine if building it into this
   release will destabilize other parts of the code.  The conjecture is
   that the Composite code is relatively self-contained and will not
   affect the functioning of the remaining code.  With the XAA changes
   proposed, it will also not cause an ABI change.
2. Disable the Composite extension by default, and add an experimental
   extensions section to the configuration file that would allow
   interested parties to explicitly enable it at run-time.

There is strong sentiments that it should be included in the release and
I'm inclined to agree with the provisoes above.  XEVIE will also be
considered experimental and will default to off.  If it is determined
that XFixes and/or Damage should also be considered experimental, then
they will also default to off; however, at this point, the general
consensus was that they should be enabled.

> 3. Mac OS X 10.4 "Tiger" (http://www.apple.com/macosx/tiger/) will 
> ship with an updated version of Apple's X11. Apple is going to be 
> faced with the question of what release they should base Tiger's 
> version of X11 on. The next X.Org release might be timely if it is a 
> solid release on Mac OS X.

If having our release out by 25 Aug 2004 will allow Apple to base their
X11 release on ours, that would be extra motivation to make this release
solid on Mac OS X.

Kevin