Video Decode and Presentation API for Unix (VDPAU) [PSARC/2009/059 FastTrack timeout 02/05/2009]

[James Carlson]

John Martin writes:
> James Carlson wrote:
> >   
> > I don't follow that "however."  Open source doesn't mean "volatile,"
> > nor does hosting by X.Org.
> >
> > The interfaces should be given the _right_ stability level.  If NVIDIA
> > intends to maintain full backward compatibility (and thus will treat
> > deviations as bugs), and if others are building software atop these
> > interfaces, then "Volatile" just isn't right.  That actually sounds
> > like "Committed" ... or perhaps "Uncommitted" if we need to remain
> > fearful.
> >
> >   
> It depends upon timeframe.
> 
> NVIDIA told me last week "full compatibility should be maintained in the
> future".  If the ARC document should reflect only the intent of the owner
> at the time the document is written, then I agree it should be "Committed".

That's exactly what I'm saying.

> However, further discussions with NVIDIA revealed the intention to move
> this to an open source project hosted by X.Org.  My experience here does not
> match your optimism for interface stability.  If the ARC document should
> reflect a conservative or worst case belief of the interface stability 
> over the life
> of the project, I cannot in good faith say it should be "Committed".

Why does moving it to an open source project automatically mean that
it becomes unstable?  The two things should be unrelated.

If you feel you must do so, then call it "Volatile" or worse.  We'll
just be back here again, when we deliberately torture every project
team that makes the mistake of using the interface with a cloud over
its head.

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677