Joerg Schilling wrote: > Rainer Orth <ro at techfak.uni-bielefeld.de> wrote: > >>> document again. Uncommitted could probably be OK as well but we at Sun >>> actually have no control over these interfaces. >> Sun control is not the point here and never was; this is a common >> misunderstanding. This is all about the actual stability of the >> interfaces, which is all the user cares about, not who controls them. If >> the project has a track record of keeping interfaces stable, non-Sun >> projects can easily be Uncommitted or even Committed. > > You are correct, this unfortunately still is missunderstood by Sun people. > Interfaces they cannot control are declared "unstable" although they may > deliver a lot more stability than Sun projects do. > > My conclusion is that OpenSolaris is not yet a really open system as Sun likes > to control things that are controlled by the community.
You're jumping the gun. It's not a matter of "openness" in any sense. It's a matter of project teams being confused and not wanting to promise more than they can deliver. You and Rainer are both correct that stability has nothing whatsoever to do with control or authorship, and that using one to determine the other is just nonsense. That's not the issue that these project teams are looking at, though. It's not some nefarious plot. It's just that the teams who are doing the work to port this software into the build system do not want to promise that future updates won't break things, and they don't want to be held accountable to "fix" any breakage that originates upstream. They don't know those people and can't figure out how to monitor or control them. It's an understandable concern, if unjustified. It's the job of the team that does this sort of delivery via OpenSolaris to figure out what the real stabilities are, and how they'll be managed to deliver reasonable bits to OpenSolaris users. Yes, it does take a bit of work in some cases. No, this responsibility can't just be tossed into the wind by claiming "Volatile" everywhere. If that's what you want, then you really want /contrib without benefit of architectural review, or something like it. You're seeing incomplete work, not "closed" behavior. The two are quite different. -- James Carlson 42.703N 71.076W <carlsonj at workingcode.com>