John Plocher writes: > James Carlson wrote: > > 1. Appeals Path > > My thoughts on this are rather simple:
Mostly agreed with those. The point I was making is that we don't yet have a suitable agreed-on appeals process. > More thoughts on this whole topic are in the ARC community: > http://www.opensolaris.org/os/community/arc/handbook/arc-dev-process/ I don't see anything there on how appeals will work. Am I missing something? > > 2. Contracts > > > > ARC contracts currently require 'signatures' from managers that > > 'own' the technology components within Sun, and these normally > > tie into bugster category/subcategory. What's the equivalent for > > OpenSolaris? (Perhaps communities can sign contracts now ... but > > what happens when a community dissolves?) > > What happens when there is a Sun Reorg? Sometimes what happens is that the contract is left dangling in space, and we potentially run into serious problems much later. I suppose I shouldn't try to assert that either is a more or less stable reference point than the other. > IMO, contracts between components/consolidations are a BAD THING and > generally should be forbidden/avoided if at all possible. Instead, > interfaces that have a greater commitment should be explored and > developed such that we don't end up lock-stepping everything together > with contracts. I agree they're often just a horrible hack compensating for a lack of engineering effort. That's not really the issue. The issue is *if* we still support contracts (say, for example, between the OpenSolaris Install and ON consolidations for sharing Zones and WANboot interfaces), how does this process work? Who "signs?" Should we redesign all of our code to avoid contracts before releasing it as open source? Do Sun managers get pressed into service as representatives for things they no longer really own (such as community projects -- say, contracts between ksh93 libraries and other consumers)? Do we squish it all into one big consolidation? ;-} I'd love to be able to say "no contracts, ever." I don't think that's realistic, though, so if they still must exist in the extreme cases, how do we do them in OpenSolaris? -- James Carlson, Solaris Networking <james.d.carlson at sun.com> Sun Microsystems / 1 Network Drive 71.232W Vox +1 781 442 2084 MS UBUR02-212 / Burlington MA 01803-2757 42.496N Fax +1 781 442 1677
