Alan Burlison writes: > James Carlson wrote: > > > Yes, I saw that, but in terms of the force-fit problem, the User > > Groups are the only fix I see. We also know that consolidations, > > architecture, and distributors are in the cold and are not well > > represented by the model. > > As I've said, I fully agree that we will need new types of structures > in the future. However I don't think that's a reason not to get on and > implement the bits we already know we need. Waiting until everything > is known and everything is decided is equivalent to saying wait forever.
OK; I see the point. I suppose I don't see the outcome as the same -- the likely result of just "doing something" is that other things will be piled on later without much regard to whether or not they fit or make sense, because, human nature being what it is, nobody sane will want to reopen this box. I'd expect ARC-and-consolidation-as-project-with-benefits to continue, which argues against bothering with User Groups. May as well not fix that one, either. > > (Oddly, in all the upheaval, we still have "communities," which few > > seem to understand -- they were, to my eye, apparently seeded in some > > part by Sun's internal management structure [every director needs his > > own community], but some are active, some are in the weeds, and a few > > are tombstones. In any event, they seem to be the least historical of > > the major parts.) > > Personally I don't find them hard to understand, but I suspect they are > one of those types of thing that are easier to define by example. And > the same criticisms you've just made of CGs can also be made of > Projects, and of UGs - in fact *any* groupings that we might choose to make. In comparison to the pre-OpenSolaris model, communities had no formal basis. Perhaps they were management organizations or department numbers. Or they might have been mostly-open mailing lists. They had neither the structure nor formal role that they do in the OpenSolaris model, which is why I'm calling them ahistorical here, and hard to understand with respect to long-existing structures. I'm not saying that they're unwelcome or that they don't serve a function, but rather that they sort of dropped from space, leaving OpenSolaris participants to fumble over what they mean. > > I agree. But backing up just one step: we need to have a clear set of > > goals (e.g., are consolidations relevant?) before we can judge whether > > we're meeting them reasonably. > > Why? > > To steal the .sig of a friend of mine: > > There is this special biologist word we use for 'stable'. It is 'dead'. > -- Jack Cohen Then why bother with the User Groups proposal? It seems a bit random to me to fix one known problem and give a pass on the others. "Eh, good enough" can't be an OGB motto, either. :-/ -- 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
