On Wed, Apr 16, 2008 at 12:02:07PM -0700, John Plocher wrote: > Shawn Walker wrote: > > I'm rather uncomfortable with the attempts here to seemingly codify > > "as a rule", the capabilities of a software product > > As am I. > > I would feel much better if the ips team was doing this - it is really > *their* job, not mine. The problem is that the ARC is stuck, having to > review things that depend on IPS/Packaging's unspecified new abilities > without first having an architectural commitment for them in the first > place.
I think we've gotten to this statement by a circuitous, and not altogether sound, chain of reasoning. The actual problem, as I see it, is that the ARC has started to get inundated with a whole bunch of cases that don't fit the rules that we've been working by for the last twenty years, and the output has generally been "Syntax Error". The source of the flood (let's call it Sun Management, though it needn't have been them) has basically said, "Don't care, make it work" (so I guess the source is Tim Gunn), leaving the ARC members flailing for a solution to make both sides happy. Somewhere along the way, someone suggested that IPS had a bunch of magic goodies that might be used to help make the problem disappear. But IPS wasn't being designed to make the problems between ARC and Sun Management disappear, so if it's actually the magic bullet, it needs some more design / thought / architecture, which I guess is what this thread is all about. Problem is, the IPS team (and Stephen and Bart can chime in if I don't speak for them) doesn't see itself as this particular droid. We're perfectly happy to help, but it's not our priority right now, and we're liable to reject externally imposed design decisions to solve a problem that we don't see as ours. We see this as a problem that exists entirely outside IPS. As Stephen and David said a few weeks ago: forget IPS -- imagine that our DVD just got infinitely big and we wanted to fill that space with all the SVr4 packages we could. How do you solve *that* problem? Is that solvable without presuming IPS? I haven't seen anyone really try, at least not out loud, in public. As in Stephen's response to your original message on this thread, it's not clear that we've even nailed down the architecture of the solution, never mind implementation details like switching to a new packaging system. If it's not possible to solve the problem without IPS, and IPS isn't ready to be the knight in shining armor, or unwilling to suit up, then perhaps "Syntax Error" needs to, as Keith suggested, turn into "Deny". I don't think everyone would be happy with the consequences of that, but maybe it's another phase we have to go through before we get to our end-state. Danek
