John Plocher writes:
> The "proper use of abstractions" mantra says that as long as you
> release a self-consistent bundle of things, you can do what you want
> inside the bundle.

This mantra doesn't save you if different consolidations choose to
have Major releases.  The taxonomy allows (with justification)
incompatible change in Committed interfaces in a Major release.  In
fact, the desire to create deliberate incompatibilities is the driving
force for having such a release -- not just because we "can" but
because we "must."

I realize it must seem that I'm harping on Major releases in
particular.  It's because I think the go-with-the-flow approach to
managing the releases works only if you assume an unbroken string of
Minors.  Unfortunately, I don't think that matches with everyone's
expectations well.

> Non-Committed cross-consolidation dependencies quickly devolve into
> a single intertwined WOS that must be release-managed as a single unit,
> which is why we pay attention to them when developing and reviewing
> architecture.

Yes, and to a large degree, that's what we have now, in part because
we've been treating ARC contracts as mere paperwork.

The issue with release management still seems murky to me.  I assume
that distributions do not get to manage uname -r, unless they're
willing to fork.  If this is the ON community's task, then how
(reasonably) do distributions identify their releases?

-- 
James Carlson, Solaris Networking              <[EMAIL PROTECTED]>
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
_______________________________________________
opensolaris-code mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code

Reply via email to