Date: Wed, 28 May 2008 21:45:12 +0200
From: Roland Mainz <roland.mainz at nrubsig.org>
Subject: Re: 2008/344 [ksh93 Integration Update 1 Amendments 1]
Glenn Skinner wrote:
> Glenn Skinner wrote:
> > Date: Tue, 27 May 2008 16:19:44 -0700 (PDT)
> > From: Alan Coopersmith <Alan.Coopersmith at sun.com>
> > Subject: ksh93 Integration Update 1 Amendments 1 [PSARC/2008/344
> > FastTrack timeout 06/03/2008]
> >
...
> > Are we to infer from this nomenclature description that ksh's
> > stability level is decreasing as part of this case?
>
> I read it as "we don't integrate '-' versions into Solaris". Did
> I guess right?
>
> That's my underlying concern. Lower quality and higher probability of
> exposed bugs would seem to imply higher risk that we can't meet the
> guarantees implied by the (ARC taxonomy) stability level we've
> assigned to ksh93.
As said in
http://mail.opensolaris.org/pipermail/ksh93-integration-discuss/2008-May/006127.html
- we don't want to change the ARC stabilty level. We just wanted to
point out that we grab a newer version of the ksh93 upstream sources
with more changes than appropriate for a "stable branch" codebase
(http://mail.opensolaris.org/pipermail/ksh93-integration-discuss/2008-May/006127.html
explains why we did that).
> If the project team wishes to retain the existing stability
> classification, that's their prerogative; they'll have chosen to
> assume the risk of violating the stability classification's guarantees
> (and presumably will have a strategy for mitigating the risk).
Erm... what does that mean ?
It means that you can choose to leave ksh's stability classification
(for ARC purposes) unchanged, and (I think) have done so. To go along
with that choice, you've decided that extensive testing is sufficient
to uphold the guarantees that are implicit in that choice of stability
level. (That's the mitigation strategy I was talking about.)
That's enough to address my concern.
-- Glenn
