On 4/21/06, James Carlson <james.d.carlson at sun.com> wrote:
> Alan Coopersmith writes:
> > I. Szczesniak wrote:
> > >> A suggestion from a PSARC member was to keep /bin/ksh
> > >
> > > This is a suspicious suggestion. First PSARC suggested a transition
> > > period and now they want a 'all or nothing' solution? Is PSARC
> > > actually WILLING to allow the integration of ksh93 as /bin/ksh?
> >
> > PSARC is a committee with a number of members - as is normal for any
> > collection of humans, they don't all agree on every topic and different
> > members will give you different suggestions. Until a formal proposal
> > is brought to PSARC for discussion and vote, you cannot and will not know
> > what "PSARC's opinion" is, and what they are willing to allow.
>
> Even when that's done, PSARC isn't a person and won't have an opinion.
> There'll be a written opinion, though, that reflects the opinions of
> the ARC members.
>
> So, as long as you ask just one member, all you'll get is his (in this
> case, my) opinion. It won't necessarily match what the rest of the
> members will say.
>
> I'd rather see a single ksh, based on this reasoning: if ksh93 is
> compatible enough that we're comfortable with inserting it as
> /usr/bin/ksh, then this means to me that there's no further purpose to
> the old Sun ksh except as a museum piece. If there is _any_ purpose
> to having access to the old Sun ksh, then I'd question whether we're
> all quite so confident that we're ready to do the replacement at all.
>
> Moreover, if there are any "failure cases" that require switching back
> to oksh, then that's a disaster. Those failing scripts already refer
> directly to /usr/bin/ksh, and thus will be unfixable without changes.
> And if you're going to make changes, you might as well port the script
> to ksh93. Thus, keeping /usr/bin/oksh around doesn't even serve to
> mitigate problems.
>
> In other words, the solutions that make sense to me are:
>
> 1. /usr/bin/ksh stays the same, and /usr/bin/ksh93 is shipped as
> the new ksh93. Perhaps someday the two converge again, but not
> now.
I hope you are aware of the consequences of such a step? The inability
of Sun to deliver a decent version of ksh as /usr/bin/ksh already
DAMAGED the reputation of Sun and Solaris. How much further should
this go?
--
_ Felix Schulte
_|_|_ mailto:felix.schulte at gmail.com
(0 0)
ooO--(_)--Ooo
