unzoo integration [PSARC/2008/183 FastTrack timeout 03/14/2008]

[James Carlson]

Joseph Kowalski writes:
> James Carlson wrote:
> > John Plocher writes:
> >> Do we do unzoo and unustar and unasc and unvdb and untnef and ....
> >>
> >>      OR
> >>
> >> do we do pax, which does them all.
> >>     
> >
> > There's no such architectural question if this is a mere repository.
> > Or, rather, the answer is a simple "yes."  You do them all.
> >   
> That probably is the right response from the ARC.  I'd hope the answer
> wouldn't be as simple as that from the business side.

Indeed.  But for the architectural point, I want to bring up the
question: what are the rules for integrating into this repository?
The rules appear to be different from the ones we've dealt with in the
past.  At least the assumed cost of carrying bloatware is apparently
quite low versus its value in making us the robust, Linux-beating,
full-featured OS that we need to be.

Are there perhaps other differences?  How about a blanket exemption
from (or rewrite of) 1991/061?  Leaving that standing would prohibit
branched implementations from reaching the repository -- and would not
allow a dim-sum "pick any one of a, b, or c" kind of approach to
package selection.

I haven't seen any repository rules yet, so it's hard to say anything
for certain, but I suspect that the above issue may well be a
showstopper.

-- 
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