Gary Winiger writes:
>       This one is mine, not the project team's.  In clarifying the
>       output, Unstable -> Not-An-Interface, I suggested
>       Evolving -> Uncommitted.  It seemed to me that even with

Note that in the updated taxonomy, both ON and PSARC have treated
"Evolving" as essentially "Committed."  Some others historically
considered "Evolving" to be what we now call "Uncommitted," but I
think doing something different requires some explanation.  See:

http://sac.sfbay/cgi-bin/bp.cgi?NAME=interface_taxonomy.bp#EXPSARC

>       Uncommitted, full notification of incompatible change/EOL/EOF
>       processing is needed and the boundary is Minor.  So, just like

No, Uncommitted does not require "full notification."  It requires
"impact assessment" for incompatible change, and none at all for other
change.  Note also that it's intended for experimental, transitional,
or otherwise undependable interfaces.  Is that really what we're
describing here?

The "impact assessment" in this case is exactly what a contract would
document.

>       non-Sun consumers, unbundled Sun consumers would be adequately
>       notified.
> 
>       Of course, if members believe it's necessary/desirable, a contract
>       can be drawn.  As I'm presently away -- and Ashish is likely
>       optimistic about my return date -- let's not hold things up
>       only for a contract.

Given the lack of visibility into Sun Ray that we have on the ARC, I
think I need to insist on a contract to nail this down.  We really
have no way of knowing what the impact might be.

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
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

Reply via email to