[arc-discuss] Requesting PSARC/2007/124 details

[James Carlson]

Dale Ghent writes:
> On Mar 28, 2007, at 2:47 PM, James Carlson wrote:
> 
> > Who are "we?"  Your message is addressed to arc-discuss, but it's
> > really not the ARC's role to determine what a given consolidation will
> > accept or not accept.
> 
> "We" are people who aren't SWAN-enabled. "We" generally take the  
> stance that if code or a policy is committed to the otherwise open  
> portion of the Nevada codebase, the reasoning and deliberations  
> behind that change (ie; the PSARC in this case) should be made  
> available as well, if anything but to document that change (which one  
> of the reasons behind an ARC case, yes?) so 3rd parties who use these  
> interfaces can adjust accordingly.

And I said I agree with you.

The problem is that this isn't the right forum.  Instead, I think you
need to address the ON Community itself.  The rule you're effectively
proposing -- "no ON integration into the open source part of the tree
based on closed ARC cases" -- makes eminent sense to me.  It's just
not an architectural issue.

> Case in point is this particular PSARC: It's something that changes  
> interfaces that are public in both the effective taxonomical sense  
> and the unencumbered code sense. I'd argue that in cases like this  
> the ARC documentation should be part and parcel to the actual putback  
> at the very least...

That's actually not relevant and not true.  These interfaces are not
at all public in the ARC taxonomy; they're private to the ON
Consolidation.

It's not relevant because the issue here is that the code being
modified is itself open source, and thus the case must be (should be,
ought to be) open as well.

"Public" in the taxonomy sense has nothing whatsoever to do with "open
source."  They're orthogonal issues.  "Public" in the taxonomy means
that there is reference documentation (such as man pages) supplied and
we provide compatibility information for others depending on the
interface.  The source code itself is emphatically not documentation.

If you're trying to argue that interfaces without documentation are
somehow stable, then that's another matter.  I'd like to see an ARC
case filed to change the explicit stability first.

-- 
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