[AFS3-std] Re: A call for consensus on draft-brashear-afs3-pts-extended-names-07

[Andrew Deason]

On Fri, 17 Dec 2010 10:21:28 -0600
"Douglas E. Engert" <deengert@anl.gov> wrote:

> So I would like to ask Andrew and Jeff in particular if they feel that
> we need to add these items to the document?

A few words just saying that the implicit fallback mapping is
implementation-defined but may be defined by a future spec or something
may be helpful. I don't know if I consider it required, but I don't
think we need any _more_ than that.

As to the question of defining implicit mappings, I feel that that can
wait until a future spec. Additional interfaces can be designed to
add/inspect/modify implicit mappings, and everything else in this spec
stays the same. And such a thing would certainly be optional, complex
enough and need very mechanism-specific definitions that I think it
warrants its own separate document.

I defer to others on how that would be best represented in the current
doc. For right now, I don't think we need any configurability or
specification for implicit mappings beyond "the server has some
hard-coded rules", but in the future we may want more than that.

-- 
Andrew Deason
adeason@sinenomine.net