[AFS3-std] Consensus Call - AFS3-Standardization Charter

[Marcus Watts]

   Jeffrey Hutzelman <jhutz@cmu.edu> writes:
...
> IMPORTANT:
> This has gotten fairly lengthy, but please read through to the end.  This 
> message contains important information on the future of AFS protocol 
...

I've read draft-wilkinson-afs3-standardization-00.txt .
It seems rather long, other than that mostly fine.

I have one quibble, or maybe it's a bit more than a quibble.
That has to do with the documentation / licensing process.
1. Internet drafts are text documents.  I have a huge amount
   of respect for text documents.  Having said that, I found
   it incredibly valuable to include *pictures* in the rxk5 doc.
2. The process makes a big thing about "achieving" consensus before 
   issuing "code points" and proceding to "implementation".  I
   understand the reasoning, and it is wrong.  2 words: Abilene
   Paradox.  Experimental implementations are vital to achieving
   correctness, and should be encouraged, early and often.  The later
   wording on the registrar is different and better.
3. The copyright thing seems overly onerous.  It also seems largely
   derived from section 5 of BCP 78.  Except for E, which seems to
   be unique to this document.  Can this really be an invitation to
   document features with i386 machine language?  Regardless, I think
   this is overly prescriptive.

I think reducing the size of this will address some of Ken's concerns
regarding indians and complexity.

					-Marcus Watts