[AFS3-std] First order of Business - AFS Standardization Process

[Jeffrey Hutzelman]

--On Monday, December 06, 2010 02:45:57 PM -0600 "Douglas E. Engert" 
<deengert@anl.gov> wrote:

> Section 2.3.3 also states it needs to be confirmed that the RFC editors
> are happy with this process, and submitting this document as
> an individual document to the RFC editors would confirm that they are
> happy.

I believe Jeff Altman had a conversation with RFC-Editor folks before we 
started down this path.  Perhaps he'd like to confirm that?

> Speaking for myself, not as a co-chair, feel changes should include:
>
> That we have now bootstrapped the process and have elected 2 chairs.

Which means we ought to be able to simply remove the section on 
bootstrapping, though we still need the bits about recovering from a 
situation in which we have no chairs or no registrars.

There were a number of specific comments on Simon's document which were 
raised in response to my original consensus call on the bootstrap election 
process.  I believe these all appeared on the mailing list; I summarized 
them at the time and repeat the summary here:

- Tom Keiser says the IPR boilerplate needs to be updated.
- Andrew Deason asks about a process for changing the process.
- Andrew also asked whether RFC-Editor had been consulted.
- Marcus Watts wanted the ability to include pictures in documents.
- Marcus also raised an issue about when code points should be assigned.
- Marcus also wanted to reopen the IPR boilerplate discussion.

There were also some points mentioned in the Call for Votes issued on 
August 24, related to gaps the vote-takers found in the elections process. 
I suggest that those points be reviewed and taken into account in the next 
version of the document.

Finally, there was some discussion post-election of additional edge cases 
in the elections process, particularly with regard to what happens if a 
vote-taker becomes unavailable or unable to fulfill his/her duties partway 
through an election.  Tom Keiser made a concrete proposal for a process 
change which would result in there normally being three vote-takers rather 
than one, and Jason Edgecombe proposed (in the context of a model having 
three vote-takers) that reports from only two vote-takers be required.

I suggest that both Tom's and Jason's suggestions be taken into account in 
the next version of the charter.  Naturally, the "two of three" provision 
will need to be worded so as to work even when the number of vote-takers is 
not exactly three.


> Section 2.3.1.1 refers to BCP 78 section 4. But the BCP 78 changes from
> time to time http://www.apps.ietf.org/rfc/bcplist.html
> and I am assuming it is referring to RFC 3978, Section 4, "Rights in RFC
> Editor Contributions" But the next BCP 78 is RCF 5378, and Section 4 is
> "Non-IETF Documents"
>
> So should the RFC 3978 be listed in the references, and in section
> 2.3.1.1?
> rather then "Section 4 of BCP 78"?

No.  The first sentence of 2.3.1.1 is descriptive, not prescriptive; it 
points out that the IETF and RFC-Editor processes already require granting 
of certain rights.  Those rights are defined in whatever version of BCP 78 
happens to be current at the time of submission (and in the TLP then in 
force), and not specifically in RFC 3978.  The section number mismatch is 
annoying, but we can avoid that issue by removing the reference to a 
specific section or (preferably) referring to the appropriate section by 
name.

Similarly, the remainder of the text should be modified to include only 
that boilerplate which is required _in addition to_ whatever language is 
required by BCP 78.  We should not include the IETF-mandated boilerplate 
verbatim in our charter, because it may change.  Even the IETF itself has 
learned that lession -- the required text is no longer contained in BCP 78, 
but instead is available from <http://trustee.ietf.org/license-info>, where 
it is maintained by the trustees of the IETF Trust.

-- Jeff