Minutes of XRI TC Calls Thursday 8PM PT / Friday 9AM PT USA

["Drummond Reed" <drummond.reed@cordance.net>]

Following are minutes of last night's Asia/US call and today's US/Europe
call.

ASIA/US CALL THURSDAY 8PM PT 2004/09/15

Present (at different times on the call): Kunal, Peter, Drummond, Wil, Nat

This call was disjointed because different people could make it at different
times. However the following was covered with everyone on the call:

* Reaffirmation of the TC process outlined in the agenda below. (Consensus.)
* Proposal to advance the Syntax spec by Oct. 15 even if the Resolution and
Metadata specifications are not ready for advancement by that point.
(Consensus.)
* Discussion of the need to refactor the Resolution spec to remove the
non-normative text. (Consensus.)
* Discussion of some of the requirements driving the Resolution refactoring
proposals. (Consensus that incorporating proxy resolution was necessary.)


US/EUROPE FRIDAY 9AM PT 2004/09/16

Present: DaveM, Gabe, Mike, Chetan, Marty, Steve, Aman, Kunal, Les, Victor,
Peter, Drummond

AGENDA

1) PROCESS
We discussed the overall process for development and change management of TC
specs. There was consensus that normative content needs to follow this
general series of steps:

	Step 1: Submit proposal w/ use cases/requirements (design optional)
	Step 2: Reach consensus on requirements and scope
	Step 3: Submit design proposals
	Step 4: Reach consensus on design
	Step 5: Draft specification text/artifacts
	Step 6: Reach consensus on spec text/artifacts and vote

It was agreed that non-normative content does not necessarily require this
full process, but it should be always be carefully vetted (and in general
should avoid non-normative content in specs as much as possible.)

2) PRIORITY
We discussed a proposal to make our priority to advancing the Syntax spec to
an OASIS vote first and if necessary following with Resolution and Metadata
at a later date if we do not yet have consensus.

There was a very long discussion about this proposal. Major points of the
discussion included:

* Syntax should not have normative references to Resolution or Metadata (if
they do they should be removed).
* There was consensus that there are significant adoption advantages to
moving at least Syntax to OASIS Standard level.
* It is the strong preference of many TC members to also advance Resolution
and Metadata at the same time.
* There is not yet consensus about Resolution, including scope, design, and
documentation issues. There are strong feelings about building on CD01 with
as few changes as possible, and there are also strong feelings about
refactoring.

In the end the consensus was to make it a priority to advance Syntax to
Committee Draft 02 by Oct. 15 in preparation for an OASIS vote, and to try
as hard as we can to reach consensus on doing the same for Resolution and
Metadata by that date.

3) SYNTAX
We were able to close three of the four current normative Syntax issues.

* Issue 1: X.500 Directory Attribute Appendix 

Based on input from Dave and Marty, there was consensus that this is better
tackled independently from the Syntax spec. It is an important deliverable,
but it can be a standalone document or spec, and it would benefit from
involvement of multiple X.500 directory experts.

* Issue 3: ABNF/Add HXRI and QXRI

There was consensus that this does not need to be in the Syntax spec, but
can be specified independently in the Resolution spec. It will be moved to
the Resolution change management page.

* Issue 4: XRI Normalization

Dave was able to stay long on the call and discuss the conclusion documented
in the minutes of the special call on this topic at
http://lists.oasis-open.org/archives/xri/200509/msg00045.html. He agreed
that requiring NFKC normalization at the time of the encoding into XRI
normal form was a good decision and involves only changing the current
references from NFC to NFKC.

The game plan for the other current normative issue is:

* Issue 2: ABNF/Allow Empty Path Segments 

Dave and Mike both agreed they would research why this decision was made
last winter. They will also review the analysis and proposed new ABNF from
Drummond that fixes this (see
http://www.oasis-open.org/committees/download.php/14487/draft-xri-2-cd02-abn
f-v1.doc). 

ALL TC members are encouraged to review the proposed changes to the ABNF
carefully, since this is the only key issue that affects the overall ABNF.

Lastly, it was agreed that Wil Tan's post to the list today at
http://lists.oasis-open.org/archives/xri/200509/msg00057.html was an
important new issue that will be added to the Syntax page. Drummond will
check with to Wil to ask how many characters this involves and thus what the
potential impact is on the IRI ireserved character set.

4) RESOLUTION/METADATA REQUIREMENTS
We agreed that the fastest way forward on Resolution and Metadata is first
to review the use cases/requirements for the proposed changes. TC members
are requested to add detailed requirements as quickly as possible, and then
pursue discussion on the list or wiki so we can be as far advanced by next
Thursday's/Friday's call as possible.