OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

xdi message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: Minutes:Joint XRI & XDI TC Telecon 10AM PT Thursday 2007-09-27

Following are the minutes of the joint unofficial telecon of the XRI and XDI
TCs at:

Date:  Thursday, 27 September 2007 USA
Time:  10:00AM - 12:00PM Pacific Time

Event Description:
Weekly unofficial joint call of the XRI and XDI Technical Committees.


Les Chasen
Wil Tan
Drummond Reed
Gabe Wachob


John Bradley
Kermit Snelson



We began with John Bradley and Kermit Snelson introducing themselves. John
is working with Steve Churchill on a C++ XRI resolver implementation that
will become part of several OpenID libraries. Kermit is working with Victor
Grey on a Ruby XRI proxy resolver project that will be called Barx. Both of
them plan to join the TC this coming week as individual members.

This led us to talk about the emerging practice of deploying XRI proxy
resolvers similarly to DNS name servers, i.e., running a proxy resolver on a
LAN to serve local applications enables efficient caching of XRDS documents
much like running a DNS nameserver on a LAN. We agreed this is likely to be
the most common deployment scenario.


At Digital ID World there was dicussion about the fact that XRI Resolution
2.0 Working Draft 11 Editor's Draft 05 (ED05) does NOT declare namespaces
for XRD attributes, however it DOES show them in examples.

After further research, Gabe recommends that we keep XRD attributes
unqualified, and specify that extensions using attributes should be
namespaced. There was consensus on this approach.

# DRUMMOND to: a) update examples in the spec, b) add a reminder that
attributes defined in extensions should be namespaced to prevent conflicts.


John brought up the challenge that many OpenID XRDS documents may not be
properly observing case sensitivity. He asked what implementers should be
doing regarding malformed XRDS documents?

The consensus was that this is an XML issue and thus out of the scope of the
spec, but implementers could choose their own practices.

# DRUMMOND to make in clear in the spec that XML standards require case
sensitivity in XRDS documents.


We discussed that since OpenID Authentication is in its final implementer's
draft, and since when it goes final it will be referencing XRI Resolution
2.0 Committee Draft 02, we should have several TC members read through it to
make sure there are not any issues.

# LES, WIL, JOHN, and KERMIT to read through and send any feedback to the

Based on the recent active email thread, followed by in-person discussions
between XRI TC members and XRI resolver implementers at Digital ID World, a
page summarizing the proposed revisions to ref processing was posted to:


We spent the balance of the call discussing this proposal.

* There was consensus that moving all ref processing to the Service level is
both simpler and more conceptually accurate. There was concern that this
could break XRDS documents that use Refs today, however we are not aware of
implementers making use of Refs to any significant extent yet. Kermit also
suggested that implementations could treat an authority ref as a default
service ref, thereby ameliorating backwards compatibility concerns.

* There was some concern that Service Refs had special semantics that would
not be duplicated by moving Authority Refs to the service level.

# JOHN to discuss this further with Steve Churchill and report back.

* We discussed the need for Ref processing to always result in fully nested
XRDS documents, even when completing XRI authority resolution, in order for
CanonicalID verification to work properly. Les was concerned that this might
changed the way authority resolution functioned, and that this needs further

* We ran out of time on the call, and agreed we needed another dedicated
telecon to close on Ref processing. It was scheduled for 1PM PT on Monday,
October 1.

# DRUMMOND to schedule this special telecon.

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]