Minutes: XRI TC Telecon 2-3PM PT Thursday 2010-01-21

[Drummond Reed <drummond.reed@xdi.org>]

Following are the minutes of the unofficial telecon of the XRI TC at:

Date:  Thursday, 21 January 2010 USA
Time:  2:00PM - 3:00PM Pacific Time (21:00-22:00 UTC)

ATTENDING

 

Breno de Medeiros

Will Norris

Scott Cantor

Drummond Reed

Markus Sabadello

Bob Morgan

Nika Jones

Nat Sakimura

 

REGRETS

 

John Bradley


1) FINAL REVIEW OF XRD 1.0 WD 12

            http://www.oasis-open.org/committees/download.php/35954/xrd-1.0-wd12.html

 

Will said only two additional sets of feedback he received were from Nika and Drummond. Those have been incorporated. Of those, the only ones needing discussion were:

 

* Title element: we agreed to change its description to simply state: "The <Title> element contains a string value that provides a human-readable description for the link.”

 

* Timezone: After much discussion and a review of the XML Datatypes spec, we agreed on the sentence: “Time values MUST be expressed with the UTC designator ‘Z’.”

 

# ACTION: Will to tag WD 13 and send a message to the list that we will open a vote Monday both to approve it and hold a 30-day public review unless anyone objects. Note that we will also vote on the revision to our XML namespace policy for XRD. (So that’s three votes in all.)

 

# ACTION: Drummond or Peter to then open the vote.

 

 

2) PROPOSAL FOR AN X.509-BASED XRI TRUST PROFILE

 

See the thread starting with this message from Breno (which contains an entire spec!):

 

            http://lists.oasis-open.org/archives/xri/201001/msg00083.html

 

We have set a record for proposal-to-spec time: Will has already drafted it in the XML template and sent it as an attachment to the list. (Thank you!)

 

            http://lists.oasis-open.org/archives/xri/201001/msg00101.html

 

We discussed whether it should be/needs to be “sync’d” with XRD 1.0. The feeling was that they are very complementary, but do not need to be sync’d.

 

Scott mentioned a few minor items of feedback:

 

* “subjectAltName” needs to be spelled that way throughout.

 

* The term “Authority Name” was not intuitive – Scott suggests the term “Claimed Subject”.

 

Breno brought up the following points:

 

* Section 2.3 still needs work to make sure extensibility works smoothly.

 

* Section 1.3 about the root certificates needs to make it clear that this set is dynamic.

 

* The trust model is based on correlation between a triple: the trust anchor in the XRD, the trust anchor in the certificate, and the resource that the XRD describes. If the binding to any one of these three does not verify, the descriptor is not valid. There was agreement that it is necessary to be confident about all three of the inputs to the verification process.

 

* The matching of the certificate matching rules in section 2 may need more work. Bob’s feedback is that it should, in theory, be the same processing steps specified in PKIX. Scott agreed, and yet he said SAML had to profile that further as well.

 

# ACTION: Breno to review this draft and then he and Will will commit it.

 

3) XRI 3.0 SYNTAX

XRI 3.0 Syntax is finally content complete. Drummond posted Working Draft 03 on Monday:

 

            http://www.oasis-open.org/committees/download.php/35972/xri-syntax-3.0-wd03.pdf

 

He said is ready for a close read-through, particularly in the final sections which are the only new ones. In particular he would like a close read on:

 

* The Relative XRI Reference resolution section. Since in XRI 3.0 architecture, all XRIs are relative URIs, this means a relative XRI can be relative to either: a) another bound XRI (which means a URI or IRI), or b) another unbound XRI. The reason this is important is that when the relative XRI is relative to a bound XRI, the URI/IRI rules of relative reference resolution apply, whereas when a relative XRI is relative to an unbound XRI, the spec calls for direct concatenation of the two. This is necessary because XRIs have subsegment structure that URI/IRI architecture does not.

 

* The Conformance section. Two conformance targets were designated: XRI Parsers and XRI Libraries. The purpose of the latter was to specify a broader set of conformance requirements around XRI transformations, comparisons, and relative reference resolution.

4) NEXT CALL

 

The next call is next week at the regular time.