[Geopriv] (draft) meeting minutes of the GEOPRIV working group at IETF 58

From: Andrew Newton ^lt;anewton@ecotroph.net>
Date: Thu Nov 13 2003 - 11:13:58 EST

The following are draft minutes from the meeting.

-andy

Meeting Minutes of the Geographic
Location/Privacy (GEOPRIV) Working Group
-----------------------------------------
Monday, November 10, 2003, 1930-2200

Chairs:
Allison Mankin
Randall Gellens
Andrew Newton

Scribe:
Ashir Ahmed

Published Agenda:
1) Agenda Bashing
2) WG Administrativia
3) LO and GML Issues (Hannes Tschofenig)
4) PIDF-LO (Jon Peterson)
5) POLICY
    Permissions and Rules (Henning Schulzrinne)
    Changes and Open Issues (Hannes Tschofenig)
6) Any Other Business

1) Agenda Bashing
    No agenda modifications requested

2) WG Administrativia
    Two drafts in RFC Editor queue
    One draft under review of IESG
    No additional comments.

3) LO and GML Issues

Presentation given by Hannes Tschofenig

Concerns expressed from the room ranged from size of the documents,
performance issues, and usage details for implementors.

4) PIDF-LO

Presentation by Jon Peterson on draft-peterson-geopriv-pidf-lo

It was asked that the permission flags be separated out of the PIDF-LO
schema for use in other applications. After a minor discussion on the
merits of using a new namespace vs. global scoped elements in the
current schema, the draft author indicated he would create a separate
namespace and schema for the flags.

The issue of the indicating the device source of the presentity was
raised. There was no decision that this was needed.

The room discussed avoidance of interoperability issues by declaring a
MIME type for the location object. It was noted that using protocols
will know the format anyway.

Concern was raised regarding the "must understand" requirement in PIDF.
  This was proceeded by a discussion of XML validation.

The room then discussed civil location in PIDF-LO. The chair noted the
working group's consensus regarding the desire for civil location in
PIDF-LO vs. just POLICY. A question was put to the floor: should
PIDF-LO incorporate the civil location elements from POLICY. The
consensus of the room was positive.

It was noted that the X.509 certificates for S/MIME would require a
subject alternative name of URI type of a pres: URI.

4) POLICY

Presentation by Henning Schzulrinne regarding permissions and rules
philosophy for draft-ietf-geopriv-policy.

An issue was raised by Jonathon Rosenberg, but he and Henning decided to
take the issue to the list for detailed discussion.

There were no other questions for this presentation.

Hannes Tschofenig gave a presentation on changes and open issues in
draft-ietf-geopriv-policy.

The issue of logging was raised. It was noted that the service provider
will be doing logging of the using protocol and that this issue could
not be solved by the working group.

It was mentioned that the current draft did not address identities and
authentication types. One of the co-authors indicated that this would
be placed back in the next version of the draft.

The room discussed notifications and the level of detail and size of
information involved. A question regarding authorization in the using
protocol was asked. Hannes indicated that the authors were working
through those details.

A question was asked of the room: should the POLICY document contain a
section regarding the using profiles. The room consented.

The next issue raised involved the URI's for authentication and
identity. Henning noted that there are two ways to proceed: using an
opaque string or declaring using protocol specific elements. This led
into the discussion about domains and individuals and the convergence
with XCAP. It was noted that XCAP has notion of domain match and that
not all authentication schemes have the notion of "user@domain".

This next led to a discussion on exception lists to rule targets. After
much discussion, the room was asked the following question: are
exceptions NOT needed. The room did not consent. It was then asked of
the room if only additive permissions without exceptions were
acceptable. The room was evenly split and not consensus was found.

Convergence with XCAP and the work of the SIMPLE working group was then
raised again. The desire to avoid divergence was expressed. After much
discussion on how to proceed, the following proposal was presnted on a
method to go forward: if the SIMPLE working group agrees to retain
domain-based authorization but exclude exception-based rules in XCAP,
then the GEOPRIV working group agrees to use XCAP. This proposal noted
that exception-based rules would not be included in version 1 of the
specification, but would be considered in subsequent versions with the
hind-sight of deployment experience. The room consented.

6) Any Other Business

Brian Rosen (and Jon Morris) presented a topic on Emergency 911. He
noted the danger in not specifying rules for crypto with regard to
emergency call routing. He proposed supporting TLS (or IPSEC) as normal
security mechanism for emergency calls. While the room was not formally
asked to indicate approval, many participants favored the idea. The
chairs asked for more discussion of the topic on the working group
mailing list.

_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv
Received on Thu Nov 13 11:15:25 2003

This archive was generated by hypermail 2.1.8 : Thu Jan 22 2004 - 12:32:24 EST