---------- Minutes - GEOPRIV - IETF82 Summary (prepared by Richard Barnes): 1. Chairs' Introduction Brian Rosen intends to patch loc-filters in AUTH48 to remove conform with edits to the rate control document. The authors of the policy-uri document will update it in response to feedback from the AD. 2. GEOPRIV Policy draft-ietf-geopriv-policy There was agreement in the room that the current "fuzzing" text in this document adequately captures the problem, and that the document is ready for WGLC. 3. Local Civic Addresses draft-ietf-geopriv-local-civic A resolution for the last open issue on this document has been agreed by the authors, and a revised version should be out soon. Roger Marshall agreed to send his editorial comments to the list. 4. Relative Location draft-ietf-geopriv-relative-loc This document is substantially complete, but is waiting on the local-civic draft. Roger Marshall agreed to send his editorial comments to the list. 5. Next Steps The chairs announced their intent to issue WGLCs for all of the WG's current documents shortly after this meeting, and invited discussion about possible additional work items. There were a few suggestions, but none that clearly had enough energy to justify adding them to the WG charter. A couple of minor documents may be adopted. The chairs will work to complete the WGLCs on the remaining documents, at which point the question of what to do with the group will be revisited. Raw notes from the tools.ietf.org etherpad follow ---------- Tuesday, November 15, 2011 GEOPRIV Working Group Session 1520 - 1700 : Afternoon Session II, Room 201DEF Administrivia (Presentation by Chairs) http://www.ietf.org/proceedings/82/slides/geopriv-3.ppt ------------ Brian Rosen: There is a normative reference from location filters to event rate control. But it was an old version of event rate control and so there is now an inconsistency. We will make the change to location filters to conform with rate control as part of AUTH48. The authors, chairs and AD believe that the intent of location filters was always to conform to rate control. Text will be sent to the list for the working group to review prior to being sent to the RFC editor. Authors of the policy-URI draft (Richard Barnes) aggreed to spin a new version of the draft for the Area Director. This will happen ASAP! Geopriv Policy (Presentation by Hannes Tschofenig) http://www.ietf.org/proceedings/82/slides/geopriv-0.pptx ------------ Hannes will put on the web supporting material to aid in explaining the limitations documented in the current draft. This material will be linked in the Geopriv wiki. Additionally, Joe Hildebrand has offered to help put these materials into a useful format. Local Civic (Presentation by Brian Rosen) http://www.ietf.org/proceedings/82/slides/geopriv-1.pptx ------------ The only open issue is a minor change to IANA considerations. Brian will put out an edit to fix this very soon. Roger Marshall will send some editorial comments to the list. Authors request a working group last call. Relative Location (Presentation by Brian Rosen) http://www.ietf.org/proceedings/82/slides/geopriv-2.pptx ------------ This document has been done for a while, but it has a normative dependence on local civic (this document performs registrations in the local civic registry) Roger Marshall will provide editoral comments on the list. (In particular, there is concern that the document may overload the term "reference" unnecessarily.) This document has no open issues. Authors request a working group last call. Geopriv: What Next? (Presentation by the Chairs) ------------ Ray Bellis: The UK emergency services architecture uses Identity extensions for looking up the geolocation for an IP address. However, the current Identity extensions seems to be insufficient to meet this use-case. Ray suggested that it would be useful to update the identity extensions to provide a complete set of TCP flow identifiers. The RIPE NCC is looking at adding Geolocation information to the RIPE (whois) database. Note that this database is information about networks (address blocks) and not individual addresses. Kaveh Ranjbar(from RIPE) presented a brief demonstration of the current RIPE prototype. Terry Manderson: It would be good if your prototype included the option to provide a HELD URI instead of a fixed geolocation. It may be easier to have the HELD server to provide up-to-date information without having to repeatedly update a whois record. Richard Barnes notes that providing indirection to a HELD server also provides a point of geopriv policy control, which has desirable privacy properties. Brian Rosen: One of the things that we are definitely missing is a way to tell that the location that you have been given is likely to change. We need the notion that I have given you this location, but it may change. Brian Rosen: Additionally, we really need a way to explicitly state confidence (and corresponding uncertainty) in the PIDF-LO Brian Rosen: HELD only permits you to repeatedly poll if you want to get an update on a changing location. Perhaps we need some type of push mechanism for location updates in HELD. (Or else, maybe we need to say "Do SIP" if you want to get updates to location over time.) Cullen Jennings: What were our original goals? It's been many years, have we achieved what we set out to do? This would be a good thing to discuss on the list. Jon Peterson: I think we should figure out a path to closing the working group which allows those last oddly shaped pieces of work (such as things Brian mentioned) that need to get done, to get done.