Title Subtitle

1

GEOPRIV IETF-70 Meeting Vancouver, Dec 2007
HTTP Enabled Location Delivery (HELD) draft-ietf-g
eopriv-http-location-delivery-03.txt Editor
Mary Barnes (mary.barnes_at_nortel.com)
Contributors James Winterbottom
(james.winterbottom_at_andrew.com)
Martin Thomson (martin.thomson_at_andrew.com)
Barbara Stark
(barbara.stark_at_bellsouth.com)
2
Contents

• Status
• Changes since last version (s)
• Issues
• Way Forward
• Comments/Questions

3
Status

• 2 revisions since IETF-69
• Good progress in resolving issues
• Current version has been simplified as compared
  to the -01 at IETF-69.

4
HELD changes since version -01

• Updated description and definition for
  responseTime to include type of service
  (emergencyRouting or emergencyDispatch).
• One important note here is that the current
  definition is not easily extensible.
• Removed jurisdicationalCivic and postalCivic from
  the locationType, leaving just civic.
• Simplified the error processing.
• Rather than the error and locationResponse
  sharing the same type, a new errorType has been
  defined, thus there's no longer a code for
  "success" nor does the locationResponse contain a
  code.
• Clarified that locationType is optional with
  "any" being the default - there were some
  inconsistencies between table 1 and sections 5.2
  and 6.2.

5
HELD changes since version -01

• Updated terminology (i.e., using "LIS" rather
  than "LCS") and removed all duplicate
  definitions, which resulted in no new terms
  defined in this document, since they're defined
  and/or commonly used elsewhere.
• Updated the schema and examples based on all the
  other changes.
• Updated the appendix A to reflect the last
  requirements and problem statement document
  (specifically A.1, A.3 and adding A.10).
• Lots of miscellaneous editorial
  changes/clarifications based on detailed WG
  feedback (Hannes, Barbara, Carl and Shida).
• Ended up missing a set of comments from Guy
  Caron.

6
HELD changes since version -02

• Added text to address concern over use of IP
  address as device identifier in VPN scenarios.
• Removed WSDL.
• Added extensibility to baseRequestType in the
  schema (an oversight from previous edits), along
  with fixing some other nits in schema.
• Moved discussion of Location URI from section 5.3
  (Location Response) to where it rightly belonged
  in Section 6.5 (Location URI Parameter).
• Clarified text for "expires" parameter
• it's an optional parm, but required for
  LocationURIs

7
HELD changes since version -02

• Clarified responseTime parameter
• If missing, then the LIS provides most precise
  LI, with the time required being implementation
  specific.
• Clarified that the MUST use TLS when using HTTP
  as a delivery mechanism is a MUST implement.
• Updated references
• removed unused/added new.

8
HELD Minor issues/discussion

• Terminology inconsistencies
• attribute versus parameter versus element
• Propose to use element throughout
• HTTP binding versus HTTP transport versus
  HTTP delivery
• Propose to use HTTP transport
• Editorial
• Change wording that describes basic use of IP
  address as device identifier
• The LIS uses the source IP address of the request
  sent from the Device as the identifier in
  determining the location of the device.
• Section 4 Add subsections Location by-value
  and Location by-reference to match the new
  sections added for devices IDs, NAT and VPNs
• responseTime clarify the type for the
  responseTime in section 6.1.

9
HELD Issues/discussion

• responseTime type
• Currently, type is decimal
• Propose to make this a nonNegativeInteger (in
  milliseconds)?
• Section 4.1.1 Should we add text to address
  devices that move after the VPN is up?
• Error Code extensibility
• Option1 Define an IANA registry for error codes.
• Pros will help to avoid future collisions
• Cons limits extensions to standards track
• Option2 just use basic XML extensibility
  mechanisms
• Pros doesnt require standards track document
  to extend
• Cons collisions have to be caught manually
• Proposal Option 1

10
HELD Issues/discussion

• HELD URI
• Issue discussed in Chicago resolution was it
  was not necessary
• Issue re-surfaced and conclusion is that it is
  necessary.
• Proposal add IANA registration for HELD URI
• Normative document dependencies requiring WG
  adoption/progression
• draft-thomson-geopriv-lis-discovery

11
Way Forward HELD

• Update document
• Editorial changes
• Issue resolution per discussion.
• Doc (-04) should be ready for WGLC after next
  update