RE: [Geopriv] Location objects

From: James M. Polk ^lt;>
Date: Sat Feb 07 2004 - 17:16:11 EST

At 01:36 PM 1/27/2004 -0500, Abbott, Nadine B. wrote:
>An excellent suggestion.
>In discussions with the National Emergency Number Association last fall, the
>idea of having the source of the location information identified had strong
>It might be useful (although perhaps not practical, given the limitations of
>the DHCP signaling) to include an identifier/contact information for the
>provider of the source location information, e.g., XYZ Wireless or ABC Map
>Co, or maybe a URI for the same.

This would necessitate an accompanied digital signature from who is
asserting this valid location information - which needs to be checked
somewhere - before believing it. This makes things a bit more complicated -
and I can come up with a list of questions to this idea:

- If this signature fails, is the call completed?
- Is the location trusted?
- Is anyone dispatched before some other location information can be verified?
- How do all the PSAP databases get records of these digital signatures to
match the inbound location organization assertion to?
- How often do they get them (meaning: how often is the local database
- How does this work internationally?
- How does this work with an non-US SIP device (traveler with a laptop)
connected to a US network (maybe a Starbucks or in an airport)?

This is not meant as an attack on the idea, just opening up a lot of
questions that ought to be looked into for this initial request above.

>Nadine Abbott
>-----Original Message-----
>From: [] On Behalf Of
>Henning Schulzrinne
>Sent: Tuesday, January 27, 2004 1:20 PM
>Subject: [Geopriv] Location objects
>In many scenarios, there will be multiple sources of location
>information. I think it would be useful to label location data with an
>optional enumerated (not free text) source identifier, such as
> 802.11 triangulation
> manual configuration
> location beacon
> geocoded (e.g., translated from geo to civil or vice versa)
>There will be odd-ball scenarios that don't fit into any list of
>categories, but I suspect that the vast majority will be drawn from a
>relatively short and IANA-extensible list.
>This allows systems processing this information to make better decisions
>when they get conflicting information, for example. By making the label
>explicit, we can then also make this part of the privacy rules.
>Geopriv mailing list
>Geopriv mailing list


                 Truth is not to be argued... it is to be presented

Geopriv mailing list
Received on Sat, 07 Feb 2004 16:16:11 -0600

This archive was generated by hypermail 2.1.8 : Sat Feb 07 2004 - 17:19:05 EST