RE: [Geopriv] Re: Review of draft-linsner-geopriv-relativeloc-01.txt

From: Thomson, Martin ^lt;>
Date: Wed Nov 28 2007 - 16:39:47 EST

I have a few major concerns with this draft.

The first is that the relative offset fields are part of a civic address. These are not civic fields - it seems that the civic format is only used because it is convenient. Convenience is not a good technical reason to use those fields.

The second is similar to what others have expressed. A relative location applies to both civic and geodetic locations. This follows from the first - the offset should be specified separately.

In addition, there is no formal way to describe a specific point using a civic address. A civic address describes a region of uncertainty at all levels, except where informal fields like "LOC" are used. This makes establishing a reference point difficult. i.e. Given a building name, which point within the extents of the building is the actual location. Keep in mind that informal fields are extremely difficult to interpret programmatically.

Finally, consideration needs to be made to how relative location can be controlled with policy. In particular, the obfuscation method requires extra thought - obscuring the reference location could invalidate the location entirely, but only obscuring the offset could still reveal too much information via the reference location (for instance, if the reference was an AP location, this reveals the means by which the location was derived and places certain bounds on the result).

My minor considerations are more numerous, but these need to be addressed first.

I have some thoughts on how to resolve these problems using geodetic forms. However, I think that we should focus on PIDF-LO before updating the DHCP formats.


This message is for the designated recipient only and may
contain privileged, proprietary, or otherwise private information.
If you have received it in error, please notify the sender
immediately and delete the original. Any unauthorized use of
this email is prohibited.

Geopriv mailing list
Received on Wed, 28 Nov 2007 15:39:47 -0600

This archive was generated by hypermail 2.1.8 : Wed Nov 28 2007 - 16:39:58 EST