Re: AW: [Geopriv] Quickrandomcommentsondraft-ietf-geopriv-pdif-l o -profile-00

See below, marked with [PTT]

Brian Rosen wrote:
> Two thingo:
> 
> There are many uses for location information.  Some of them require 
> calculations, and some of them require human interaction.  For the 
> latter, you always need a representation a human can deal with. 
> Specifically, ifthe floor is labeled some way, you need to know that. 
> The example is that you are given a location that is floor 22. You go in 
> an elevator and press 22.  If you arrive, all the markings say floor 22.
> 
> Now, if your pidf has floor 22, I think you want that to be the same 
> floor, even if the building does not have a floor 13, which is often the 
> case.
> 
> What you do with a geo with mu=floors?
> 
[PTT] Good point -- this needs to be specified.
> 
> The other issue is getting both a civic and a geo in a pidf.  If you 
> have both, you have converted one from the other. It must be possible to 
> determine which was the original, and which was derived from it.
> 
> If you route, specifically for emergency calls,you must route on the 
> original.
> 
[PTT] The assumption of conversiuon has been implicit in a number of 
your postings.  I don't think it's necessarily true.  DHCP (to take an 
example) could have information from different sources.  Of course, the 
problem then is to decide which source is more reliable if they disagree.

> Brian
> 
> _______________________________________________
> Geopriv mailing list
> Geopriv@ietf.org
> https://www1.ietf.org/mailman/listinfo/geopriv
> 
> 


_______________________________________________
Geopriv mailing list
Geopriv@ietf.org
https://www1.ietf.org/mailman/listinfo/geopriv