Re: [Simple] Updated RPID

[Henning Schulzrinne <hgs at cs.columbia.edu>]

As Jonathan noted, alignment efforts are under way. However, the issue 
you raise is somewhat more general, namely distinguishing between the 
source of information (which element derived this information) and the 
scope (all the elements that are affected by it). This becomes relevant 
if different devices may arrive at conflicting information. For example, 
my cell phone service may know that I'm driving (since it can detect 
movement based on my cell coordinates), while my PDA might consult its 
calendar and determine that I'm in a meeting (which I'm actually late 
for...).

We agreed, I believe, that in many cases, only the watcher can be 
sufficiently intelligent to deal with conflicting information and thus, 
there need to be facilities for each element to report such information, 
so that the watcher can determine whom to believe or just be confused. 
In some cases, a composer can merge the data and remove contradictions, 
but since the RPID format is also used for publishing, we need to be 
able to carry this information in the element that it belongs to.

Thus, activities can be reported by a (service) tuple, a person or a 
device, depending on the particular circumstances.


Dag Ekengren wrote:
> I think the relationship between the latest RPID document and the presence
> data model document is a little unclear.
>
> In RPID, it says that the "<activities> element describes what the
> presentity is currently doing, expressed as an enumeration of <activity>
> elements". I think this clearly belongs to the <person> element in the
> presence data model. In the examples in the RPID draft, the activities
> element is applied to service tuples. Maybe it should be mentioned that
> these should be used on the <person> element?
>
> Wouldn't it be a good idea to include the definitions of <person> and
> <device> in the RPID draft?

_______________________________________________
Simple mailing list
Simple at ietf.org
https://www1.ietf.org/mailman/listinfo/simple