Re: [Simple] What are tuples?

[Jonathan Rosenberg <jdrosen at dynamicsoft.com>]

inline.

Henning Schulzrinne wrote:


> > The think the watcher benefits because it provides what I think is 
> > useful information for the watcher to make a choice. For example, "oh, 
> > i know bob is travelling and has his cell phone with him, look at 
> > that, he has sms on his phone, i will send him an sms".
>
>
> No different than a service.

What is no different than a service?

> > I also think that people tend to think about communications in terms 
> > of devices. People say, "call me on my cell", such that the device is 
> > the center point for choice and preference.
>
>
> People buy cell phone service to be reached. The device is a necessary, 
> but not particularly interesting, implementation detail. This will 
> become more obvious as we start separating numbers from individual 
> pieces of plastic.

I think we want solutions which model the concepts of interest that 
exist today. I don't see how one can deny that a "cell phone" doesnt 
exist or is meaningless as a concept in communications systems?

> > Also keep in mind that watchers are not the only consumers of presence 
> > documents. I think that the device information is incredibly useful 
> > for presence servers in the process of composition.
>
>
> I disagree.

Henning, I have given several examples of how the device identity is an 
extremely useful correlator. Can you please explain why you disagree 
with each of those cases? Its hard to rebut against "I disagree".

> > 1. if a device is idle, like my PC, it means I have not accessed the 
> > device or any services on it for some time. This is the classic 
> > meaning of idle in todays IM systems
>
>
> Not really true. If I watch a video on my PC or have it cycle through a 
> set of PPT slides, the IM will show "idle", although the PC is doing its 
> job and I'm watching it. In practice, it seems to mean "haven't touched 
> the mouse or keyboard"...

Thats fine. We can define "idle" for a device as not having touched the 
input sources of that device (keyboard, mouse). There is still clearly 
an "idle" for the device and an "idle" for the service that are distinct.

> > 2. if a a service is idle, like my IM client, it means I havent used 
> > the service (i.e., havent sent/receive an IM) for some time
> >
> > 3. if a person is idle, it means that they are bored and not doing 
> > anything
> >
> > Inheritance makes sense only when there is a is-a relationship between 
> > the thing that is inhertiting and the thing that is the source of the 
> > inheritance. Since services and devices and presentities are not the 
> > same thing, I dont think we should be inheriting amongst them.
> >
> > I think that, the fewer places an attribute can appear (being 
> > associated with a service, device or tuple), the less confusion there 
> > is about what that attribute means. When it does appear in multiple 
> > places, it should be because it is truly meaningful for a devices, 
> > presentities and services. For example, location is meaningful for a 
> > device and for a presentity.
>
>
> It is also useful for a human, for example (which is not the same as the 
> presentity). Is the "human:" URI a service or a device?

In the presence data model I have suggested, the "presentity" models the 
human, and so location is a characteristic of the presentity.

-Jonathan R.


--
Jonathan D. Rosenberg, Ph.D.                600 Lanidex Plaza
Chief Technology Officer                    Parsippany, NJ 07054-2711
dynamicsoft
jdrosen at dynamicsoft.com                     FAX:   (973) 952-5050
http://www.jdrosen.net                      PHONE: (973) 952-5000
http://www.dynamicsoft.com

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