Re: [6lowapp] Device profiles

["Don Sturek" <[e-mail]>]

Hi Jon,

Here is a link to the problem statement for CoAP:
http://www.ietf.org/id/draft-shelby-6lowapp-coap-00.txt

I think most of the points you asked about are to be addressed.

Don


-----Original Message-----
From: Jon Smirl [mailto:[e-mail]] 
Sent: Tuesday, December 29, 2009 6:47 PM
To: [e-mail]
Cc: [e-mail]
Subject: Re: [6lowapp] Device profiles

On Tue, Dec 29, 2009 at 9:29 PM, Don Sturek <[e-mail]> wrote:
> Hi Jon,
>
> Right now, the CoAP envisions an attribute protocol that follows REST
(GET,
> SET, PUT and DELETE).   The structures and attribute content formats are
not
> addressed (and probably would not be).  It would then be up to individual
> application developers to define these structures.

This approach works fine for me too.  It would make sense to define
standard names for the http parameters and responses in order to
achieve interoperability.

How are you going to integrate a constrained node that only xmits
periodically like a temperature sensors? UDP with a cache and a TTL?
The cache would need to respond to http requests.

What about service discovery?

>
> In this regard, I would assume that CoAP would not end up looking like
> either existing ZigBee device profiles or have the application specifics
of
> the ZigBee Cluster Library.

Is the idea to provide a parallel solution minus the the Zigbee
licensing baggage?

-- 
Jon Smirl
[e-mail]