Re: [core] Protocol Constants section of coap-08

[Zach Shelby <[e-mail]>]

As it seems we had good consensus on this new text, I have made the following addition to Section 9 (Protocol Constants), replacing the current text about future mechanisms with:

"The values for RESPONSE_TIMEOUT, RESPONSE_RANDOM_FACTOR, and
   MAX_RETRANSMIT may be configured to values specific to the
   application environment, however the configuration method is out of
   scope of this document.  It is recommended that an application
   environment use consistent values for these parameters."

Zach

On Feb 22, 2012, at 7:53 PM, Carsten Bormann wrote:

> On Feb 22, 2012, at 18:44, Thomas Herbst wrote:
> 
>> Suggested Text :
>> The values  for RESPONSE_TIMEOUT,  RESPONSE_RANDOM_FACTOR, and MAX_RETRANSMIT may be configured to network specific values, but the configuration method is out of scope of this document.  It is recommended that a network use consistent values for these parameters.  
> 
> Thanks!
> We probably want to replace "network" by something more general, as the network characteristics are not the only factor in choosing good values -- one also has to look at application requirements for timeliness and delivery probabilities etc.  E.g., some applications' networks will "the Internet", and there won't be a single set of values right for those applications...
> 
> maybe:
> … to values specific on the application environment …
> … that an application environment employ consistent …
> 
> In the long run, I expect many implementations will be able to employ adaptive schemes for configuring the values.  But, as you said, this is out of scope for this document.
> 
> Thanks again for bringing this up; it is important we get this right.
> 
> Grüße, Carsten
> 

-- 
Zach Shelby, Chief Nerd, Sensinode Ltd.
http://www.sensinode.com
http://zachshelby.org  - My blog "On the Internet of Things"
http://6lowpan.net - My book "6LoWPAN: The Wireless Embedded Internet"
Mobile: [phone]