Re: [earlywarning] (no subject)

[Art Botterell <[e-mail]>]

On Mar 25, 2010, at 3/25/10 6:52 PM, Winterbottom, James wrote:
> That is a nice function.
> If the device is using both mechanisms though, because it has both access technologies won't it still receive both messages, even if it only reports the first one to the user of the device?

That frequently happens in mesh systems like (on its better days) the U.S. broadcast Emergency Alert System (EAS), especially in areas where it's well integrated with the federal Weather Radio network.  One can think of it as a long-interval form of multipath.  

Which is arguably a good thing from a reliability point of view.  Perhaps a bit inefficient, but then, efficiency is frequently the enemy of resilience.

- Art



>> -----Original Message-----
>> From: [e-mail] [mailto:[e-mail]]
>> On Behalf Of Art Botterell
>> Sent: Friday, 26 March 2010 12:50 PM
>> To: [e-mail]
>> Subject: Re: [earlywarning] (no subject)
>> 
>> On Mar 25, 2010, at 3/25/10 6:40 PM, Winterbottom, James wrote:
>>> ...it is quite possible that alerts could be received by either or both
>> means at the same time.
>> 
>> FWIW that problem already arises in a variety of transport domains, so CAP
>> addresses it at the application level.  Alerts have unique identifiers for
>> that purpose, and also to allow the mapping of subsequent update and
>> cancellation messages.
>> 
>> - Art
>>