Re: [earlywarning] Some further scoping issues

["Hannes Tschofenig" <Hannes.Tschofenig at gmx.net>]

Hi all, 

I did not receive a lot of feedback on the important aspect of scoping the
work in this upcoming group. 

Please take a briefly look at it. 

Based on the mailing list discussion in a different thread I would also like
to add: 

* Are we really interested in focusing on a particular security deployment
model? I don't think so. 
For example, alerts that are sent from a particular entity, such as a
government, only. Technically this does not make a lot of sense and is too
far into the deployment choices of certain entities.

* We also do not make specific restrictions regarding the type of alerts
(e.g., alerts about a tornado vs. early school closing) as the semantic
largely comes from the content of the alert -- something we will not define
anyway. There may be some technical requirements that some folks might want
to discuss, such as delivery mechanisms to deal with potential congestion
and so on. That's fine for me. 

* Excluding discussions about the type of device and the available user
interface also seems to make sense to me. This appears to be more issues
with a specific architecture that, as stated below, would not be in our
scope. 
 
Ciao
Hannes 

PS: Another pragmatic way to scope the work is to focus on the things that
have actually been proposed already rather than speculating about things.
From what I can tell there isn't a huge amount of work being proposed in
terms of documents. When new proposals come along then one can make a
judgement again to determine whether this is something the group is
interested in. 

>-----Original Message-----
>From: earlywarning-bounces at ietf.org 
>[mailto:earlywarning-bounces at ietf.org] On Behalf Of Hannes Tschofenig
>Sent: 02 May, 2009 12:50
>To: earlywarning at ietf.org
>Subject: [earlywarning] Some further scoping issues
>
>For the future work in this group I furthermore suggest that 
>
>* we do not focus our efforts on defining an architecture used 
>for distributing of alert messages (like we did it in ECRIT 
>with the PhoneBCP/Framework document). Instead, we develop 
>building blocks (like conveying CAP in SIP). I believe we will 
>run into lots of problems with the different usages, different 
>solutions that have been specified elsewhere already, etc. 
>
>* we do not pick a single alert document format (as a 
>mandatory-to-implement), like CAP or GeoRSS, but instead give 
>those who deploy these mechanisms freedom. 
>
>* we do not define our own alert document format (because 
>there are other formats available already). 
>
>We can always decide later to document ways how people use our stuff. 
>
>Ciao
>Hannes
>
>_______________________________________________
>earlywarning mailing list
>earlywarning at ietf.org
>https://www.ietf.org/mailman/listinfo/earlywarning
>