[net2-wg] future direction for net2

[Martin Turon]

Hi Om,
 
I think considering a ZigBee implementation is still a good idea.  Only the ZigBee 2004 spec is open right now, but the ZigBee Pro 2007 specification is going to be released soon -- perhaps as early as April this year.
 
Martin

________________________________

From: net2-wg-bounces at millennium.berkeley.edu on behalf of Philip Levis
Sent: Sun 2/18/2007 7:13 PM
To: Omprakash Gnawali
Cc: net2-wg at millennium.berkeley.edu
Subject: Re: [net2-wg] future direction for net2



On Feb 16, 2007, at 7:03 PM, Omprakash Gnawali wrote:

>
> Dear net2 members,
>
> As you know, the net2 agenda for the last few months has been
> dominated by work related to the Collection Tree Protocol. Now that
> CTP design, specification, and implementation is nearing its
> completion, we need to decide two things:
>
> 1. Topics net2 should pursue.
>
> In the past we have had proposals on Zigbee compliant stack and
> specification of a better data link layer. I am sure there are many
> other items you would like net2 to address. This is a good time to
> bring all those on the table, prioritize the tasks, and allocate the
> volunteers.

After SenSys, I think it would be good to revisit dissemination. 
Kaisen's been doing some good work on increasing the efficiency of 
being able to disseminate many items. Getting the JHU people involved 
(they've been doing the network side of Deluge) would also be good. 
The only part left in the middle then is dissemination of medium-
sized objects that are more than one but fewer than ten or so packets 
in length. It might be that the Deluge spray+NACK approach works well 
enough, but figuring this out would be great.

That doesn't mean this should be the only thing we look at. 
Generally, it's good to have 2-3 things going on so the agenda can be 
active even if one topic is stuck on a tricky problem.

>
> 2. Organization of the group.
>
> Our work on collection has not only dominated the agenda of net2 but
> also made it difficult for other efforts to get any traction. As a
> result people uninterested in collection were not able to
> contribute. Moving forward, I propose:
>
> - Recruiting researchers who publish papers on network protocols and
> help them publish user-ready protocols in TinyOS. This would mean net2
> would evolve into a group of people experienced in networking that is
> ready to provide encouragement and feedback to external contributors
> to produce more networking options for TinyOS. For example, if someone
> wrote a paper in networking and wants to provide it to other
> researchers, that person would think of net2 as the place to turn to
> to get pointers and feedback. There is no reason why this model would
> not be applicable even if it is software that would eventually go into
> contrib and not lib/net.

This sounds like a great idea to me. Examples I could think of off 
the top of my head are IFRC and S4.


>
> - If we come across a project such as CTP that is long-term and
> involving a lot of people in the future, spin off a sub group or a
> different working group so that rest of the members are able to engage
> in activities that interest them.

I think that makes sense too.

Phil
_______________________________________________
net2-wg mailing list
net2-wg at millennium.berkeley.edu
https://mail.millennium.berkeley.edu/cgi-bin/mailman/listinfo/net2-wg


-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.millennium.berkeley.edu/pipermail/net2-wg/attachments/20070218/3f93451c/attachment.html