Re: [Roll] WG call to adopt draft-brandt-roll-rpl-applicability-home-building-03

[Don Sturek <[e-mail]>]

Hi Michael,

Having implemented and certified 6 vendor solutions using ROLL RPL in
ZigBee IP, I doubt very seriously that this is true:

"......because these are questions that need to be
answered if we are going to have interoperability between muyltiple
vendors without resorting to yet another level of vendor forums)"

It would be best to keep the scope of the applicability statement to just
RPL P2P, not add assumptions on the use of outside RFCs/drafts and let
commercial groups complete certification by choosing configuration
parameters that address their needs.

I would doubt we are going to get to a place where an ad hoc group of
vendors are going to pick up 6LoWPAN, 6LoWPAN-ND, ROLL RPL, ROLL RPL P2P,
ROLL Mr. HOF, etc. ....  ,   implement all these and find their solutions
are all interoperable.   Just looking at all the configuration options in
these various drafts, it is unlikely for interoperability to just happen
(ignoring for a minute the security bootstrapping topic which has its own
set of complications)

Don





On 2/21/13 10:16 AM, "Michael Richardson" <[e-mail]> wrote:

>
>>>>>> "Ralph" == Ralph Droms <[e-mail]> writes:
>    >> Therefore, in order to use P2P RPL, there must first have been an
>    >> RPL that provided stable addresses to all nodes.  (I wrote
>    >> "rooted", intending to mean grounded, but probably that's too
>    >> much)
>
>    Ralph> Why is RPL needed for address assignment?
>
>On a number of media,  I guess that it isn't.
>On the media where classic RA/ND or DHCPv6 works, you don't need it.
>On those media, you might not need P2P RPL either.
>
>On some 6LowPAN mesh under systems, we have an RA/ND that does work.
>
>As far as I can tell, none of these are the situation contemplated by
>rpl-applicability-home-building.
>
>(And this is why I want our applicability statements to be very very
>narrow in their scope, because these are questions that need to be
>answered if we are going to have interoperability between muyltiple
>vendors without resorting to yet another level of vendor forums)
>
>-- 
>Michael Richardson <[e-mail]>, Sandelman Software Works
>IETF ROLL WG co-chair.    http://datatracker.ietf.org/wg/roll/charter/
>
>_______________________________________________
>Roll mailing list
>[e-mail]
>https://www.ietf.org/mailman/listinfo/roll