Re: [Roll] Working Group Last Call:draft-ietf-roll-protocols-survey-02

Hi Emmanuel,

I will let the authors of the draft answer the excellent comments that  
have been raised on the list during WG LC (and I'll add mine) but I  
would like to comment on your statement and hopefully clarify our  
objective.

On Dec 9, 2008, at 10:02 AM, Emmanuel Baccelli wrote:

> Hi all,
>
> I think this draft is far from being ready to become an RFC, and  
> such for several reasons, some of which are, unfortunately,  
> structural.
>
> 1 - The draft is evaluating moving targets such as OLSRv2, Dymo,  
> NHDP, which are potential solutions that are not finalized yet.  
> These protocols could meet more requirements in the end (maybe all  
> of them?). For this reason, the "pass/fail" table of section 5  
> cannot honestly list anything else but a series of "question  
> marks" (or "pass"), for now, for these protocols. This makes this  
> table rather useless, and thus this whole draft rather useless,  
> presently.

NO. Let me clarify something really important.

The WG cannot afford to wait until all protocols are finalized to get  
a routing solution for LLN. The industry is waiting for a routing  
solution for such networks and time is absolutely critical. No we have  
to analyze the state of the art as of today.

>
> 2 - The ROLL charter mentions that " Existing IGPs, MANET, NEMO, DTN  
> routing protocols will be part of evaluation." I did not see  
> anything mentioned about DTN protocols, or Nemo in the draft, not  
> even a justification why they are not mentioned.
>

Would you want to make a contribution ?

> 3 - The draft is not constructive in its current shape.

I do not think that your comment is fair. The draft has been discussed  
in the WG for quite some time and extensive discussion took place  
during the interim WG and on the ML.

> It is on the other hand extremely controversial with its judgement,  
> and does not reflect a consensus within the community, not even a  
> rough one.
>

Bear with us: we are not *judging* protocols: we are evaluating the  
existing routing protocols *in light* of our specific requirements.  
Yes one can argue for ever on a pass/fail criteria but we have to be  
driven by rough consensus to be able to move forward. Of course, if  
there is a mistake when evaluating a protocol it has to be fixed.

Thanks.

JP.

> Emmanuel
>
>
> On Mon, Dec 8, 2008 at 7:50 PM, Thomas Heide Clausen <ietf@thomasclausen.org 
> > wrote:
>
> On Dec 8, 2008, at 18:58 PM, Dearlove, Christopher (UK) wrote:
>
>
> Your point about hierarchical addressing is interesting, and I think
> it relates to comments others have made on fisheye routing and other
> state suppression approaches.
>
> Not quite. Hierarchical allocation is indeed a state suppression
> approach. However fisheye is a control message suppression approach.
>
>
> And, it should be added that such are, of course complementing each  
> other.
>
> Thomas
>
>
> _______________________________________________
> Roll mailing list
> Roll@ietf.org
> https://www.ietf.org/mailman/listinfo/roll
>
> _______________________________________________
> Roll mailing list
> Roll@ietf.org
> https://www.ietf.org/mailman/listinfo/roll