Re: [Roll] Review and concerns

["David E. Culler" <[e-mail]>]

That's correct.  It says"- Survey the applicability of existing protocols to LLNs. The aim of
this document will be to analyze the scaling and characteristics of
existing protocols and identify whether or not they meet the routing
requirements of the applications identified above. Existing IGPs, MANET,
NEMO, DTN routing protocols will be part of evaluation."

My understanding is that the authors have done that review and have reached out in various was to engage in discussion with those groups to identify concrete instances that should be examined in detail.  My question is a concrete one.  In the Nemo IDs is there a protocol that should have been addressed in the survey in the manner of the ones that are there?  I didn't see a specific candidate.  Did they miss it?  Which?  Where?


Emmanuel Baccelli wrote:

On Thu, Dec 11, 2008 at 9:52 PM, David E. Culler <[e-mail]> wrote:

Emmanuel,
   We discussed the list of rows in each of the IETF meetings and associated discussions leading up to those.   It sounds like you agree that if you have some criteria for limiting the search space the discussion will never complete.   That is necessary, according to the charter, to consider the space of solutions that might actually solve the problem.  To do any analysis there has to be something specific to analyze.  You mention two addition WGs that have done relevant work.  Indeed, from the very beginning it was stated that Roll should work in conjunction with those groups.  It is key another thing to say "this specific protocol as described by this RFC or I-D ought to be analyzed in draft-ietf-roll-protocols-survey-03". 

 Here's the NEMO list.  Which are you arguing should be included and why?

Sorry to be blunt, but your charter says that DTN and NEMO are to be reviewed as well as the rest. There is not a word about either of these "fields" in the draft, why? You need to answer this question. Not me.

Emmanuel

 

http://tools.ietf.org/wg/nemo/draft-ietf-nemo-basic-support/" target="_blank" rel="nofollow">draft-ietf-nemo-basic-support	http://tools.ietf.org/html/draft-ietf-nemo-basic-support" target="_blank" rel="nofollow">-03		2004-06-07	http://tools.ietf.org/html/rfc3963" title="" target="_blank" rel="nofollow">RFC 3963
http://tools.ietf.org/wg/nemo/draft-ietf-nemo-home-network-models/" target="_blank" rel="nofollow">draft-ietf-nemo-home-network-models	http://tools.ietf.org/html/draft-ietf-nemo-home-network-models" target="_blank" rel="nofollow">-06		2006-02-21	http://tools.ietf.org/html/rfc4887" title="" target="_blank" rel="nofollow">RFC 4887
http://tools.ietf.org/wg/nemo/draft-ietf-nemo-multihoming-issues/" target="_blank" rel="nofollow">draft-ietf-nemo-multihoming-issues	http://tools.ietf.org/html/draft-ietf-nemo-multihoming-issues" target="_blank" rel="nofollow">-07		2007-02-07	http://tools.ietf.org/html/rfc4980" title="" target="_blank" rel="nofollow">RFC 4980
http://tools.ietf.org/wg/nemo/draft-ietf-nemo-requirements/" target="_blank" rel="nofollow">draft-ietf-nemo-requirements	http://tools.ietf.org/html/draft-ietf-nemo-requirements" target="_blank" rel="nofollow">-06		2006-11-09	http://tools.ietf.org/html/rfc4886" title="" target="_blank" rel="nofollow">RFC 4886
http://tools.ietf.org/wg/nemo/draft-ietf-nemo-ro-problem-statement/" target="_blank" rel="nofollow">draft-ietf-nemo-ro-problem-statement	http://tools.ietf.org/html/draft-ietf-nemo-ro-problem-statement" target="_blank" rel="nofollow">-03		2006-09-18	http://tools.ietf.org/html/rfc4888" title="" target="_blank" rel="nofollow">RFC 4888
http://tools.ietf.org/wg/nemo/draft-ietf-nemo-ro-space-analysis/" target="_blank" rel="nofollow">draft-ietf-nemo-ro-space-analysis	http://tools.ietf.org/html/draft-ietf-nemo-ro-space-analysis" target="_blank" rel="nofollow">-03		2006-09-18	http://tools.ietf.org/html/rfc4889" title="" target="_blank" rel="nofollow">RFC 4889
http://tools.ietf.org/wg/nemo/draft-ietf-nemo-terminology/" target="_blank" rel="nofollow">draft-ietf-nemo-terminology	http://tools.ietf.org/html/draft-ietf-nemo-terminology" target="_blank" rel="nofollow">-06		2006-11-09	http://tools.ietf.org/html/rfc4885" title="" target="_blank" rel="nofollow">RFC 4885

There are many topics that as a working group we'd like to move on to once this process completes.  Much important work has been done outside IETF WGs.  Some of it in the IRTF.  Others in other places.  All the more reason to complete this well formed, bounded effort.   Again, I do not hear any suggestion that any protocol within the domain of analysis meets the criteria.  I do hear that there are protocols or modifications to protocols that do or could.  We need to complete the process on this draft so that we can move on to consider such enhancements, candidates, or new designs. 

On Thu, Dec 11, 2008 at 4:09 PM, Emmanuel Baccelli <mailto:emmanuel.baccelli%20at%20gmail.com" target="_blank">emmanuel.baccelli at gmail.com> wrote:

On Tue, Dec 9, 2008 at 11:28 PM, David E. Culler <mailto:culler%20at%20eecs.berkeley.edu" target="_blank">culler at eecs.berkeley.edu> wrote:

Some concerns recognize that there are protocols outside the defined universe that should be considered and that appear to be far better suited to meeting the criteria than the set of IETF protocols that were considered - since those protocols were developed without these criteria.  Again, that suggests that the criteria are a reasonable guide, that they do not appear to be unobtainable, and that the analysis has provided a suitable guide.

None of the comments seem to be claiming that there is a solution within the universe that we were chartered to consider that meets the criteria.

 

Even admitting that the universe you defined is wide enough (a point which is actually agued since months), we could conclude to this only if this universe (defined in the charter) is covered entirely. Again, it is not the case: I have seen nothing said about NEMO or DTN, so reports of investigations in these fields need to be included in the draft anyways.

Second, if we are not going into a long research phase, I am still wondering what existing protocols/algorithms, IP compatible or not, are being "hinted at" here, which could meet the "dream" criterii that are evoked in the draft. Chris Dearlove and other people have asked this question many times, and it is yet to be answered.

Emmanuel

_______________________________________________
Roll mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/roll" target="_blank" rel="nofollow">https://www.ietf.org/mailman/listinfo/roll

_______________________________________________
Roll mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/roll" rel="nofollow">https://www.ietf.org/mailman/listinfo/roll
  

_______________________________________________
Roll mailing list
[e-mail]
https://www.ietf.org/mailman/listinfo/roll