manet@conference.ietf.jabber.com - 2002/11/21
[11:08] %% d has arrived.
[11:14] %% Joseph has arrived.
[11:15] %% starsu has arrived.
[11:15] %% oletroan has arrived.
[11:15] %% starsu has left.
[11:16] %% hanzzz has arrived.
[11:19] %% rafdalb has arrived.
[11:21] <d> no-one doing logs?
[11:23] <Joseph> Ugh, I do some casually
[11:23] * Joseph has changed the subject to: MANET Strategy
[11:23] %% starsu has arrived.
[11:23] <Joseph> First question, Why?
[11:24] <d> alex zinin: very useful work
[11:24] <starsu> wide problem statement
[11:24] <Joseph> Which is a good question btw ;)
[11:24] <d> (cos i was gonna do some if anyone felt the need)
[11:24] <Joseph> please do ;)
[11:25] <d> wide problem statement, competing protocols, lcak of consensus, no signs of convergence
[11:25] <d> == research mode
[11:25] <d> but, this is i*e*tf
[11:25] <d> GOALS
[11:26] <d> convert wg to ietf mode:
[11:26] <d> clear/focused problem statement
[11:26] <d> focused charter, well-defined milestones
[11:26] <d> wg members must work towards consesus
[11:27] <d> final goal: define IETF MANET standard
[11:27] <d> more research-based work must move to IRTF
[11:28] <d> choice of what goes where is subjective
[11:28] <d> IRTF has new life, and has useful stuff going on now (as opposed to a while ago)
[11:28] <d> STEPS
[11:28] <d> create MANET group within IRTF
[11:29] <d> move LANMAR FSR ZRP TORA (suggested)
[11:29] <Joseph> 'bout time MANET moved to the IRTF
[11:29] <d> push DSR AODV OLSR TBRPF to experiemental
[11:30] <d> nothing goes std track
[11:30] <d> the IRTF protocols can also be published as either informational or experimental
[11:31] <d> redefine IETF MANET group wiht *engineering* prob statement
[11:33] <d> if wg agrees on a problem statement, then we can re-charter and develop std-track protocols
[11:33] <d> PLAN of ACTION
[11:33] <d> wg chairs + ads will make the decision
[11:33] <d> discussed here
[11:33] <d> announcement of decision within a month of this mtf
[11:33] <d> s/mtf/meeting/
[11:33] <d> OPEN MIKE
[11:34] <d> james kempf: wants list of specific & engineering uses of tech
[11:35] %% acbell_acbell has arrived.
[11:35] <d> ??: is interface between MANET and internet research or engineering? being done now for simple (stub) uses. transit might be research
[11:37] <d> charles perkins?: any further delay in publishing protocols as experimental? ADs have AODV and OLSR(?). intention is to move on ...
[11:38] <d> pete shorelander?: apps inside DoD networks. not one protocol for all uses. encourage hybrid routing work. updates coming for SF meeting. -> basically, bring it on
[11:39] <d> ??: final DSR draft not submitted yet; not sure re: AODV
[11:40] <d> mention of new drafts status tracking tools
[11:40] <d> on IETF page, then IESG page, I-D tracker ...
[11:41] <d> look at procedings of routing area meeting at yokohama for explanation of publishing process, etc
[11:41] <d> ??: is multicast research or wg?
[11:42] <d> -> some simple aspects could be wg, but lots of it is obviously research
[11:43] <d> ??: [ missed the question details, something about wireless or ??? OSR and path-mtu]
[11:43] <d> -> what impact from IPv6? to list
[11:44] <d> no hum! raise hands if this is reasonable?
[11:44] * Joseph raises hand
[11:44] <d> no votes for this being a bad thing. none for don't care. consensus is that this is sensible
[11:45] <d> [ back to the agenda ]
[11:45] <d> strawman re-charter details ...
[11:45] * Joseph has changed the subject to: strawman re-charter
[11:46] <d> [ big problem statement -- see slides ]
[11:47] <d> revised charter will limit scope and focus on promoting some reasonably mature routing specs to experimental
[11:47] <d> need more ops experience
[11:47] <d> protocols represent a basic coverage of proactive and reactive design spaces
[11:48] <d> phase 2: design and develop common group protocols for standards track
[11:48] <d> strawman reduced scope [ see slides again ]
[11:49] <d> ie. approx 100 nodes, unicast only for a start, possible later work on config/mgmt, flooding or multicast
[11:51] <d> bunch of completed milestones, present work to restructure and push some work off to IRTF.
[11:51] <d> JAN 2003 progress reactive protocols
[11:51] <d> and re-charter done
[11:51] <d> MAR 2003 progress proactive protocols (to exp)
[11:52] <d> JUN 2003 focused wg problem statement and scoped engineering goals
[11:53] <d> next presentation: IRTF subgroup
[11:53] <d> scalability is theme
[11:53] <d> routing
[11:54] * Joseph has changed the subject to: IRTF MANET Research Subgroup
[11:54] <d> understand limits of existing schemes; propose/develop better schemes
[11:54] <d> security ...
[11:55] <d> encouragement to look at big picture and novel approaches in IRTF forum.
[11:55] %% jm has arrived.
[11:56] <d> proposal to meet with IETF maybe once a year, and instead meet at conferences, etc (infocom, ...). mailing list, of course.
[11:56] %% timo has arrived.
[11:56] %% jm has left.
[11:56] <d> goal is to have good understand of space before bouncing back to IETF
[11:57] %% jm has arrived.
[11:57] <d> ??: multicast? where does it live?
[11:57] <d> -> possibly two kinds of multicast? one simple (in ietf) and one more complex in IRTF?
[11:58] <d> revised charter proposed rough limits of 100 nodes, but no limit on scalability of mobility.
[11:58] <d> -> other scalability metrics include traffic rates, etc
[12:00] <d> charlie: scalability -- every other IETF protocol is designed to scale, would hate to see design eliminate scalability as a goal. flooding works well for small numbers, so we could define the problem away entirely. would prefer to see initial scope limited, but not entire charter. eg disaster scenarios where scalable manets are needed.
[12:01] <d> -> experiments have been done with 100000's of nodes. OSPF (etc) doesn't really scale indefinitely. point is taken, but need is to keep to engineering.
[12:02] <d> charlie: wants charter to explicitly allow future extension/scale.
[12:03] <d> -> 10k node manets are possible, but might not be immediate engineering. IRTF work would look at extending limits.
[12:04] <d> charlie: 2nd point. IRTF proposal will only succeed if handled carefully. would like to see a close relationship between IETF and IRTF groups (more so than is common). would like to see meetings at all IETF meetings (i think).
[12:05] <d> -> close coordination is required. IETF can ask hard questions of IRTF.
[12:05] <d> charlie: prefer not to limit to one co-located meeting per year.
[12:05] <d> -> logistics to be arranged with secretariat (sp?)
[12:06] <d> ??: specifying large scalability goals in charter is slippery slope. doesn't mean we cannot scale up, but need to make goal achievable.
[12:07] <d> ??: second point: where does security belong?
[12:08] <d> -> routing security group not thinking about wireless; need to coordinate. possible slippery slope (eg. scalable, dynamic key servers is separate research problem).
[12:09] <d> ??: have to solve security for routing protocol. make simplicity of protocol a goal (to limit creature feep). simplicity also helps security.
[12:09] <d> -> right, has to be done. what is the minimum feature set we care about.
[12:09] <d> ??: not route spoofing
[12:09] <d> -> basic authentication
[12:11] <d> another ??: DoD would like RFCs for 50-100 node networks within 12 months and funding, etc.
[12:11] <d> -> ok
[12:11] <d> ??: more security stuff
[12:11] <d> ??: more security
[12:11] <d> -> no more security please ... back to agenda
[12:11] <d> [ starting draft presentations now ... ]
[12:11] <d> [ and now Jabber is dead ... ]
[12:12] <d> [ ok, maybe it's back? ]
[12:13] * Joseph has changed the subject to: TBRPF
[12:14] <d> these slides are text heavy, so i'm gonna limit myself to comments from people
[12:17] <d> some discussion of relay priority
[12:17] <d> allow nodes to decline to route, but still participate in discovery, etc. allows less transient nodes, etc.
[12:20] %% _ruffi_ has arrived.
[12:22] <d> [mic] timing of neighbour discovery and hello messages to be talked about offline.
[12:23] <d> -> take it to the list, please
[12:23] <d> -> clean up, no new features, goodness.
[12:23] * Joseph has changed the subject to: DSR
[12:23] <d> [ david johnson -- DSR ]
[12:24] <d> mobicom CFP due early march
[12:24] <d> [ ouch ]
[12:29] %% jm has left.
[12:30] <d> david is done -- no comments ...
[12:30] <d> OSR next ...
[12:30] * Joseph has changed the subject to: OSR
[12:31] * Joseph has changed the subject to: OLSR
[12:32] %% timo has left.
[12:35] * d has changed the subject to: AODV
[12:36] <d> there were no questions or comments for OLSR.
[12:40] <d> wg-chairs: request that some of the AODV implementors list forward wg-relevant traffic to the manet list
[12:43] <d> no questions. thanks to all.
[12:43] %% _ruffi_ has left.
[12:44] %% hanzzz has left.
[12:44] * d has changed the subject to: [ finished ]
[12:45] %% starsu has left.
[12:45] %% d has left.
[12:45] %% Joseph has left.
[12:46] %% acbell_acbell has left.
[13:02] %% rafdalb has left.
[13:33] %% oletroan has left.
[13:43] %% Joseph has arrived.
[13:43] %% Joseph has left.