LWIG @ IETF 87 Berlin (Wednesday 31 July 2013, 13:00 Potsdam 3) 1300-1315 Introduction, Agenda Bashing and Document Status Status Terminology draft: finished WGLC and send to IESG draft-ietf-lwig-ikev2-minimal-00, ready for WGLC? Volunteers asked for review of document, 3 hands raised. Volunteers: Hannes T., Robert C., Xuan H. Milestones: Hannes Tschofenig: argues that the IKEv2 draft is fairly finished and ready for WGLC. Should review milestones however. Carsten Borrmann: last time it was proposed to slice up the implementation guidance document into separate documents. Hannes: Why isn't this split up already? Conclusion: let's discuss other documents first and then review the milestones. No remarks on agenda. 1315-1325 Zigbee IP Finalized (Robert Cragie) Objective: - Briefly introduce how Zigbee choose and use IETF protocols Questions *Hannes Tschofenig: is the presentation a tutorial about Zigbee IP or will it actually help us with the WG documents? Robert: next slides deal with modifications to ietf protocols where necessary *Hannes: would be nice to get into contact with some of the developers of ZigBee IP. This kind of input could be usefull. Robert: would be happy to send out email to query developers. *Sandeep S. Kumar: security for MCast? Done via L2 key. *How is the global MAC key distributed? Done by PANA as part of the initial bootstrap. 1325-1340 CoAP Implementation Guidance (Matthias Kovatsch) http://tools.ietf.org/html/draft-kovatsch-lwig-coap-01.txt Goal: raise some feedback about usefulness and improvements Questions: *Zhen Cao: is FSM used just to ensure that the implementation behaves correctly or is there any link with energy efficiency? Matthias: just for the first thing. *Carsten: is that ASCII art going to become any more complex? Don't spend too much effort if the result in unreadable. *Matthias: thinks that the machines are complete, but asks the WG to take a look at it if he has missed anything. *Zhen: does this document collide with the CORE Coap document? Zach Shelby: core CoAP document contains alot of tutorials, but doesn't see overlap with lwig-coap *Zhen Cao: shall we do a milestone dicussion for this document? Carsten: removing the CoAP part from the terminology draft was defintely a good idea (?) Hannes: thinks the document is a reasonable starting point for a new milestone 1340-1355 A Hitchhiker's Guide to the (Datagram) Transport Layer Security Protocol (Sandeep S. Kumar) http://tools.ietf.org/html/draft-tschofenig-lwig-tls-minimal-03.txt Goal: get feedback on the existing material Remarks from Hannes: looked at papers about security for constrained devices, papers often don't provide enough detail to include their findingd in the document. Harder to provide data over energy consumption. Matthias: what's mostly missing in academic papers: how do these scenarios look like? Would be usefull for the academics to benchmark their implementations. *???: scenarios would be usefull. Wonders about the scope of the document, what do you want to tell people with the document (and how is this different from Dice) Sandeep: Should contain guidance for others on how to implement DTLS on constrained devices. *Carsten: later we can draw the line between LWIG and DICE. generic TLS lib is as big as everything else, what is in there? Why is it so big? Hannes: can upload source code somewhere. *Carsten:What can we do for a future version of TLS that wouldn't require so much code? Hannes: Cross-compilation and no HW support add to increased size of library. *Matthias: are both the client and generic library necessary for a client? Yes *Richard Kelsy: code size numbers are always suspicous. Most crypto code is highly optimized for speed. Reducing speed can usually give flash gains. Sandeep: want to know the relative costs in terms of size for the different components (no absolute values). Richard: general libraries can be much bigger than what you actually have to do. e.g. x509 certificates code was very big. Hannes: this approach might not be optimal, but it is the best way that I know how to do it. Welcomes suggestions. Sandeep ends with stating that this document is complementary to the work in DiCE? WG chair asks how many people think that WG should adopt this document? +- 12 hands WG chair asks how many people think that WG should not adopt this document? No hands 1355-1410 Energy Efficient Implementation of IETF Protocols on Constrained Devices (Zhen Cao) http://tools.ietf.org/html/draft-hex-lwig-energy-efficient-01.txt Goal: Document update and discussion Chair asks show of hands for document adoption. about 10 supports, none objection Objections? Carsten: unsure if the document has found its final shape yet. Matthias: would like to see differentiation between RDC and sleeping nodes in this document. Hui Tian: would like to contribute some test results. 1410-1425 Link Layer Implication of LWIG (Yuanchen Ma) http://tools.ietf.org/html/draft-ma-lwig-3gpplink-imply-00.txt Goal: call for feedback Questions: *Zhen: How many attendees have experience with 3GPP two hands raised *Zhen: only CoAP client needed, not server? 1425-1435 http://tools.ietf.org/html/draft-hong-lwig-sleepynode-problem-statement-00 (Yong-Geun Hong) Goal: Discuss the use case of sleepy nodes and problem statement Questions/remarks: *Matthias Kovatsch: overlap with energy-efficiency document. Consider RDC interms of sleepy nodes? *Zhen: considers to also look at the sleepy nodes document from doc. *Robert Cragie: thinks RDC nodes and sleepy nodes are two very different cases and it would be worth distinguishing between them. 1435-1440 6Lo BOF announcement/preview (5 min) Objective: - briefly motivate the BOF - explain how this relates to, and differs from, the LWIG work on 6LoWPAN etc. implementation guidance Carsten Bormann /Ralph Droms 1440-1500 DICE BOF announcement/preview (10 min) Objective: - briefly motivate the BOF - explain how this relates to, and differs from, the LWIG work on DTLS implementation guidanc - discuss, agree on a clean division of work Zach Shelby