IETF TICTOC Working Group Meeting Monday, March 23, 2009 San Francisco Hilton, Continental 3, 1740-1900 Co-chairs: Stewart Bryant and Yaakov Stein The jabber log can be found at: http://jabber.ietf.org/logs/tictoc/2009-03-23.txt The meeting was called to order by co-chairs Yaakov Stein and Stewart Bryant. Karen O'Donoghue took the minutes. There was no volunteer to act as jabber scribe; however, Stewart Bryant and Karen O'Donoghue monitored the room and provided comments and questions from remote attendees. There were significant audio issues that impeded the efforts of the remote individuals to participate. The blue sheets were distributed, and the agenda was presented with no additions. Status Update, Yaakov Stein ===================== Slides: tictoc-1.ppt Yaakov led a discussion on the status of the current working group work items and documents. Requirements Draft, Silvana Rodrigues ============================= Draft: draft-rodrigues-lindqvist-tictoc-req-02 Slides: tictoc-0.ppt Silvana Rodrigues provided an update on the requirements draft. The current draft was updated based on the last meeting, various email exchanges, and the December conference call. Remote telco and power application areas were removed based on the December conference call. There are still application areas with missing requirements including Industrial Automation, Networking, Legal Time, Metrology and Sensors. Yaakov will coordinate with Silvana to complete the Sensors and Legal Time applications, Stewart will provide input on networking. Silvana is working with a colleague from the IEEE 1588 community to get Industrial Automation requirements. Yaakov will send a message soliciting requirement to the IETF general discussion mailing list. The next version of this document will be released as a working group document. Architecture Draft, Karen O'Donoghue ============================ Draft: draft-stein-tictoc-modules-03 Slides: tictoc-2.ppt Karen O'Donoghue provided a brief update on the status of the architecture document. Very limited progress has been made on the document to date. It is still in the process of transitioning from the modules document to an architecture document. It was expected that work would proceed on this document over conference calls, but these were never scheduled. There was some confusion over the use of the term application classes, especially in light of the requirements document. The intention in this document is general categories of clock applications. Thus, application here refers to the type of clock versus the application that utilizes the clock. This will be clarified in future versions. Stewart asked about the utility of using conferences calls to help progress the work. Karen stated that she thought regularly scheduled conference calls could help gain traction in the working group. It was agreed that the WG would start having one hour conference calls the first and third Thursday of each month in an attempt to progress the work a little faster. Stewart agreed to provide the facility for the calls. NTP Document Status, Karen O'Donoghue =============================== Documents: draft-ietf-ntp-ntpv4-proto-11 draft-ietf-ntp-ntpv4-mib-05, draft-ietf-ntp-autokey-04, draft-ietf-ntp-dhcpv6-ntp-opt-03 Slides: tictoc-3 Karen O'Donoghue went over the basic steps for document publication once a document leaves the WG. The NTPv4 protocol specification and MIB have both completed IETF last call and are in IESG evaluation. Comments received on the protocol specification include questions on IANA registries needed, normative references, and security considerations including relationship to the autokey document. The issue of how to address security considerations was discussed at some length with no change in direction resulting. Several minor clarifications and corrections were received on the MIB. Responses to the comments and editing instructions need to be prepared for both these documents. The autokey document has gotten delayed in the AD Evaluation state, and we need to engage the AD to see what we need to do to progress this document. Yaakov and Stewart clarified that an Informational document does not need to go through the IETF Last Call process. Finally, the NTP server option for DHCPv6 document has been updated based on comments from Danny Mayer. It is ready for a WG Last Call to be issued. Report on SG15/Q13 interim meeting, Jean Loup Ferrant ========================================= Slides: tictoc-4 Jean Loup Ferrant provided an overview of the results of the SG15/Q13 interim meeting held the previous week. There were a large number of contributions for the meeting, and it was a very productive meeting. Synchronous Ethernet was consented in February and will be published shortly. For packet based timing, they plan to address transport of frequency followed by the transport of time. The details of the progress are provided in the slides. Yaakov thanked Jean Loup for the detailed report with such limited time to prepare. Classless Timing Distribution, Yaakov Stein ================================ Slides: tictoc-5 Yaakov Stein provided an overview of some of his ongoing work on classless timing distribution. Yaakov believes that this algorithm provides significant benefits over current NTP and IEEE 1588 approaches. Details are provided in the slides and in the IEEE Transactions paper referenced in the slides. Wrap-up ====== Yaakov put up a list of items and asked who planned to work on which topics. The concern is that the working group seems to be slow getting traction on the technical efforts. Greg asked if we were putting the cart before the horse by choosing from this list now. The approach was to have a problem statement, requirements, and an architecture document. From this we were to generate a gap analysis and determine what needed to be done. We don't want to prejudge what the results of the gap analysis should be. There does need to be additional discussions to clarify the relationship between the ITU and IETF efforts. Greg Dowd pointed out that the areas that he thinks are the key areas in which the IETF can contribute are security, provisioning, and redundancy. There was some further discussion on the meaning of redundancy in this context and how it relates to the protection work being done by the ITU. Yaakov asked if the general consensus was to put all these topics on hold while we finish the requirements and architecture documents. Karen suggested that individuals should begin developing drafts for discussion. An internet draft in the IETF sense is analogous to a contribution in the ITU. Not all drafts are destined to become RFCs. The conclusion of the discussion was that the working group should proceed with putting individual drafts on the table for discussion because this will help clarify the scope of our work. Should the working group end up with too many drafts, they can be prioritized at that time. The meeting was adjourned.