[6tisch] Minutes, 26 February 2016 interim, 6TiSCH WG

["Pascal Thubert (pthubert)" <[e-mail]>]

Note: timestamps in PST.

Connection details

  *   Date: 7-8AM Pacific: http://www.worldtimebuddy.com/?qm=1&lid=100,12,5392171,1850147&h=100&date=2016-02-26&sln=15-16
  *   Webex recording: https://cisco.webex.com/ciscosales/lsr.php?RCID=e4a9c0366aa64ee5b2009dc2e462bed0
  *   Meeting Slides: https://bitbucket.org/6tisch/meetings/src/d11560128ee11658756b78fd9746c45b0d0918f5/160226_webex/slides_160226_webex.ppt?at=master

Taking notes (using Etherpad)

  1.  Xavi Vilajosana
  2.  Michael Richardson
  3.  Pascal Thubert

Present (alphabetically)

  1.  Daniel Nepomuceno
  2.  Diego Dujovne
  3.  Dominique Barthel
  4.  Jonathan Munoz
  5.  Lavania H M
  6.  Maria Rita Palattella
  7.  Michael Richardson
  8.  Pascal Thubert
  9.  Qin Wang
  10. Remy Leone
  11. S.V.R. Anand
  12. Sedat Gormus
  13. Tengfei Chang
  14. Thomas Watteyne
  15. Xavi Vilajosana
  16. Zhuo Chen

Action Items

  1.  6top-sublayer Authors to rename 6top-sublayer to 6top-protocol.
  2.  6top-sublayer Authors to rename the field container to metadata.

Agenda

  *   Administrivia [2min]
     *   Approval agenda
     *   Approval minutes last call
  *   Next steps for 6P and SF0 [40min]
  *   draft-thubert-6lo-inner-compression [10min]
  *   6LoRH to last call? [5min]
  *   AOB [1min]

Minutes

  *   [07.05] Meeting starts

Recording starts

*        Administrivia [2min]

o   Approval agenda

        *   Meeting started, agenda bashing
        *   Michael suggests to add a discussion on ethertype at 6lo during the 6LoRH section

o   Approval minutes last call: Agenda and last call minutes have been approved.

o

*        IETF 95: chairs asked for 90 minutes;

     *   preliminary agenda by 4 of March
     *   Monday March 21st is draft cut off, Including 00
     *   New charter was posted for external review
     *

*        charter posted Friday 26th Feb. Online to be checked.

     *   Milestones presented
     *   submission the minimal draft. In AD evaluation state.
     *

*        Next steps for 6P and SF0 [40min]

o   Diego presenting

        *   details of the Evolution of 6p and SF0 drafts
        *   started with the OTF draft and 6top sublayer draft
        *   6top sublayer evolved to 6P
        *   OTF evolved to the SF0
        *   6top sublayer monitors the cells and move if they are underperforming.
        *   the monitoring process is described in the SF
        *   Pascal suggests to rename draft 6tisch sublayer to 6top protocol.

ACTION rename it to 6top-protocol. Authors agree. Do it before adoption.

o   6P open issues:

        *   parent preference, child proposes slots that cannot be used (black list), the parent selects and child acks
        *   use a data packet to piggy back the IE with the confirmation ?
        *   details on why this approach. The idea is that the parent select. Not clear yet if this is the best approach. Discussion about low latency approaches where the child really wants one specific cells.
        *   Qin Wang: using the container as the base to determine the cell or the slot. We need a mechanism to reference a cell in a chunk to the global schedule structure.

?  consider to redefine the name of the container to metadata so the SF can use it for any purpose, e.g indicating the container or indicating who has the responsability to select the cells (parent or child)

ACTION: to rename the field container to metadata.

?  Conclusion: take 2 bits from the NumCells field

        *   Discussion about the need for a token during the transactions. THe idea is to identify a request and a response with a token. Need to understand if there is the need for that item.
        *   Sedat Gormus: indicates that from an implementation perspective this is useful and needed.
        *

*        draft-thubert-6lo-inner-compression [10min]

     *   New draft, problem to solve is that in IP in IP we cannot use the outer header to compress the inner header if some addresses are taken from the same prefix.
     *   Enables stateless address compression for the inner header.
     *   MR: why do not use a context to compress that inner header?
     *   PT: you do not need to use RAs to distribute the context.
     *   PT: you avoid distributing the same prefix twice.
     *   MR: why we should do it in that way and not using a RPL context.
     *   PT: this requires a discussion, let us take it to the mailing list
     *

*        6LoRH to last call? [5min]

     *   PT: updated it after the PlugTest.
     *   PT: almost done. Check if people agree and ask for Last Call at 6lo.
     *   TW: Makes sense to ask for presenting at Buenos Aires and then ask for last call there.
     *   TW: Michael asked to discuss Ethertype
     *   PT: Michael just left, this is top of the hour. This is a discussion at 6lo
     *   TW: Ethertype to be done in the next call?
     *   PT: Best is to read the 6lo thread and discuss there
     *

*        AOB [1min]

     *   no other B.

*        [08.06] Meeting ends