Minutes interim-2017-6tisch-07: Fri 07:00
minutes-interim-2017-6tisch-07-201703170700-00

## DRAFT ##

# Minutes, 17 March 2017 interim, 6TiSCH WG #

Note: timestamps in PDT.

Connection details
------------------

* Date: 7-8am Pacific:
http://www.worldtimebuddy.com/?qm=1&lid=100,12,5392171,1850147&h=100&date=2017-03-17&sln=14-15
* Webex link:
https://cisco.webex.com/ciscosales/j.php?MTID=mcdbbe3a4e38d97d986b507ec12a1f9b1
* Webex Recording:
https://cisco.webex.com/ciscosales/lsr.php?RCID=83818f25e86144849f121e60e49bd090
    * Recording password: SveKaRy7
* Material link:
https://bitbucket.org/6tisch/meetings/raw/master/170317_webex/slides_170317_webex.ppt

Taking notes _(using Etherpad)_
-------------------------------

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

Present _(alphabetically)_
--------------------------

1. Tengfei Chang
1. Pascal Thubert
1. Michael Richardson
1. Diego Dujovne
1. Xavi Vilajosana
1. Sedat Gormus

Action Items
------------

1. Michael to send feedback on security piece of the agenda

Agenda
------

* Administrivia [2min]
    * Approval agenda
    * Approval minutes last call
* Status of drafts [Pascal] [5min]
* Update on security [Michael] [15min]
* Update on 6P [Xavi] [10min]
* Update on SF0 [Diego] [10min]
* Preparation for IETF 98[Pascal] [QS]
* AOB [3min]

Minutes
-------

started 10:07.

* Administrivia [2min]
    * Approval agenda
    * Approval minutes last call

Note well. Reminder.

* Status of drafts [Pascal] [5min]

    * 6tisch-minimal-21 status, email sent to Tero to confirm that he is happy.
    No reply yet (Tero AFK). * no other hangups in status yet. Expect draft to
    advance during the meeting.

* Update on security [Michael] [15min]

    * Major change is that there are more documents:
        - created another version of EDHOC-OSCOAP called EALS
        - tries to do secure transport
        - zero touch security within the EDHOC protocol
        - question is how the WG will articulate the JRC centraly driven.
        - Devices do not need to stay awake the whole time. They need to listen
        for traffic for them. - Shared schedule is used for everybody to join
        (e.g minimal). - having indication on how long it should wait the JN. -
        trade-off between energy consumption and energy congestion during the
        joining phase. This is related on how long a JN has to wait to be
        authenticated. - GOAL: figure out the compromise between energy and
        congestion.

* Update on 6P [Xavi] [10min]
    * finalize the missing pieces: reordering of the text, remove duplicated
    text. * refine section 4, define first the headers, and then the details. *
    renamed status to count now that we can allocate ny type of cell * removed
    xcell suggestion that was complex * structure now ready. plan for last call
    at the meeting * will do a pass and on Monday of the IETF will publish and
    ask for feed back

* Update on SF0 [Diego] [10min]
  * some changes to summarize the changes since last IETF.
  * Cell Estimation Algorithm, three alternatives.
  * picked alternative 3, most accurate to what people have implemented.
  * Xavi:  what about churn because the application has a cycle different than
  the slot frame. * Xavi: what if 3 time slots needed every 2 slot frames, or
  if I have bursts, many needed in one slotframe and none at the next. Need
  mechanism to avoid continuous re-scheduling * SF0 has a hystersis so that it
  does not deallocate until the need is below a *lower* threshold. * Diego:
  Added PDR computation. Need to discuss Timeout calculation, requires 6P
  collaboration.

   * Requires 6P collaboration.
   *
* Preparation for IETF 98[Pascal] [QS]

    * security questions:
    * do we need centralized control over enrollment to avoid congestion in the
    network? * can we give a (secured?) hint to nodes waiting to be contacted
    as to how long to sleep? * do we want to use EALS to do vouchers within the
    key agreement protocol? (round trips vs size of packets) * CoMI for
    management of rekeying * Do we need K1 and K2? Or just one key set? * open
    issues with pledge initiated version * * 6P/SF0 ask for Last call. * * *

* AOB [3min]

    * No OB.