Skip to main content

Minutes interim-2020-cbor-06: Wed 17:00
minutes-interim-2020-cbor-06-202004081700-00

Meeting Minutes Concise Binary Object Representation Maintenance and Extensions (cbor) WG
Title Minutes interim-2020-cbor-06: Wed 17:00
State Active
Other versions plain text
Last updated 2020-04-14

minutes-interim-2020-cbor-06-202004081700-00
CBOR WG Meeting - Interim 20-06 (IETF107)
Wednesday, April 8, 2020, 17:00 - 18:00 CEST
Chairs: Francesca Palombini, Jim Schaad

Recordings: https://www.youtube.com/watch?v=AZNVw91gZp0&feature=youtu.be

* Introduction [5'] : Chairs
  Agenda bashing and WG status update

WG documents:

* CBOR specification status : Carsten
  https://tools.ietf.org/html/draft-ietf-cbor-7049bis
  Issue 177:  Do we want more control over the 1+2 allocation space for the DE?
   1+2 space is FCFS Cartsen suggests moving the 1+2 space to DE and leave the
  1+4 space alone FP:  Need to look at the list to see why the discussion was
  setup that way. CB:  Could potentially must do a portion of the range to
  Expert review leave the rest alone FP:  ACTION ITEM: Review the list for
  history
       Move the discussion to the list for a short review
  ID: Also would like part of 1+2 range to be expert review.

  Issue 176:  Really an implementer consideration.
  MR: Is this on the wire or only internal
  CB: THis is only internal, but what happens if it gets on the wire is an issue
  MR: If on the wire - then silently ignore - it would be appropriate to ignore
  and detecition of leakage would be fast CB: Can be an issue if different
  implementations treat differently.  Could be done separate from this
  document. FP: Why deal with it here? CB: It would be more shown to people
  reading the document if here. JS: Suggests that the no tag field might be
  used for the purpose of saying field not present - previously brought up
  issue JS: ACTION ITEM: Make sure to respond to this on the mailing list when
  Carsten sends the mail FP:  Summary - good support on the call for this to be
  added.  Need more discussion on how it is used.  Bring to mailing list.

  Issue 178: Security considerations on hashes
  CB: Should add text to say that other strategies exist to address the issue. 
  Suggests point to JAVA method to solve the problem.

  FP: Close to done.  Queries CB about reaching out to implementers
  CB: I have, but I can push out the net a bit farther.   Starts getting a be
  on the spam side in some venues.
     Suggest waiting a bit longer before closing.
   Submit new version in 10 days - give time to disscuss.
  FP:  Ask for more reviews or ??
  CB: Not really a good time to ask for attention due to external factors.  By
  end of month we should make the decision on shipping. FP: Started shepherd
  review, but should be able to ship baring problems end of month.

Other:
* Date and time tag document: Mike, Carsten
  https://tools.ietf.org/html/draft-jones-cbor-date-tag-01
  https://tools.ietf.org/html/draft-bormann-cbor-time-tag-03

  Mike Jones not present so CB takes a stab at this.
  CB: Need to be able to define just a day rather than a daytime which is time
  zone relative. CB: Argued againist 1+1 for daytime, but for the epoch version
  this is a short value. CB: If merge, is there a single tag for both?  Sense
  of list was to do the same thing as tags 0/1 HB: Fine with using distinct
  tags - but confused by the term epoch based. CB: Don't have to give it a name
  - CB: Related items include time of day and time zone CB: From email - Mike
  Jones will have an update draft published this week FP: Last interim agreed
  to merge the documents. --  some confusion over which pair of documents was
  being merged, Carstens or ??? (Jerge?) -- Jörg and Mike's proposal will be
  merged in Mike's document (adding Jörg as author), Carsten's document will
  still be separate.

  CB: Looking at the CB draft, which adds more extensions to the current tag 1.
  Already registered. Can continue working on this but it is pretty ready. Do
  we want to have a common strategey here HB: Reguarly gets question of how to
  create a timestamp.  Global usable interoperable is not easy. CB: Gotten
  queries on time scales, not necessarily for percision. CB: Could do a single
  document on some more useful tags to prevent draft proliferation. FP: Thinks
  multiple documents might be easier BL: General preference to consolidate but
  wants to review discussions before offering an opinion. CB: Proposes doing an
  every 5 year document where new tags can get registered and a stable document
  exists. CB: Lots of contributers, but not necessarily a large number of
  authors HB: Agrees that consolidation makes life easier.  Having it as a
  living document is fine as long as the process works.

* CDDL freezer [if time allows] : Carsten
  https://tools.ietf.org/html/draft-bormann-cbor-cddl-freezer-03
  CB: Please read section 4 before the next meeting for discussion at next
  meeting.

* Wrap-up [5'] : Chairs

*******************************************************************
Bluesheets
*******************************************************************

webex: 12 people
jabber: 7 people

Francesca Palombini, Ericsson
Michael Richardson, Sandelman Software Works
Jim Schaad, August Cellars
Alexey Melnikov, Isode Ltd
Peter Yee, AKAYLA
Michael Breuer, ilSF
Ira McDonald, High North
Carsten Bormann, TZI
Barry Leiba, FutureWei
Max Nansen, ilSF
Marco Tiloca, RISE
Klaus Hartke, Ericsson
Tim Cappalli, Microsoft
Henk Birkholz