CBOR WG Meeting - Interim 20-15
Wednesday, Sept 16, 2020, 17:00 - 18:00 CEST
Chairs: Francesca Palombini, Jim Schaad

Recordings: https://youtu.be/2G4vR3awnT4

Attendees:
1. Francesca Palombini
2. Jim Schaad, August Cellars
3. Carsten Bormann
4. Michael Richardson

Agenda:

* WG documents status
    - Update on CBOR bis
        [CB] Paul and I are usng github to do pull requests - please follow what we are doing there.
        
        - PR #208 (merged) - Changes to psuedocode and related statements (via Ben Kaduk){https://github.com/cbor-wg/CBORbis/pull/208}
        - https://mailarchive.ietf.org/arch/msg/cbor/txIKHMXRFzNo7oH-eHZigO1L47w/
        [CB]  This is a discuss - we don't get a get of jail free card from previous
        Also tag 36 may need addressing
        Getting rid of it means that an orphan registration would exist.  Don't really want to point to obsoleted document for this.
        Could move to notable tags document
        Preference is to get it out of the way.  Just need to shape expectations somehow.
        Even deletion needs some descriptive language.
        [JS] How strongly do they feel that we need a usable reference on this?
        [MR] Understood that if go beyond a specific subset.  Specs may not know what the user's do.
        Point to copy of the webpage for PCE?  Stable reference.
        BIS cannot easily specify common subset - could use the ECHA document but for PCrE stable reference.  Surprised at how bad PCE is defined.
        [CB] ECMA 2018 almost good enough to stop referring to PCRE
        [MR] Maybe just change the reference might be good.  Does it contradict PCRE?
        [CB] Also issue with perl changing so that would be an issue as well.
        Need to specify that UNICODE flag is on in order to make it work.  No flags in ECMA
        [MR] Never used tag 45 so don't really care if there is a change in behavior.
        [FP] Have not seen people complaining about the tag so either it is OK or nobody is using.
        [CB] Suggestion:  don't use tag 35 - if you need a specific version of RE then you should have a new tag for that specific RE lanaguage and version.
        ACTION - write up pull request on this for review.
        
        - https://datatracker.ietf.org/doc/draft-ietf-cbor-7049bis/ballot/

        - NAN discussion
        [CB] Not an IESG comment - need to have some type of decision on this.
        Still having arguments on the specific syntax for this.  For empty string on indefinite it is ambigious if text or byte string.  I suggested using the single/double quote representation to deal with this.
        [JS] Personal opinion - shove it off to a new document. 
        [CB] Need a NAN payload document anyway.  May be an issue if IESG memeber notices that it is ambigious and cares.
        [FP] Ok with Jim's purposal to move to a new document
        
        [CB] Still lots of comments from Ben - for example what is the rounding proposal is on numbers.  Just point to ECMA is a good answer
        Similar issues that will keep use busy, but no problems expected.
        
        [JS] Do we need to addresses Laurence's comment on mime tags?
        [CB] Not sure what would change in the text.

- Update on tags OID
    - draft-ietf-cbor-date-tag in RFC Ed Queue
    [CB] Nothing has happened here.

* Adoption calls
    [FP] Had support during the last interim but very low traffic - no objections came up.
    [FP] Documents adopted and 
    
    - draft-bormann-cbor-cddl-control-02
    - draft-bormann-cbor-packed-01

* AoB
   [CB] Need to write up a proposal for dictionaries.  Brendan also has a proposal.
   Suit WG is not going to be able to use packed at this point.  So no hurry on that point
   [JS] Yes, but CoRAL is blocked on this.
   [FP] Should add to agenda?
   [CB] Should be OK for next meeting.
   
   Next interim agenda:
   - OID tags
   - new proposal dictionaries
   - rFC7049
   
    AP Chairs: request 60 minutes meeting
        add to conflicts: ASDF, JSONPath

*[CB]: Carsten Bormann
*[FP]: Francesca Palombini
*[MR]: Michael Richardson
*[JS]: Jim Schaad