Minutes for NETMOD at interim-2015-netmod-3
minutes-interim-2015-netmod-3-1

Meeting Minutes Network Modeling (netmod) WG
Title Minutes for NETMOD at interim-2015-netmod-3
State Active
Other versions plain text
Last updated 2015-02-26

Meeting Minutes
minutes-interim-2015-netmod-3

=============================================================
NETCONF Data Modeling Language WG (netmod)
12th YANG 1.1 Virtual Interim
Wednesday, February 4th, 2015, 16:00-17:00 CET
Minutes Juergen Schoenwaelder
=============================================================

* Participants:
    
  JS = Juergen Schoenwaelder
  LL = Lada Lhotka
  BL = Balazs Lengyel
  MB = Martin Bjoerklund
  IB = Ignas Bagdonas
  KW = Kent Watsen
  SH = Susan Hares
    
* Status:
    
  Charter: Mar 2015 - Submit YANG 1.1 to the IESG

  IETF 92: March 22 - 27, 2015
  IETF 92 I-D cutoff: 2015-03-09
    
  Virtual meetings: 2015-02-04 (today)
                    2015-02-18 (focus on conformance)
                    2015-03-04
                    2015-03-18
    
  - VRFY :Y09: introduce optional keys <>
    -> pending action JS (verify moving to dead on mailing list>)

  - VRFY :Y16: module advertisement optimization <>
    -> pending action JS (verify issue resolution on mailing list)

  - VRFY :Y34: remove/deprecate/replace the 'anyxml' statement
    -> tons of emails due to its interaction with JSON encodings

  - OPEN :Y18: fix "when" expression context problem
    -> pending action MB done, ready to discuss again

  - OPEN :Y25: make enum numbering purely informative and optional
    -> pending verify action JS, new position Andy Bierman

  - OPEN :Y26: allow mandatory nodes in augment
    -> we need concrete proposals how to relax the current rules

  - OPEN :Y45: better conformance mechanism <>
    -> we need to schedule a meeting with just this topic, could be
       2015-02-18; MB has written up an I-D, there is AB's I-D

  - OPEN :Y57: non-unique leaf-list
    -> ready to be discussed?

  - OPEN :Y58: associate actions with a data node
    -> JS to verify with NETCONF chairs that NETCONF is doing a NACM
       revision, then verify again on mailing list

  MB: We proposed to add a new issue (1.0 and 1.1 coexistence)
  JS: Yes, but it is not yet added to the issues list.

* Y34 remove/deprecate/replace the 'anyxml' statement

  JS: Solution Y34-05
	
      Same as Y34-02 plus two guidelines adopted from Y34-04:
	
      - For YANG 1.0 backward compatibility, allow anyxml to be used,
        except implementations MAY restrict the XML; document that
        anyxml is not considered interoperable.
      - The guidelines document should say that YANG Doctors will review
        each use of anyxml in IETF modules when YANG 1.1 is adopted to
        avoid its use whenever possible
	
  Proposal: try to run Y34-05 through the verify phase

  We will move the JSON document based on YANG 1.0 forward, the plan
  is to revise the JSON document once YANG 1.1 is done. LL will check
  whether a new version is needed before we go to WG last call.
	
* Y18 fix "when" expression context problem

  LL: I did send a question to the list:
      http://mailarchive.ietf.org/arch/msg/netmod/haP-w-KYM0QkxzYtUBeA2E3BNE0
  MB: It may be desirable to have the same algorithm, regardless
      whether the node exists or not.
  MB: I think my solution works even in cases where you have
      leaf-lists.
  LL: But it might be possible to construct cases where it does not
      work. The issue is what we do in the case with multiple
      instances. The must statement does not have this problem.
  MB: Perhaps we need a warning that certain expressions may not be
      meaningful or interoperable.
  JS: Is it possible fill out "if the when expression contains (xxx),
      then the interpretation is implementation specific"?.
  
  Action: LL and MB to work out a proposal for this corner case

* Y57 non-unique leaf-list

  BL: I will revise the proposal and post an updated solution to the
      mailing list.
	
* Y58 associate actions with a data node

  JS to get final confirmation from NETCONF chairs, then try to close
  the VRFY phase on the mailing list.