Skip to end of metadata
Go to start of metadata

Date

17 October 2017

Location

https://global.gotomeeting.com/join/240843765

Attendees

Apologies

Goals

  • Record the meeting (smile)
  • Discuss ONF Interim Meeting
  • Determine if there are any Liaisons
  • Discuss MEF feedback (Andrea)
  • Walk through the TAPI FRD (Lyndon)
  • Further discuss TAPI for CORD Build event (Nigel)
    • Science Fair
    • Meetings
  • Discuss the Ethernet Spec (Andrea)

Discussion Items

Link to recording to be added shortly.

Time
Item
Who
Notes
5minReview AgendaNigel DavisOK
 ONF InterimNigel Davis

Worked through the actions:

  • Highlighted the interim agenda
    • Requested assistance in generation of the TAPI agenda formation
  • Briefly discussed CORD Build agenda (http://cordbuild.org/schedule)
    • Requested assistance in defining the demonstration of TAPI running
 LiaisonsNigel Davis
  • No liaisons identified
 MEF FeedbackAndrea Mazzini
  • We need to get the comments in MEF fed back to ONF via liaison
  • Andrea summarised some of the concerns highlighted
    • One of the challenges is purpose of the model
      • ONF is strongly interface focused
        • Neither the core nor TAPI provide a model that must be implemented in a DB
    • There are concerns with terminology such as pruning and refactoring
    • Usage of specification approach is being questioned
  • Nigel asked whether the CORD build event P&R example would help
    • Andrea emphasised that priority 1 is getting the spec approach demonstrated across the MEF-ONF boundary
      • Tapi structure is priority 2
  • Andrea emphasised that the challenge with TAPI is it is more than just information, it is a formal runable definition
    • We need to close the loop on the formal part of MEF modeling

From previous meeting:

MEF concerns

  • Pruning and Refactoring
    • Concern over the terms, so NRM will be considered as being derived
  • Spec approach
    • Only adding at the moment
  • Recognized pace of approach
  • Some conceptual aspects still open
  • Does not block progressing to letter ballot
  • MEF Core model should provide a common ground for all models in MEF and hence there needs to be a MEF approach to ONF modeling environment
    • How to approach the federating of models?
  • We need to maintain sufficient consistency to prevent us from diverging
  • Perhaps MEF should provide a formal input on the specific issues such that ONF can respond
  • Distinctly different usages of UML cause part of the issue
  • We need to identify which aspects of the modeling approach are oriented towards the interface implementation (client-server)
  • The MEF core model has a different purpose to the interface model
  • Navigability discussion
    • We need to unpick the rationale
    • There are a number of separable concerns that are tangled up with navigability
    • Consider the purpose of the model
 Ethernet specAndrea MazziniMEF NRP SDK to be publicly announcement very soon - It is based on TAPI.
 TAPI FRDLyndon Ong
  • Lyndon noted that the changes are highlighted in section 12 of the FRD
  • Italo noted that the multi-domain example needs to be 3 domains not two (onf2017.041 https://www.dropbox.com/sh/lmra1b06iwps77l/AABbqMS5cHJVWicLAJyQPJ6Aa/onf2017.041.04.pptx?dl=0)
    • Italo suggetsed that 041 should be public
    • Nigel
  • Need a section on OAM. Nigel suggested that Andrea could help. Andrea agreed.
  • Lyndon requested that the team review the document by 24 Oct 2017
  • Bernd noted that terms for projects etc need to be updated
  • Lyndon
 CORD BuildNigel
  • TAPI demo... screen capture?
  • Nigel noted that for the booths it was intended that
    • The Core and TAPI models would be as posters on the booths
      • The models can be discussed on a laptop or main screen (one screen per booth)
    • The P&R tooling will be demonstrated on a laptop or main screen
    • There will be an animation of the spec approach
    • On the TAPI booth there should be a demonstration of UML-Yang
    • Ideally there would be a demo of the TAPI interface but some screen capture or video etc with a code sample would be sufficient.
  • Lyndon noted that there may be recordings from the OIF that may be usable
    • Lyndon will also ask if Karthik if there are any recordings of running TAPI that could be used for the CORD build event

Action Items (Previous)

Action Items (New)

  • All: Consider what needs to be covered in the TAPI portion of the interim face to face meeting
  • All: Propose approach to demonstrating TAPI at CORD Build event
  • Nigel Davis: Validate with David that we do have two booths at the CORD Build Science Fair 
  • All: Review the FRD by 24 Oct 2017
  • To-be-removed: add slot to TAPI calls to cover FRD.

1 Comment

  1. I have updated the FRD chart with the different projects and removed any remaining "OTWG"s.