Re: [Netmod-ver-dt] [netmod] NETMOD 106 Schedule and Call for Presentations

["Rob Wilton (rwilton)" <[e-mail]>]

We might like to ask you (the chairs) to reconsider the discussion as to whether we can do a call for adoption on some of these drafts?

I know that the chairs have previously wanted to adopt the drafts as an entire set to cover the whole solution.  And we are obviously still missing the draft of schema comparison tooling.  Arguably this draft isn't required for the solution, only strongly desirable to have because it is only intending to address requirement 2.2 (which has always been a SHOULD rather than a MUST):

       2.2  A mechanism SHOULD be defined to determine whether data
            nodes between two arbitrary YANG module revisions have (i)
            not changed, (ii) changed in a backwards-compatible way,
            (iii) changed in a non-backwards-compatible way.

Note, there is no intention of dropping this last draft, and we already have some pyang tooling/scripts that perform this task, but I think that it could also realistically follow on a bit after the other drafts.


All the "MUST" requirements are intended to be addressed by the set of drafts below.


Personally, I think that these 4 drafts are all in good enough shape that they could be considered for an WG adoption call:
  draft-verdt-netmod-yang-solutions-02
  draft-verdt-netmod-yang-module-versioning-01
  draft-verdt-netmod-yang-semver-01
  draft-rwilton-netmod-yang-packages-02

I think that this draft is also well on its way to be ready for WG adoption:
  draft-wilton-netmod-yang-ver-selection-01

  - IMO, the key bit that we still need to figure out for this draft is exactly how packages are sensibly put together (e.g. for optional feature sets and bugfixes) in a way that doesn't defeat one of the main benefits of packages for clients (i.e. being able to know the exact device schema offline without having to download/check the entirely of YANG library on connection).  But equally I see no reason why that couldn't be figured out after the document as been adopted as a WG item.


Finally, doing an adoption call on these drafts may act as a forcing function to get the WG to discuss/agree whether this solution is on the right path.

Of course, it is your decision.

Thanks,
Rob



> -----Original Message-----
> From: Lou Berger <[e-mail]>
> Sent: 05 November 2019 15:11
> To: Rob Wilton (rwilton) <[e-mail]>; NetMod WG Chairs <netmod-
> [e-mail]>
> Cc: [e-mail]
> Subject: Re: [netmod] NETMOD 106 Schedule and Call for Presentations
> 
> You're the headliner -- so should have the time ;-)
> 
> On 11/4/2019 4:49 AM, Rob Wilton (rwilton) wrote:
> > 1) DT Update overview & status update (10 mins, Rob?):
> > YANG Module Versioning Requirements
> > draft-ietf-netmod-yang-versioning-reqs-01
> >
> > YANG Versioning Solution Overview
> > draft-verdt-netmod-yang-solutions-02
> >
> > 2) Updated YANG Module Revision Handling (10 mins, tbd?):
> > draft-verdt-netmod-yang-module-versioning-01
> >
> > 3) YANG Semantic Versioning (10 mins, Joe?)
> > draft-verdt-netmod-yang-semver-01
> >
> > 4) YANG Packages (25 mins, Rob)
> > draft-rwilton-netmod-yang-packages-02
> >
> > 5) YANG Schema Version Selection (25 mins, tbd?)
> > draft-wilton-netmod-yang-ver-selection-01