[netmod] regarding draft-bierman-netmod-yang-data-ext-00

[Kent Watsen <[e-mail]>]

I'd like to start a discussion about adopting this draft...or something like it (see below).

The primary driver for wanting to expedite this draft is that it is being discussed as a required aspect of a chartered NETCONF WG effort to define a new encoding for YANG's 'notification' statement.

But I wonder if it would be better to define something like yd:uses-yang-data that can have both 'augment' and 'refine' as substatements.  The motivation for this is because the ANIMA WG wishes to define a "voucher-request" yang-data artifact that is essentially a "voucher" yang-data artifact that has had a leaf changed from "mandatory true" to "mandatory false" (via refinement) while also adding some new fields (via augmentation).  The current ANIMA solution defines a common grouping used by two yang-data statements, but this approach is neither intuitive nor lends itself to further downstream consumption.

Lastly, I wonder if it would be better to have a draft that [re-]defines a 'yang-data' statement outside of RFC8040.  This way drafts wanting to define yang-data wouldn't have to explain why they have an otherwise unexpected normative reference to RESTCONF.

Thoughts?

Kent // pick a hat