Re: [yang-doctors] Yangdoctors last call review of draft-ietf-rtgwg-yang-rib-extend-06

["Acee Lindem (acee)" <[e-mail]>]

Hi Martin, 
See inline below.

On 4/23/21, 3:19 AM, "Martin Björklund" <[e-mail]> wrote:

    Hi,

    Thank you for addressing my comments.  Pruning to the one remaining
    question.


    Yingzhen Qu <[e-mail]> wrote:
    > Hi Martin,
    > 
    > Thank you for your review, and we've published version -07 to address your
    > comments.
    > 
    > Please see my answers below inline.
    > 
    > Thanks,
    > Yingzhen
    > 
    > On Wed, Apr 14, 2021 at 4:51 AM Martin Björklund via Datatracker <
    > [e-mail]> wrote:
    > 
    > > Reviewer: Martin Björklund
    > > Review result: Ready with Nits
    > >
    > > Here is my YANG doctors review of draft-ietf-rtgwg-yang-rib-extend-06.
    > > This is a well-written draft, and my comments are minor.

    [...]

    > > o  module description
    > >
    > >      This YANG module extends the generic data model for
    > >      RIB by augmenting the ietf-routing model.  It is
    > >      intended that the module will be extended by vendors
    > >      to define vendor-specific RIB parameters.
    > >
    > >   I don't think I understand this description.  Here's my understanding,
    > >   but I don't think it is correct:
    > >
    > >     1. This module extends the existing RIB data model by using
    > >        augmentations.
    > >     2. The existing RIB data model is defined in the YANG module
    > >        ietf-routing.
    > >     3. The purpose of this new module is to allow vendors to extend the
    > >        the existing RIB data model with vendor-specific parameters.
    > >
    > >   It seems 3 is at least incomplete, since this module defines some
    > >   additional config param for static routes, and addtional state and
    > >   statistics for ribs.
    > >
    > >   It is not clear how vendors are expected to extend this model; the
    > >   word "vendor" doesn't show up anywhere else.
    > >
    > > [Yingzhen]: This module does define additional parameters and is
    > augmenting the existing RIB model. The module can be further augmented. Any
    > suggestions for a replacement of "vendor-specific"?

    All models can be augmented so I don't think this needs to be spelled
    out.  When it is spelled out like this I expect some discussion about
    how it differ from the "normal" augment that always can be done.

    Some modules define some generic common structure, but are not
    very useful unless they are augmented; they need to define how vendors
    (or sdos) should extend the module.  However, I think that this module
    is useful on its own, and thus I would remove the sentence "It is
    intended ...".  

I'm not sure the source of this text but it has been almost boiler plate for protocol YANG models. Having said that, I don't have any problem removing it.

Thanks,
Acee


    /martin