Re: [yang-doctors] Looking for YANG doctors'helping for a Mboned draft

[Juergen Schoenwaelder <[e-mail]>]

It seems the authors do§ what Benoit would call a service model. The
text clearly needs work. For example:

   Network operators can input this model to a controller who is
   responsible to translate the information and invoke the corresponding
   protocol models into configurations to configure the network elements
   through NETCONF/RESTCONF/CLI.

What does 'input the model' mean? I note that the document says MBONED
and then lists [e-mail] as contact. In YANG, we use revision
statements for published models, not for posted I-Ds. And yes, I agree
with Martin that definitions of the type

    typedef si {
        type uint16;
        description
          "The type for set identifier";
    }

or somewhat opaque (no way to tell whether uint16 is an adequate
choice).

What I generally do not understand about these service models is how
we deal with the fact that the mapping to the underlying device models
is entirely undefined. But this is a model content issue and hence
something to look into by the subject matter experts.

And yes, this is a data model and not an information model as far as I
can tell.

/js

On Thu, Nov 30, 2017 at 08:51:40AM +0100, Martin Bjorklund wrote:
> <[e-mail]> wrote:
> > Hi Martin, Acee,
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > Thank you for your response. 
> > 
> > 
> > The suggestion from Singapore Mboned WG meeting is the draft needs YANG doctors' review. So I sent this email.
> > 
> > 
> > In this draft, we also provide a executable YANG model except UML. 
> 
> It seems to me from a quick glance that your document - despite its
> title - does *not* define an information model, but in fact a data
> model in YANG.  Then you have a "UML-like" picture to help people
> understand the model.  If this is correct I suggest you change the
> title and the text that uses the term "information model".
> 
> I don't have the domain expertise to have any opinions about the
> model itself.  From a pure YANG pow, I think the model would be more
> clear if you collapsed most of the groupings, since it seems that they
> are not defined for external re-use.
> 
> Maybe this is obvious if you have the domain knowledge, but a
> description like this seems a bit terse:
> 
>         leaf source-wildcard {
>             type uint8;
>             description "The wildcard information of source.";
>         }
> 
> (just an example)
> 
> 
> 
> /martin
> 
> 
> > Though this YANG model has been used in our project in ODL, we think it still needs YANG doctors' suggestion.
> > 
> > 
> > Appreciate for your helping. :-)
> > 
> > 
> > 
> > 
> > 
> > 
> > Thanks,
> > 
> > 
> > Sandy
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > 
> > 原始邮件
> > 
> > 
> > 
> > 发件人： <[e-mail]>;
> > 收件人： <[e-mail]>;
> > 抄送人：张征00007940; <[e-mail]>; <[e-mail]>; <[e-mail]>; <[e-mail]>; <[e-mail]>;
> > 日 期 ：2017年11月29日 21:52
> > 主 题 ：Re: [yang-doctors]   Looking for YANG doctors'helping for a Mboned draft
> > 
> > 
> > Hi,
> > 
> > "Acee Lindem (acee)" <[e-mail]> wrote:
> > > Sandy,
> > > Early reviews aren’t done until, at a minimum, the document is adopted
> > > by a WG and the early review is requested by the chair. We don’t have
> > > the bandwidth for individual requests.
> > 
> > Actually, the YANG doctor's page says that we do (may) review
> > individual documents, but they are listed as:
> > 
> >    3.  non-WG document: best effort (really)
> > 
> > 
> > So it's not surprising if people ask for reviews.
> > 
> > In any case, I think such reviews are not done "offically", but rather
> > by individual YANG doctors, if they feel they have the time.
> > 
> > > Also, I didn’t think we were doing more information models in the
> > > IETF.
> > 
> > Is this a documented policy?
> > 
> > In any case, *if* you're doing an information model, I don't think you
> > should use YANG for that.  Use UML or text or something else.  Then if
> > you want to do a *data model* that is supposed to be implemented in a
> > server (at some level), write a YANG model.
> > 
> > 
> > /martin
> > 
> > 
> > > 
> > > Thanks,
> > > Acee
> > > 
> > > From: yang-doctors
> > > <[e-mail]<mailto:[e-mail]>>
> > > on behalf of "[e-mail]<mailto:[e-mail]>"
> > > <[e-mail]<mailto:[e-mail]>>
> > > Date: Wednesday, November 29, 2017 at 2:36 AM
> > > To: YANG Doctors <[e-mail]<mailto:[e-mail]>>
> > > Cc: "[e-mail]<mailto:[e-mail]>"
> > > <[e-mail]<mailto:[e-mail]>>,
> > > "[e-mail]<mailto:[e-mail]>"
> > > <[e-mail]<mailto:[e-mail]>>,
> > > "[e-mail]<mailto:[e-mail]>"
> > > <[e-mail]<mailto:[e-mail]>>,
> > > "[e-mail]<mailto:[e-mail]>"
> > > <[e-mail]<mailto:[e-mail]>>
> > > Subject: [yang-doctors] Looking for YANG doctors' helping for a Mboned
> > > draft
> > > 
> > > 
> > > Dear YANG doctors,
> > > 
> > > 
> > > Could you please help us to do early review for
> > > "draft-zhang-mboned-multicast-info-model-02" in Mboned working group?
> > > 
> > > And could you please give us advice about if it is better that using
> > > mount points in this model?
> > > 
> > > Thank you very much!
> > > 
> > > 
> > > Best Regards,
> > > 
> > > Sandy
> > > 
> > > 
> > > 原始邮件
> > > 发件人： <[e-mail]<mailto:[e-mail]>>;
> > > 收件人： <[e-mail]<mailto:[e-mail]>>;
> > > 日 期 ：2017年11月18日 05:22
> > > 主 题 ：[MBONED] Singapore MBONED meeting mins
> > > 
> > > Draft mins of the MBONED meeting in Singapore have been posted:
> > > 
> > > https://datatracker.ietf.org/meeting/100/materials/minutes-100-mboned/
> > > 
> > > Please take a look and let us know if you have any corrections or
> > > things
> > > to add.
> > > 
> > > _______________________________________________
> > > MBONED mailing list
> > > [e-mail]<mailto:[e-mail]>
> > > https://www.ietf.org/mailman/listinfo/mboned
> > >
> _______________________________________________
> yang-doctors mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/yang-doctors

-- 
Juergen Schoenwaelder           Jacobs University Bremen gGmbH
Phone: [phone]         Campus Ring 1 | 28759 Bremen | Germany
Fax:   [phone]         <http://www.jacobs-university.de/>