Re: [netmod] I-D Action: draft-ietf-netmod-yang-tree-diagrams-02.txt one or many

["t.petch" <[e-mail]>]

----- Original Message -----
From: "Andy Bierman" <[e-mail]>
To: "Lou Berger" <[e-mail]>
Sent: Friday, October 27, 2017 8:56 PM

> On Fri, Oct 27, 2017 at 12:40 PM, Lou Berger <[e-mail]> wrote:
>
> > On 10/27/2017 01:20 PM, Juergen Schoenwaelder wrote:
> > > Why do we come up with such rules in the first place? It really
> > > depends on the modules and their relationship and it is the
> > > responsibility of the WG, the authors, the reviewers to produce a
> > > reasonable document.
> > >
> >
> > My personal view - Because having a few knowledgeable people doesn't
> > scale and not all model writers are at the IETF.
> >
> It is easy to add a guideline that says "a YANG tree diagram MAY be
pruned
> if it is too verbose, in order to improve readability.".
>
> Not so easy to define "too verbose" or define what should be pruned.
> This is where I agree with Juergen. The authors and reviewers have the
> responsibility to make these decisions.

A personal take on the size of tree diagrams, based on experience.

One page, as the I-D suggests, fine.

Over five pages, probably of little or no value.

Two or more pages; take steps to reduce, the steps mentioned in the I-D
already, looking for logically separate parts (as per RFC7317), reducing
tree depth, collapsing groupings (probably with a separate tree for each
grouping, which may in turn need to have these rules applied) ....

I think that the tree diagrams I see from the Routing Area are pretty
dire, on account of their size, and that this WG can see that already
whereas other WGs may take a year or two to cotton on.

I am an engineer - I want something that works!

Tom Petch

> > Lou
> >
>
>
> Andy
>
>
> >
> > > /js
> > >
> > > On Fri, Oct 27, 2017 at 06:08:31PM +0100, t.petch wrote:
> > >> Lou
> > >>
> > >> Suggested text
> > >>
> > >> NEW
> > >>
> > >> 3.3 One Document Several Modules
> > >>
> > >> When a document contains several YANG modules, all the tree
diagrams
> > >> should be placed together, before all the modules.  Each tree
diagram
> > >> should be preceded by a brief introduction to highlight where one
tree
> > >> diagram ends and another starts.
> > >>
> > >> If a document contains a single module which is logically a
number of
> > >> distinct components, the same strategy should be followed;
RFC7317
> > >> provides a good example of this approach.
> > >>
> > >> /NEW
> > >>
> > >> Like Juergen, I am conflicted as to at what point details like
this
> > >> should be part of rfc6087bis; I think a paragraph like this does
belong
> > >> in 'tree-diagrams'.
> > >>
> > >> Tom Petch
> > >>
> > >> ----- Original Message -----
> > >> From: "Lou Berger" <[e-mail]>
> > >> To: "t.petch" <[e-mail]>; <[e-mail]>
> > >> Sent: Friday, October 27, 2017 1:38 PM
> > >>
> > >>> Tom,
> > >>>
> > >>>
> > >>> On 10/27/2017 7:08 AM, t.petch wrote:
> > >>>> Lou
> > >>>>
> > >>>> On a slightly different tack, so a slightly modified Subject:
line,
> > >>>> when an I-D contains multiple modules, some place all the
models
> > >>>> together and then all the modules, e.g.
> > >>>> draft-hares-i2nsf-capability-data-model-04 while others
intersperse
> > >> the
> > >>>> models and the modules, e.g. draft-ietf-lisp-yang-05 .
> > >>>>
> > >>>> I think the former is superior and should be recommended,
especially
> > >>>> when, as Sue has done, there is a brief paragraph of text
before
> > >> each
> > >>>> model, so it is very clear where a model ends and another
begins.
> > >> With
> > >>>> the latter, models can be hard to find.
> > >>>
> > >>>> I see this as dovetailing into Juergen's comments on RFC7317
which,
> > >> to
> > >>>> me, is really several separate modules packaged as one, so the
> > >>>> separation makes excellent sense to a reader.
> > >>>>
> > >>>> Where the I-D is already several modules, then do as RFC7317
has
> > >> done.
> > >>> Sure, Do you have text you'd like to propose?
> > >>>
> > >>>
> > >>>> Tom Petch
> > >>>>
> > >>>> ----- Original Message -----
> > >>>> From: "Lou Berger" <[e-mail]>
> > >>>> To: <[e-mail]>
> > >>>> Sent: Wednesday, October 25, 2017 2:13 PM
> > >>>>
> > >>>>> Hi,
> > >>>>>
> > >>>>> This version addresses all known / open issues in the draft
known
> > >> to
> > >>>>> the authors.
> > >>>>>
> > >>>>> The changes are as follows:
> > >>>>> - Added groupings and yang-data descriptions
> > >>>>> - Added Comments, Long Diagrams and Security Considerations
> > >> sections
> > >>>>> - Clarified representation of schema mount points and
> > >> representation
> > >>>> of
> > >>>>> modules exposed using schema mount.
> > >>>>> - Miscellaneous editorial changes
> > >>>>>
> > >>>>> Lou (for draft authors)
> > >>>>>
> > >>>>> On 10/25/2017 8:49 AM, [e-mail] wrote:
> > >>>>>> A New Internet-Draft is available from the on-line
Internet-Drafts
> > >>>> directories.
> > >>>>>> This draft is a work item of the Network Modeling WG of the
IETF.
> > >>>>>>
> > >>>>>>         Title           : YANG Tree Diagrams
> > >>>>>>         Authors         : Martin Bjorklund
> > >>>>>>                           Lou Berger
> > >>>>>> Filename        : draft-ietf-netmod-yang-tree-diagrams-02.txt
> > >>>>>> Pages           : 11
> > >>
> > >> _______________________________________________
> > >> netmod mailing list
> > >> [e-mail]
> > >> https://www.ietf.org/mailman/listinfo/netmod
> > >
> >
> > _______________________________________________
> > netmod mailing list
> > [e-mail]
> > https://www.ietf.org/mailman/listinfo/netmod
> >
>