Re: [Json] Fwd: Blog: YANG Takes Off in the Industry

[Phillip Hallam-Baker <[e-mail]>]

On Fri, Aug 1, 2014 at 8:35 AM, John Cowan <[e-mail]> wrote:
> Carsten Bormann scripsit:
>
>> That vacuum is filled by draft-ietf-netmod-yang-json-00.  (Nature
>> abhors vacuum.)
>
> Neither here nor there: that is merely a JSON expression of the XML.  It
> has nothing to do with modeling JSON.
>
> Phillip Hallam-Baker scripsit:
>
>> But you could use it to create a data model that was represented in
>> JSON.
>
> I haven't read the RFC in any detail, but the model given in 4.2.2 does
> not seem to be a JSON model: it is far more complicated.
>
>> And that community refuses to make any changes to XML or XML schema to
>> support protocol work.
>
> Or for any other purpose.  Ghu knows I tried.
>
>> So XML is a bad choice for protocol design.
>
> That conclusion I believe to be unwarranted.  The problems with XML are
> mostly the hard-to-use APIs, of which one of the worst is that embedded
> in all browsers.  If anything is carved in stone these days, it is the
> DOM.

True, but JSON is (or certainly should be) just a serialization format
of the DOM. So it is actually a short distance between the two points
than XML.

Which is why a binary version of JSON should be a binary encoding of
the DOM and not some different data model.