Re: [RTG-DIR] RTG-DIR QA review of draft-ietf-babel-rfc6126bis-04.txt

[Juliusz Chroboczek <[e-mail]>]

>> for example, saying that BABEL implementations SHOULD support NETCONF,
>> a YANG model will be in a separate document,

> I am fairly sure that such a statement would not reflect WG consensus.
> Requiring a full YANG implementation (which includes, for a start, an XML
> parser) in every Babel implementation would be a sure way to kill the
> protocol.

I realise that I should clarify this.

The simplicity and implementability of Babel is one of its main features.
Babel can be explained to a lecture hall of fourth-year students in 40
minutes.  Babel can be implemented in less than 1000 lines of Python code,
and a stub implementation of Babel has been written that compiles to 12kB
of text.  Babel has been independently reimplemented three times, and in
one case it took the author (Markus Stenberg) just two nights to do the
implementation.

Given how important for us it is that Babel is a conceptually simple and
highly implementable protocol, I believe that it would be a tragic mistake
to make Babel dependent on the arguably complex protocol stack that is
NETCONF/YANG/XML.

What is more -- Babel development has been driven by the needs of our
users.  While we have been actively discussing management issues with our
userbase for the last years (I can send some pointers to the list if
you're interested), none of our users have every requested support for
YANG.  Hence, by making Babel dependent on YANG, we would not be serving
our constituency well.

-- Juliusz