Re: [Isis-wg] Adam Roach's Discuss on draft-ietf-isis-l2bundles-05: (with DISCUSS and COMMENT)

[Adam Roach <[e-mail]>]

On 5/24/17 7:10 AM, Alia Atlas wrote:
> Adam,
>
> First, I would greatly appreciate it if you and others could read the 
> ballot write-up.  I did
> try to explain what I did to mitigate the concerns in the shepherd's 
> write-up; as far as I
> can tell, this is a case of one person being in the rough.

Thanks for the pointer. For clarity, the person you're describing as "in 
the rough" is the document shepherd, right?

For avoidance of doubt: I did read both the shepherd's write-up and the 
ballot write-up prior to entering my position. The problem I'm running 
into is that the explanations in the ballot writeup seem to be 
contradictions rather than clarifications. For example, the shepherd's 
answer to question (9) paints a different picture than the "clear WG 
consensus to publish" in the ballot writeup.

> Please read my write-up - which I did spend time on to explain some of 
> these
> aspects.   The existing IPR claim - by a non-author- was brought up to 
> the WG
> and very briefly discussed, as is typical.

Thanks for the clarification; "briefly discussed" disagrees with the 
answer in the shepherd's writeup: "no WG discussion around IPR." I do 
see that the Ericsson disclosure was posted to the list on May 12, 2016; 
and a number of assertions in June of 2016 that other individuals were 
unaware of additional IPR, but I find no emails that directly contradict 
the shepherd's assertion, so I must be missing the "brief discussion" 
you refer to. Do you have a pointer? Note that I don't consider this 
lack of discussion to be blocking in and of itself, but it does add to 
the "this smells kind of funny" situation around this draft.

I concede that many of the issues I'm seeing may be the result of a 
reluctant shepherd rather than actual process issues. Is there a chance 
we could get a different shepherd assigned (e.g., the other ISIS chair)? 
In addition to noting that the current shepherd, if opposed to the 
document in the way that is being implied, is unlikely to be proactive 
in moving it forward, I think shepherd-writeup-style input from a 
different party who was involved in the working group during the 
progression of this document would be very helpful in clearing the air here.

>     These overarching process problems seem large enough that any
>     comments I
>     may have on actual content -- such as an apparent lack of IPv6 support
>     (or, at least, a complete omission of IPv6 from the examples) -- would
>     seem like rearranging deck chairs on the Titanic.
>
>
> If you have specific technical concerns, they are, of course, easier 
> to just deal with
> at the same time.  The document does, of course, support IPv6 - as is 
> clearly indicated
> in Section 2.1 where, for instance,  use of " IPv6 Interface Address 
> (sub-TLV 12
> defined in [RFC6119])" is mentioned.  The appendix doesn't include an 
> IPv6 example,
> but the primary purpose there is to explain the complex sub-TLV 
> structuring since
> the standard bit-fields ascii art was challenging to do in a 
> meaningful way.

Well, they're more editorial than technical, since IS-IS is a bit 
outside of my wheelhouse. If the draft moves forward, I would like to 
see the IPv4 examples use addresses from the RFC5737 blocks, and perhaps 
an alternation between IPv4 and IPv6 in the examples.

/a