Re: The note to Tony Li about draft-ietf-isis-traffic

[Alex Zinin <zinin@psg.com>]

Harald,

> here's my current draft...comments?

In general--very consistent with the numerous personal
conversations I had with Tony. Though I know his answers,
sending this seems like a logical step...

> From: Harald Tveit Alvestrand, IETF chair
> To: Tony Li

Cc: hhwsmit@zonnet.nl, prz@net4u.ch, iesg@ietf.org?

> Subject: draft-ietf-isis-traffic-04 and copyright

...

> From what Bill and Alex tell me, you are taking a related position as
> Bill's in the ISIS working group - that you trust your personal
> judgment more than you do the jugment of the IETF "some time in the
> future", and that you want the ability to block changes to the
> protocol - the difference to Bill Simpson's situation being that at
> this time, apparently nobody in the WG or the IESG want changes to
> your draft.

Agree with Thomas here, will reply to his e-mail separately.

> The document is a specification of extensions to the IS-IS protocol,
> which is formally the realm of ISO/IEC JTC1.
> If it had been solely an IETF protocol extension, it would clearly
> have been a candidate for standards-track processing.

> Alex Zinin and the IESG have been
> working on draft-zinin-ietf-jtc1-aggr, which tries to make explicit
> the relationship between JTC1 and IETF work in this area. But this
> document isn't signed yet, so I'll leave that out of the discussion
> for the time being.

His answer will be that there isn't any interest in putting this
on the STD track, and such interest is not expected either.
(background: they just want to get things done, not overload
themselves with the STD process.)

> I'm assuming that you don't care about ownership of the exact words -
> what you care about is changes to the technology.

> In order to block technology changes, I think you have to:

> - Block someone introducing amendments in the ISIS WG using words not
>   yours.
> - Block someone introducing amendments in SC6 using words not yours.

> I think this is possible. But I think it will be easier if you work
> within the process to build consensus that this technology should not
> change, or should only change within sharply limited bounds.

My suggestion to him was to exercise his role of the WG chair in the
technical expert mode (as opposed to just a managerial one), where he
should ensure that the WG does not build a consensus around a broken
idea. And believe me, he can do this very effectively. Of course, WG
chairs and ADs come and go...

Another idea that we briefly discussed was the possibility to freeze
specifications, like it is possible in JTC1... He says it would also
solve his problems, but we don't have such a tool...

So the question still remains: what other tool can we offer to Tony...
WG chairmanship looks like the best compromise to me...


Alex

> My main worry with this document is not this particular document, but
> the impact its publication will have as a precedent setter.

> If the IESG were to grant your request to have this document published
> with a no-derivative-rights clause, we have established a precedent.
> Next time someone wants a core IETF technology to remain under their
> personal control, this precedent will be cited.

> And if we find errors in such published documents, or the ownership of
> the right to produce derivative works grows unclear (through corporate
> mergers, bankruptcies or other maneuvers; personal illness, death or
> extended vacations; other causes one cannot imagine now), we can
> easily find ourselves in a situation neither of us want the IETF to be
> in: Where we have protocol standards with known, important errors that
> we are unable to fix.

> Out of 2145 current internet-drafts, there are 44 that use the
> no-derivative-works clause at present. I would like that proportion to
> remain low.





> What do you think?

>                   Harald Alvestrand