Re: [spring] Request to close the LC and move forward//RE: WGLC - draft-ietf-spring-srv6-network-programming

[Ted Lemon <[e-mail]>]

On Feb 27, 2020, at 1:59 PM, Robert Raszuk <[e-mail]> wrote:
> It is very unfortunate that IETF does not have a good way of retrieving judgement from real group of folks who understand given proposal. 

We do.   It’s called “substantive comments.”

> "+1" is just only one demonstration of it. Humming is another. Raising hands one more. We say there is no voting but while there is no formal ballot box nor even e-ballot version of it all of the above ways to gather "consensus" are examples of voting. 

Actually, the purpose of humming is not to make a decision, but to figure out whether there is general consensus.   If you ask for a hum and you get a 50-50 response, there probably isn’t consensus, and you might just say “we don’t have consensus” and go on to figuring out how.   If the “no” hum has no loud participants, you might say “looks like we’re good to go, we’ll confirm on the list.”   If there’s someone humming loudly no when everybody else is in favor, and you don’t know why they’re humming that way, that’s a good time to ask them if they are willing to explain.

But bear in mind that humming does not take place on the mailing list, and that consensus is called on the mailing list, not in the room.

On the mailing list, people pretty much have to raise objections verbally.  No amount of +1s should be considered meaningful at all.   The work is chartered; the wg is supposed to do it.   If there are no objections, and people feel the document is ready, then it should move forward, whether there are +1s or not.   If objections are raised, and they are substantive (that is, not opinion or conjecture), then they have to be addressed.   They can be addressed by saying “we considered that, and the working group as a whole agrees that the problem exists, but it doesn’t need to be addressed because this document is only applicable in a situation where the objection raised doesn’t matter.”  Or it can add text to address the objection, as Brian I think has suggested.   Or it can do additional work to address the problem, as Brian has also suggested.

But the WG can’t simply ignore the objection.  That is not what “rough consensus” means.