[Mod-discuss] Re: Is there a problem?

[Ted Lemon <[e-mail]>]

On Sep 12, 2025, at 05:18, Brian E Carpenter <[e-mail]> wrote:
> If the draft doesn't make that clear, then the draft needs more work...

We should avoid magical thinking here. No amount of clarity in the text can have the result that everyone who has a clear understanding of what the document says that is contrary to what it actually says will stop having that mistaken understanding. Best is the enemy of good enough.

The only thing that can help with this is to run the code and let the people who are afraid things will go wrong point out actual cases of it going wrong. This is an important part of the consensus process. Imagining that a document will accomplish this is in fact magical thinking. Only people can accomplish a good process. You can't write a document that, in itself, prevents all possible abuses. We achieve success through successive approximation, not by solving the halting problem. I have absolute faith that John and Keith et al. will in fact step up if this happens, and I promise to do so as well.

As has been mentioned many times here, we have been running a version of this process on the IETF mailing list for a long time. We have not heard specific examples from that body of experience documenting failures of the type that are being discussed here.

I will point out that there have been times in the past, which most of the people objecting to this remember clearly, I think, where in fact working group chairs of certain now-defunct working groups have in fact behaved in the way that those objecting to this proposal fear. The particular working group I have in mind is now defunct, which to me says that the process worked: the chairs who were committing the abuses are no longer chairs. I am sure there are other examples of this that I don't know about.

One of the goals of this document is to have a clear process for preventing that from happening again in the future, and I think the document does a good job at that, and we actually have experiential evidence that it is possible to accomplish this goal.

So while on the one hand these objections are completely understandable, on the other, they have been addressed. The rough consensus process does not demand that every problem be definitively solved. It demands that every technical objection be addressed. And they have been.

A large number of people from the community participated at the beginning of this process, and I'm confident are still watching and participating, but are not engaging anymore because their objections have been addressed. It's natural for the folks who are in the rough to be the last ones talking, because they're the ones are dissatisfied.

This is not an indication that we have a problem. This is normal and expected. This is why we call it "rough consensus" and not "consensus."

I'm not claiming to call consensus here, but FWIW, I am content with the current document, and request that the chairs move to the next step in the process once the WGLC expires.