Re: [websec] handling STS header field extendability

[Tobias Gondrom <[e-mail]>]

On 19/08/12 01:56, Barry Leiba wrote:
>>>> I'd also noted that we need to decide on a IANA policy to declare.
>>> Do we need to do this?  Assuming the proposed resolution achieves
>>> consensus (and there have been no nays yet), we're not setting up a
>>> registry. I don't think we get to set a policy for a registry we're not
>>> setting up.
>> <hat="WG chair">
>> AFAIK from an administrative perspective, Yoav is right.
>> In general we set the IANA policy for registry updates at creation of the
>> registry. So no need to do it here without the registry (assuming we don't
>> create an IANA registry).
> The point in this case is that this is a response to Ben's GenART
> review, which suggested that we at least nail down the registration
> policy at this point, lest someone come along later with an extension
> and create the registry with, say, an FCFS policy (probably too
> light), or perhaps with a Standards Action policy (arguably too
> heavy).
>
> The WG can validly disagree with Ben's review, and unless Russ (or
> another AD) strongly agrees and uses it as a DISCUSS point, we're
> done.  Even in that case, the WG can still argue against it and try to
> convince the AD that it's OK not to do this... and, as Tobias point
> out, we usually don't.
Thanks for the clarification.
>
> So maybe we should modify what Jeff said, thus:
>
>>>> I'd also noted that we need to decide on a IANA policy to declare.
> "We need to decide on an IANA policy *or* explicitly decide that we
> don't want to choose that now, and leave it to whoever creates the
> registry later."
>
> Apparently, we have at least two votes for the latter, from Yoav and Tobias.
>
> Barry (with AD hat off, but handily tucked at the ready under his arm)

<hat="individual">
Yes, I would vote not to specify an IANA policy for a registry we don't define here. I would be in favor of leaving that to when or if we create such a registry later. Though I don't have a strong opinion on this.

Tobias