[Ssr2-review] FW: Imminent SSR issue

k claffy kc at caida.org
Thu Sep 13 16:52:11 UTC 2018



my audio problems prevented me from hearing whatever Scott
might have added to his question on the call today, and i won't
have time to listen to that call again this week, so i'm hereby
asking Scott explicitly whether we covered his concerns adequately?

there is one more thing i wanted to mention but didn't want
to take up more time on the call. i encourage (really, please)
anyone with some time or some students to go through and analyze
the public comments for this KSK roll announcement  -- there are
only ~22 of them) 
https://mm.icann.org/pipermail/comments-ksk-rollover-restart-01feb18/
and to consider the size and scope of the stakeholders who
made them, and see if you come up with the same comclusion
ICANN drew that there is a "preponderance of support to roll".
i personally drew the opposite conclusion from the public
comments. i predict that if "the wrong things go wrong",
many people will go back and analyze these comments and
come to a different conclusion than ICANN did.

this i think is a serious SSR (and AT, and CCT, and even
RDS-WHOIS!) issue: that icannORG has responsibility for
evaluating feedback on its proposed policies before launching them.  

as an academic, this felt to me like authors submitting a paper, 
getting a bunch of critical feedback about the paper,
and then the authors deciding whether the paper gets published.
(or in ietf-land, whether a new RFC is standardized.)

i think icannORG is in an untenable position here, i am not
suggesting a deep dive into all the things they could have done
differently for the KSK roll. but our recommendations should
recognize CoIs inherent in the process of peer review of SSR issues,
which i suspect it will come up as we evaluate SSR1 recommendations.
i suspect several of us are already considering a recommendation
that includes support for someone independent from icann 
determining whether icann has implemented all SSR2 recommendations.

k



More information about the Ssr2-review mailing list