[message really to RSSAC members who find themselves on such a podium, but for full transparency on this list] The same old questions. The same old imperfections in root-ops/RSSAC answers. My favorite two: Root zone distribution times: root ops are understandably proud of the short zone update times. And now we even publish them. However it is important to manage expectations here. When answering this question it is important to mention what the worst case is and that it is a consequence of the very distributed nature of the root server system; that is an engineering choice made for the resiliency benefits it brings. I always add that even 24h delay is not really a problem *for TLDs*. This is not about end-user domains. TLD operators oerate in a much more stable environment; most of the delay is in the update process itself and its required checks. A TLD op that depends on a-la-minute updates is doing something wrong. Accountability: It is important to always mention that root ops are accountable to their respective organisations. So there is no lack of accountability. It is just not centralised. That is often just assumed. It is only after that existing accountability has been mentioned that one should talk about reputation and loss of it and that the organisations are significant enough to care. I always mention that centralised accountability is a single point of failure, which is what we want to avoid. Resiliency again. It is important to slowly and carefully walk thru this argiment each and every time. Daniel On 8.03.16 14:48 , Alejandro Acosta wrote:

Hello, Yesterday during ICANN was given a presentation called: "How It Works: Root Server Operation", it given by some folks of RSSAC

The presentation is right here: https://meetings.icann.org/en/marrakech55/schedule/mon-how-root-server-ops/a...

Great presentation, thanks for it.

Alejandro,

_______________________________________________ rssac-caucus mailing list rssac-caucus@icann.org https://mm.icann.org/mailman/listinfo/rssac-caucus