[NCAP-Discuss] Risk terminology

[Jeff Schmidt]

It seems we are having some terminology issues. Let me define some terms:

RISK MANAGEMENT PROGRAM

In the 2012 round, ICANN adopted a programmatic approach to collisions that included: (1) identification of potential Black Swan strings from the list of applied-for strings; (2) a notification and “cooling-off period” known as Controlled Interruption; and (3) an emergency response capability should an unanticipated Black Swan scenario materialize.

These components are discussed in detail in the JAS report, on which ICANN based their program. Taken in their entirety (1-3 above), this is the Risk Management Program ICANN deployed to address the risk of problematic Black Swan collisions during the 2012 round.

RISK ASSESSMENT

Only Step 1 of the above is the Risk Assessment: determining which applied-for strings present unacceptably high risk of Black Swan collisions problems. Both Interisle and JAS performed a Risk Assessment in 2012 (Interisle Sections 7 and 8) and JAS Recommendation 1 which assessed c/h/m as presenting unacceptable risk and therefore not continuing. The remaining 3,719 pages of the JAS report are really just us showing our work on how we came to make that judgement and describing our thinking, analytical techniques, etc. Ditto the 200 pages of the Interisle report. It was a judgement based on data, experience, and expertise.

As one can see by reading the Interisle and JAS reports, both firms used a full range of data and analytical techniques to form their risk assessments. The approaches used by Interisle and JAS leveraged *all* the metrics presented yesterday *and more* (except the ad network).

The correctness or incorrectness of Interisle and JAS judgements about high risk strings is fair game for analysis and criticism. No data indicates clearly that we were wrong and a decade of experience indicates that we seem to have gotten a lot right. NCAP Study 1 confirmed that, as did SubPro.

ACCEPTABLE RISK PATH

Step 2 is the “acceptable risk path.” Even if a string is assessed to pose acceptable risk in Step 1, prudence dictates some risk mitigation - including the notification and cooling-off period known as Controlled interruption. It is inspired by collisions that occur in any large dynamic addressing system like telephone “wrong numbers” and misrouted postal mail and allows operators become aware of and react to the change. In 2012, ICANN implemented this as JAS recommended (Recommendations 6,7,8,9).

The effectiveness of Controlled Interruption as a notification technique is fair game for analysis and criticism. There are other options all with tradeoffs. These were exhaustively discussed in JAS and SSAC documents at the time. However, no approach objectively superior to Controlled Interruption has emerged, a decade later. NCAP Study 1 confirmed that, as did SubPro.

EMERGENCY RESPONSE

In case we were wrong, prudence dictates that ICANN be ready with a pre-determined emergency response program. In 2012, ICANN implemented this as JAS recommended (Recommendations 3,4,5).

IMPROVING THE NEXT ROUND

Of the above, what can be improved upon?

Jeff

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mm.icann.org/pipermail/ncap-discuss/attachments/20220609/a64856bf/attachment.html>