Re: [Eligibility-discuss] On 3797 alternatives

[Michael Richardson <[e-mail]>]

Salz, Rich <rsalz=[e-mail]> wrote:
    >> It's not the random sources that are the problem, I think.  We could
    >> say that we will take the XYZ Lotto on June 12th, and then if we need
    >> further randomness, take the XYZ Lotto from June 19th, or June 26th,
    >> right?

    > In general the lotteries that have been used for NomCom choices are
    > once or twice a week.  That means on average 2-4 days delay just
    > waiting for the numbers to be drawn.  (E.g., in your example, suppose
    > someone is disqualified on June 20?)

On June 13, one collects the random numbers from June 12th
The list is published, and the lotto of June 19th is identified as being (a)
re-initialize source.  We wait for challenges and refusals.
On June 19th the new lotto numbers come out, and they get collected.
If this person is disqualified before the process is re-run on the 20th, then
maybe it's all good.   Or maybe it's this EKR-detailed attack where people
are strategically withdrawn?

Worse case, of course, is that they are disqualified after the new numbers,
and one has do it again.

    >> The two weeks are because we have to give people a chance to review.
    >> That's a people process, not a technical one, right?

    > Not just.  The whole process.

So, what I'm not getting is why the new random sources are such a major
impact to the timeline, when I think it's really the people process.
(I agree that they are a pain, and I'm not convinced they are worth it)



--
Michael Richardson <[e-mail]>   . o O ( IPv6 IøT consulting )
           Sandelman Software Works Inc, Ottawa and Worldwide