Re: [Eligibility-discuss] NomCom selection Fwd: Notification for draft-eastlake-rfc3797bis-00.txt

[Paul Hoffman <[e-mail]>]

On 29 May 2023, at 15:40, Joel Halpern wrote:

> Whether or not the procedure in that (draft-hoffman-...) is useful for other people I can't say.  But it removes may elements that the community felt were important in defining the nomcom process and rfc 3797.  For example, you removed all of the challenge periods and challenge criteria. 

Wait, wait. All this draft does is specify how to pick from a group. It is not meant to be a drop-in replacement for RFC 3797, and nothing in the draft says it is meant to be. If folks here like the idea of using an easier-to-understand mechanism that doesn't have the pitfalls of RFC 3797, a 3797bis that keeps all the important parts of the process could just slip in this mechanical part.

> And your educed the random numbers to one source, without specifying anything about the required degree of randomness in that source.

That is covered in the Security Considerations.

>   While I am not expert on the matter, the reason that was stated for having multiple sources was to get enough reliable randomness into the mix.

For the method used in RFC 3797, yes. For the method in this draft, a few bits is just fine.

--Paul Hoffman