Re: [ICANN-EU] of general interest

[Thomas Roessler <roessler@does-not-exist.org>]

On 2000-08-17 14:19:27 +0000, Lutz Donnerhacke wrote:

> Personally I dislike the current scheme, it causes dependent
> endorsments. Mail your proposal to Andrew.

A more statistical scheme with mostly independent votes would have
considerable disadvantages:

- Several candidates prefer to resign from the list of nominees at
  some point.  While in most cases realistic estimates about their
  own chances may be sufficient, at some point the number of votes
  (pardon, endorsements) collected so far may be important for this
  decision.

- This collapse of candidates - as well as the dependent voting -
  helps to actually get @large member-nominated candidates onto the
  ballot, by avoiding random splitting of votes.

Actually, what happens can be thought of as a continuous version of
normal multi-pass voting.

-- 
Thomas Roessler                         <roessler@does-not-exist.org>