On Thu, 2 Oct 2008 19:52:31 -0400 Michael Allan wrote:
Dave Ketchum wrote:
Cycles happen, and perhaps should be reported, but are NOT a reason for he
system to do anything special beyond normal analysis and reporting.
Of course reporting should e based on total voting, thus updated as soon as
practical after any vote. Big point is that cycles happen and nothing gets
done to encourage or discourage their existence.
Assume the ideal Condorcet resolution is no resolution at all. If
reality intervenes and you would have a resolution, the closest to the
ideal is a hands-off method.
I do not understand 'no resolution':
By time N1 there have been 10 votes in the poll - to analyze as a complete
Condorcet election.
By time N2 there have been 2 more, for a total of 12 to analyze as if a
complete election.
Any such election may produce a CW.
Those that do not produce a CW result in a cycle. I suggest at least the
ability to implement multiple cycle resolution formulas, to support
comparison of the resolutions provided by various formulas.
If it is a hands-off method, it ought to be transparent to other
hands-off methods. No need to restrict to a single one. Allow
multiple parallel resolutions and approach even closer to the ideal
(Condorcet, phantom, and test bed or proving grounds) of no resolution
at all.
In terms of technical supports, Votorola's core is a continuous
medium. It never reports a winner at all. So it meets the ideal.
But the design allows for parallel analyses and massaging of the raw
data stream. So external sites can report their own resolutions in
more-or-less real time. (But this might not be implemented till the
production release, depending on need.)
Here I see Votorola offering a useful, though incomplete, service. What I
see desirable for Condorcet is an external site using that service.
In terms of my own interest, I want a rough understanding of how
external signals will cross with other events in the real world, and
influence the ideal (core, Condorcet, phantom). This discussion has
me thinking that cascade decision rings are not a resolution mechanism
after all, but some kind of defence formation (wagon circle) or
protective response against (at least in part) external pressures.
Possible values of such as wagon circles seem minimal to me for the current
discussion.
I do see external pressures possibly influencing later votes based on
earlier results - all a human loop.
These polls vary from real elections:
Current content of polls can be analyzed while voting continues.
Real elections do not get analyzed until after voting ends.
--
[EMAIL PROTECTED] people.clarityconnect.com/webpages3/davek
Dave Ketchum 108 Halstead Ave, Owego, NY 13827-1708 607-687-5026
Do to no one what you would not want done to you.
If you want peace, work for justice.
----
Election-Methods mailing list - see http://electorama.com/em for list info
