[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Condorcet Voting and Supermajorities (Re: [CONSTITUTIONAL AMENDMENT] Disambiguation of 4.1.5)



On Tue, Nov 28, 2000 at 04:01:14PM -0500, Raul Miller wrote:
> Let's try it this way:
> A.6(1)  If the vote has a quorum, this number of ballots are
>         cast for the default option (these ballots are in addition to
>         those votes cast by people).

This isn't meaningful. You don't cast ballots "for" an option, you use
a ballot to express preferences between options. We're trying to get
rid of ambiguities, not add them.

One possible way of doing this would be to add Q (the quorum) fake votes
of the form "D", which are to be treated as D (the default option) is
preferred to all other options, but no preference is expressed amongst
the other options.

In a simple majority vote, this could result in something like:

	N = 600 (number of developers)
	Q = 3 / 2 * sqrt( N )
	  = 37

	310 vote Yes, Further Discussion, No  [132]
	290 vote No, Further Discussion, Yes  [312]

Which would normally be interpreted as:
	Yes dominates No and Further Discussion, 310 : 290
	Further Discussion dominates No, 310 : 290

But if you add:
	37 fake votes Further Disucssion      [--1]

You end up with:
	Further Discussion dominates Yes (290 + 37 = 327 to 310)
	Further Discussion dominates No (310 + 37 = 347 to 290)
	Yes dominates No (310 to 290)

and, in spite of every single developer voting, that "quorum" requirement
makes the resolution fail. Which makes it not a quorum requirement,
but rather an increase in the majority required for a vote to pass.

Two better requirements for quorum would be:

	The number of ballots received must be greater than the vote's
	quorum.

or

	The number of ballots that prefer the winning option to the
	default option must be greater than the vote's quorum.

(the latter is almost directly from the constitution as it stands.
Note that this does *not* take into account the number of ballots that
prefer the default option to the winning option. If there're Q votes
that rank W over D, it doesn't matter if there are 0, Q-1, Q, Q+1, or
even 2Q votes that ranked D over W, at least as far as reaching quorum
is concerned) 

I assume this is where I'm meant to say something like ``I guess I'm
uncomfortable with that approach, because I don't have the idea that you
fully understand what [the constitution says right now]. You can't do a
very good job of fixing something if you can't distinguish what aspects
of it are working and what aspects are problems.''

Cheers,
aj

-- 
Anthony Towns <aj@humbug.org.au> <http://azure.humbug.org.au/~aj/>
I don't speak for anyone save myself. GPG signed mail preferred.

     ``Thanks to all avid pokers out there''
                       -- linux.conf.au, 17-20 January 2001

Attachment: pgpUFf_lzd9lG.pgp
Description: PGP signature


Reply to: