[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 Mon, Dec 04, 2000 at 05:03:25PM -0500, Raul Miller wrote:
> On Tue, Dec 05, 2000 at 12:32:12AM +1000, Anthony Towns wrote:
> > The Condorcet criterion says that if there's a single option that
> > pairwise beats every other option, it should win (assuming there's no
> > supermajority requirement, and quorum is met).
> That's a relatively weak criterion, all things considered.

Yeah, well. Most of the time, it's still a good one, though. (It's
applied to all the votes we've actually had.)

> > It's what the current A.6(2) and A.6(3) are for. (The Condorcet criterion
> > doesn't say anything about the ambiguous cases we've been talking about)
> I thought we'd agreed that they are to ensure that the Smith criterion
> is met (which is more specific than the Condorcet criterion).

What I meant was that there was no reasonable way of misinterpreting A.6(2)
and A.6(3) so that the Condorcet condition wasn't met.

> > Ignore all the procedures established in the appendices, and just
> > note that a consitutional amendment requires a `3:1 supermajority'
> > to succeed. I can only see two things this can be attempting to require:
> > 	* For every person who wouldn't accept it as a resolution, there
> > 	  must be three who will for it to pass.
> Agreed.
> > 	* For every person who would prefer some other solution, there
> > 	  must be three who think this is the best solution possible.
> Agreed.
> > The way the N+1 style of voting behave, you're effectively requiring the
> > former.
> My point of view is that these two are essentially equivalent: in the
> N+1 style of voting, a person who thinks that the option isn't the best
> would vote for "further discussion".

Well, they might do that, yes. Or else they might think to themselves, "well,
I'm never going to get my preferred choice, and this isn't bad, so yeah,
let's do that". I think that's a valid point of view to hold.

In the current voting system, you'd express that by voting
	[ 1 ] My First Preference
	[ 2 ] My Second Preference
	[ 3 ] Further Disucssion
then, when "My Second Preference" wins, you'd say "Oh well" and vote
	[ 1 ] Yes
	[ 3 ] No
	[ 2 ] Further Discussion

Similarly, in the counting scheme I proposed, you'd vote:
	[ 1 ] My First Preference
	[ 2 ] My Second Preference
	[ 3 ] Status-quo (No change, but further discussion possible)
and whichever of First/Second won the majority count would implicitly
be treated as a "Yes" vote for quorum and supermajority purposes.

If you wanted to only help "First Preference" achieve its supermajority,
you'd instead vote:
	[ 1 ] My First Preference
	[ 3 ] My Second Preference
	[ 2 ] Status-quo
which would have the same effect as voting NFY (No, Further Discussion,
Yes) on the final vote.

> I'm guessing that you're saying my view of supermajority isn't
> particularly useful is based on your idea where "further discussion"
> is equivalent to a "no" vote, even though they're disctinct options
> on the ballot, and even though the matter is dropped if "no" wins,
> but further votes will be conducted if "further discussion" wins.

No, your view of supermajority isn't good because it affects other
options than the status-quo options, and raises the odds that they'll win.
An alternative way of looking at it is that there's no way to express
a preference towards a non-supermajority vote without that preference
counting for three preferences the other way. What if I *don't* feel
that strongly about it?

As far as "no" versus "further discussion" goes, consider what would
happen if the non-free vote had gone something like: my amendment passes,
so then it's voted on and fails, with "No" winning. I, personally, would
suspect that John and friends would immediately make a new proposal so
they can get their issue actually voted on, although there would be good
odds that it'd fail a Yes/No vote, just as it did against my amendment.

> I suppose a literal interpretation of 3:1 "supermajority" would be that
> over 75% of all of debian's developers must be in favor of an item before
> it can be passed. I think that my intepretation of "supermajority"
> is more relaxed than this, but that it's equivalent for the case where
> everyone is voting.

A supermajority enforces a proportion of yays versus nays, a quorum
enforces a reasonable proportion of the voting membership is represented.

Requiring 75% of developers to vote in an election would basically make
it impossible to ever change anything.

So, I don't think you're being particularly liberal there at all :)

> > > The way I see it, when determining which votes Dominate which others,
> > > votes which prefer an option with a supermajority are reduced to an
> > > appropriate fraction (in this case 1/3).  But, 40 Dominates 33 1/3,
> > > so M wins.
> > Yes, that's how I understand your interpretation too; it just doesn't
> > seem sensible to me to have the voting system be biassed towards options
> > that don't require a supermajority (or require a lesser one).
> Well.. the constitution seems to explicitly state this.

Well, again, my reading explicitly requires a vote with Yes/No/Further
Discussion as the only options, and only applies supermajority and quorum
to those options.

> > That is, given two options, A and B, both of which pass quorum, both of
> > which are acceptable to their respective supermajorities, then the one
> > that's picked should be the one that's preferred by more people, simple
> > as that.
> Do you agree that "preferred by more people" really means "preferred by
> a majority of relevant votes"?

Sure. If people want their preferences to count, they have to vote. If they
choose not to, we have to assume the votes we do get are representative
(unless we miss quorum).

> > There's an obvious justification for choosing A: more people prefer
> > it. I can't see one for being biassed towards B (anywhere up to a
> > 2.99:1 majority for A:B would still make B the winner), merely because
> > it's easier, constitutionally, to implement.
> By "easier" do you mean "closest to what the constitution says", or
> do you mean something else?  [If you mean something else, I don't
> see that "easier" has anything to do with it.]

Doing something that doesn't require a supermajority is easier as far
as the constitution is concerned. It's easier, for example, to overrule
the DPL or a delegate than the technical ctte.

> > [1] As opposed to a single ballot that lets you say:
> > 
> > 	[2] Remove non-free [NYF]
> > 	[1] Support non-free [YNF]
> > 	[3] Further Discussion
> > 
> >     which is effectively just incorporating the N+1 votes into a single
> >     message (which is how I'd have interpreted A.3(3)).
> I don't know what you mean by NYF and YNF, above.

It incorporates the two votes into a single message. The above would
be interpreted as an initial vote for keep non-free, or remove it, or
discuss more, and depending on which of those options won, a final vote
of either:
	[2] Yes, [1] No, [3] Further Discussion
or	[1] Yes, [2] No, [3] Further Discussion

> On Tue, Dec 05, 2000 at 12:36:08AM +1000, Anthony Towns wrote:
> > Nor are "Debian will have nothing to do with non-free ever again"
> > and "Debian will continue to maintain non-free, but will do it on
> > unofficial.debian.org instead of ftp-master". So I don't see what
> > relevance the above has at all.
> The example we were discussing was "Debian will have nothing to do with
> non-free, and we're pulling it out of the constitution", and "Debian
> will gradually phase out the current non-free ftp archive".

Yes, and it doesn't seem meaningful to try to do both of these at once.
You can't continue to support non-free but gradually phase it out while
having nothing to do with non-free at all.

> > I'm much more interested in how we should conduct votes in future,
> > anyway.
> I guess I'm not comfortable replacing the constitution with anything
> that I'm not convinced is unambiguously better.

Sure, I don't see why anyone would be.


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: pgpGyWR03HcIC.pgp
Description: PGP signature

Reply to: