[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 Sun, Dec 03, 2000 at 08:49:38AM -0500, Raul Miller wrote:
> On Sun, Dec 03, 2000 at 10:13:58PM +1000, Anthony Towns wrote:
> > I will assert that the options "no" and "further discussion" aren't
> > usefully different.
> They're procedurally different, however.

Not particularly so. If "No" wins the final vote, does that actually
stop people from continuing discussion, and proposing a new vote? If
"Further Discussion" wins, does that actually require people to continue
discussing things, or does it ensure the newly reset vote doesn't expire?

> > > If that's not biasing towards options which don't require a supermajority
> > > (which only need to exceed 1 voter in favor for every one voter against),
> > > I'd like to know what is?
> > I'd like you to show one other voting system in use where supermajority
> > is interpreted in this way.
> Why?  This is debian-vote, not congress-vote.

To give some evidence that your interpretation is reasonable. It doesn't
have to be something an American government uses.

> I claim that "my first preference is yes on option A", is a yes vote
> for option A.  And, if A requires a supermajority, then A.6(7) applies.
> Do you claim this is not an actual reason?  Why?

You're also claiming that "my second preference is a yes on option A"
is a no vote for option A.

I'm claiming that "I prefer option A to the status quo" is a yes vote
for A, and "I prefer the status quo to option A" is a no vote for A, at
least as far as quorum and the supermajority are concerned.

> > > >     and M has unanimous support over N, and then add a third option:
> > > > 	* Evade constitution (E)
> > > If M really has unanimous support, you wouldn't add a new option --
> > > no one would want a new option.  Adding a new option only happens:
> > No, M has unanimous suppoer over N, that is everyone prefers it to
> > the current situation. That is, everyone wants to get rid of non-free.
> > A minority of those people would rather do it by exploiting loopholes in
> > the social contract ("We'll continue to provide non-free for download,
> > but we won't necessarily allow people to upload to it!! Muahahaha").
> > Why, exactly, is it so undesirable to modify the social contract that
> > we should just ignore the expressed preferences of half/three-quarters
> > of the developer body, and take the "easier" solution, instead?
> I don't actually know, this is your fantasy.

Well, it's also the result of your interpretation of the constitution, and
your proposed changes to the constitution.

The *sole* reason that "E" wins in the above is because the alternative
preferred by the majority modifies the social contract. It's not because
that alternative doesn't have supermajority support over the status-quo.
It's not because most people dislike. It's not even because most people
prefer some other option.

I assert that the purpose of supermajority requirements (like the purpose
of a quorum requirement) is simply to ensure that the character of the
organisation doesn't change without due consideration by the members of
the organisation.

Your scheme seems to give it a meaning more like "try to choose any other
possible alternative that doesn't involve modifying the constitution or
overturning a decision of the tech ctte or whatever else might require
a supermajority".

> > Certainly, we should make sure that everyone prefers the modified
> > social contract to what it says at the moment, but there's no good
> > reason to accept *any* other option in favour of one requiring a
> > supermajority.
> You don't put just *any* option on the ballot.  Only alternatives.

"any other *alternative*" then.

> > > That's my understanding of how the constitutional voting system is
> > > supposed to work,
> > Considering, under the most liberal interpretation, the constitution
> > describes two methods, one of which works that way, and one of which
> > works the other way, it seems a little bold to say that the constitution
> > is *supposed* to act that way.
> > Certainly, I think the constitution pretty clearly indicates that's
> > *not* how votes are meant to be handled or interpreted.
> Eh?  If there's a sequence of votes, and people don't like the way
> that sequence went, they can vote against the final option.  They
> can even keep the sequence going at the same time by voting for
> "further discussion".

Even on an A.3(3) vote in your interpretation?

Could you please give an example of some possible ways of conducting
a vote that includes at least "remove non-free" and "move non-free to
unofficial.debian.org", assuming that "remove non-free" modifies the
social contract, and that that requires a supermajority?

> > If they then get their supermajority support, why then ignore the stated
> > preferences of the voters, and choose the unnecessary compromise position
> > anyway?
> Because people voted for a different option from the one which required
> supermajority.  Do you want to discount someone's vote on the basis
> that you find their position disagreeable?

Again, you don't vote "for" or "against" an option when you're expressing
preferences. You only say which options you prefer to which others.

This isn't just a quibble about wording: it's a fundamentally different
way of voting. Just because you list an option as second, doesn't mean
you dislike it. It doesn't even mean you particularly care whether your
first or second preference is chosen, it may just mean you prefer either
of those to your third preference.

This is in stark contrast to first-past-the-post votes, and "aye/nay"


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: pgpQDhmtl9AU7.pgp
Description: PGP signature

Reply to: