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

[Raul Miller <moth@debian.org>]

On Mon, Nov 27, 2000 at 02:20:04AM +1000, Anthony Towns wrote:
> Not helped by me making up my own terminology now and then, by the looks.
> 
> What I've been randomly calling the "schwartz" set, is actually meant
> to be called the Smith set, and the Smith condition is that whoever wins
> the vote should be a member of the Smith set.
> 
> The definitions link from http://www.fortunecity.com/meltingpot/harrow/124/
> defines what I've been calling "circular ties" as cycles.

Looks fine.  I'll assert that the constitution's concept of "Dominates"
is intended to exclude every vote which is not in the smith set.

> I'm not sure what you mean by a "strict" preference. The only use in
> the constitution is A.6(2), which is just referring to there having to
> be *more* votes preferring A to B than prefer B to A, rather than the
> same number.

I'm using the dictionary definition of the word "strict".

However, I'm happy with formal definitions of the concept "strict
preference" for example,

   An individual strictly prefers one option to another if and only
   if that individual accepts one side of a tradeoff and rejects the
   other side.
   -- http://www.src.uchicago.edu/depts/polsci/research/american/hansen96.htm

[As an aside, you might consider installing dict-web1913 -- wordnet's
definitions are more for research purposes and are at best a supplement
to a real dictionary.]


                            * * * * *

[the constitution, in particular A.3(3)]
> > ... loosely *specifies* N votes on most issues, where the last vote
> > *might* be a simple yes/no vote.  It also *allows* those votes to
> > be collapsed into fewer votes -- no fewer than 1.  In no place does
> > it *require* any any specific number of votes take place.  That's
> > a judgement call (or, a series of judgement calls).
> 
> It describes a method of voting on an issue where there are first N
> (presumably concurrent) ballots on each of the independent aspects of
> the issue under discussion, and then a final ballot on whether the final
> form of the resolution should be passed or not (which has only Yes,
> No and Further Discussion as options).

There is nothing in A.3 which requires concurrency among independent
votes, though that is an option.

More importantly, A.3(3) specifically allows for the combination of
amendment votes with the final vote.  You seem to be contradicting this?

[I've deleted the rest of your message, in the belief that I've addressed
every major point you raised.  If you feel this isn't the case, please
feel free to restate those points.]

Thanks,

-- 
Raul