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

Re: Request for comments [voting amendment]


Branden Robinson:
> > Yes, it must. The default option is "further discussion" in most GRs, and
> > a "none of the above" that equates to further discussion in DPL elections.
> > If the default option wins, we have another vote, where proposers of that
> > option can either argue their case better, or realise that they're the
> > appropriate majority of developers don't support the opinion at that time.
> In the case of elections for an office, doesn't this give the incumbent
> an unfair advantage if he is also on the ballot as a candidate?
Personally, for votes where there simply must be some sort of result
(you _can't_ leave the office vacant), I would not use any supermajority
or quorum, and I'd replace the "drop if defeated by the default-option"
rule with one that states that if the default option wins, then it is
eliminated, the whole vote is considered again, and the term of office of
the 'winner' is reduced to a maximum of three months or so.

> > In a way, the default option is there specifically so that you can
> > vote strategically. 
> Is that wise?  Susceptibility to strategic voting is generally
> considered a flaw.
These terms are understood differently here. We mean by 'strategic voting'
things like "I prefer A, and I can live with B, or C if must be, but I'd
have serious problems with E or F", and I can actually _say_so_ on my

With other election methods, this term means things like the well-known
"I'd really prefer N, but I'll vote G because otherwise B will win"
situation (you all know what the letters stand for) which is Not A Problem
in a Condorcet voting.

> > Another alternative might have been to have the default option win if
> > it's _ever_ a member of the Scwartz set, rather than if it's a member
> > of the Schwartz set after the sequential dropping phases are complete.
That rule might be equivalent to the "drop all options defeated by the
default option" rule, or it might not be. I think it is, based on gut
feeling, but I don't know.

I'd rather run the algorithm with the full set of votes first, and _then_,
if the default option wins, have a separate rule on what to do next.
Dropping some alternatives entirely is exactly the reason why IRV is a bad
voting method, and we shouldn't introduce that to Condorcet without a very
good reason.  :-/

Matthias Urlichs     |     noris network AG     |     http://smurf.noris.de/

Attachment: pgpgTGDsVxrtL.pgp
Description: PGP signature

Reply to: