PEP 308 vote type (was Re: Update to PEP308: if-then-else expression)

[Andrew Koenig]

Aahz> If I run the voting, and it's a straight "pick between these
Aahz> alternatives" vote, I'll probably use Condorcet.  OTOH, after taking a
Aahz> quick look at more web pages, approval voting is easier, so maybe I'll
Aahz> use that instead.  <shrug>

By approval voting, I assume you mean allowing people to vote for as
many alternatives as they like.  In which case I'm strongly in favor
of the idea; I posted another note in this discussion explaining why.

Aahz> Does anyone have actual experience with both Condorcet and approval
Aahz> voting?  (I'm *not* going to use IRV (aka Australian/Hugo).)

Here's a personal experience with approval voting.

During the organizational meeting of the C++ standards committee, in
December 1989, I proposed approval voting to the committee.  There was
a lot of discussion while people tried to sort out what it meant.  At
the end of that discussion, most people didn't care much but there was
a general trend in favor of it -- except for one person who seemed to
have decided that he would oppose any suggestion I made regardless of
its merits.

Anyway, after much discussion, we voted on the idea of using approval
voting in future multi-way votes and there was a small but comfortable
margin in favor of it.  Then we broke for lunch.

I had lunch in a restaurant with surprisingly slow service, on account
of which I returned 20 minutes late to the afternoon session.  By that
time, the guy who opposed anything I did had already introduced a motion
to repeal my original proposal, and gotten the committee to accept it.

I never did find out what arguments he used.

-- 
Andrew Koenig, ark at research.att.com, http://www.research.att.com/info/ark