Re: [ga] Voting mechanisms

[Alexander Svensson <alexander@svensson.de>]

Michael Froomkin wrote:
> I think STV should be available for votes with multiple alternatives.
Eric Weisberg wrote:
> I strongly recommend the use of the IRV voting system.

Some background info:
IRV (Instant Runoff Vote) is also called Alternative Vote (AV)
and e.g. used in Australian House of Representatives elections. It
is a *majoritarian* electoral system for *single* seats. Like every
electoral system, AV has some advantages and some disadvantages,
both of which are explained at length in a study for the Australian
Electoral Commission. One of the drawbacks according to this
study is that it "can result in the election of the least unfavoured
rather than the most popular" choice.
http://www.aec.gov.au/pubs/electoral_systems.htm#majoritorian

In contrast, STV (Single Transferable Vote) is a *proportional
representation* electoral system for *multiple* seats. The main
rationale behind it is not to waste votes.
http://www.aec.gov.au/pubs/electoral_systems.htm#proportional

Harald Tveit Alvestrand's proposed voting rule 4 (the N top
candidates win) is called "Single Non-Transferable Vote" and
used in Japanese elections. It's a *plurality* electoral system for
*multiple* seats; a variation of First-Past-The-Post.
http://www.aec.gov.au/pubs/electoral_systems.htm#plurality

--> So:
Obviously, there is no perfect (fair and simple) electoral system,
but in the case of selecting N candidates from a slate of M,
Alternative Vote (=IRV) is not a suitable choice. (And DNSO shouldn't
have *several* electoral systems at the same time -- let's make
it as straightforward as possible!) Please let's not recap the
19th and 20th century historical electoral systems discussion in its
entirety. Both STV and SNTV (+ eventually runoff) would do fine for
DNSO purposes, and both are easy to understand.

Regards,
/// Alexander Svensson

www.icannchannel.de

--
This message was passed to you via the ga@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga" in the body of the message).
Archives at http://www.dnso.org/archives.html