On Thu, Jul 3, 2008 at 3:06 AM, Scott Kitterman <[EMAIL PROTECTED]> wrote:
> On Wednesday 02 July 2008 18:31, Cody A.W. Somerville wrote: > > On Wed, Jul 2, 2008 at 5:29 PM, Scott Kitterman <[EMAIL PROTECTED]> > > > <snip> > > In my experience Condercet methods work quick well with multiple selection > votes too and retain their resistance to strategic voting. In either case > we > wouldn't count it by hand, so the complexity of counting isn't a > significant > factor. > > Yes, Debian does like the fancy charts to prove they got a correct result. > Here is an example (another election I was in) that shows how easy it is: > > > http://www.cs.cornell.edu/w8/~andru/cgi-perl/civs/results.pl?id=E_8e5a1ca7f86a5d5d<http://www.cs.cornell.edu/w8/%7Eandru/cgi-perl/civs/results.pl?id=E_8e5a1ca7f86a5d5d> > > I'd rather just use the best system to begin with rather than have to worry > about when we need to change. How is a Condorcet system superior exactly? According to the Gibbard-Satterthwaite theorem, tactical voting is possible in all non-dictatorial deterministic voting systems. I fail to see how it is more resistant to strategic voting (I can describe a number of methods for you). Furthermore, the small number of methods of tactical or strategic voting that does exist for elections conducted using STV in general only are effective in marginal district... but do we really need to worry so much about strategic voting? Strategic voting really comes into play when there are "parties" involved (we don't have that). And do we want an individual least un-preferred (Condorcet) or most preferred (STV)? The biggest fault that I can see with STV would be that it fails the monotonicity criteria but it is important to note that *no* preference voting system satisfies all the criteria described in Arrow's impossibility theorem. For example with Schulze (the actual system I think you're proposing) it fails to meet independence of irrelevant alternatives, participation, consistency, invulnerability to compromising, invulnerability to burying, and later-no-harm criteria. And to hit the simple point home a bit more, don't you think we'd all have a bit more faith in a system that we could actually figure out the result on our own? Personally, I don't want my vote going into some black box which magically produces a set of winning candidates. I'm not suggesting that STV is the choice we go with but I don't think Condorcet is either "just because". > > > Scott K > > -- > Ubuntu-motu mailing list > [email protected] > Modify settings or unsubscribe at: > https://lists.ubuntu.com/mailman/listinfo/ubuntu-motu > -- Cody A.W. Somerville Software Engineer Red Cow Marketing & Technologies, Inc. Office: 506-458-1290 Toll Free: 1-877-733-2699 Fax: 506-453-9112 Cell: 506-449-5899 Email: [EMAIL PROTECTED] http://www.redcow.ca
-- Ubuntu-motu mailing list [email protected] Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-motu
