Sunday, January 30, 2011

More on New Hampshire Approval Voting Bill

The story has made it all the way around my corner of the internet and found its way to Slashdot (which links back to how I ended up writing about this topic in the first place), so now would be a good time to summarize some of the arguments and counter arguments I've been having with people about this:

First, look over to the right and take note of the pink-highlighted link, What Do You Mean By Best?. There I talk about Arrow's Theorem, the debate over voting systems and voting system criteria, and how computer simulations can help us resolve it. Approval voting and range voting stand head-and-shoulders above other alternatives. This is especially important to the Slashdot crowd, because many Linux distributions, as well as Wikimedia, all perform their elections using a Condorcet method, and so many have come to accept that Condorcet methods are the way forward (although the Fedora project uses range voting). Regular readers will of course recall that Condorcet's ideal is better-met by approval voting than by any actual rank-based "Condorcet method".

The bill is having committee hearings next week; I'll continue to keep you posted.


  1. Sounds far superior to the current system.
    I think voting papers would need a "number of candidates approved box" to prevent voting paper tampering.

  2. That's an interesting idea.

    I wonder though if the gains in tamper-resistance would be offset by the increase in accidentally spoiled ballots.