[Noisebridge-discuss] Voting experiment.

[Christoph Maier]

On Fri, 2010-09-10 at 10:51 -0700, Jacob Appelbaum wrote:
> On 09/09/2010 07:23 PM, Sai Emrys wrote:
> > On Thu, Sep 9, 2010 at 9:08 PM, Jacob Appelbaum <jacob at appelbaum.net> wrote:
> >> It's rare in a voting system to give out extra ballots to people who say
> >> they're representing some other people who aren't at the polling place.
> > 
> > ... except in corporations, where proxy voting at meetings is
> > standard. And there are plenty of ways to relatively well authenticate
> > the proxy claim.
> 
> Yep. In profit oriented, privacy unfriendly, non-equal, non-democratic
> systems - nearly everything is possible. Especially when you want to
> have a majority overrule a minority and force them into doing something,
> rather than failing to achieve cohesive and unified action.
> 
> I know - lets design a new evoting machine. Perhaps we'll be the ones to
> make it un-hackable! We can ask India, they seem to have no problems
> with their system at the moment. :-)
> 
> All the best,
> Jake

"It's not the people who vote that counts. It's the people who count the
votes." - J Stalin

I.e., the key part of new evoting machines aren't so much the machines
themselves, but an improvement of double-entry bookkeeping, namely,

1. IMMEDIATE dissemination of any voting transaction to independent,
distributed transaction logs in multiple locations under DIFFERENT
jurisdictions, 

2. redundant communication channels between voting and transaction log
locations that cannot be interrupted, delayed, or falsified, without
leaving noticeable traces.

3. open sourcing ALL the technical details of the voting infrastructure.


Of course, there's always René Goscinny's fallback solution:
First, you hold a vote, then you throw all the ballot boxes in the
ocean, and the strongest one wins.