[liberationtech] Looking for collaborators for free-range voting project

[Michael Allan]

Ruben and Rich,

Ruben Bloemgarten said:
> It seems I might have jumped the gun, assuming the discussion was
> about voting systems for use in political elections. Disclosing all
> voter data, including voter identity would solve much if not all
> issues regarding verifiability, however would that not also restrict
> the use of such a system to topics that have no political or social
> consequences ?  Otherwise it seems that the removal of
> secrecy/anonymity would be extremely problematic if not out-right
> dangerous.

Rich Kulawiec said:
> I'm with Ruben on this one.  There are serious problems (in many
> cases) with disclosure of how someone voted; there are even problems
> disclosing *if* they voted or possibly if they were *eligible* to
> vote, even if that disclosure only (putatively) is done to the
> voter.

I guess the main concern is coercion and vote buying.  I've discussed
this with others and we foresee some important mitigations.  (These
aren't obvious BAM, and it took us some time to see them.)  *

  (a) Continuous primary voting: Vote sellers can shift their votes
      after taking the money, perhaps re-selling them to other buyers.
      This makes vote buying a poor investment.

  (b) Full disclosure: Buyers, sellers and systematic pressure by
      others (employers, unions, churches, and so forth) are
      detectable by statistical pattern analysis of vote shifts and
      dispositions in correlation with facts (known buyers and
      sellers, workforce structure and dynamics, and so forth).

  (c) Separation of primary from decision systems: Public and private
      voting may be interrelated through separate electoral systems: a
      public vote in the run-up (primary system) culminates in a
      private vote on election day (decision system).  The final
      private vote (secret ballot) filters out instances of individual
      vote buying and coercion.

      A similar strategy may be applied to normative decisions.  Here
      the decisive vote is often not private, but instead restricted
      to a small number of people, such as elected assembly members.
      Concerns of coercion and vote buying are thus *also* restricted
      to that smaller group of people, who may therefore be closely
      monitored and scrutinized.

These should at least prevent skewing of decisions and other serious
harm.  Or have we overlooked something?

I used to point to the harm caused by our faith in the secret ballot,
but now I feel it's the wrong approach.  Whatever we suffer on account
of our political arrangements (we in the West, who have so much else
to be thankful for) is our own fault.  We have the wherewithal to fix
things, and could even proceed a little faster if we wished.


  * From this footnote, which also links to discussions
    http://zelea.com/project/votorola/d/theory.xht#fn-2

Mike