[kwlug-disc] Lower voter turnout?

[Chris Frey]

On Tue, Apr 05, 2011 at 07:21:39AM -0400, Russell McOrmond wrote:
>   What do KWLug folks think about online voting?  It was included in
> the Liberal platform as a "solution" to low voter turnout.

As long as the old paper method never disappears, I guess I don't mind
too much if a *workable* system is found.  (Big "if")  This may be a naive
assumption though.  I view the whole Diebold experiment in the US to be a
failure, and would not want that repeated here, in any fashion.

I'm not at all sure that low voter turnout is a convenience issue.
You might get a few more voters, but I'm guessing probably
not enough to justify the system.

On the other hand, if there were a voting system like this available for
citizens to provide regular "referendum" style feedback on specific
current decisions, both binding and non-binding, this might spark
more political interest.  And viewing the security concerns, a non-binding
system might be the best first attempt.


>   Do you feel voting has the same security threat model as banking,
> and thus simple point-to-point encryption solves the problem?  What is
> the threat model for voting, and is only third parties we don't trust
> (IE: do we trust the voter and the body calculating the votes).

*groan*  Old memories of these discussions are coming back to me,
with some friends in the US. :-)  This is definitely a headache in
the making.

You can't trust the voter, because he could be connecting from anywhere
online, and could be anyone.  And you can't trust the body calculating
the votes, unless you have a paper trail.  And you can't have a solid
paper trail unless you have a way of connecting a citizen's vote with
the confirmation he received online.  And you can't make such a paper
trail connection and yet still have a secret ballot. :-)

Or so I remember the debate at the time.


>   Do we trust the individuals/companies that write the software for
> these voting systems?

Only if the result is FLOSS.


> Would making all the software FLOSS solve all
> the potential security threats?

No.  You need to have some way of verifying, publically, that the
servers are running the published code.



>   What if the same "Security" solution was used for voting as was used
> for the Online Census, which only ran on specific platforms and for
> which no third party security audit was allowed?

Probably would get the same kind of support: an outcry of scandal? :-)
I thought that was one of the big complaints about the census being
mandatory a few years ago... with the software and systems being run
by a foreign country.


>   What do people think about the Verified Voting campaign in the USA,
> and do we need one here now that the Liberals have placed online
> voting as a possibility within their platform?  Given what we saw in
> C-32, do we expect the Conservatives to follow suit with similar
> goals?

I'm starting to think paper ballots look pretty good, again.
And if this is all just to solve the voter turnout problem, it is
solving the wrong problem, I think.

- Chris