[kwlug-disc] Say No To Electronic Voting ...
Mikalai Birukou
mb at 3nsoft.com
Tue Aug 25 19:50:38 EDT 2020
> Actually, computerized != centralized
>
> this is why technologists start to look for an online solutions, as
> distributed needs online, which not centralized.
>
>
> It is centralized one way or the other.
> If you have computers in each polling station feeding into a central
> computer
> that tallies the results, then the programmers for the central
> computer can
> be bribed or coerced into flipping every nth vote for some candidate,
> and it
> is game over.
>
> If there is no central computer, there still is a program that has to
> be loaded
> on tens of thousands of distributed computers.
Legitimacy comes from edges. And in the talk we'll see how counting can
be done on edges and so that edges (people) come to consensus.
> That program has to be
> identical on all machines.
Right now I have a program of behaviour in my head. I am 100% sure that
it is not the same as anything in heads of others. At the same time we
follow same protocol!
Just say no to mandated electoral program(s)!
We should have a protocol, following which edges are able to correctly
observe and participate in voting.
When, as a developer, I have a protocol, I develop program for my own
party. I will open source it. But you don't have to trust my code. You
can write your own.
> Again, it requires a specialist to create the program
> and a specialist to audit it, and those can be bribed/coerced. What we
> lost
> here is that the lay person has no way of knowing that a program is fair,
> or tampered with, other than "trust me, I am a specialist". Even if we
> pass
> that point, there is no way to guarantee that what was audited is actually
> what was loaded into the machines.
Just say no to mandated electoral device(s)!
I don't understand use of machines in booths electronic or mechanical. I
don't get. If there is a special place, people came in, someone can
count paper.
When it is voter's device, scheme will be equivalent to an online one. I
subscribe to mantra: I provide a device, and I provide a program that
plays according to voting protocol. Common is only protocol.
> Paper has none of those problems, given enough eyeballs watching the
> whole process.
Paper is not end to end verifiable. Simple online scheme, which I'll
present is end to end verifiable. Verifiability is what people look for.
If there were an end to end verifiable voting, then dictator in Belarus
would've been removed after 4-8 years, no problemo. May be in 12 with
coercion on edges. That is why there is a huge appetite for online
voting tech, its an appetite for verifiability.
This is a spoiler alert for a part of my talk.
Before start of August elections in Belarus, broader society have
pledged money and organization to setup continuous video surveillance of
ballot boxes. There was an official suggestion and an official f-off by
electoral commission. Ultimately, what worked were measures that
increased transparency of the process.
In other words, in places with possible coercion (vote buying) on ends,
we should add surveillance tech to ensure correct counting. See, tech is
a tool, and there is room for improvement even with paper!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://kwlug.org/pipermail/kwlug-disc_kwlug.org/attachments/20200825/09190ea4/attachment.htm>
More information about the kwlug-disc
mailing list