[MG] Pollserver Federation - Crosstalk between voting systems

[Alex Rollin]

On Wed, Jan 5, 2011 at 7:49 PM, conseo <4consensus at web.de> wrote:
> It shouldn't not be personal though.

My point is exactly that.  I hope I can be clearer about it.  Let me try again:

try not to take it personal if I come off as critical; it is not meant
to be personal; I am doing my best to focus on the objectives and
issues, in my own human and limited ways; I am doing my best to remove
anything personal from what I'm  saying.

On Tues, Jan 4, 2011 Alex Rollin wrote:
>> Based on this need, then, I would say that there is a clear need for
>> at least 1 standard that allows all polling facilities to implement
>> some form of identity check (that can be verified) to reach a certain
>> level of "certainty" that adequate steps, to a certain level of
>> certainty, are being taken to eliminate duplicates.
>
> That is a *very* serious problem in my perspective which has been tackled
> by Mike with a streetwiki/trust-network approach already.

I am aware of this approach and I find it to be inadequate to my needs
and use case as my polity crosses extended geographic territory.

> The approach that Thomas has proposed to me, was to create "admins" in
> communities which are both trustworthy to the network and the member and
> can prove the identity of every voter when queried. You would have a proxy
> then who protects your official identity from the network. In a trust
> network you can likely create valid admins out of the network as well.

I am going to be implementing some form of process for identification
verification within my own system that will be open to third party
verification of some form.  This will make me an "admin" of this type,
and a "central authority."  A standard is the key to spreading this
power out, though, as I am not interesting in monopolizing identity
verification, and I don't think a monopoly serves anyone.  I think
that we should work together on something like this.
http://en.wikipedia.org/wiki/Third-party_verification
http://www.idology.com/
At which point I will issue something like an RSA key:
http://www.rsa.com/node.aspx?id=3347

Like a bank, basically.  In Europe this type of authentication is very
common for online banking, and I was issued a complex system with a
USB key for my computer when I lived in China.  It is growing rather
commonplace; there are reasons it is not perfect, but I cannot put
everyone in a single neighborhood right now so they can knock on
doors.  My system involves people voting for the disposition of real
assets.  I will already have a need of their identity documents in
order to enable their legal participation.

I will write a standard for what I do and open myself to an external
audit of my procedures and hope that other polling providers would do
something similar.
>
> Still I don't know how you can avoid people registering with different
> admins. Maybe the admins should visit each other regularly and compare addresses and identities?

I'm primarily concerned about the people in "my system" who have a
legal right to vote.  If other people think my standard is gold, then
perhaps it will be sufficient to grant my users access to their "other
systems" as well.  I do not have need of address verification yet, but
at some point I will, I am sure.  In the meantime RSA with central
authority to reset and reclaim accounts is sufficient, and auditing
this system will come into play.

> +1 But we still need to implement it. A standard without a reference
> implementation is pretty useless.
If several projects are working along the same lines drawing the
conversation together towards a standard and a consensual "mandatory
minimum implementation" the standard is Definitely not useless; far
from it because, as Owen has pointed out, it is something to which we
could all sign our names as a shared strategic plan and draw our
communities together on.  Of course I want to implement it and will.
My point was that my implementation may exceed the mandatory minimum,
but I'd like to start the conversation there and connect what I'm
doing with the work of others on this list.


Alex



Originally posted to the mailing list of the Metagovernment Project:
http://metagovernment.org/mailman/listinfo/start_metagovernment.org