Free Range Voting and Network of Trust projects

[Hans Brucker]

Mike,

Thanks, these approaches are very interesting. Just wanted to make the 
point that they seem to be a tad complex overall for an initial 
implementation   :-)

Cheers
Hans



On 17.01.2010 07:51, Michael Allan wrote:
> Hi Hans,
>
> If there are misunderstandings, then I'm mostly to blame.  I neglected
> to document the technical details, especially for the voter register.
> I'm just starting on that, here:
> http://metagovernment.org/wiki/Streetwiki
>
> It's a work in progress, and all are welcome to contribute.
>
>    
>> It looks like you are trying to do too many things at once. For example in
>> another discussion you are talking about voters with addresses and voters
>> without addesses. Next step would be voters without names, like a Wikipedia
>> IP-only edit or so, I guess.
>>
>> Here it makes sense for an early implementation to restrict the problem
>> domain by (for example) defining that a voter always has to register with
>> some kind of objectively verifyable personal information, being it his
>> address, a copy of her phone bill, her cell/mobile or her social security
>> number. This scenario would be much more realistic, in my opinion.
>>      
> Thomas was probably talking about separating the registration and
> voting identities through a "proxy blind" (I think that's the correct
> term).  We haven't documented that, yet.  This is just a stub:
> http://metagovernment.org/wiki/Streetwiki#Proxy_blind
>
> It would be too much to attempt (as you say) for the first cut of the
> streetwiki.  We'll need at least an additional week to prototype a
> proxy blind.  (Then the possible exploits and weaknesses are
> mind-boggling!  But we can at least float something.)
>
>    
>> Another issue is that if you require n+1 people to grant level n trust
>> which is sufficient for voting, this automatically means that you can game
>> the system with n+2 people, because at that level you can create phonies
>> which can create voting phonies. To avoid this a much more elaborated
>> system of trusting dependencies is required than the one
>> described (at least as I understand it right now).
>>      
> http://habermas.liqd.de/W%C3%A4hler-Registrierung
>
> I think what Thomas was saying here, is that in order to increment
> your trust level (currently at M) you need to find at least M+1
> trusters, each of whom is already at a level of M+1 or higher.  Then
> you'll rise to M+1, too.
>
> I've documented the use case for detection of phonies here:
> http://metagovernment.org/wiki/Streetwiki#Policing_for_local_sock_puppets
> (Sorry for the misunderstandings.  I should have documented it
> earlier.)
>
> Comment and critique are welcome.  If anyone sees additional problems
> - or if the explanations are unclear - please take a moment to point
> it out.
>
>    
-------------- next part --------------
-- 
You received this message because you are subscribed to the Google Groups "Votorola" group.
To post to this group, send an email to votorola at googlegroups.com.
To unsubscribe from this group, send email to votorola+unsubscribe at googlegroups.com.
For more options, visit this group at http://groups.google.com/group/votorola?hl=en-GB.