Voting, procedures, costs, and privacy. Was: Re: [governance] Innovation

[Karl Auerbach]

Norbert Bollow wrote:

> Sounds like it might be wise to establish the internet governance
> equivalent of genuine, independent unions.
> 
> Given that such a new structure aiming to genuinely represent
> internet users would probably have to be operated on extremely
> limited financial resources (since otherwise it would quickly
> lose its independence through being financially dependent on
> its donors), how could it be set up to make it genuinely
> democratic and robust against forgery of votes etc?

Let's not try to solve the internet voting problem in its general case. 
  In that general case one tries to permit everything from registration 
to voting to occur over the net.  That means having means so that each 
human gets no more and no less than one ballot and one vote.  That's a 
tough problem to solve - in fact it may not be solvable.  And there is 
the separate problem of privacy of the votes so that voters can't be 
coerced.  That's another problem that is extremely difficult to solve.

So, any election process needs to have its requirement softened and/or 
to tie itself to some external, hopefully already pre-existing, mechanisms.

One such mechanism was used in the proposal made to ICANN for a 
constitutency for individual domain name owners.  In that case we tried 
to ride on the coattails of the credit card companies that most people 
use to pay for domain names.  That wasn't completely satisfactory - 
because while it largely solved the problem of fictitious created people 
(for the most part, but clearly not completely, fictitious people don't 
have credit cards that they've used to pay for domain names) but it did 
not solve the problem of one person having many cards and thus 
potentially getting multiple votes.

Paypal's system of making small deposits to bank accounts is also an 
interesting method.

At the end one may end up with a sieve of techniques that reduce, but 
not eliminate the problems of fictitious people or one person with 
multiple votes - in this case the question is when the risk of such 
problems becomes acceptably low.  Trying for perfection pretty much 
means that we will never get anywhere.

The other problem - coercion of the internet voter as he/she casts 
his/her ballot - is a problem that perhaps we can solve by declaring it 
a non problem.  Maybe the issues faced in internet governance will be 
such that the drive to coerce does not become a real problem.  Maybe - 
I'm merely suggesting that this question should be asked.

There's also the notion of voter privacy during the canvasing process, 
i.e. the vote counting process.  Again, for the kinds of matters that 
arise in the context of internet governance we should ask whether this 
is a real issue or whether we can simply hire some people to do the 
counting and impose upon them some solid contractual obligations not to 
disclose what they see (and perhaps add a bonding requirement as well.)

As for the costs - In the case of the company unions, the company paid 
for the meeting halls and the beer.  Of course that tended to influence 
the opinions of the members - particularly when more beer was provided.

So, for a body to really be independent it needs to cover its own costs 
from a source that is disinterested in the outcome.  That sort of 
disqualifies the body of internet governance that the election is 
associated with.

Assuming we've solved or are dismissing the coercion/privacy question, 
then the main cost is the registration - identifying who is whom - the 
actual voting itself isn't all that expensive.

For example, when I vote my shares in a shareholder election the main 
cost is in making sure that I have a unique voter ID number.  The actual 
dissemination of voting materials and the actual voting is done by a 
relatively simple web mechanism.  Again that's for corporate stuff where 
the privacy requirements are less than in a political election - and I'm 
hoping that internet governance voting has privacy concerns more on par 
with what is the norm in the corporate shareholder election rather than 
the typical political election.

The bottom line is, however, that the costs, whether high or low, need 
to be borne by the body itself.  Otherwise there is systemic compromise 
of the integrity - it does not matter whether that compromise is real or 
perceived - of the system, i.e. an ALAC.

		--karl--




____________________________________________________________
You received this message as a subscriber on the list:
     governance at lists.cpsr.org
To be removed from the list, send any message to:
     governance-unsubscribe at lists.cpsr.org

For all list information and functions, see:
     http://lists.cpsr.org/lists/info/governance