Re: [Membership] [Comment-Mac] Re: Secret ballots

[Kent Crispin <kent@songbird.com>]

On Tue, May 18, 1999 at 10:43:23PM +1200, Joop Teernstra wrote:
> At 00:10 18/05/1999 -0700, Kent Crispin wrote:
> >Re Joop's point about electoral fraud coming from the top: That is
> >not an argument for secret ballots.  Indeed.  That is precisely the
> >problem *with* secret ballots -- they nourish all kinds of fraud at
> >the top.  In particular, the danger of a dishonest election process 
> >is very much lessened by roll call voting.
> >
> I hear no argument here, just an unsupported statement.

It's too obvious to need any support: I was referring to the actual
process of conducting an election.  That does not include means
(legitmate or otherwise) that might be employed to influence people's
votes.  The actual election procedures, you know: the counting of the
votes, etc.  With secret ballots you have the possibility of a
dishonest process.  With roll-call, you essentially have no
possibility of a dishonest process -- it is all visible at all 
times. 

The things you talk about -- vote buying and coercion -- happen 
outside the election procedures themselves.

> Vote buying is done by having a means of verification of the vote and the
> voter.
> In third-world countries, that involves a carbon copy of the ballot handed
> over to the paymaster.

You are making my case, Joop.  Obviously, if someone is in a position
to simply demand a cc of the vote, then secret ballot does not help. 
If a company tells its 3000 employees to join ICANN and vote, it is
almost certainly the case that the substantial majority of them will
vote the way the company wants, secret ballots or not.  The employees
will know nothing of ICANN, or the issues at stake.  However, with
secret ballots we will not be able to detect such block voting. 

> Do you  want to reduce the procedure to plain data matching?
> Intimidation and harassment are also a lot easier when you know whom to
> target.  In a digital world, you are talking nightmare.
> I am astonished that you can advocate against the secrecy of the ballot.
> Do you know nothing of open-vote abuse under dictatorial regimes? 

Please -- You are hyperventilating.

Every voting method has trade-offs between expense, security,
complexity, and other factors -- for example, the desire for
verification of voter identity vs the cost of such verification. 
There is no perfect voting method; every one has possibilities for
abuse. 

There are many many different voting models, and many different
situations in which votes are conducted.  An election in a 30 person
chess club is not the same as an election in a country ruled by an
oppressive dictator; the votes of a corporate board are not the same
as the votes carried out in the IETF.  There are many environments
where roll-call voting is clearly acceptable; there are some where it
is not.

Therefore, the choice of voting method depends greatly on the actual
characteristics of the organization in question.  My advocacy for an
open voting protocol is based on my view of the *actual*
characteristics of ICANN, not on hysterical and odious comparisons 
with dictators.  (*)

To review those characteristics:

1) The membership of ICANN will be distributed internationally
[therefore, barring unusual circamstances that will require other
safeguards anyway (**), no dictatorial regime will control a substantial
fraction of the votes]

2) The international character of the ICANN membership makes it
hard to create a audit procedure for secret ballot that all the 
members can trust:  what reason does a person in, oh, say, Namibia, 
have for trusting an audit conducted in Los Angeles?  The 
distributed nature of the membership simplifies the the task of a 
corrupt voting procedure...

3) The ICANN Board has no realistic means of retaliating against 
individual members for their votes -- for example, the ICANN board 
is in no position to take away a domain name.

4) The subject matter of concern to ICANN is restricted and arcane,
and ICANNs powers are limited.  Despite the ravings of a few, ICANN
is *not* a general Internet government.  It is not a dictatorship,
and it can't, for example, threaten my family. 

5) Online voting is much to be preferred, but security in online 
voting must be considered carefully.  You mention the "digital 
nightmare" above -- the digital nightmare is much worse when you 
have a single secret point that all votes must filter through.  It 
would be straightforward, for example, to simply alter the votes as 
they arrived.


> >Re Diane's point: I would be interested to know why she thinks the
> >individual members privacy is so important, while the Board members
> >privacy is not.  All the potential abuses that secret ballots
> >presumably guard against are present in spades at the director level
> >-- vote buying and coercion are likely to be most effective when
> >applied to directors.  Why don't the demands for openness and
> >transparency come down to the individual member level? What,
> >precisely, is the value of being able to conceal your vote, and why
> >is it valuable at the individual level, but not at the director
> >level?
> >
> Diane can speak for herself. 
> >From me just this: ICANN directors are servants of those who elect them to
> public office. They may forget that after the elections.
> You are not even aware of it *before*.

Tsk, tsk.  Always with the knee-jerk ad hominems.  I'm surprised you 
can walk ;-)


(*) Note that, unlike your implied equivalence between ICANN and
dictatorships, Dr Quaynor's point was that a dictatorship could
control the votes of *local* ICANN members.  To repeat, my reply to
that was yes, it's possible, but the ICANN membership is
international in scope, the reach of a local dictator will be
limited, and thus control of a few local votes will not translate to
any significant power base in ICANN.  Diane mentioned that regional
voting pools might affect this: But the proposed regional voting
pools are all large, and include many countries.  Thus the payoff for
controlling local votes is problematic, at best.  Furthermore, I
offered the possibility of providing secret voting id's in extreme
cases. 

(**) The circumstance of interest is where some regime or religious
group or terrorist group or large corporation tries to flood the
membership of ICANN with a huge number of new members.  Note that
secret ballot does nothing to safeguard against such actions, and in
fact may help them, by concealing voting patterns.  This problem is,
IMO, a *far* more realistic and serious concern than intimidation,
and secret ballots do nothing to solve it. 

Note that the MAC solution (hoping for a very large membership so
that capture becomes difficult) is not a solution, but a hope.  There
is a *fundamental* difficulty with a self-selected electorate that is
a tiny portion of the total population... 

-- 
Kent Crispin                               "Do good, and you'll be
kent@songbird.com                           lonesome." -- Mark Twain