[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[ga-full] Re: [ga] Matters to discuss ... Am I a member?



On Wed, Feb 23, 2000 at 08:25:19AM +0100, Harald Tveit Alvestrand wrote:
> At 15:12 23.02.00 +1300, Joop Teernstra wrote:
> >At 17:47 22/02/00 -0800, Kent Crispin wrote:
> >
> > >
> > >It is not necessary to publish the names for any purpose related to
> > >voting.
> > >
> >
> >It is. People need to see if they are not omitted from the roll.
> 
> This purpose can be served by sending out individual email ballots; if you 
> didn't get yours, you're not on the roll (or you have to contact the admins 
> to correct your email address).
> 
> The other purpose (to assure people that the rolls are fairly kept, with no 
> "strange" additions) cannot, I think, be met without publication.

Actually, there are other ways to do that, as well -- all that is really
required are auditors.  If you have a set of mutually suspicious
auditors the level of trust required for any given auditor can be quite
low.  (In fact, one can almost think of a public list as a just a
limiting case, where the set of auditors is equal to the set of 
voters.)  [The votebot supports an arbitrary set of auditors.]

I may be conveying the wrong impression, though -- I *prefer* open
publication as the simplest way of accomplishing integrity in elections
-- other methods quickly become complex and/or insecure.  

More: I believe that publication at some point and in some form is a
necessary component of fair (email) elections.  One can imagine, for
example, a voting system where a list of randomly generated voter ID's
is published, each voter knows their own ID, and the auditors are the
only ones who can see the full association between the IDs and the
actual voters... 

kent

-- 
Kent Crispin                               "Do good, and you'll be
kent@songbird.com                           lonesome." -- Mark Twain
--
This message was passed to you via the ga-full@dnso.org list.
Send mail to majordomo@dnso.org to unsubscribe
("unsubscribe ga-full" in the body of the message).
Archives at http://www.dnso.org/archives.html