[MLUG - DISCUSSION] electronic voting systems

[Mike Miller <EMAIL:PROTECTED>]

On Tue, 4 Apr 2006, McIntosh Jason wrote:

> On Apr 4, 2006, at 10:12 AM, Mike Miller wrote:
> ---
> > It's pretty amazing and wonderful to think that a change to a fairer 
> > voting system, one that allows more real choices and let's Rick Buford 
> > have a shot, can actually completely alter our political landscape.  It 
> > would happen overnight as soon as we implement a Condorcet system.  Why 
> > hasn't this been done before?  It is used in small elections, but you have 
> > to use a computerized system if there is a large number of voters.  We 
> > have to get a good secure, open-source voting system working.  After that, 
> > we can change to Condorcet and reap the benefits.  Read about it and 
> > promote it in small elections, e.g., in your workplace, in high schools, 
> > etc.
>
> I was thinking of this a while back, trying to find a good "open" voting 
> solution using commputerized/electronic voting methods.  The only 
> solution I've come up isn't a diebold system, but something else like 
> this:
>
> Use electronic voting, which upon completion prints out two receipts. 
> One receipt goes in as an "official" ballot.  You can check this out 
> yourself, see the results.  This is where "officials" are elected from. 
> However, your electronic vote gets stored in a database, and is used to 
> validate the paper vote.  This electronic vote also MUST be able to be 
> verified online in an open way. Call it a generic, unique ID for your 
> vote (i.e. concept of a sequence number).  You can go online to a 
> website, and type in the sequence number off of your receipt to make 
> sure your vote is valid.  There would be no identifying information on 
> the vote (a CRITICAL feature in my view), nor would the number be tied 
> to a person in any way.  It's just a generic "vote number XYZ".

What use would there be in that?  One problem with it is that some people 
could force other people to vote certain ways.  Right now we have a system 
where that can't happen -- you can vote for anyone and lie about it.


> This system would support a number of checks - including the basic check 
> of a) making sure the vote, both paper and electronic were counted 
> correctly, b) reducing some of the overhead and making it easier for a 
> person to vote, and c) allowing a way to publicly verify your own vote.

Having a copy of my voting ballot would not do very much to confirm that 
my vote was counted if there were more than a few votes counted at my 
polling place.  I'm not sure that public verification is a good thing (I'm 
really not sure because I can see a downside).

The electronic system wouldn't have to be much different than what we have 
-- you go to a computer (not networked), fill in your votes, confirm, 
print one copy, bring that copy to the scanner, put it in.  That's like 
what we do today except that we don't print.  It might be possible to 
avoid the computer and just scan forms, but there are many more ways that 
can fail.  Getting the equipment may be an issue, but computers are easier 
to get than voting booths.  The printers could be the tricky part.  It 
could be done a little differently, but I'm not sure how.  We have to 
remember to maintain privacy.


> Thoughts on this?  (As a note, I came up with this idea while back, but 
> the two paper receipts, one for the official tally, the other for your 
> own records is a new idea). Jason

Thanks for replying.  I think this is the most important thing we geeks 
can do for our country.  We have to get the electronic voting systems 
working securely.  It would really be a great advance with wide-ranging 
effects.

Instant runoff voting has already been used in some fairly major elections 
(e.g., in San Francisco).  Instant runoff isn't as good as Condorcet, but 
it is better than what we have and, like Condorcet, it does use ranks. 
So I see widespread Condorcet voting as a real possibility for our future, 
maybe even within my lifetime.

Mike

_______________________________________________
discussion mailing list
EMAIL:PROTECTED
http://mlug.missouri.edu/mailman/listinfo/discussion