Sequoia, Diebold Voting Machines

2 Nov 2004 - 9:35am
9 years ago
5 replies
792 reads
Timothy Shey
2004

So, when anyone voted today, or in the past weeks, did you use either
the Sequoia or Diebold electronic voting machines?

I voted in D.C. this morning, where each polling place has at least (or
maybe exactly) one Sequoia AVC Edge voting machine (
http://www.sequoiavote.com/productguide.php?
product=AVC%20Edge&type=Introduction ) and found the user experience
left a lot to be desired; it wasn't even equal to the farecard machines
at the D.C. metro.

Most disconcerting seemed to be an outright bug, or a poorly designed
feature: several times, when attempting to select a candidate on the
ballot by pressing the bullet next to his or her name, I was then
presented, for no apparent reason, with a write-in candidate keyboard
input screen (the write in option was at the bottom of the onscreen
ballot, in most cases nowhere near my choice), with the only way to
return to the choices a small "cancel" button at the bottom of the
screen, far below the touchscreen keyboard. I can only imagine the
havoc it is wreaking for people less familiar with submit/cancel
buttons, or who then tried to write in the name of the candidate they
just tried to select.

At best, it slows down the process, as it did for me, when it happened
four times in my case. At worst, it's misrecording votes. Even in
D.C., where the results of the Presidential and Congressional Delegate
tickets are all but foregone conclusions, there are still important
races, like Council-at-Large and the School Board, which could be
skewed by poor voting interfaces.

I've seen the recent articles about Sequoia's code being leaked (Wired
News, Oct 29: http://www.wired.com/news/privacy/0,1848,61014,00.html )
and issues with Diebold machines, but was wondering if any critiques
have been posted of the UX itself, or if anyone else can share their
experiences voting.

There's a demo of the Sequoia e-voting experience here:
http://www.sequoiavote.com/democenter.php

Two other things that were disconcerting: a printout is produced after
you vote, spooling out of a unit below the screen, but nothing human
readable that you can review to confirm (for prvacy reasons), and no
receipt for the voter to keep to prove how they voted. Just a
confirmation onscreen - then you must simply trust the vote was
recorded. Additionally, the magnetic strip card that one is given to
vote with, when handed off at the ballot box, was just left in a stack
on a chair. Now, likely, there's no damage done if those cards are
lost, or attempted to be reused, and this was explained to me when I
asked about it; but for a voter, it's unnerving to see the card you
just used for your private vote left lying around, when everyone else's
paper ballots are being fed into a secured reader.

___________________________
Timothy Shey
Creative Director, Co-founder
Proteus, a Telitas Group Company
http://proteus.com
http://telitasgroup.com

Comments

2 Nov 2004 - 9:57am
whitneyq
2010

Thanks for the great report.

If anyone else encounters usability/design-related problems with voting, we
have created an online form to collect those notes (Timothy, may I add
yours?) Other groups are focusing on security...we are focusing on what we
do and the human aspect of voting.

The collection form: http://www.surveymonkey.com/s.asp?u=17812688584

Just a comment on one of the things that Timothy said:

At 09:35 AM 11/2/2004 -0500, Timothy Shey wrote:
>Two other things that were disconcerting: a printout is produced after
>you vote, spooling out of a unit below the screen, but nothing human
>readable that you can review to confirm (for prvacy reasons), and no
>receipt for the voter to keep to prove how they voted.

No one ever gets a "receipt" to prove how they voted. This is one of the
big misconceptions about verified voting. The point is to not only allow
the voter to confirm their vote, but to produce a tangible record that
could be used for either a recount or an audit.

The word receipt started being used because it has a strong metaphor for
most people, but the metaphor is wrong.

The Mercuri Method, which is in use in Nevada, prints out the vote entries
on a paper, but displays it to the voter under a glass panel. They can
inspect, but not take it. The paper is then cut and dropped into a box -
this protects anonymity by randomizing the order, so no one can tie "voter
#525 on this machine" to a connected entry on the paper spool.

There are other solutions to both allowing both electronic vote selection
and an auditable paper trail. There are several systems which use the
computer to mark a ballot, which is then read with an optical scanner. This
eliminates many of the errors in opscan (generally caused by a
less-than-perfect mark in the bubble or arrow). One of the systems prints
the entire ballot. Another uses a pre-marked ballot. In either case, one of
the advantages is that all ballots end up in the same medium and are all
counted the same way.

>Additionally, the magnetic strip card that one is given to
>vote with, when handed off at the ballot box, was just left in a stack
>on a chair. Now, likely, there's no damage done if those cards are
>lost, or attempted to be reused, and this was explained to me when I
>asked about it; but for a voter, it's unnerving to see the card you
>just used for your private vote left lying around, when everyone else's
>paper ballots are being fed into a secured reader.

I agree that this is unnerving - but the card does not actually record your
vote in any way. All it is is a one-time key that activates the machine.

By the way - the reason there is probably only 1 electronic machine in the
precinct is that it is there to meet HAVA mandates for providing accessible
voting systems to people with disabilities.

I'd be interested if anyone has voted with an audio ballot, or used any of
the visual enhancements on any of the machines.

Whitney Quesenbery
Whitney Interactive Design, LLC
w. www.WQusability.com
e. whitneyq at wqusability.com
p. 908-638-5467

UPA - www.usabilityprofessionals.org
STC Usability SIG: www.stcsig.org/usability

2 Nov 2004 - 4:55pm
Thomas Vander Wal
2004

Thank you for the UPA survey link. I just posted my poor experience
with a Diebold machine. I could not figure out how I would change my
vote should I make an error. I am completely uneasy with the fact
that there is not a paper trial for my vote either. If the interface
is as poorly designed as it was, I really am not trusting what is
beneath the unhelpful screens.

All the best,
Thomas

On Tue, 02 Nov 2004 09:57:21 -0500, Whitney Quesenbery
<wq at sufficiently.com> wrote:
> [Please voluntarily trim replies to include only relevant quoted material.]
>
> Thanks for the great report.
>
> If anyone else encounters usability/design-related problems with voting, we
> have created an online form to collect those notes (Timothy, may I add
> yours?) Other groups are focusing on security...we are focusing on what we
> do and the human aspect of voting.
>
> The collection form: http://www.surveymonkey.com/s.asp?u=17812688584
>
> Just a comment on one of the things that Timothy said:
>
> At 09:35 AM 11/2/2004 -0500, Timothy Shey wrote:
> >Two other things that were disconcerting: a printout is produced after
> >you vote, spooling out of a unit below the screen, but nothing human
> >readable that you can review to confirm (for prvacy reasons), and no
> >receipt for the voter to keep to prove how they voted.
>
> No one ever gets a "receipt" to prove how they voted. This is one of the
> big misconceptions about verified voting. The point is to not only allow
> the voter to confirm their vote, but to produce a tangible record that
> could be used for either a recount or an audit.
>
> The word receipt started being used because it has a strong metaphor for
> most people, but the metaphor is wrong.
>
> The Mercuri Method, which is in use in Nevada, prints out the vote entries
> on a paper, but displays it to the voter under a glass panel. They can
> inspect, but not take it. The paper is then cut and dropped into a box -
> this protects anonymity by randomizing the order, so no one can tie "voter
> #525 on this machine" to a connected entry on the paper spool.
>
> There are other solutions to both allowing both electronic vote selection
> and an auditable paper trail. There are several systems which use the
> computer to mark a ballot, which is then read with an optical scanner. This
> eliminates many of the errors in opscan (generally caused by a
> less-than-perfect mark in the bubble or arrow). One of the systems prints
> the entire ballot. Another uses a pre-marked ballot. In either case, one of
> the advantages is that all ballots end up in the same medium and are all
> counted the same way.

3 Nov 2004 - 6:07am
Martyn Jones BSc
2004

Not sure if you've seen this article yet, Wired has run a little piece
concerning difficulties with the machines:
http://www.wired.com/news/evote/0,2645,65579,00.html?tw=wn_tophead_2

Of particular interest to IxDers:
> Michelle Shafer, spokeswoman for Hart InterCivic, said the problem that
> occurred in Texas with her company's machines were caused by voters rather

> than by the machines. The Hart machines are not touch-screen machines but
> instead use a wheel that voters turn to make their selections. Shafer said

> after choosing the straight-party option, many voters turned the wheel to
> manually go through the races and click their choices individually to
> emphasize them, not realizing that in doing so they de-selected their
> choices. Shafer said they probably then mistakenly moved the wheel to
> select a candidate from another party.
>
> "It's not a machine issue," Shafer said. "It's voters not properly
> following the instructions."

Try using that statement in your next design evaluation!

Martyn

----------------------
Martyn Jones BSc
Interaction Designer
Kode Digital Ltd.
----------------------

3 Nov 2004 - 4:57pm
M S
2006

Did anyone else use the WINvote touch screen machines, as we had in
Arlington, VA? I had a mixed experience.

Good experience (elements roughly in order as I encountered them):
1. Good match between screen size and the "chunks" of ballot
information shown on each screen
2. Big targets to touch
3. Forward and back buttons and a summary at the end (no paper trail)
4. A "lock" on selections. Once a selection is made any other touches
do not change the selection (except touching the same selection, which
unselects it, then another selection can be made).

Bad experience:
1. My wait in line gave me a chance to watch a tech repairing a machine
for at least 15 minutes. Of course I was assigned this machine.
2. The name: "WIN" anything reminds me of an unreliable desktop OS (an
all-too-easy explanation of #1)
3. No apparent audit trail.

Tom Selsley
Arlington, VA

> So, when anyone voted today, or in the past weeks, did you use either
>
> the Sequoia or Diebold electronic voting machines?

__________________________________
Do you Yahoo!?
Check out the new Yahoo! Front Page.
www.yahoo.com

4 Nov 2004 - 2:00pm
Listera
2004

Tom Selsley:

> 3. No apparent audit trail.

We don't get to vote on these fancy schmansy machines in New York City, I
guess, we love our mechanical stuff from the 20s. But I never understood the
paper -> audit trail argument. Paper/paper-trail can be easily forged,
perhaps more easily than a properly designed digital one. The future of
securing ballots may not be paper after all.

Ziya
Nullius in Verba

Syndicate content Get the feed