Re: WHAT THE DEMO WILL DEMONSTRATE -- V.001

From: Matt Shomphe <mshomphe_at_comcast_dot_net>
Date: Sun Aug 24 2003 - 10:11:52 CDT

There are only a few things that I want to address:
(1) Will we actually tally the votes in the end? Has anyone thought about
how to do this? This may be fairly trivial, but it may be the most
important part!

(2) From below:

13. WRITE-IN VOTING WITH ELECTRONIC SYSTEM

Selecting "Write-in" will bring up an on-screen typewriter. A QWERTY layout
will appear on the top half of the screen (three rows plus space bar below).
Under the space bar, we'll have two (or three rows) of letters arranged
alphabetically.

Why will there be both a QWERTY layout and an alphabetical layout?

(3) Using commodity printers: I have nightmares about printer jams,
etc. I don't see any way around it, but it's just something to be aware of.

(4) Security. I'm no security expert, but I think we need to assume that
the system is hackable. Maybe it's beyond the scope of this project, but
perhaps we should consider defining how to detect hacks.

In addition, Alan has stated that there should be clear boundaries between
QA and Dev. If it's acceptable to the group, I'll withdraw from any
development to work exclusively on QA. Does anyone have any issue with this?

Matt

At 03:31 PM 8/20/2003 -0700, Alan Dechert wrote:
>WHAT THE DEMO WILL DEMONSTRATE
>
>I have been somewhat derelict about producing this document. There has been
>some confusion about just what the demo is about, and I hope to clarify that
>now.
>
>This is NOT the definitive word on this. I ask for your input, and I expect
>that this document will go through a few iterations before we are all in
>agreement. Now is the time to speak up! We need to reach a consensus on
>what features will be there.
>
>I think this is already a lot. We need to carefully consider the costs of
>adding things -- especially if they will wind up pushing completion much
>further away.
>
>
>1. ELECTRONIC VOTING
>
>This may seem a little obvious to people close to this project, but most
>people have not witnessed electronic voting and they know very little about
>it. According to Brady's Counting all the Votes paper (9/01), about 10% of
>the counties in the USA used electronic voting machines, with about 12 % of
>the ballots cast on such machines in 2000. These numbers were higher in
>2002 and will be even higher in 2004, but still the vast majority of people
>have never seen one of these new-fangled 'lectronic vote machines.
>
>A more subtle aspect of this subject has to do with the fact that every
>other voting machine also involves a certain level of electronics -- from
>the way the ballots are produced and read to the way that votes are
>tabulated.
>
>2. COMMODITY PC BASED VOTING
>
>This has been demonstrated in Australia, but only in a pilot project. The
>percentage of people that have cast votes on such a system is very small in
>Australia and near zero elsewhere in the world. We have to show how the CPU
>will be situated so that it looks reasonably secure. For example, we don't
>want the user to see the CPU and say, "I wonder what happens if I push this
>button?" or "I wonder what will happen if I hit Ctrl-Alt-Delete?" We don't
>want media popping out. Only the display and mouse (or stylus) should be
>accessible. Most likely, the monitor will be situtated horizontally (in
>some custom made wooden cradle) sitting on a table. The CPU will be under
>the table securely situtated.
>
>Most people are not aware that the touch screen DREs being sold today are
>really just repackaged commodity PCs. So, our system is not really that
>different. It's just that the repackaging is a lot simpler and the PCs can
>go back to becoming regular PCs right after the polls close.
>
>3. CHEAPER ELECTRONIC VOTING
>
>DREs tend to run around $4,000. They usually charge something like $3,000
>for the unit and another $1,000 or so for warranties, service, etc. We will
>show that the cost of our unit could easily run less than $1,000 (on
>average -- the touch screen will cost most but most machines will use a
>mouse) to purchase from a vendor. We need to make a comparison chart that
>we can hand out -- maybe have a large chart for display at the press
>conference.
>
>We will also explain the possibility that vendors will just rent the systems
>to the counties. In this scenario, a PC remarketer could take existing PCs
>from inventory -- sanitize them, test them, deliver them to the polling
>places and then, after Election Day, put them back on the road to wherever
>they were headed. In this case, a PC might only be used once as a voting
>machine. This scheme obviates the need for counties to store (and protect)
>their voting machines. We need to produce some flow chart showing how this
>would work and provide some breakdown on the economics -- get some numbers
>on what it costs Riverside county to store their 5,000 (whatever) DREs.
>
>4. VOTING WITH HI-RES LARGE SCREEN
>
>Most DREs have pretty small displays -- some older ones are monochrome. We
>will utilize large hi-res color displays. At least 17 in. diagonal at 1280
>x 1024 resolution. This means that voters can see more of the ballot more
>clearly and they won't have as many steps or page turns as with systems that
>have smaller displays. This is in accordance with new FEC guidelines for
>voting equipment that calls for minimizing these steps and page turns.
>
>
>5. ELECTRONIC VOTING SYSTEM THAT PRODUCES A PAPER BALLOT
>
>This is a profound concept! Almost no one has ever seen or even thought of
>such a thing (I said "almost," rememeber. Actually, the Populex system does
>this but it's a crappy system -- not sure if it's used anywhere). Our
>system is not just a DRE that produces a paper receipt. The official
>authentic ballot will be produced by on an ordinary PC with a laser printer.
>
>We'll probably use an inexpensive compact laser printer, like maybe an HP
>LaserJet 5L that can be had on eBay for 50 bucks. The printer will have to
>be situated in the voting booth so that someone walking by will not happen
>to be able to view the printout.
>
>Besides providing the paper trail that David Dill and hundreds of other
>technologists demand, it eliminates the need for a pre-printed ballot. This
>is a large cost savings.
>
>The printout will only show the selections made -- not all the choices for
>all the races. This improves clarity and economy. Imagine the ballots for
>the runoff election we're having in California in October. If our system
>was available now, the printed ballot would only show maybe 5 items (recall,
>replacement candidate, and 2 or 3 ballot measures). But the printed ballots
>we'll use as it is will list the names of 135 candidates -- a waste of
>paper.
>
>Overvotes will be impossible while undervotes will be highlighted with "NO
>PREFERENCE INDICATED," or similar wording -- but not "none of the above"
>(too controversial -- raises questions about actionable none of the above
>plurality).
>
>6. ELECTRONIC SYSTEM THAT CAN RENDER LARGER TYPE ON SCREEN
>
>Some DREs already do this, but many older machines cannot increase the font
>size. Our demonstration will display a normal sized ballot on one screen in
>a font size that normally sighted people will be able to read without
>difficulty. But we will include a button that will increase the type size.
>The re-sized on-screen ballot will have to be on multiple pages and we'll
>provide a navigation button (actually 3 buttons <- prev - return to normal -
>next->). We will only provide 2 sizes in the demo: nomal and large
>(probably twice the size).
>
>The production system may have more options for changing the display to
>accommodate people with some vision problems. We can figure out what those
>other options will be and state that we will have them in the production
>system.
>
>7. ELECTRONIC VOTING MACHINE FOR THE BLIND THAT INCLUDES A PAPER PRINTOUT
>BUT WILL NOT COMPROMISE SECRECY AND STILL PROVIDE VOTER VERIFICATION
>
>The voter will hear the selection through headphones and indicate selections
>with a keypad or buttons. Other DREs have this feature. Some claim that
>adding a printer will compromise the blind voter's right to a secret ballot.
>It's essential that we demonstrate how our system will protect voter secrecy
>and still provide for voter verification of the printout.
>
>This will be achieved by the use of a bar code on the long edge of the
>printout (duplicated on opposite edges -- say within a half inch of the
>edge). The blind voter will be able to take the printout, put it in a
>privacy folder (a manilla file folder cut to 12 X 8 inches). This means
>that the barcode (not human readable) will be exposed but the printing on
>the ballot will be covered. There will be another PC set up with a scanner
>and headphones so that the voter can put on the headphones and hear the
>selections read back.
>
>8. PROCEDURE FOR DEPOSITING BALLOT IN BALLOT BOX
>
>We will have a locked ballot box with a slot and a ledge on which the ballot
>(while inside the privacy folder) will be place so that the printed side is
>down (the exposed bar code also makes it easy to identify which side has
>printing on it). The pollworker will then slide the ballot through the slot
>in a way that the printing is never exposed to view.
>
>9. TOUCH SCREEN VOTING WITH DENSE BALLOT AND STYLUS
>
>Most touch screen systems literally have the voter touching the screen with
>their finger. Fingers are too broad to precisely locate a position on the
>screen. Fingers also transfer grease to the screen necessitating frequent
>wiping. Our touch screen printer will work with a stylus allowing for more
>precise location of the touch and eliminating the need to wipe the screen.
>
>10. VOTING UTILIZING A MOUSE
>
>Some people don't use a mouse so we have to have the touchscreen available.
>However, most people can use a mouse so it's a waste of resources to make
>all the systems with the more expensive touchscreens. Our demo set up will
>have at least one voting machine with a touchscreen and one with a regular
>PC monitor. The application will look and work exactly the same with
>either.
>
>11. GRAPHICALLY BEAUTIFUL ON-SCREEN BALLOT
>
>The on-screen ballot should be very pleasant to the eye. The tiles for each
>contest should be clearly separated utilizing an attractive color scheme.
>We also need to test some with a color-blind user to make sure the colors
>provide acceptable shading differences for them.
>
>12. MULTIPLE LANGUAGES
>
>We will have a button that has a label in several languages. Selecting the
>button will bring up a menu of choices for language. We only make one other
>language selectable for the demo. Selecting the other language will return
>the voter to the on-screen ballot but with the text in the selected language
>(Spanish, probably, for the demo).
>
>13. WRITE-IN VOTING WITH ELECTRONIC SYSTEM
>
>Selecting "Write-in" will bring up an on-screen typewriter. A QWERTY layout
>will appear on the top half of the screen (three rows plus space bar below).
>Under the space bar, we'll have two (or three rows) of letters arranged
>alphabetically. The string appears in large upper case type as the letters
>are selected. Selecting the "DONE" button returns to the ballot with the
>write-in inserted for the relevant contest.
>
>14. SECURITY FEATURES OF OUR PRINTED BALLOT
>
>We need to guard against counterfeit ballots -- counterfeits that could be
>brought in by voters (who knows why -- maybe just to throw a monkey wrench
>into the process) or counterfeits that could be substituted later.
>
>The printout will utilize "special paper." One measure could be dipping a
>corner of the ream of paper into an ink that no one would know in advance
>the exact color and composition. So authentic ballots at the precinct will
>all have this mark. Another feature will involve markings on the ballot
>that will be unique to the PC and voting session. A sample printout would
>be stored so that authentic ballots from this PC will all have these
>markings (readily visible differences -- maybe a watermark or something like
>that)
>
>Probably, we'll have multiple features analogous to, say, multiple features
>on printed currency.
>
>15. ONE PERSON ONE VOTE
>
>People often ask what's to stop a voter from printing and depositing more
>than one ballot. The voter can print more than one but, in our system, the
>only one that counts is the one they deposit in the ballot box. After the
>ballot is deposited, they sign on the roster next to their name.
>
>16. WEB BASED DEMO
>
>Besides the standalone demonstration system, anyone with Internet access
>will be able to try out the system. We won't be able to exactly replicate
>the standalone system but should make it look and feel much the same.
>
>We can explain that the production system would not permit remote unattended
>Internet voting, but would include Remote Attended Internet Voting (RAIV)
>that would look much like the Internet version seen in the demo. The voter
>would proceed in a similar fashion but the printout would be placed in an
>envelop and dropped in a mail slot by the "pollworker" at the remote
>location (e.g., public library, court house, etc.)
>
>17. RANKED PREFERENCE VOTING
>
>The County Commissioner race will demonstrate ranked preference voting. We
>should include an option so that all eight candidates (if write-in included)
>could be ranked, or n of eight could be ranked. For example, some rules
>only allow the voter to indicate their first two preferences.
>
>18. N of M VOTING
>
>The Cat Catcher race will allow three to be selected.
>
>19. CONTEST DATABASE
>
>Names of the candidates will be read from a "database." However, this
>database will be trivial. A real contest database is quite complicated and
>there is no way we will have time to program such a thing. The contests
>will pre-determined and the number of candidates will be the same although
>we will be able to change the names in the "database." The ballot measures
>will be pre-termined and we won't change the text.
>
>20. TABULATION
>
>We will not tabulate the votes that users cast during the demo. However, we
>will set up one PC with some dummy data that shows the table of ballot
>images and shows how the votes get counted. We will show how several
>scoring methods present the results in the ranked preference contest (County
>Commissioner).
>
>21. EASY TO TRANSFER FROM SAMPLE BALLOT
>
>We should printout the ballot on tabloid paper and show how this could be
>used as a sample ballot that the voter could take into the voting booth and
>see the same image on the screen.
>http://home.earthlink.net/~adechert/ballot-mockup3.gif
>
>22. OBVIOUS VOTER INTENT
>
>The display for a contest will change dramatically when selections are made
>leaving no doubt about who was selected and who was not selected. The mock
>up implies all radio buttons. Actually, I want to have boxes that will
>contain a large red X when selected.
>
>The radio buttons for the County Commissioner race should remain as radio
>buttons. In this case the radio buttons would not be individually
>selectable but would indicate the position of the order of selection.
==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Sun Aug 31 23:17:14 2003

This archive was generated by hypermail 2.1.8 : Sun Aug 31 2003 - 23:17:18 CDT