Caltech & MIT Urge Wait On Net Voting

Booker writes "According to this article, a study by the Voting Technology Project (a joint venture of the California Institute of Technology and the Massachusetts Institute of Technology) is recommending that Internet voting be significantly delayed for further study. "The teams expressed even less enthusiasm for Internet voting, which `has all the problems of absentee voting and adds problems of security,' said MIT computer scientist Ron Rivest. `At least a decade of further research on the security of home computers is needed before Internet voting can come in.'" They do recommend better use of technology in voting, just not on the net - yet. They also report that between 4 million and 6 million votes were lost last November due to faulty equipment or other snafus. Read the report for yourself for all the info."

Comments

Problems w/ Net Voting

[mlc]

The problems with net voting do not just extend to the "obvious" (to the /. crowd, anyway) authentication, security, and other technical challenges. The real problems, IMHO, are much much bigger. I see two key problems: [my comments apply specifically to US but may have parallels in other places.]

• In current society, access to computers is not evenly spread throughout all socioeconomic and racial classes. Poorer people are less likely to own home PCs than richer people. Even ignoring this effect, white people and asians of any given income level are more likely to own PCs than black and latino/a people of the same income. For statistics and pointers to sources, google turned up this; plenty more info exists on the 'net.

  It's no brilliant observation to note that people of different classes and races tend to vote for different candidates. So, any voting mechanism that makes it easier for some given type of people, who are likely to vote in a certain way, will have an "unfair" bias on the election's results.

• With Net voting, the anonymity and "sanctity" of the ballot box can be entirely lost. Imagine either of the following scenarios:
  • Since many of its workers are low-income and do not own home PCs, Company X sets up terminals for its workers to use to vote when they come to work. However, there are posters around for the favorite candidate of X's CEO and managers. The bosses clearly make it known who their favorite is. Do we have a fair election here? (What we have is, in effect, the privitization of the polling place. This frightens me.)
  • [blatent gender stereotypes ensue. switch the genders or make the people the same gender if you like in this example.] Mr. & Mrs. Y have been married for n years; however, recently there has been some trouble in the relationship and Mr. Y has been beating Mrs. Y. Mr. Y favors some kind of conservative, "traditional values" candidate. Mrs. Y would really rather vote for a candidate more supportive of women's rights. However, Mr. and Mrs. Y sit down to vote together, perhaps because Mr. Y thinks he has to show Mrs. Y how to use the computer or whatever. As Mr. Y stands over Mrs. Y's shoulder, perhaps after having had a drink or two on his way home from work, who do you think Mrs. Y is going to vote for?

Now, I love technology. I'm working this summer to bring technology to people who might not otherwise have access. But the problems with Net voting extend far beyond the ones that technology alone can solve, and my desire to see a fair election far outweighs my desire to avoid trekking over to the polling place. Net voting is not a good solution. The study refernced in the article makes some better ones, as does the Center for Voting and Democracy.

--
// mlc, user 16290

Re:Problems w/ Net Voting

[quintessent]

My thoughts exactly. Instead, why don't we just set up computers at the voting booths. Everyone gets an ATM-type card used for identification, and then they vote on a closed network of high-security computers found at the same places we used to vote. Instant results. No arguments (assuming the network is running right [knock on wood]). No cheating (ok, less cheating).

Re:Another reason for delay ...

[Overt Coward]

If legislation gets bound tighter to these poll-results we have the purest form of direct democracy, in which political leaders are forced to react.

Which may work well for broad issues, but would be an unmitigated disaster for the details. What percentage of the population is truly politically active? Even among them, how many follow the issues closely enough to cast an informed vote on every issue?

The vast majority would simply vote the way their media source of choice would tell them to (though of course, not be "told" in so many words), because they don't have the time or experience to folow the legislative minutia.

Direct democracy is dangerous in that regard, especially as we continue to complicate our everyday lives more and more through technology. The reason representative governments were created in the first place may have been for simplicity, but the fact is that given the amount and level of detail of modern legislation, very few except those whose full-time jobs are to deal with it can keep up with it all. And probably not even then -- that's one of the main reasons why Congressmen have staffers: to help them understand what they're voting on.

To illustrate, go to Thomas and look up 10 random bills and see how long it takes you to not only be able to summarize them, but to be able to answer any question about them.

--

Re:Internet voting will lead to more incorrect vot

[Rupert]

Actually, computerizing the voting process (be it in a polling place or over the internet) would actually help people who can't punch a hole in a card, because you could give them immediate feedback that they had spoiled their ballot. In my mind, the major problem with the Florida system was the lack of user feedback.

--

Re:On the internet...

[pmc]

When you register to vote in the UK, you are required to prove your identity using a passport/photo-id drivers license[1], and to prove your address by means of utility bills or tax documents

Wrong. The system in the UK is that once per year a form is delivered to each household which you are legally required to fill in and return. This is used to compile the electoral register. Electoral registers are compiled and maintained by the local council.

The electoral register is "rolled forward", which means that an entry for a household stays until a different one is received. Needless to say this is a bit of a flaw.

From February 2001 (after an amendment to the "Representation of the People Act") you can request a form at any time and you will be added within 7 weeks. This is slightly different in that it is a personal form as opposed to a household form.

The main difference between the UK and the US is that registering to vote is automatic/compulsory in the UK, and voluntary in the US.

Polling Booths Needed!

[cr0sh]

The reasoning behind needing polling booths is a simple one:

It blocks vote/voter coercion!!!

That is, by having booths in a central, guarded, public area - such that one, and only one, person may enter the booth - no one can put a gun to your head (literally or figuratively) and tell you how to vote, to further their (or the group they represent) agenda.

This is the current problem with absentee ballots - there is no real way of knowing that the vote on an absentee ballot wasn't coerced in some manner. Currently, we have a low percentage of absentee ballots, so we just shrug our shoulders and move on. However, with internet voting from the home or elsewhere, we would have, in effect, a HUGE percentage of "absentee ballots" that could be coerced ("Want your check/job this week, Johnson? Go into that room and vote - ahem - properly...").

It isn't about security, it is about voter coercion.

However, I do believe that the kiosks should contain computers running voting software - such software could show the candidate, a synopsis of what they are about (maybe with links to outside info - allowing the user to come up to speed on the candidate), and other info - with a set of buttons on the side (like an ATM) that say "Vote", "Next Candidate", "Previous Candidate" - maybe some arrow keys. Have synthesized audio with headphones (or make the booth soundproof) to aid the blind (along with braille on the buttons).

Such software would need to be simple and robust, so as not to crash - don't use touch screens, because they aren't as intuitive, and are useless (or near useless, I would suppose) to the blind. The software could report the votes back to a tabulating center for final count, etc via the internet - using a highly secure encrypted system - or maybe they should just all go back to the center over a leased line system, or maybe back to an armoured truck that uplinks the data via satellite.

Finally, allow a week or so for voting - so everyone can vote, and have the voting booth hours be flexible - heck, make the voting booths mobile or something, like they have mobile ATMS, etc.

All of this could be done today, and relatively cheaply. It isn't that hard...

Worldcom - Generation Duh!

Re:Bizarre system

[wumingzi]

Us over here in the UK cannot understand how the US system can be so broken.

I should point out that much of the perceived bizarreness in the US was related to:

1) A somewhat loopy state which did not have good hand-count procedures in place.

2) The teams of two utterly ruthless, completely unscrupulous human beings (OK. I'm not sure if Gore REALLY passes the human test, but I have to give him the benefit of the doubt) fanning the flames of this system.

Here in Washington State, we had an extremely close Senate race which forced a full recount (by law, not by lawsuit). Washington has all the issues Florida has: poor districts with lousy voting equipment, military personell sending in absentee ballots, etc. (admittedly, we don't go out much for vote-buying, and there aren't enough blacks outside Seattle to make it worthwhile to intimidate people in to not voting).

There was also a well-established procedure in place to handle all of these cases.

It took two weeks for all the absentee ballots to be considered "in", and a few days to handle a full recount of votes.

It was as boring as watching paint dry. The process merited less than two column-inches in the newspaper every day and a total of 40 seconds of coverage of the entire process on NPR.

j.

It's not just the act of voting, it's those around

[heliocentric]

It's not just the act of voting, it's those around the voter. Some employers have video cameras to watch the workers and could conceivibly see their employees votes. More needs to be done in the voter privacy area beyond sending things encrytped. This is for two reasons - one, people who don't want their vote linked to them and two, people could create proof of a vote and have a greater ability to sell their vote.

Then again, I'm probably just a paranoid freak...

Internet voting will lead to more incorrect votes

[Dr_Cheeks]

Let's ignore the Slashdot community for a sec - there's a vast amount of people out there *cough* AOL *cough* who have trouble doing the simplest things on their PCs (otherwise IDG wouldn't be making so much money off the Dummies series). If people really can't punch a hole in a piece of paper properly, then are they going to manage to vote over the net? Right now, I doubt it. But I think that if internet voting is publicised widely then they're likely to try for the same reasons that so many people without a clue (and only a slim chance of getting one) get on the net in the first place.

The process of voting needs to be made much much simpler and clearer, people need to be able to review their choice before they cast it, and they need to have an equally simple alternative in the real world. Try beta testing any new system on a bunch five-year-olds and you'll see just how many ways something you thought was simple can be done wrong.

And then there's all the arguments about security, skewed distribution of PC access, etc. etc.

I just LOVE it.

[BiggestPOS]

I go to all the trouble of voting, and there is a chance, however small it is, that for one reason or another my vote simply didn't count. WHAT THE FUCK IS WRONG WITH COUNTRY?? To steal something from last Sundays Simpsons episode "We can build a ladder to put a man of the roof but we can't count something as intergral to our society as votes correctly?" I'm scared to vote again til they get this shit worked out. A voter registration card should be like an ATM card with a magnetic stripe, and voting machines would scan them, and tally all votes on an electronic, private network with heavy security. I'm sick of all the bullshit, and after the Florida affair, I hope you are too.

The Best Damn Voting Scheme You Ever Heard

[The Gline]

1. Get everyone in the United States into a BIIIIIIG room.

2. Everyone who wants to vote for one guy, they get on THIS side.

3. Everyone who wants to vote for the other guy, they get on THAT side.

4. A trap door opens up and drops everyone who voted for the other guy into a big-ass pit fulla scorpions 'n tigers 'n lizards 'n crocodiles.

5. Now everyone in the USA votes exactly the same. Yay!

Problem solved.

Real value of paper ticking voting

[jesterzog]

There are lots of arguments against using computers, or other technology for that matter, for voting. I agree with pretty much all of it, but I also think that a lot of people overlook one of the very important reasons that ticking a ballot is better than using computers.

To put it simply, everyone can understand the process - barring the very challenged. The process of indicating a candidate on paper, putting it in a box and having it counted by people reading it is an almost 100% manual process. Virtually anyone can understand it, and therefore trust it based on their own judgement.

Compare this with a computer voting system, or even a mechanical voting system. Show even a simple computer program to the masses and 99%+ of the general population won't understand it. Up this exponentially for something based on a distributed system.

Nearly everyone has to rely on and trust a minority of the population to verify that their vote is actually being recorded correctly. Even for those who understand it, actually formerly proving that it works is one total bastard of a job. Then there's verifying the proof over and over, verifying the compilers, making sure the hardware is 100% accurate, and there's always room for lawsuits when there's big corporate money sponsoring government involved.

Alongside arguments of people having guns stuck to their heads in secret and so on, this is one of the big reasons that I think I'll always support cumpulsory anonymous polling booths run in a very manual, and very understandable way.

In my experience to date, a lot of the people who want net voting are people who don't have a clue about the risks or problems involved with the software development process, and are simply more interested in convenience. It makes sense for things like internal corporation votes, but general public elections shouldn't change, IMHO.

===

MY idea

[rabtech]

Here is my idea:

Each voting booth has a LCD touch-screen in it. There is a list of names for the specific office (also listed on the screen in bold letters.)
When you touch the name, it highlights in large red letters. There are only two other buttons on the page, back and forward. Thus making it very simple to use.

Each of these booths are connected to a printer, or series of printers. These printers spit out square pieces of paper, with what appear to be random clouds of dots. But these are actually a form of "barcode" that indicate what choices were made. These papers can be fed into reading machines for the final vote.

This system preserves anonymity, and yet makes it much harder to counterfeit votes, at least in any new ways.

The problem with Internet voting is that how could you tell if someone hacked the system and were entering invalid votes? How would you do a recount? It adds a very large element where corporate or government interests can really screw over the numbers.... or even a disgruntled programmer or system operator.
-- russ

Another excellent site about electronic voting

[plcurechax]

Rebecca Mercuri is an authory on electronic voting, and her site is an excellent source of information on the subject.

The usual good source of thoughful and insightful comments, is RISKS / comp.risks, and in particular Vol 21 Issue 14.

There's going to be a problem with this in Chicago

[pyramid termite]

It's going to be expensive to hook up all the cemetaries ...

4-6 million votes!

[burke3gd]

Hey stop the press for a second, ok.

I live in Finland, a country whose entire population is about 5 million people. The US lost more votes than there are voters in my country. And noboy thinks there is anything wrong with that?

The USA, a country where about 50% of the people vote and they loose 4-6 million votes, a small countrys worth.

And nobody even blinks. You just talk about the security in voting via the net. Now how fucked up is that?