Personal tools


From Wij vertrouwen stemcomputers niet

This page was translated from Vraag en Antwoord (in Dutch), which at times may be slightly more up to date.

Who are you?
"Wij vertrouwen stemcomputers niet" (or "We don't trust voting computers") is a coalition of citizens concerned about the use of voting computers during Dutch elections. It started as an initiative of Rop Gonggrijp, Peter Knoppers, Anne-Marie Oostveen and Barry Wels, but now involves many more people. Check our masthead (Dutch) for the latest state of affairs.
What exactly do you want?
"Wij vertrouwen stemcomputers niet" wants all elections for local council, provincial government, Parliament and European Parliament to happen in such a way that it is possible in principle for every citizen to monitor whether or not the election has been fair. The current situation - in which even the polling station officials do not monitor the actual vote count - is completely unacceptable. All software used in the election process needs to have its source code published. But much more importantly: there has to be a voter-verified paper copy of every vote cast. And these paper votes need to be regularly counted. For the time being, a return to the use of the paper ballot seems easiest.
They say "voting machines" where I'm from, and not "voting computers"
This is deliberately deceptive, by pretending it is not a computer. People who know better (for instance because they read this site) say "voting computers". We wrote a longer piece (in Dutch) about it.
So what's wrong with voting machines, ehm, I mean voting computers?
Whether it is the actual voting process or the counting of the votes: in the era of the paper ballot one needed a phenomenally large conspiracy to steal an election. Vote fraud is a phenomenon of all ages, but the procedure was watertight assuming a polling station had one or more honest officials. And given the sheer number of people in polling stations during a national election things would have to be mighty crooked for any manipulations to have a large-scale effect.
The introduction of the computer changed all this. The vote-count is a closed process that happens out of sight of election officials. All over The Netherlands, people at the end of the day walk to a computer they do not understand to press a button which causes the computer to print the election results. The verification of the election has been pulled from the hands of the many thousands that used to do it, and placed in the hands of a handful of anonymous techies and bureaucrats at manufacturers, certification institutes and government bodies.
Despite what the bureaucratic language may say: a voting computer runs plain old ordinary software. Which, if all is well, should have been looked at during the certification process, but what does that mean? And how do the officials at a polling station ascertain that the tested software is indeed running on their computer?
Are you against modern technology? Do you want to halt progress?
Quite the contrary! There are many programmers, networking specialists, computer security experts, system managers and other techies among us. We enjoy the benefits of modern technology every single day of our lives. Some of us even actively contributed to its widespread use. But we also realize the risks of blindly trusting technology.
As far as the issue of voting computers is concerned, we think the advantages (faster and easier to count the votes) are outweighed by the disadvantages (count no longer verifiable), and that as such the changes have not brought progress. Of course it's a lot of work if we start counting paper ballots again. But it's a small price to pay for verifiable elections, if you think about it.
I use ATMs regularly. Why would I distrust a certified voting computer?
Simply put: it may not be nice if you are short-changed by an ATM, but you do notice. With the current generation of black-box voting technology, fraud would have to be extremely non-subtle for anyone to even notice. With an ATM you can get a receipt and check your balance (even online) to see if all is well.
Is this all necessary? Aren't you overly paranoid?
As far as we can tell, the framers of the original Dutch electoral law had a healthy distrust towards everything and everyone. The entire process is riddled with people all checking on each other. If you read the law you can see for yourself: this was no accident and not a by-product of the limited means of the time. Elections are always about power: a better motive to cheat is hardly imaginable. And in the era of disappearing ideologies, undecided voters and permanent opinion polls the results will be closer together than ever, so a successful fraud can be subtle.
To be honest we think there has not yet been any large scale fraud using voting computers in The Netherlands. But we know enough about computers and security to worry. An election has to be honest even if the government isn't. As a member of EU, OSCE and the UN, the Dutch government demands other countries to have verifiable elections. What makes us think we would we be exempt?
I'm really into conspiracy theories. This whole voting computer thing is obviously a plot. Just like the Mossad killed Pim Fortuyn. Can I join you before the aliens on Area 51 take over?
No. We realize full well that the topic of voting on computers attracts people also working on brain manipulation from satellites, crop circles, new theories for the JFK assassination and God knows what. We're not into that, and we don't want to be associated with it either.
Guys, do you really believe in democracy. I mean
voting is useless, isn't it?
Yes, we believe in democracy. Not because we think the current form of government has no flaws, but simply because we cannot think of a better alternative.
But what does this mean for all these great improvements? Voting for all sorts of stuff every month, direct democracy and all that?
We do not have an opinion with regard to electoral reforms other than that they may never affect the verifiability of the election results.
Do you really think this can be won?
It won't be easy. But we have a good story to tell, and if we want to sleep at night we have no choice, really. We do notice that people from many different walks of life "get it". (Try...)
Aren't you far too late? All of The Netherlands is voting on computers now.
Yup, we are a little late. A number of the people behind this site are Amsterdammers, and they did not get to vote on computers until 2006. And maybe you have to vote on a computer yourself to realize the seriousness of it all. Furthermore, this would have been difficult to explain to people 15-20 years ago when this all started. Whatever the case may be: just because things got so far out of hand means we have to take action. And we need your help.
Can't secure voting computers be built?
We thought about this a lot, we do have a lot of bright techies running around here after all. Wonderful cryptographic algorithms have been invented for elections, and elections are of course a great time to be using Open Source software. Simply put: yes, things can be much more elegant, more verifiable and more advanced that the well-meaning junk that currently litters our polling stations. But whichever system is used, the problem of people depending on software that only a few understand remains. Without the manual counting of pieces of paper, the majority will have to trust a minority that says everything is fine.
What about Open Source? What if the source to the computers is released?
Unfortunately there is not yet a method for non-technical people to verify that a given computer is only running a given piece of software. Even for technical people this can be quite a challenge. So after publication of the source, we'd all be staring at this program of which we still don't know whether or not it is inside our voting computers. Therefore a "paper trail" is always needed so that we can count (and re-count if need be) whether everything was honest. We wrote a slightly longer piece (in Dutch) on this.
So if the computers print a ticket all is well?
Possibly yes. But then we do demand that the papers are counted during regular random spot checks (and not only if there is doubt) and that the paper count determines the election results if there is a discrepancy.
But then we don't really need the computers anymore, right?
Probably not. But then we never claimed society had a problem that needed to be solved in the first place. We do not remember any news items about the great ballot-counting crisis. Voting computers are at best a problematic non-solution of a non-problem.
But people get more involved with the political process when the results are known quickly, right?
Says who? We didn't research this, but it would seem to us that elections are more exciting if the results take a little while to come in. The 'close call', and the odd cliffhanger until the following morning: they all add excitement which in our minds almost certainly leads to more public involvement. The instant-results at the very least spelled the end of a night of exciting television.
The introduction of voting computers has led to less doubt over valid and invalid ballots, and counting is much easier. This is all good, isn’t it?
We certainly do not claim paper ballots solve all the world's problems: it is merely the best crappy solution we can come up with. Asking a computer to print the totals is much easier than counting by hand. But if speed and ease would prevail over verifiability we could also just have a good opinion poll done, and declare the results binding: even easier and faster still.
But voting by computer is much cheaper isn't it?
No, it is Pdf icon.png more expensive.
So what about voting via Internet?
The fact that this is considered a serious option in some circles of government scares us to no end. For starters, voting over the Internet is plagued by the same problems that the current voting computers suffer from. On top of that there are sizable security problems with regard to the Internet. Attacks like 'phishing' and dDoS can be controlled. But there is a whole array of new problems, most of which are fortunately generally recognized. One of them is the fact that voting via Internet means there is no means of verifying whether or not a voter votes independently and in secret. Some circles of government are now making it seem as if the secret ballot isn't all that important because they'd love to introduce something new.
The most important issue for us remains that during an Internet vote, the results can never be verified by the population. In short: all (hopefully) well-meaning initiatives need to be nipped in the bud.
The DUTCH government is doing an i-voting experiment for voters abroad, who otherwise should vote by letter. The system that is used, is called RIES. A security analysis can be found Pdf icon.png here.