A quick round-up of coverage on the 2007 e-voting pilots announcement.
In other news, I'm afraid it looks like Ireland are going try to use their e-voting machines again:
The Times reports today that the UK will be investigated by the Council of Europe for electoral fraud and alleged breaches of human rights. The announcement currently relates specifically to postal vote related fraud particularly the aftermath of the infamous Birmingham postal vote fraud case which caused Labour serious embarrassment and lead to outspoken comments from presiding Judge Richard Mawrey:
“Anybody who has sat through the case I have just tried and listened to evidence of electoral fraud that would disgrace a banana republic would find [government assurances about postal voting] surprising” (Source: BBC)
Postal voting is extremely problematic, particularly when an election is all-postal. However electronic voting creates the opportunities for fraud on a much wider scale whilst being considerably harder to detect. Additionally remote electronic voting continues to breach human rights treaties demanding a secret ballot. So, given yesterday's announcement of pilots in 2007, I hope this investigation is widened to include electronic voting.
(Thanks to Ian Brown for the tip-off)
Sorrow of sorrows but in spite of all the evidence showing that e-voting is A Bad Idea, today the Department for Constitutional Affairs has announced plans for electronic voting pilots to be held in May 2007 and are inviting proposals.
The prospectus states:
Research for the Electoral Commission released in 2003 shows there is significant
demand for electronic voting and that it may help stem the declining turnout at elections.
Yet the Electoral Commission’s own figures for the 2003 pilots showed that e-voting had no significant impact on turnout (more info)
It would seem that a year makes all the difference as in September 2005 Harriet Harman MP, a minister responsible said to Radio 4 that:
“We just think that the time is not right for it (e-voting) at the moment. We talked to a lot of people, we listened to a lot of views including from the Conservative Party. The general consensus seemed to be that the time is not right for it at the moment. So we are not going ahead with the pilots that we were planning to run otherwise in the May 2006 council elections.” (Source: The Independent)
What has changed since Ms Harman said those words, I do not know. Since 2003 The Electoral Commission have been repeatedly calling for a clear framework or road-map to be set out by government but nothing has emerged. Still they seem to be happily going along with these pilots. Why?
Some good news is that SMS, digital TV and all-postal voting have all been ruled out for 2007. Sense prevailed there thank goodness!
Despite repeated criticism from The Electoral Commission, suppliers, councils and qualified observers such as myself, the timetable is again tight. Applications close 17th November leaving the usual scant time for supplier selection, contractual agreements and technical work to be completed. Indeed the prospectus timetable leaves only February and March for development and testing – madness!. This process should, at the latest, have been begun in the summer. Also with the central electronic voter register project stalled, I wonder if now is a good time to be piloting anything?
Immediate Action
Please write to your local councillors asking them to ensure that your local council does not waste tax payer money and risk the integrity of elections by applying to run a pilot.
Use WriteToThem for quick access to your local councillors.
Learn more:
The uncertain nature of elections to come [PDF] – detailed report on the serious failings in the planning, management and implementation of the 2003 pilots
Electronic Voting Doesn’t work – written for Electoral Reform Society
LEARN: I want to understand the issues – learn about the pros and cons of voting methods
Silicon.com are reporting that no e-voting will be used in this year's UK local elections due to security fears. A Department for Constitutional Affairs spokeswoman is quoted as saying:
“Because of concerns of the security of votes we need to make sure people have confidence in the system. We need to be sure it is completely tamper-proof.”
This is good news indeed.
Anyone who has visited the Middle East even a little (my grand total out there is 8 days so far) won’t be surprised by the gushing enthusiasm of Bahrain E-Voting Forum’s concluding recommendations. They love their reports one-sided and upbeat. So allow me to provide some balance…
1 It has been proven beyond any doubt that technology exist today to effectively enable and run e-voting.
Not true at all. Most computer scientists still see fundamental theoretical barriers in current technology which prevent the proper implementation of secure and anonymous e-voting systems. One of the best examples of this view is expressed by Bruce Schneier in this article which concludes:
“Building a secure Internet-based voting system is a very hard problem, harder than all the other computer security problems we’ve attempted and failed at. I believe that the risks to democracy are too great to attempt it.”
Avi Rubin also has an excellent paper on why remote e-voting in particular isn’t possible with current technology.
Back to the Bahrain recommendations…
2 e-voting is doable and has many benefits some of which are enabling security, increasing accuracy, saving cost, enabling wider participation and catering for people with special needs like frequent travelers and disabled citizen.
I thought we’d killed the cost saving red herring years ago. When e-voting first began to be sold commercially the cost savings were all people spoke about, especially the vendors. But although securely printing paper ballots is costly, it pales in comparison to the costs of building and running a distributed highly scalable and secure e-voting system. Which is why no major vendors now trumpet cost savings in their marketing and why e-voting ‘pioneers’ like the UK government stopped talking about saving money through e-voting years ago.
It’s a fairly open secret that many of the suppliers to the UK pilots swallowed huge cost overruns just so that they could stay in the preferred supplier list for when e-voting went national and they could then recoup costs and some profits. Oops, not happened yet.
That e-voting can increase accuracy is utterly debatable, some would say yes because they believe computers make everything better. Others, especially usability experts would argue that computer interfaces are going to introduce whole new forms of errors. And of course software as well as hardware bugs can create fun unexpected inaccuracies too!
We also know that e-voting does not widen participation significantly. Repeated pilots have shown negligible boosts in turnout, as shown in my analysis of the 2003 UK pilots. We also know that many of the key reasons why people don’t vote are not related to convenience.
Don’t just take my arguments, many smart people agree, including former Labour minister Michael Meacher MP.
Finally while there is no doubt that helping disabled people to vote on their own is an important goal, I have to strongly question whether e-voting for everyone is the best way to meet that goal.
3 Kingdom of Bahrain enjoys many edges over other countries for successful implementation of e-voting such as; solid communication and IT infrastructure and a highly educated population.
Yeah, yeah. I’m sure that point has nothing to do with His Excellency Sheikh Ahmed bin Ataytallah Al-Khalifa State Minister to the Cabinet of Bahrain being patron for the forum.
4 The success of e-voting depends on the involvement of all stakeholders (Public, Government, Societies, professional and political parties)
Good point but impossible to ensure it’s not tokenism.
5 Governments have a vital and focal role to play in the e-voting process as; facilitators, enablers, promoters, legislators, organizers.
Well, maybe, but it does assume e-voting is a good thing. You know I don’t think e-voting is a smart move but either way should governments be championing this?
6 Increasing awareness, especially amongst public is one of the key success factors for e-voting to gain the trust of all concerned parties.
In my experience raising awareness in countries around the world has led to the public becoming increasingly suspicious of e-voting. Their first thoughts on e-voting tend to be positive but after 5 minutes of chatting their against it. Or is that just my sales technique?!
7 Kingdom of Bahrain’s smart Card project strongly supports e-voting due to its security and transparency features.
Yawn.
>8 The Forum highly commends the initiative of the organizers of the upcoming Bahrain Youth Parliament of implementing e-voting technology.
Get them while they’re young.
>9 The Forum highly commends Bahrain Government’s initiative of forming an advisory panel of experts form all relevant stakeholders including external relevant entities such as UN and independents to discuss and steer the e-voting initiative for Bahrain. This will also be a step in further establishing e-democracy.
I have no idea how this gets us to e-democracy but of course expert panels always seem like a good thing. They can be excellent but it all depends on those invited to participate and which voices are heard.
A quick look at the [speakers for the Bahrain forum][7] show us a bunch of people either:
There is nobody there with a deep technical or theoretical understanding of e-voting and not a single voice which might be considered slightly dissenting. This is not a surprise for Bahrain but still shameful that Avi Rubin, Barbara Simons, Bruce Schneier, David Dill, Rebecca Mercuri, Peter G Neumann, Doug Jones, Bryan Pfaffenberger or one of the other deeply intelligent people with concerns over e-voting wasn’t invited.
That such a limited group was invited to the forum does not bode well for the expert panel.
10 Implementation of e-voting should not eliminate traditional methods of voting. The voters should have a choice of selecting more than one convenient method.
Multi-channel voting is great in theory, certainly better than all e-voting at least, as it allows more robust channels of voting to take the strain if e-voting fails. However the challenges in terms of managing the electoral roll and voter authentication are significant indeed – it’s no easy task whatsoever to keep it all securely synchronised between channels.
11 It’s recommended to have an independent auditing body that consists of members from Government, private sector, political parties, professional to increase trust and credibility of the entire process.
Excellent – give them full access to the source code.
12 The forum shall state clearly that there is no 100% risk free system in either traditional or modern technology devised systems for voting. However, there are many proven ways and means of minimize such risks to an acceptable level.
Last but not least, eh? Well it’s a fair point but it would be nice to see them look at the risks from e-voting (potential for all votes to be manipulated electronically) vs. paper voting where there is a cost to each manipulation of physical ballots making wide scale fraud much harder.
The Khaleej Times report that Microsoft have signed a memorandum of understanding with regards to implementing Bahrain’s e-voting. Now Microsoft do have some good products, (only some ok?) but their reputation for security is deservedly poor and they’re not an e-voting vendor. So what on earth are they doing with Bahrain? I’d rather there was no e-voting but at least go with someone like VoteHere who actually has some experience of this field.
It looks like another country is being drawn by the siren cries of e-voting’s attractions. But sadly it’s all a mirage.
As always, a detailed non-technical explanation of the pros & cons of e-voting is available at http://www.free-project.org/learn/
The Register is the absolute best read on e-government type stuff at the moment – informed, incisive and critical where it needs to be (which sadly is rather often).
John Lettice has a cracking article on CORE, the bedevilled centralised electoral register project which was once known as LASER and will be known as SPECTRE when Prescott is done with it. Ok I made up the SPECTRE bit but this idea of a project has dragged and dragged ever since I was first involved in the e-voting area back in 1999.
The Electoral Roll system has its flaws but CORE, while potentially offering some benefits, is also opening up a bumper sized Pandora's box of nasty issues.
It's not surprising at all that insiders have taken advantage of their privileged access to information in order to raise a few bob through online betting.
They only got caught (well suspected as no charges have been pressed) because a betting site got suspicious. I doubt there are audit trails in places that would spot any cheating on the outcome of the result etc. I've noticed that on programmes like Strictly Come Dancing they don't tell us who got how many votes – it's all left rather woolly.
BT is quoted in the Register piece as saying:
However, one thing we would like to make clear is that this does not affect the integrity of the result. It is impossible to tamper with the results to affect the outcome in anyway – the viewers' choice will win.
I chortled heartily at that. Impossible to tamper? I don't think so BT but I understand why you're saying it. They're just one byte away from being totally discredited and the game show voting is proving much more profitable than the UK's e-voting pilots ever were.
The Usable Security blog from UC Berkeley has posted the first part of it's complete examinations of optical scan and DRE systems from the big 4 e-voting vendors Diebold, ES & S, Sequoia, and Hart Intercivic.
First up is Diebold. Reading the details of how one votes and the problems they've identified I don't know how anyone can possibly get it all right and cast an accurate ballot.
Particularly scary is how the optical scanner gives you no inkling if your individual votes have been counted correctly. The complexity of the DRE's interface is daunting.
A very useful read for the e-voting interested.
Those who know me may not realise that I’m actually half French-Canadian, but I am, proudly so. I don’t sound it due to years of English education, though fat good that does me – every time I’ve been abroad recently people have commented that, on the basis of our english-speaking accents, I must be American and my Polish wife must be English sigh.
I digress. The news is that on Sunday 6th November the Quebecois voted in municipal elections for councillors and mayors. In many areas this was done with electronic voting kiosks within polling stations (there was no remote e-voting). DRE and optical scan machines were used, details of all the systems used can be found in this forum post.
Paper Vote Canada has a summary of some Canadian press coverage [French]. Essentially the reports argue that very little has been put in place to ensure the security of the vote and even the president of one of the suppliers is quoted as saying, I translate, “There isn’t really a way to prove to a voter that their vote was stored exactly as they wanted. One needs blind faith in the integrity of the local election officials.”
Two days after the election, the province’s Chief Electoral Officer is reported as saying that the computer glitches were due to a network crash and a few defective machines. As a result the official states that electronic balloting may not best suited for bigger cities and so rules out e-voting from provincial elections or referenda for the near future. CBC report
Last Friday a leading Montreal politican (who lost his race to become mayor) called for the election results to be cancelled due to the technical problems [French] This was after results, supposed to be ready in minutes, took hours to appear but it soon emerged 45,000 votes had been counted twice, a few days later the results changed again in three districts, a new winner being named in one race. Details
The elections had very low turnout, I’ve seen figures as low as 35% mentioned. Canada’s paper-based electoral system is excellent. There seems to be no good case for the expense and risk of introducing e-voting in Canada. The current system is simple, note that they also seem to lack any certification process for these new systems. I hope that the current doubts over the recent election results can be cleared up – for everyone’s benefit – and that they then leave this stupid machines alone. The damage to turnout from uncertain results is far worse than any benefits e-voting could bring.
More Links
French site recul-democratique.org has a bit more on the Quebec situation [French]
The English-speaking Montreal Gazette has more, but for subscribers only, and I didn’t have time to get in but the stories are here
That e-voting train keeps steaming on… Press reports in Poland are discussing the possible introduction of e-voting there.
Their basic argument is that e-voting would boost turnout, which is rather odd considering that it hasn't done so anywhere else. Also what with high unemployment (20% officially) and a largely agricultural population Internet access isn't exactly rampant. Certainly Internet voting would be attractive to the large expat population, but how many of them would be eligible to vote I wonder?
Pressure is being applied for Poland to jump on the e-voting bandwagon because “the UK, Belgium, Switzerland, US and Estonia all vote via the Internet”. Not true.. They've had trials but no general elections have taken place with Internet voting. In the UK e-voting is currently dead. In the US remote e-voting is a long way off after the SERVE project was killed. Belgium uses polling place computers, not remote voting. Switzerland and Estonia have both been involved in limited testing with low usage figures.
It's bad enough having to counter false hopes over what e-voting can deliver, but when untrue claims are made about remote e-voting's prevalence one has to worry.
Press report on Onet.pl [Polish]
if you understand Polish the discussion at the bottom of the article is entertaining
Jason Kitcat 