Jason Kitcat

The Electoral Reform Society have for a long time failed to make a clear statement regarding their views on e-voting. In the past reports have mentioned some of the weaknesses inherent to e-voting but nothing definitive has ever been published by the ERS.

This has now changed as the ERS Council's latest report states:

Following a resolution at last year's AGM that was remitted to the Council, a policy paper on e-voting and e-counting was approved. The Council accepted that e-counting can have benefits, provided there are sufficient checks and safeguards, but that internet, text and telephone voting can, like postal voting, make elections more vulnerable to fraud.

The policy paper is due to appear on the ERS website soon. I warmly welcome the Society's willingness to add their voice to the consensus over e-voting's significant risks.

UPDATE: The policy paper is now online as a PDF.

DCA have published the statutory orders for the May 2007 pilots on their website. These orders are the changes to law required to support the new technologies and processes that will be used.

I've only had a brief look so far but two sections caught my interest.

The order for South Bucks states:

43A. Voting procedure for internet voting
…
(10) The voter must not be able to access the internet voting facility for an election once his vote at that election has been cast.
(11) A voter voting using the internet must vote without undue delay and must exit the polling website after he has voted.

Extremely odd. I guess for (10) when they say 'access' they mean actually log in because as it reads now visiting the voting website again after having voted would be a breach. As for (11)… why put this in and how on earth would they enforce it?

Stratford will be using e-counting and have the following in their statutory order:

46 Re-count
…
(4) In so far as the votes have been counted using the electronic counting system, the returning officer may treat a request for a re-count as unreasonable unless he has reason to believe that the system has not scanned the votes correctly.

I find this mind boggling. Of course there's plenty of wiggle room – 'returning officer may treat a request for a re-count as unreasonable'. So he doesn't have to but the guidance is clear, unless there's good reason to believe a malfunction occurred then requests for re-counts can be disregarded. There should be mandatory sample re-counts at all times along with mandatory complete re-counts if the victory margin is slim. Will anybody be checking the paper ballots at these e-counting pilots?

I also have to say that it's extremely poor that there was no notification of these orders being published. They just appeared on the DCA website, nothing on their homepage or elections homepage (last latest news is dated 29th January). The Electoral Commission's site has nothing on these orders being published either. I only found out about them by talking to electoral administrators.

There's an elections community that should be kept informed of such things, how hard would it be for DCA or the Commission to email us? Furthermore as these documents are law for the period of the May elections, there's a responsibility to make them available to candidates, party agents and the general public. The sharing should be better folks.

All the statutory orders on the DCA site

The Department for Constitutional Affairs' Contract Notice to supply e-voting and e-counting solutions for the pilots has some interesting bits of information.

The notice was dispatched 1st August '06 and published by an EU journal on 3rd. The closing date for a response was 1st September '06 and DCA were to notify candidates for negotiation on 18th September.

A month later on 17th October the notice requesting applications to run pilots from local authorities was published with a closing date of 17th November.

So, vendors had a month to prepare their initial proposals as did local authorities. I'm sure DCA notified favoured vendors and authorities before the official publication but strictly nobody should have been able to prepare their applications until the announcements were published. These were absurdly tight timescales for no good reason whatsoever. Either work should have started much sooner or they should have contracted for the next local elections, not this May.

DCA were very honest about their goals in the contract notice (emphasis mine):

The Department for Constitutional Affairs (DCA) is exploring changes in electoral procedures to make elections straightforward, efficient, secure and, above all, readily accessible to all electors.

So, here we have it in black and white. Sir Alistair Graham was right – DCA put participation ahead of all other considerations. It's written in their own words, here's how it looks in the original:

This Contract Notice was released after a request was made to the DCA, however DCA have not published it. So it is available to download here [PDF]

OSCE’s Office for Democratic Institutions and Human Rights (ODIHR) have an excellent Election Observation Handbook in it’s fifth edition.

The handbook’s guidance on e-voting is really not bad at all, my comments in double brackets inspired by Bruce Sterling’s Viridian stylings:

DIRECT RECORDING ELECTRONIC (DRE) VOTING SYSTEMS: New election
technologies using direct recording electronic (DRE) equipment
(electronic voting, automated voting) have the potential to facilitate
voter participation. However, such new technologies may also pose challenges to the transparency and accountability of an election process. Furthermore, they may influence perceptions on the security of the vote, with a potential impact on voter confidence. It is therefore important to assess the introduction of DRE equipment, where relevant, and whether its introduction will have a direct impact on the possibility to observe the voting process compared to conventional voting. Such issues should be followed closely by an [observation mission] where relevant. International observers do not certify DRE voting systems. However, they should have full access to the certification and independent domestic verification process ((if there was any certification – none in the UK)). The criteria electoral authorities use when choosing DRE voting systems, and the manner of introduction, including voter education and pilot testing, are also relevant issues for an [observation mission] ((barely any meaningful information on this has been released)). The following measures could prove essential with a view to enhance voters’ confidence in such new voting technologies:

(i) Permission for competent individuals, academic institutions or civil society groups to comprehensively and independently test automated voting equipment subject to reasonable limitations related only to patent or copyright law. However, such testing should not be perceived as a substitute for the establishment of inclusive and transparent certification procedures;
((Nobody inside or out of government has done such tests that we know of. If they have been done then nobody has published any findings, which means it’s as good as if they hadn’t been done.))

(ii) Regulations that ensure against possible conflicts of interests of the vendors; ((Nothing like that in the UK that I know of))

(iii) Facilities that produce a permanent paper record with a manual audit capacity, and serious considerations to ensure a voter-verified auditable paper trail; and ((No such records are going to be produced, of course they’re rather problematic with Internet voting, but they’re not going to recount the e-counted ballots to check system accuracy))

(iv) Establishment of a clear division of responsibilities between vendors, certification agencies and election administrators to fully ensure accountability and an effective response in the case of failure of DRE equipment. ((They do seem to be getting better at clarifying responsibility, but of course there are no certification agencies)

Possible problems to be aware of: ((Keep your eyes peeled folks!))

• Unduly long or complex ballots that may confuse voters;

• Inadequate safeguards or accountability at any stage of the ballot production or distribution process;

• Ballots circulating outside of polling stations on or before election day;

• Polling-station procedures that may compromise the secrecy of the ballot; and

• Electronic voting systems with no voter-verified auditable paper trail or other manual audit capacity.

This week there have been four written answers in Parliament that attracted my interest:

Oliver Heald continues to be very curious about the pilots and asks for correspondence between the Electoral Commission and the Department of Constitutional Affairs. This was correspondence I was also interested in, I would love to see what was put in the library by the Commission.

Jonathan Djanogly is another Conservative asking ever more questions about the pilots. He asks Bridget Prentice how DCA will prevent malpractice and fraud and receives a mish-mash of an answer. We are told of 'independent experts' but they aren't named and no promise of their findings being published is made. Who are they? When did or will they check the systems? What does independent mean?

We are to be reassured because “All e-voting systems will also include audit processes and records to allow returning officers to have confidence that the number of votes cast electronically tallies with those cast.” This means very little indeed, how it could do such a check and what value it would be is questionable. It would be much better (and much harder) to check that votes are counted as voters intended.

Then Ms Prentice tells us that because electors will be able to provide a password of their choice the potential for credentials being stolen or misused will be reduced. Which is rather doubtful in my view. The password picked by the elector has to be sent to the election office, an obvious address for cherry-picking from the post. Furthermore these passwords will need to be entered into the system manually – so they'll be lying around, typed in by who (we don't know) and then stored in a central database ripe for the picking. Human picked passwords are going to be easier to brute-force guess also.

Finally we're told about the Electoral Commission's statutory duty to evaluate the pilots. Which is great and very welcome but completely unrelated to how DCA will prevent fraud.

Mr Djanogly also asks how DCA will prevent coercion and intimidation during Internet and telephone voting pilots. Good question. Once again Ms Prentice's answers seem slightly… adrift. Firstly she reminds us that electors can still vote in polling stations, but she doesn't specify if you can vote in a polling station if you think your e-vote has been compromised. E-voting isn't compulsory yet, but this doesn't stop those who do register to use remote voting methods from being open intimidation. So this doesn't really answer the question.

Next Ms Prentice helpfully tells us that electors will be using identifiers unique to them and that there will be no receipt to show to others how they have voted. But print screen, a photo of the computer screen or even an audio recording of the telephone vote would work as a receipt – there is no protection and receipts are possible. The unique identifiers don't protect from intimidation either – they just limit it to one vote that can be stolen at a time by intimidation.

Finally we are told of good links with local police being fostered to ensure allegations are properly investigated because they could result in prison or a fine. I do think that local authorities are already pretty closely linked to their own local police forces. Perhaps the threat of prison and a fine count as prevention through fear but nothing else in Ms Prentice's answers can be seen as prevention. This is because it's impossible to prevent coercion and intimidation of people voting remotely – which is why we invented monitored polling stations.

Mr Djanogly finishes up by asking if we could perhaps use the Northern Ireland postal vote application process for the rest of the UK. Ms Prentice tells us how popular postal voting has proven to be, which is the problem as it puts a large number of votes out there for stealing. Essentially the Government position is that, even after numerous recent convictions of postal vote fraud and a Council of Europe investigation, we're not as bad as Northern Ireland so let's carry on with just the minor tweakings of the 2006 Electoral Administration Act. Sir Alistair Graham is right when he says that DCA haven't even tried to balance participation with electoral integrity – they've just gone for participation at all costs.

• BBC News: Electronic count is 'way forward'
  Northern Ireland's chief electoral officer, Douglas Bain, has said that he hopes to introduce electronic counting to speed up his counts. For Mr Bain this May's Scottish elections, with nation-wide e-counting, will be the litmus test.

• OSCE: Election observer report on Dutch Parliamentary Elections, November 2006
  The Dutch elections in November 2006 were the first following 'We Don't Trust Voting Computers' demonstrations of the serious vulnerabilities in Nedap voting machines. The OSCE 'Election Assessment Mission Report' provides interesting reading including unusually strong words about the inappropriate secrecy and inadequate testing of e-voting systems.

• DCA: CORE – Response to Consultation
  I've been sitting on this one for a while but I've now finally had time to read most of it. The Co-ordinated Online Record of Electors (CORE) is interesting for many reasons but when it comes to e-voting, two in particular: Firstly CORE will provide for a national electoral register database which the government explicitly lists as enabling e-voting in the future. Secondly the key data format for CORE will be Election Markup Language (EML). My paper on EML's development can be read here.

• Open Rights Group: Election Watch 2007
  The Open Rights Group (disclosure: I'm their e-voting campaign co-ordinator) are calling for volunteers to help observe the e-voting and e-counting pilots this May. ORG and our volunteers will be submitted for Electoral Commission accreditation which offers greater access (and responsibility) than previously possible.

• Wired News/AP: Diebold may dump their e-voting unit
  Diebold weren't always in the voting business, they bought into it fairly recently, and given the trouble it's caused I'm not surprised to hear that they're considering dumping the unit to a private investor or competitor. Diebold's core businesses of ATMs and safes are not well served by the bad PR (self-inflicted, I might add) the election services unit keeps attracting. Diebold aren't the only e-voting provider with questionable practices and poor technology, they've just been much worse at dodging the bullets than some of the others!

• The Sun: Phone voting computers crash and don't count incoming calls
  When you phone or text a vote for one of these TV competitions an insider reports that the computers often crash. The calls are still received, and callers charged, but votes are not counted particularly at peak times. Because there's no voter verification there's no way of knowing that your vote has actually been counted. It's also very much not in the supplier's interest to admit that something has gone wrong. I'm not surprised that problems have happened but I am mystified that after so many years of such shows being popular that the computers still can't cope with peak numbers of callers.

• House of Commons: Home Affairs Select Committee evidence on secret ballots (1998)
  A rather interesting short transcript of evidence by academics to the committee regarding the UK's practice of non-secret ballots. The academics recommend going to fully secret ballots as the negative perception of numbered ballots alone is not worth the minimal benefits they offer, which are rarely used. (via Ideal Government)