Anyone who remembers John Gummer feeding beef to his daughter during the mad-cow scare will never forget that politicians are more interested in appearances than complex technical arguments. Gummer sought to prove that beef was safe but instead provided an iconic example of how very low his profession is prepared to stoop.

So I don't find it surprising that e-voting is being served up in a similarly distasteful manner.

We all know it's impossible to make a networked computer system secure. Even non-networked systems are vulnerable to unscrupulous operators. It seems unlikely that the government is proposing a human-free, non-networked electronic balloting system, so any e-voting system will be in danger of being attacked or subverted.

You wouldn't know this from The Implementation of Electronic Voting in the UK, a government report issued by the DTLR, Office of the E-envoy, Electoral Commission et al and based on research at De Montfort University. It reviews experiments in new forms of voting undertaken in the recent local elections, ranging from SMS mobile phone voting to all-postal ballots.

Perhaps 95 percent of readers will never get past the 11-page summary, so this boiled-down version of events dutifully notes that denial-of-service attacks, viruses, confidence attacks, disruption of power supplies and system overloading are the main dangers faced by e-ballots. But the summary then unexpectedly states that, "Technological solutions can be found and, indeed, are already in hand. Creating and maintaining public confidence in the voting process should lie at the heart of these solutions."

The meat of the report is, however, buried in the "Research Annexes" at the back. The laudably thorough Technical Options Report by Ben Fairweather and Simon Rogerson at De Montfort, describes a reality far removed from the cosy summary. The solution "in hand", sketched out by Fairweather and Rogerson, is a thought experiment not a proposal. If e-voters would not mind reformatting their hard disks and installing yet-to-be-designed encrypted open-source programs to cast their vote, then the solution is indeed in hand.

In other areas the summary does, thankfully, convey the correct picture. "E-voting on its own is unlikely to stimulate greater democratic engagement," it states, overturning the government's main reason for launching e-voting trials in the first place. "The assumption that e-voting will appeal to younger voters and boost turnout among certain groups of the population is not supported by the research," it adds. "While a small proportion of voters are inconvenienced by current voting methods and thereby prevented from voting, the majority of non-voters, and especially young people, cite other reasons for not voting." Dissatisfaction with the posturing of politicians, perhaps?

For whatever reason, the government seems intent on pushing through electronic voting. Perhaps problems with today's paper-based system are more severe than they appear. Perhaps, like the banks, the government is willing to operate a system full of holes - smothering the scale of the leakage in order to maintain public confidence.

But I feel we should see a more open and frank discussion of the technical merits, instead of being buttered up. Unfortunately, that option does not appear to be on the menu.

Have your say: contact IT Week

More IT Week Comments