E-voting project fails due to a lack of adequate testing

Analysis: the technology: It is no surprise a rushed attempt to bring in e-voting failed, writes Jamie Smyth , Technology Reporter…

Analysis: the technology: It is no surprise a rushed attempt to bring in e-voting failed, writes Jamie Smyth, Technology Reporter.

The Government should not be too surprised that it has taken such a roasting over its rushed attempt to impose an electronic voting system.

The introduction of new technology has proved controversial throughout history, in particular during the early 19th century when disenfranchised workers smashed new machinery in factories in several British cities.

Even nowadays the public can be loath to accept new inventions. After all it took 10 years for people to embrace mobile phones, and even now there is a vocal minority in rural areas that, on occasion, chops down a mobile-phone mast in protest.

READ MORE

But Government attempts to dismiss the opponents of its electronic voting system as Luddites fell apart in spectacular fashion yesterday as the Commission on Electronic Voting published a scathing report on its strategy.

The commission found that the proposed computer system had not been tested sufficiently to ensure its reliability for use in elections.

In particular, it highlighted that no final version of the software used in the Nedap/Powervote computer system had been available to test.

This goes against one of the most basic rules in the software business: beware new releases.

Technological gremlins are most often found in new releases of software or products that have not been adequately tested.

Perhaps the best example of this is the release of software made by the technology giant Microsoft, which often has to rush out software patches to ensure its systems remain secure.

It is no surprise to note that weaknesses in the electronic voting systems were identified by the commission.

After all, if a firm with the financial muscle of Microsoft cannot guarantee absolute security, no company can.

This is one of the reasons that the commission unsuccessfully sought access to the source code - the basic software building blocks which make up the computer system - to enable it to subject the system to rigorous testing.

Yet despite the late offer of indemnity by the Government in case of a leak of the valuable code, it was not provided in full to the commission.

This is surprising given that even Microsoft, which previously jealously guarded its proprietary source code, is increasingly making it available to national governments to alleviate security concerns.

It may also have been the final straw for the commission, which unsurprisingly did not sanction an electronic voting system it could not test properly.

However, it would be wrong to interpret the commission's finding as the death knell for electronic voting. After all, several hundred million Indians are currently voting electronically at the polls.

Californians will also vote electronically in the coming presidential elections, although it looks like they will get the choice to cast a ballot if they want to.

Significantly, the commission emphasised that its conclusions were not based on a finding that the current electronic system did not work, but on a finding that it had not yet been fully proved.

Voting electronically has some benefits. It would more accurately count votes and potentially reduce the number of spoiled votes. It could also result in big cost savings over time.

But clearly the Government must listen to public concerns more and allow time for adequate testing of the voting systems. Otherwise paper ballots will be around for a time yet.