Thursday, 2 August 2007

Do it right or don't bother

The Electoral Commission's report into the various internet/telephone/advanced voting and electronic counting pilots in the English local government elections has been published and they have concluded that it isn't currently worth the effort.
"we do not see any merit in continuing with small-scale, piecemeal piloting where similar innovations are explored each year without sufficient planning and implementation time, and in the absence of any clear direction, or likelihood of new insights." — Peter Wardle, Electoral Commission chief executive
So the official body for keeping an eye on all things electoral in this country agrees that this set of pilots were rushed, lacked for sound planning and didn't really teach us anything new, especially given that they were the sixth set in seven years.
There were one or two key worries highlighted about security and perception of the possibility for fraud. In Sheffield (the election I was an observer for) the risk management and accreditation document set (RMADS) had no evidence of having been reviewed or approved by the Security Accreditor or indeed having made it beyond the draft stage. There were also concerns that the risk assessment was a standard one for generic web based services and not specifically for an e-voting system.
The lack of observers at the electronic voting count, the sealed envelope nature of the electronic voting, the perceived slight of hand of the electronic votes being added into the totals and the lack of a recount facility were all causes for concern by those involved in the political process and observers alike. It is important to addresses these perceptions even I they are built on a shaky understanding, the concept of a recount for example, given that any system which could give different answers at the end would have to be binned immediately.
"there was insufficient time available to implement and plan the pilots, and the quality assurance and testing was undertaken too late and lacked sufficient depth. The level of implementation and security risk involved was significant and unacceptable." — The Electoral Commission Key issues and conclusions
Other assessments by the Commission are on a very generous basis, the bullet points on encouraging voting and facilitating counting are based on benefits to a very small number of people or by a tiny margin. The really important conclusion that needs to be remembered for when the government are doing their best to ignore this report to go ahead with what they and not the electorate want is this: The pilot scheme had a negligible impact on turnout. There was some controversy in the run up to this set of pilots about how successful previous pilots had been in increasing turnout, so we need to be sure when, in the future, the subject comes up we can point strait at this and tell them they are wrong.
The bottom line of course is that for once a government technology project came in on time, but at a cost. Telephone voting in Sheffield cost £425 for each person and this was after a discount of a third of a million pounds. While this cost would be reduced in a more widespread scheme than this pilot, it does seam an awful lot of money to try out technology that isn't ready yet, doesn't have the support of a good chunk of key stakeholders and reduces the transparency and security of the voting system while making fraud and vote selling easier.
For more information on the electronic voting pilots read the Open Rights Group briefing paper.

No comments: