Friday, May 07, 2010

E-voting is not the solution to poll chaos

It has only taken a few hours for e-voting to be proposed as the "answer" to last night's chaos as polling stations closed. It is dismaying to read such badly informed commentary as this:
Andy Williamson, director of digital democracy at the non-partisan think tank, the Hansard Society, argues that "a lack of desire to change" is a better explanation of any resistance to electronic voting than security concerns.

He acknowledges the risks with electronic voting, but says "you have to put this in the context of the current process, which we mostly accept, despite the obvious flaws and risks."

Those risks, he says, include "the lack of positive voter verification, the obvious risk of moving big piles of paper around, and the fallibility of manual counting."

As has been explained over and over and over again: personal computers and the Internet are nowhere nearly trustworthy enough to conduct national elections. Even voting computers at polling stations are far too easy to hack, as Hari Prasad, Alex Halderman and Rop Gonggrijp demonstrated again just last week in India:

Would it be so difficult to employ a few more polling station staff, and pay them overtime to ensure everyone is able to cast their vote?


Cyberdoyle said...

yep, why not stay analogue forever????

Its too hard to do it today, but tomorrow is just around the corner. There are enough brains in this country to sort it. It could save millions.

I agree, currently the kit isn't up to the job, and a third of the country has such crap connectivity it isn't possible, but it must be investigated and done as soon as possible. IMHO. The current system must cost a fortune in time, money and carbon footprints.

Richard said...

The Electoral Commission's concerns over security were, as I understood it, more to do with verifying the code rather than any actual instances of fraud or hacking that took place in any of the UK pilots.

But you write as if the current system would be perfect if only there were a few more staff, and therefore there should be no attempt to look for alternative systems. Yet the current system is massively unsound - I know at least one person who had (but didn't use) multiple opportunities to vote.

The question really is not about whether internet voting would be 100% foolproof - no system ever would be - but whether the advantages of it outweigh the disadvantages.

Multichannel voting, where there are more options available than just the polling station, postal votes and proxies, has to be the way to go.

Also, research by MORI back at the time of the pilots found reasonably high public support for internet voting, and public consultations by the government have found continued support.

Tom said...

I think eVoting is an incredibly bad idea for a lot of reasons.

1. It's just unnecessary. It's necessary in the USA because their ballots are huge, including presidential/vice-presidential races, senators, congressmen, judges, local officials, school board members, judges, dog catchers, etc. In the UK, we just count one thing. Admittedly, if we were to change from FPTP, it would be more complex, but still not that hard. We count our ballots astonishingly quickly, and even if PR adds a few hours or so, I'm not that bothered.

2. Practically all eVoting systems currently invented are flawed in design, if not in implementation.

3. Remote voting is not confidential, and I specifically mean within the family. In the privacy of a voting booth you have total freedom and no-one is watching. At home, you have peer pressure from family, and someone with low confidence may feel uncomfortable asserting their right to secrecy from their spouse. That affects the result. After realising this, I went from being pro- to anti-postal-vote, and voting anywhere outside a private one-person-per booth is just as bad.

4. Practically all of the problems with this week's ballot were down to incompetence:

a) not enough ballots to cover the known list of voters;

b) lack of extra staffing if needed at the end of the day;

c) Failure of staff to read the rules wrt. the cut-off time, leading to stations being open longer than legally permitted;

d) failure of Presiding Officers to notify other officials when there were problems, so resources could be diverted, etc.;

e) lists not being updated in line with the Electoral Register. All of these problems could be fixed by having competent, trained (and tested) electoral officers who actually bother to read the manual.

I really can't see how computers can help the UK voting process... and I'm speaking as a programmer with ~20 years experience! Everything can be solved by better training and testing of officials and solid procedures.

Dave Birch said...

Tine to look at DigiCash again?