Sunday, November 16, 2008

Voting online in the ACS Elections

Some years ago, when I was on their governing council, the Australian Computer Society changed its rules to allow for electronic voting. But the ACS doesn't have many elections and many of those, such as for the council itself, were done during a face to face meeting. The ACS therefore sensibly decided not to rush into e-voting. This year e-voting is being used for positions on the new national congress (which replaces the council) and for branches. Recently I voted electronically for the ACT Government in Canberra and last year for the Federal election. Voting for the IEEE and ACM are already electronic, so it may be that I never vote with a piece of paper again.

I wasn't involved with the selection of the ACS voting system. This uses a company called Bigpulse.com. They have an Australian office in Sydney, but I don't know anything more than that about them. The process is much the same as for ACM and IEEE elections. I was sent a message by the ACS with a link to the independent e-voting company. This then brings up an introductory message from the ACS returning officer explaining the voting system. The next screen than displays a form which looks like a traditional voting paper. There is a link next to each candidate to display their supporting statement (some you don't get with a traditional paper ballot or the systems used for the ACT and federal elections). The statements include a photo (if the candiate supplied one) which is handy if you can't rememebr exactly who is who. The statment appers on the ballot paper under the candidate's name which is easier than the previous ACM and IEEE systems, where you are referred to a separate web page.

As with other e-voting systems, this system checks the ballot is valid (the ACT and federal systems check for valid ballots but still permit the voter to cast one which is invalid). The system also invite comments on the vote (after warning the voter not to identify themselves). I submitted a comment suggesting the system require some form of identification of the voter (such as their member number or having logged in via the member section fo the ACS web site. This would be to allow for the unlikely situation where someone other than the ACS member received the email message and can so cast the vote. The ACM and IEEE systems get around this problem by sending the member a paper ballot which includes a code to enter for e-voting. This sending a paper ballot is a nuisance and I often have difficulty entering the codes provided, so such a solution has its won problems.

No comments: