Candidate Web Site

North & East Ward

 

Kieran Cunnane
East of the River
 

Renmore..  Mervue..  Ballybane..  Ballybritt..  Doughiska..  City Centre..  Tirellan..  Menlo..  Sandy Rd..  Riverside..  Castlegar..

 

This section was posted before the evoting turn around.

 

In this case the expert is myself. As a Technology Consultant I help Companies including Medical Device Companies get their Software in shape for audits. Medical Devices standards are among the highest standards tested to.

So the question "e-voting - good or bad ?".

Well, the simple answer is good if you have a paper receipt of the vote and bad if you don't. The reason is that the data is randomised, encrypted and decrypted before the counting Software gets a hold of it to do the count. We're also relying on the Counting Software to interpret Ireland's electoral system exactly correctly and implement this in code. A well thought out e-vote system could be good, but the Nedap-Powervote System does not qualify as this.

If a paper receipt appears behind a glass screen before you finally cast a vote, this vote data is real (i.e. unencrypted and not manipulated in any way - you can get back each vote on paper).  This vote data can then be used in any recount. (Although with the proposed system you can get a printout or go to court to get a recount done, both of these are completely futile exercises because the vote data has been manipulated at least three times). The question becomes one of 100% trust. Software Engineer philosophers will tell you there are two types of system, those which have failed and those which have yet to fail.

The Dept. of Environment person who presented the system recently in County Hall does not understand Software and hence, could not comprehend my points when discussing them with him. This official views the system as one black box and refers to it as "the system", when in fact it is three separate software systems. Any system is as strong as its weakest link. After a few attempts to explain, he seemed to somewhat acknowledge my points. His public statement that the software is tested to within an inch of its life is untrue. What this official spoke about the different test houses is completely misleading. Only the User Interface software system was tested by three independent bodies. The other two components, one which randomises and encrypts the data, and the other with decrypts the data and counts the votes are each only tested by one independent body and the department (which is not independent). These two components are much higher risk than the User Interface. The voter will likely know if there's a User Interface problem with the voting machine, because that's the only transparent component of the system, i.e. they can see on the screen that the vote they pressed has been allocated a preference. In essence, they have tested the easiest, lowest risk and most transparent part of the system a lot and tested the rest of the system a little.

The department official said I could access the test strategy used by the test houses on the net. I did a fair bit of digging for this information and could not find any such data. As a Software Quality Expert performing an audit, I would request details of the risk management strategy and exact types of testing performed, in order to gain confidence in the system. If the Government provided this information, I would have substantially more confidence in the Nedap Powervote system. However, I would still insist on a paper audit trail as proof that democracy had taken place as you would request a receipt at a bank machine.

Conspiracy theorists also say that the count Software is done by three technicians (not engineers) in Holland and that these people are not visible enough i.e. they can be "got to" without our knowledge. This is possible, and it depends on your trust of democracy whether or not you believe this theory. I'm aware that one motivation of the Government for introducing a system is that some voters use X X X instead of 1, 2, 3 when voting. This seems to be particularly evident in supporters of the traditional parties who have tended to run more candidates. These are accidentally spoiled votes, but could be avoided through education.

Final thoughts:

I also did a press release on this issue.

--------------------------------------------------

Since evoting was dropped:

I had focussed my thoughts prior to this on the technical concerns. Since evoting was dropped, other technologists have convinced me that the paper system engages people far better and is better for democracy. I am now in favour of a full return to the paper based system.