Now the finger is pointed at Cullen
IT was intended to be a 'belts and braces' approach, to use Martin Cullen's phrase. In February, after months of defending the accuracy and reliability of electronic voting, the Minister for Environment was forced to bow to opposition pressure and introduce enabling legislation and set up a Commission on Electronic Voting.
But if it was a concession, it never amounted to a climbdown. Cullen remained as fervent in his support for the system as ever; insisted the tests had made it fool-proof, tamper-proof and judging by his prickly putdowns opposition-proof. He mocked the doubts of Fine Gael's Bernard Allen and Labour's Eamon Gilmore as "politicking", archly questioned the independence and credentials of computer boffins who expressed criticism of the system.
"I am on record as questioning some of the notional descriptions of them as experts. They were a group opposed to electronic voting full-stop. Somehow they were being [built up] as a kind of non-government organisation," he told the Irish Examiner.
Cullen continued to exhibit his hallmark bullishness and certainty. "Belts and braces," he repeated. There was an over-weaning confidence that the Commission would find no hanging chads in his system. Allen and Gilmore accused him of arrogance. He denied it. The word that came to mind yesterday was 'hubris'.
Cullen's electronic voting system collapsed under the devastating weight of a slim 12-page report. The key finding of the Commission was couched in very understated terms, that it was "not in a position to recommend with the requisite degree of certainty the use of the chosen system at elections in Ireland in June 2004".
But the import could not be understated. The moment it appeared, nobody doubted the whole project had been blown out of the water. The minister, who in fairness did not duck for cover, looked for small consolations. He pointed to another paragraph that the "conclusions were not based on any finding that the system will not work".
But he was grasping at straws. In reality the report amounted to a humiliation for the minister. Later on, it raised questions echoing just about every doubt expressed in the preceding months.
And for Cullen, the damage was made all the more complete by the take-no-prisoners manner in which he had tackled the doubters.
He had defended it trenchantly, had aggressively put down his detractors. But those he had lampooned as so wrong were so right after all.
Some of the potential gaps identified by the Commission and experts retained by it raise very serious questions about the minister's judgment.
They made a mockery of his claim that the system was comprehensively tested, that its credentials had been established beyond doubt. In the course of two pages on testing, accuracy and secrecy, the Commission identifies many of the issues, possible flaws and shortcomings, already voiced by the opposition and computer experts.
It found, for example, that the software used had been updated many times since it was piloted in 2002 and since full testing ("the full desk review") had been conducted on the source code.
However, the original desk review continued to be relied upon as the baseline for evaluation, even though ongoing changes were taking place.
The Commission's view was that each new software version needed to be reviewed and tested before it could be relied upon in a real election software. But this was not being done.
"The fact that new versions of the software continue to be issued in the run-up to the June elections is unsatisfactory," the report stated.
Neither was the Commission able to obtain full access to the source code of the system. It said such access was necessary to establish its trustworthiness.
The tests carried out to date, it further found, had been "insufficient". In particular, it criticised the very limited "end-to-end" testing (a full dress rehearsal, as it were) that had been carried. It pointed out the necessity of this, given the complex tasks the system would have to carry out in a real election situation; this June, it would have to handle, register, combine, mix and count votes for up to three polls (local and European elections, as well as a referendum).
The Commission's most eye-catching finding related to security:
"Experts retained by the commission found it very easy to bypass electronic security measures and gain complete control of the 'hardened' PC, overwrite the software, thereby in theory to gain complete control over the count in a given constituency."
These hardened PCs, it went on to state, were the "weakest link in terms of security". Moreover, no systematic testing had been done of the hardened PCs, "notwithstanding their susceptibility to either inadvertent error or deliberate manipulation by those with access to them".
This may all sound like the technical jargon of an anorak but all those potential flaws were raised by computer experts, some acting on behalf of the opposition, some independently. What's vaguely disturbing is all of their concerns were dismissed out of hand by Mr Cullen who went on to question their motives as well as their criticisms.
But now the Commission has found somebody could potentially gain access to the system, overwrite the software, and then gain control of and deliberately manipulate the system. Of all the findings, that was the gravest, the one that made a nonsense of the fully accurate/fully reliable/fully secure line that the Government persisted with.
Martin Cullen's argument that it was he who set up the Commission is a weak one. That only happened after concerted pressure. The debate was never about computers per se. It was about public confidence in the system, which will now be completely eroded by these findings. Belts and braces are useless when they have nothing to hold up.
