Samba man 'Tridge' accidentally helps to sink request for Oz voteware source code
Dr Andrew Tridgell's testimony may not have helped the cause with candid remarks on naughty devs
Dr Andrew Tridgell, creator of the Samba file server and the rsync algorithm, appears to have inadvertently helped to sink a freedom of information (FOI) request for access to the source code of software used to count votes in Australian elections.
Tridgell was called as a witness by Hobart lawyer Michael Cordover, who sought access to the source code in order to subject it to public scrutiny. Tridgell was asked, among other things, whether the release of the source code would lead to copyright violation.
Cordover sought the source code because Australia's 2013 election produced some irregularities. Australia's Senate employs a complex voting system that sees candidates elected after they receive a “quota” of votes, either one seventh or one thirteenth - plus one - of those cast depending on whether half of the chamber (as is usual) or all seats are up for election. Preferential voting means that once a candidate reaches a quota, their votes pass to a voter's second preference. The same happens for candidates who do not secure a quota. With more than 100 candidates sometimes contending Australian Senate elections, the count is therefore lengthy and complex.
Australia's Electoral Commission (AEC) has therefore, sensibly, tried to automate the count with software called Easycount.
Cordover hoped that by accessing Easycount's source code, he and the populace could reach certainty about the software's efficacy and thereby improve confidence in the workings of Australia's democratic processes.
The AEC repeatedly rebuffed Cordover, who then had to fight off the label of a vexatious user of FOI requests. The lawyer then vowed to pursue the matter in Australia's Administrative Appeals Tribunal, which handed down its ruling on the matter last Friday.
Long story short: Cordover lost because the Tribunal decided that releasing Easycount's source code would diminish its commercial value, which as the AEC uses the software to run elections around the world (for a fee) would be a bad thing.
The longer story has elements of tragedy, because Cordover's case “relied heavily on the evidence of Dr Andrew Tridgell” and placed “considerable emphasis … on questioning Dr Tridgell about his involvement and professional beliefs in free and/or open source software.” Tridgell has also worked on electoral software, so was held to be an expert on that topic and, as a result of his open source expertise, on how source code is used in the commercial software industry.
That questioning, on The Register's reading of the decision, saw Tridgell say that releasing Easycount would create a risk that unscrupulous developers, including those of commercial software that competes with Easycount, would use the code in their own work.
Here's the summary of Tridgell's testimony, as presented in the Tribunal's decision.
”While the Tribunal was impressed with Dr Tridgell’s qualifications, expertise and candour, the effect of his evidence was to the effect that their source code had some commercial value and could benefit other electoral software developers. In his view he considered it to be unwise for a competitor to access and use another’s software, but he acknowledged that this occurred, even when there was potential to be caught and subjected to litigation for copyright breach.
Dr Tridgell’s evidence amounts to a statement that Easycount has commercial value, but reputable persons in the software industry are not likely to copy or use the information in it.
Despite this, Dr Tridgell properly conceded that not all persons in the industry were reputable and that copying and use occurred, even when copyright law supposedly presented as a deterrent.”
“This concession,” the decision says, and other evidence that the AEC would lose value if the source code were released, “persuades the Tribunal that as a question of fact, there is a reasonable expectation that the value of Easycount could be at very least diminished, if it were disclosed.”
On The Register's reading of the case, the Tribunal therefore took Tridge's remarks as evidence that even controlled release of the source code could damage the AEC.
Cordover told The Reg he is "disappointed in the decision" and is "reviewing it to determine whether there would be strong prospects on appeal."
The likelihood of having to pay the AEC's costs if an appeal were unsuccessful means "I am unlikely to pursue it".
Cordover said "I will continue to lobby for the source of EasyCount to be released, and to work for more open source code from government generally."
The lawyer also said Dr Tridgell's evidence was "... fantastic, and the Tribunal noted how impressed it was with his qualifications, expertise and candour."
"The test which was applied was whether there was some chance, more than an irrational, absurd or ridiculous chance, that the AEC would suffer any commercial loss, even if that loss were incidental or trivial. The Tribunal found that the risk of some disreputable programmer might use it as a road map for their own development, or might engage in copyright infringement."
"I wouldn’t have expected Tridge to deny the existence of copyright infringement, and the degree to which the Tribunal relied on his evidence is indicative of how well he did as a witness." ®