Michael Cordover

***** STRETCH GOALS *****

I'm overwhelmed and humbled by the support you've shown so far. The target was reached within 1 hour and 40 minutes.

Even though we got to the target, more money is necessary to run this properly. To that end, some stretch goals!

TRANSCRIPTS - target $1,250 - ACHIEVED!

Everyone who backed at least $10 will receive access to transcripts of the hearing (N.B. you may not receive a copy per se, due to some possible copyright issues; but we'll work this out on the day). These provide a complete record of proceedings. They are also extremely useful to lawyers as a record of what the big issues are. If there's an appeal, having a copy of the transcript will be essential.

AN EXPERT TO SETTLE THE PLEADINGS - target $3,000 - ACHIEVED!

This extra money will go to a solicitor with expertise in administrative law who can take look at or draft the statement of facts and contentions that will go to the AAT. This kind of thing significantly improves the chance of success. The more money we get, the better this will be.

REPRESENTATION BY A BARRISTER - target $10,000

If we can get to $10,000 we can hire a barrister with experience arguing these cases before the AAT. This is very expensive but with good reason - barristers are expert advocates. A good barrister can go a long way to improving our chance of success.

***** THE STORY *****

The software used to count votes during Senate elections is kept secret.

What we know is that it's called EasyCount and that it was developed in-house at the AEC. As a result, the source code is a "document" for the purposes of the Freedom of Information Act. This should mean you can put in a request and get a copy of the source code. I did that, and was told no.

To get an independent arbiter to decide whether we should have access to this material I now have to put in an application to the Administrative Appeals Tribunal (AAT) - and that costs money. The money I'm trying to raise here.

Most of the story so far is on a mini-site dedicated to this application.

WHAT IS EASYCOUNT?
The Senate voting system is complicated. Sensibly, the process is mostly computerised. Individual humans count the first preferences on above-the-line votes and do data entry for below-the-line votes. This is done in the presence of scrutineers and by at least two independent counters, to avoid error. Then all that data gets fed into a program - that's EasyCount - and the result comes out.

WHY SHOULD IT BE OPEN?
Well, as a basic principle, we should be able to tell how our votes are being counted. For this purpose scrutineers are permitted at every stage of the vote counting process. However, there are no scrutineers inside the computer. While EasyCount has been independently verified, that is still only done by a single company. This sort of secrecy around the process of elections is unprecedented in Australia.

From a legal perspective, there's no reason that source code shouldn't be available through Freedom of Information (FOI). Everything produced by the government should be accessible by the people - who after all paid for it - unless there's a good reason to withhold it.

WHY WAS ACCESS REFUSED?
The AEC claimed an exemption on the basis that the source code is a trade secret. The essence of this exemption is that they are involved in a competitive marketplace (that is, fee-for-service elections) and that they would be less competitive if others had access to the EasyCount source code.

There are two key problems I have with this argument. Firstly, I just want the module used in Senate elections. No other election uses this model, so there's no competitive marketplace there. Secondly, the source code is still subject to copyright. Competitors could look at it, but they couldn't just grab a copy and start using it.

The AEC says that the Senate module cannot be separated from the non-Senate parts of the software. They say EasyCount is 360,000 lines -- and that it can't be separated into Senate and fee-for-service parts. If you know anything about software development, you might be raising an eyebrow about now.

The AEC also claims that the algorithm they used for counting votes is particularly efficient and makes them particularly competitive. I'm not convinced people go to the AEC for the efficiency of their computerised count, but more than that the algorithm for the Senate count is prescriptively specified in the Commonwealth Electoral Act. It can't be a trade secret if it's not secret.

ISN'T THERE A FREE APPEALS PROCESS?
There is - the Office of the Australian Information Commissioner (OAIC) provides free impartial review of FOI decisions. Unfortunately the OAIC is being disbanded at the end of December, as one of the measures in the 2014 Federal Budget. As a result they wrote to me and said they could no longer handle this matter, because it wouldn't be resolved before the end of the year. Instead I need to go to the AAT to have the AEC's decision reviewed. I have to make that application within 28 days of the OAIC decision, which probably means by Friday 25 July.

Some thanks

The main purpose of this is pay the application fee for the Administrative Appeals Tribunal. That fee is $861 because it will be paid after 1 July 2014. The target is about 10% over that figure, to account for Pozible and account fees.

Funds in excess of this amount will be used for other matters related to the application, including for example ordering transcripts where required. If I raise enough I'll look at getting some professional legal advice to help with the formalities.

All funds raised through Pozible will be kept in their own interest-bearing bank account until required. I'll publish a reconciliation of funds used on the mini-site. I'll keep funds until a final determination is made. If necessary funds may be applied to the cost of accessing the material levied by the AEC or costs associated with any appeal.

I don't believe I should gain any financial advantage from doing this. If I still have funds left over at the end of the project, everything remaining will be donated to the Open Australia Foundation. I will not claim a tax deduction for that donation (if any).

Most importantly: WE MIGHT NOT WIN.

While I believe I'm right, it's possible the Tribunal might disagree, either as a matter of fact or as a matter of law. There's some uncertainty both in terms of the facts of the matter and as to how the law applies in this situation.

Even if the Tribunal agrees with me, the AEC could still appeal that decision to the Federal Court. I would continue the fight, but being an unrepresented litigant at the Federal Court rarely ends in success.

I am not an expert in Freedom of Information law. While I'm a lawyer, I've less than a year of post-admission experience. I've never run a case in the AAT before. Most of my professional experience is in other fields: privacy, intellectual property and commercial transactions.

Once the funds are raised there is no question that the matter will go to the AAT. Once there I will certainly fight for the release of source code in accordance with my original request. However, I can't guarantee more than that. If the final determination is the source code does not need to be released, we'll end up with nothing.