Archives For infotech

ballot

Political parties don’t take well to losing. They’re in the business of winning elections because a winner attracts money and attention, money and attention they can use to grow stronger. So in the fallout from this presidential election, one wing of the Republican party is calling for a much needed and long overdue period of self-reflection in which the GOP swings close to the center and becomes much more libertarian but without the borderline anarchist overtones, and another is mourning the death of traditional America thanks to liberal freeloaders and spinning constant conspiracy theories. This reaction is not too dissimilar from what you could see after 2004, when swarms of liberal bloggers sighed heavily about the end of the America they knew to bloodthirsty, Bible-thumping theocrats, and tossed out conspiracy theories about voting machines. But in the conservative blogosphere, a conspiracy theory about voting and technology that doesn’t target voting machines is now trying to get some traction by accusing coders of political sabotage.

Basically, the theory goes as follows. Romney’s campaign created an app called Orca to track the Republican vote and give conservative voters a tool to submit what they saw as obstructions to voting on the spot via their smartphones. One of the companies employed a developer once contracted for some unspecified work with the Gore campaign, and another Orca developer was black and therefore, a likely Obama supporter. And so, they and all their likeminded friends who were working on this project intentionally sabotaged it, making it difficult to really crank up a get out the vote effort and report voting incidents and mishaps in a timely manner; the app was too slow, frustrating too many users, and you can see that in the low turnout for Republicans. That’s a little odd to say the least when you consider that hundreds of millions have been spent on ads, canvassing, robocalls, mass mailings, and every other known effort to get people to vote during the last two years. Being slammed with election talk for a year didn’t get enough Republicans to the polls but a vote-tracking app would’ve made a multi-million vote difference?

Now this is an interesting election conspiracy theory because it’s the first one I’ve heard going after developers and campaign tools rather than the classic allegation that voting machines are being rigged. It’s true that voting machines were rigged in some cities, but they were rigged for Romney and the GOP so that angle wouldn’t have worked. Going after Orca shows that there’s some original thought happening here, even though the original thought is holding a stint more than a decade old against a developer who can easily end up working on a campaign he would rather not support and whose code will be reviewed before being added into the final product, and indulges in playing the race card. The odds that a couple of developers snuck some sort of malicious code into Orca aren’t all that high because delivering a bad product means that you’ll have a black mark on your track record and the developers in question didn’t simply volunteer to work on code for a campaign. They’re employees who were assigned some units of work, not a small team of tech-savvy political activists who volunteered to create Orca for Romney.

But if the developers can’t be held liable without a lot more proof and source code to back it up, why would Orca suddenly fail on election night? The data points to a simple but pressing issue that has little to do with the code: infrastructure. Or rather a lack of it. If you’re going to collect a lot of data in a very short amount of time, you better be ready for it. When just ten servers were hit with 1,200 or so requests per minute and the mobile part of the system was housed on only one server, it was just a question of when the system would either crash or jam so badly that for all intents and purposes it appeared dead to the outside world. If Orca was built with the proper scale in mind, it would’ve lived on a hundred servers and the mobile end would take up half of all that capacity. There would’ve been special agreements with ISPs to get the most throughput on election night. None of that seems to have been done according to reports across the web. And when we pause to consider that Romney staffers could’ve counted the number of servers then ask "are you sure that’s enough?" to catch the issue, calling this sabotage seems hyperbolic.

What seems far more likely is that Romney dropped the ball and those in key positions of all his campaign activities failed to do their research and follow up with the Orca team. Even if it was a perfectly working app, it was unlikely to make all that much of a difference because it could only track who voted and where, not spring into action and get more people to the polls. When at the end of the day we’re talking about a difference of nearly 3.4 million votes, Orca would’ve needed to get more than 1.8 million Republican voters into the booths within several hours. Romney had spent almost a decade campaigning. If all the hundreds of millions he and the GOP spent, along with the barrage of exhortations to vote from talk shows, Fox News, and right wing blogs made little difference, what exactly would a tracking app do? If anything, the campaign did what many techies like me see on a daily basis in the business world. The boss went after a buzzword, then threw a lot of money and effort into a tool he didn’t know quite how to use but which he can show to reporters as something very comparable to something used by his main competitor…

Share