You’ll often find the term “open source” here, used to describe either the source code for software, or the license that allows you take that source code and use it. But “open data” is just as important. A recent New York Times article read almost like I would have said it, starting with “It’s not boring, really!” or to be precise, the title “This Data Isn’t Dull. It Improves Lives”. NYT’s Richard H. Thaler starts on exactly the right point:
Governments have learned a cheap new way to improve people’s lives. Here is the basic recipe: Take data that you and I have already paid a government agency to collect, and post it online in a way that computer programmers can easily use. Then wait a few months. Voilà! The private sector gets busy, creating Web sites and smartphone apps that reformat the information in ways that are helpful to consumers, workers and companies.
That’s exactly the approach to election open-data that we’re taking in the next steps of election data management at TrustTheVote. Right now, I have to admit that the current set of election data might actually be fairly boring unless you have an interest in ballot proofing ot electoral districting (which we’ll get to in a couple more installments in our Bedrock series). But the next step might be more interesting: combining that data with election-result information, which up to now we’ve managed only in the context of the TTV Tabulator and some common data formats that we’re working on with some help from EAC and NIST.
But by adding election result data back into the election definition data, we get the the next cool part: a new TTV component that is like the current Election Manager (which would remain deployed privately within a BoE or state), but with only the ability to publicly provide election and election result data via a Web services API. That, in turn, becomes the back end for an election night reporting system Web site and smartphone app.
But perhaps just as important, that API would be publicly accessible to any software, including 3rd party sites and apps, as Mr. Thaler points out. Right now, most election definition data and result data is locked up in the EMS of proprietary voting system products, with a sliver of it published in human-oriented reports and sometimes web content. A new TTV component for Web publication of that information would be a fine first step, but by itself, the data would still be limited to availability in whatever form (however broad) the human-oriented Web interface provides. The really key point, instead, is this:
Not only publish via a Web site, but also make all the data accessible, so anyone can do their own thing to slice and dice the data to gain confidence that the election results are right.
And that point — confidence — is where we rendezvous with the NYT’s point about data improving people’s lives. I’m not sure that open-data for elections can save lives, but I think it can help save some people’s faith in election integrity. And now is certainly a trying time, with a variety of election-related litigation news from NY and CO and IN and SC, all seeming to say that election irregularities or outright fraud — even at the top with IN’s highest election official being indicted — is all over the country.
There’s an old saying “one bad apple doesn’t spoil the whole barrel” but we should not have to take it on faith for the large barrel of honest diligent election officials and the valid results of their well-run elections. A bit of open-data might actually help.