Wired’s Kim Zetter reported on our Hollywood Hill event, in an article titled “Nation’s First Open Source Election Software Released.” I got a few questions about that “First” part, and I thought I’d share a few personal thoughts about it.
First of all, there is certainly plenty of open source software that does election-related stuff, as a few searches on github and sourceforge will show you. And there are other organizations that have had open source election software as part of their activities. FairVote‘s work on IRV software (some done by TTV’s own Aleks Totic) is a notable example. Another is Ben Adida‘s notable work (on crypto-enabled ballot count verification among other things) represented in his Helios system, recently used in a real university election. And Helios is only one of several such systems.
Now, what is “first” about OSDV’s release of a part of the election tech suite that we’re developing in the TTV project? In my own view, one first is that the software is targeted at U.S. elections specifically, and on providing automation of election operations in ways that match the existing practices and needs of U.S. elections officials. The many well-meaning efforts on open-source Web apps for Internet voting, for example, are laudable work, but not what most election officials actually need right now or can legally deploy and use for U.S. government elections.
So I think that it is a first indeed, when you combine that factor with all the other attributes of open-source, non-proprietary, open-data, operations-transparent, and so forth. It’s not exactly a great invention to do what we’re doing: pick a target for deployment; talk to the people who work there; find out what they want, and how to deliver it without asking them to also change the way that they do their work. Applied to election tech development, that approach is fundamental to what we do, and whether our work is “first.”