As I often do, I had a thoughtful Martin Luther King Day -- as you can see from my still pondering a couple days later. But I think I now have something to share. Last time I wrote on MLK, I likened two unlikely things:
- King's demand for social justice and peace, using Isaiah's prophetic words that "Justice shall roll down like water, and righteousness like a mighty stream."
- My vision of really meaningful election transparency, stemming from a mighty torrent of data that details everything that happened in a county's conduct of an election, published in a form anyone can see, and can use to check whether the election outcomes are actually supported by the data.
Still a bit of a stretch, no doubt, because since my little moment by the waterfalls of the MLK memorial in San Francisco, I've had rather mixed success in explaining why this kind of transparency is so difficult. Among the reasons are the complexity of the data, and the very inconvenient way it is locked up inside voting system products and proprietary data formats.
But perhaps more important, it is just a vexingly detailed and complicated process to administer elections and conduct voting and counting -- paradoxically made even more complex with the addition of new technology. (Just ask a New York state election admin person about 2010.) In some cases, I am sure that local election officials would not take umbrage at the phrase "Rube Goldberg Machine" to describe the whole passle of people, process, and tools.
So, among my new year's resolutions, I am going to try to communicate, by example, a large part of the scope of data and transparency that is needed in U.S. elections. It will take some time to do in small digestible blogs, but I hope the example will serve to illustrate several things:
- What election administration is really like;
- What kinds of information and operations are used;
- How a regular process of capturing and exposing the information can prevent some of the mishaps, doubts, and litigation you've often read about here.
- Last but not least, how the resulting transparency connects directly to the nuts-and-bolts election technology work that we are doing on vote tabulation and on digital pollbooks.
One challenge will be keeping the example at an artificially small scale, for comprehensibility, while still providing meaningful examples of the data and the election officials' work to use it. On that point especially, feedback will be particularly welcome!