Voting Information Project Roadtrip!

Recently, Anthea Watson, VIP Project Manager, and I hit the road to discuss the Voting Information Project with election officials in six states throughout the Northeast. The objective of this trip, and trips that we will continue to take for the next several months, was to persuade states to create and export XML feeds of voting information (e.g. polling locations, which candidates are on the ballot, how to register) to help us create an official, dynamic, and standardized dataset of voting information. In creating such a dataset, we hope to empower innovators across the country to develop awesome tools and help push that information out to voters (something that we saw throughout the 2010 Midterm Election Cycle).

Rhode Island State Capitol

It was great to have an opportunity to meet with election officials in person to discuss VIP, but we consistently came up against two sets of challenges. First, we forgot to check the weather forecast when we scheduled our trip. It turns out that a roadtrip throughout New England when a snow storm is pounding the I-95 corridor from DC to Maine is not the best plan. Among the lessons learned:

1. Salt and sand on the road may be great for traction, but salt and sand on the windshield is not so good for visibility, so it's a good idea to make sure your car is stocked with plenty of windshield wiper fluid before you set off.
2. 30 degrees seems remarkably warm after spending time in New Hampshire during a snow storm.

Connecticut State Capitol

The second set of challenges we faced were those relating to the structural impediments to open and transparent government data. On the whole, election officials in all of the states we visited were excited about the prospect of participating in VIP, but over and over again they faced structural challenges that make sharing their data difficult. Some of the most common challenges include:

1. Decentralized Election Administration: Several of the states we visited face challenges exporting a comprehensive, statewide dataset, because the localities in their states have primary control over election administration. This means that there is no formal process to require local election officials to send voting information (that is key to the VIP dataset) to state election officials.
2. A Lack of Data Infrastructure: In some states, even though the state could compel local election officials to send information to a central source, the state lacks data infrastructure to keep the data up to date and accurate. In some places, there isn't the ability to sync up the local database and state database, leaving the state data to get stale quickly. VIP can only rely on a database that is as accurate as possible
3. Static, Rural Polling Locations: In a few places, polling locations that seldom change result in a struggle to get local election officials to see the value in any sort of database. Rural polling locations often haven't changed for years or even decades. Towns in these areas are so small that one polling location serves the entire electorate, and it is assumed that everyone knows where to vote (or can ask their neighbor). In these places, there's no obvious incentive for town clerks to maintain their polling location information in a database.

As we continue to work with election officials, we hope to find solutions to overcome these structural impediments. Currently, we look forward to working with local election officials to create tools that will make the work that they do easier and incentivize creating a data management infrastructure.

Also, the next trip we plan will most definitely be to some much warmer southern states.