For lack of GOP message guru Frank Lutz, it was a project with no name. And yet it was one of the (many) precedents set during an historic election. Commissioned by Obama for America campaign, GIS consultant Karsten Vennemann used opensource components, including OpenLayers, to map voter registration in battleground states. By combining millions of public voting records with daily updates from on-the-ground troops, these interactive map helped organizers priori to map voter registration in battleground states. By combining millions of public voting records with daily updates from on-the-ground troops, these interactive map helped organizers prioritize canvassing efforts. Used in about 12 battleground states, the opensource maps proved to be a progressive tool in an already progressive campaign.
In the wake of an ESRI proposal for NationalGIS, we take a moment to reflect on how and why opensource components came to be part of the Obama mapping stack.
It started with an email. Having worked with Washington Conservation Voters (WCV) mapping data for environmental advocacy, Vennemann saw the power of geographically plotting and analyzing voter records. He emailed the Obama team on July 10 with an idea—map voter data in battleground states to capitalize on limited campaign resources.
His proposal must have synced with Obama strategy. A couple of forwards later Vennemann was talking to the Obama technology team and by August 13 he had been contracted the work. Vennemann uploaded millions of geocoded voter recorded into the open spatial database, PostGIS and setup a system where on the ground resources such as Voter Activation Network, could help update voter information. All was rendered with the MapServer engine and viewed in OpenLayers.
This project stands out for 2 reasons. First, no other presidential campaign has made such extensive use of mapping for strategy decisions. Even before Vennemann’s involvement, the campaign had up to 4 onsite GIS experts. While web mapping tools have been used by the electorate to reflect upon the status outside a campaign (e.g., Fundrace), the ease of integrating webmapping within a campaign was only possible because of a pre-existing technology and mapping infrastructure.
Second, the choice of using such open source components as OpenLayers and PostGIS for these ends is significant. In the shadow of an administration that seemed to value obfuscation, I would like to taught ideological considerations of transparency as the reason for opensource. According to Vennemann, however, it really came down to ease. . . and money.
At the time of the proposal, the Obama campaign was considering proprietary software. However, it became clear that the combination of licensing required to meet the campaign’s canvassing needs made such options cost prohibitive. According to Vennemann it was much easier to speculate the costs involved for implementing and updating an opensource tool. And that such budget projections were much lower than previously considered routes.
In this vein, OpenLayers proved to be the ideal viewing component because canvassers wanted information along with sattelite images. With OpenLayers, Vennemann was able to pull in the Google API for viewing, but was not bound to it. As he says, “OpenLayers frees you from the Google API. Basically it’s the freedom of choice.”
This value of choice is all the more pertinent as the promise of Obama comes into office. Vennemann’s map demonstrates how powerful and cost effective opensource can be when combined and tailored for a specific use. A recent proposal by the proprietary technology company ESRI recommended that the Obama administration invest over one billion dollars to build a national GIS to both serve a public need and stimulate the economy. While this need might exist, it would behoove the administration to remember the innovation that happened on the campaign trail. Supporting an initiative (only) buoyed by proprietary resources may not lead to the stimulus and tools we need.