Carl Malamud, who would win an "open data genius award" from me*, is proposing to move the boulders that stand in the way of open, machine readable, free access 990 data. Check out the proposal to the Knight News Challenge (category "data"). Look at the price tag - $600K - amazing to think what could be done for such little money.
This is a big deal. There is an important infrastructure that has been built up around 990 data (Guidestar and several of their partner platforms - Great Nonprofits, GiveWell, etc) That infrastructure is important. So is the infrastructure built around foundation grants data (Foundation Center). These organizations are critical - opening the data once is great, but maintaining and keeping these data open and clean and accessible over time is what will make them a real resource for change.
Malamud has a track record in opening data and then changing the ways systems work around them. He's responsible for getting the SEC to provide corporate filings in this way. Long story short - way back in the early 1990s government wouldn't do it, Malamud opened the data, gave it back to the government, and said now you keep it going. New firms, new website accessibility, new products have been built from the data to keep them useful. This is the "edge" of change that opening the 990 data will bring us to - public data made available in a free and useful form incites existing firms to adapt and provide higher value services on top of those data. New enterprises, products, apps, licensing partners also emerge.
Opening up 990 data is not an end goal, it is a beginning. It is a beginning that will make possible much of what I call data philanthropy and many of the ideas listed here. It will require adaptation and innovation by existing data purveyors and open up opportunities for new ones. It will change the way long term data access and data sets are maintained (but it doesn't remove the need for those activities).
Check out the proposal. Think about what you could do differently if these data were readily available, mixable, portable, presentable. And let Malamud, the Knight Foundation and me know what you think might happen. Whether or not the proposal to the News Challenge is successful this is something we should make happen.
On a related note - here's a very cool new website of mapping data possibilities that Micah Sifry found and shared on Twitter this morning - Harvard's WorldMap Collection.
*Listen to this great webinar from Tim O'Reilly about Malamud and Government 2.0.