I was fortunate enough to attend the Complexity & Content Data Journalism Symposium at UC Berkeley last weekend. The lineup of speakers was fantastic, ranging from nearly guerrilla projects to evangelists from established monsters such as Google.
That there was a division in the speakers between tech/innovation and policy (the former speaking before lunch, the latter after) was interesting, and not only framed well the challenges in getting started in data-driven journalism (*cough*#ddj*cough*), but the contextual and mission thinking required to do it right.
Even the visual guru Alberto Cairo (who totally pulled off a last-minute change to his presentation in order to fit in a python "language or giant monster?" joke) walked the attendees through what he saw were problematic and successful data visualizations – the key difference being that the successful ones use data and visuals to show a difference, to tell the story.
Catherine Bracy of Code for All (the international effort of Code for America) made it clear that you didn't have to know how to code, but you had to know how to read statistics and data and ask intelligent, rigorous questions. (Funny, I'd just had a conversation at lunch about how the push for "everyone should code" can eat up the oxygen needed to teaching critical thinking.)
It's not a humblebrag to say that getting to introduce Catherine's presentation was seriously nervous-making. She rocks, for the public good. As do the other presenters and attendees. You should have been there.