At Open Data Day, our focus was on getting people together and building momentum — rather than finishing "apps." There was no judging or prizes, no pizza or soda, and a clear welcome to anyone without a technical background.Continue reading
Last summer, Sunlight released a series of Open Data Guidelines in reaction to a surge of municipal open data policy making. In anticipation of revamping these policies this summer (to add fresh context, ideas, and exemplary language) and in reaction to a recent surge in open data policy collaboration as evidenced by the interactive Project Open Data and the newly public (beta) Open Data Stack Exchange (or maybe more accurately in reaction to the Meta Open Data Stack Exchange...), we wanted to provide a roadmap to the world open data resources and recommendations that are available to put these resources in context of their evolution over time–a guideline to Open Data Guidelines, if you will. The first step in navigating the open data guidelines out there is to examine the chronology of how they surfaced.
The timeline below provides a landscape of current open data policy guidelines, guidance, and principles that exist and showcases the chronology in which they have manifested, each guideline often directly building off of (or crafted in reaction to) its predecessor. Looking at these guidelines in context exposes the pragmatic and technical evolutions in thought that have occurred under the banner of open data pursuit: from the foundational drive to define what information is legally available (through FOIA and other public records laws) to the trailblazing concept of proactive disclosure (where "public" access means "online" access) to establishing the qualities that make data more accessible and usable (emphasizing structured, bulk data, unique IDs, and APIs). The dialogue for discussing open data policy guidelines has itself evolved from the gathering of smaller open government groups of: Open House Project, Open Government Working Group, the Open Government Initiative, and early collaborative efforts such as the Open Gov Handbook, to the editable Project Open Data and the Q&A Open Data Stack Exchange.
A couple Saturdays ago, I helped Josh Tauberer, Kat Townsend, Dmitri Kachaev, Sam Lee, and Julia Bezgacheva organize a hackathon... View ArticleContinue reading
Here are a few of the more interesting media mentions of Sunlight and our friends and grantees from this week:... View ArticleContinue reading
Here’s a little fun for your weekend. Yesterday, the Open House Google Group turned into an outright poetry slam. David... View ArticleContinue reading
The ever-amazing Josh Tauberer , creator and custodian of Govtrack.us , never runs out of really good stuff to do... View ArticleContinue reading
One theme running through what we're doing here, in my mind at least, is to blur the line between the explicit and the implicit, or, put differently, to make evident those things which were only implied. Effective data availability is certainly a case of this. Every time there is government information that is publicized in that satisfying-due-diligence, html, doing-as-we're-told, this-is-the-full-extent-of-our-authorization, only available in a reading room at 2:30 PM on Wednesday sort of way, well, that's an example of the implied. That data is only public by implication, since there is a significant barrier to it's effective use, reuse, access, or timely updating.
I'm loving the conversation about debate transcripts that Josh just posted about, because it's a great example of information becoming increasingly public, even though it was in plain sight all along. All public televised debates are, by their nature, quite public. The transcript or video/audio, however, has been less available, so much that the battle over their fair use continues even now. Despite this struggle, innovative presentations of this most hotly contested, most scrutinized of public appearances are popping up with increasing frequency. Josh's post took the speaking time from the NYT and calculated the statistical correlation between candidates' time speaking and their poll ranking. I just came across this tool (application?) that allows for all sorts of user-defined analysis of the debate transcript. You can see, explicitly, how many times the candidates said a term of your choice, and the text from the transcript is available right along with it. This is the sort of thing that we're lucky enough to witness developing, as long as the data that drives this sort of innovative presentation stays open and available. (more after the jump.)