Sure. Let me take the first question first.
On releasing all we can and protecting what we must--and I'm sure the others will speak to this as well--let me say first, on the process of just getting started and the process of releasing data, one of the reasons we indicated this concept of high-value data and made the definition quite broad was precisely in order to encourage the hard work of trying to get the data that we could.
It's not simply an issue of political will or of dealing with controversial data that may be protected by national security or may deal with private information. There is data that is sitting on paper and that is not digital. There is data that is digital but isn't searchable. There is data that is sitting on servers that are essentially so creaky that if you tried to download the data from those servers you would crash the whole office, which is the case, for example, in the patent office and the reason why the patent office did a no-cost contract with the private sector to search the data for it while it tries to redo its back-end infrastructure.
We wanted to create this culture of transparency by starting the practice of being open as a way of effecting that culture change, and that really meant beginning with information that would be uncontroversial and starting to get into the habit of putting out that data. That said, there also are processes when information goes up on Data.gov for conducting a national security review of the information that's posted, but it's really about creating that culture through practice.
That partly gets to the second question, about incentives for behaviour change. The more we do, the more we can celebrate what we do. We invited to the table not simply White House oversight of the agency data inventory process, but outside groups, good-government groups, open-government groups, to be part of the process, hopefully both to celebrate and to criticize when that work isn't going fast enough, and also to help with the very hard process of actually building data inventories, which is a very hard technical process, not just a difficult political process.