And they're all outstanding questions, every one of them.
I want to be clear about the backlog issue, to start with. When I was mentioning backlogs, I was really referring to backlogs that exist under the Freedom of Information Act, which is the core backbone for public access in the United States. In some respects, this is an outmoded model whereby the citizen has to file a request to the government, and the government, through appropriate legal channels, determines whether it meets the law's standard for disclosure of the record.
Those requests coming from the public have accelerated over the years, and as a result there is a huge backlog. The agencies were given a mandate by the President to reduce those backlogs by 10% by this year. We will see in another month, based on data coming out, whether the agencies have achieved it. That's what I was referring to by way of backlogs.
But I think your point is still a good one, which is that there is a tendency for government.... If agencies are told to disclose, the natural proclivity is to throw the doors wide open on everything. As a result, having so much may mean that you have nothing, because you can't get through it all; it's just an overwhelming amount of information. The real trick is to get to what I would call high-priority information; that is, the information that key stakeholders of the public as well as Congress—the whole range of stakeholders—have an interest in knowing about.
This is where some of the new social media tools and new technology may prove useful for the interactive process of involving the public and other stakeholders, in helping to determine what the key pieces of information are that agencies should be disclosing voluntarily, without this kind of formal freedom of information request.
The Obama administration tried to move in this direction with something called Data.gov. It has been an extremely useful tool, but it hasn't quite gotten us to that level, because agencies are throwing all kinds of data onto Data.gov. Some of the weaknesses relate to your question about data quality. On Data.gov, it is not unusual, if I open up a file, to have no headers across the row. Think about an Excel spreadsheet without any headers; all you have is numbers. It won't tell you how to use the file. You don't know what the column letters are and you don't get a data dictionary.
If you don't have the metadata and the metadata is sloppy, then the data itself is not all that useful to the public. That's one kind of data quality issue.
The other kind is what I was talking about, cases in which the data itself is just flat out wrong; it just needs to be improved.
Then there's a third kind: missing data. On the Recovery.gov website that I was mentioning, for “place of performance”—where the money was actually spent—sometimes people put in a post office box. That's no place of performance. So there's that kind of data quality.
The last thing I want to mention to you, which is a very exciting prospect that this administration is exploring, is using something called ExpertNet. It's a way of allowing interaction by the public with various experts within the agencies around specific subjects. This is a brand new enterprise that was just proposed by the administration, and they just finished about two weeks ago taking public comments on the structure of this kind of ExpertNet.
That may be a new way. In truth, we are experimenting with ways of engaging in democracy now. We're going to have some failures along the way here.