 Well, first we're trying to make data widely available, and we're going to do that by recommending policies that would increase the sharing of data. So right now, the data sharing policies at NIH are pretty low profile. What we like to see is the data widely shared, but once it's available, it needs to be usable. So the other thrust we're doing is supporting the development of standards so that this person can work with this data. And then once the data are available and usable, people have to find it, and once they find it and use it, they should cite it. And so Group 1 is doing all of these things. The examples where you see data being reused are the most powerful. There's a reason why everybody's heard of the Framing Hand Heart Study. It's been going for since the 1940s, and that's because their data is actually available to qualified researchers, and the findings they've got out of there are great. But the data themselves have resulted in many, many additional publications, and you saw the same thing with the Human Genome Project, that how well they made their data available and how rapidly and easily, just completely revolutionized research. NIH and the U.S. citizens already paid once for the best possible data, and the whole point of data sharing is that this valuable data can be reused and more benefit can come from it for researchers and for the public. For the most part, scientists are not incentivized to share their data, with very good reason. Promotion and tenure depends on the number of scientific papers and the quality of those papers and how many times they're cited and so forth. And so, of course, when a scientist spends years collecting data, especially if it's a very rich data set, he or she does not want to make that available to everybody right away. That's understandable. But there are no incentives for that, even after the fact. So the idea here is, through the ability to cite specific data sets, for example, you can give credit to the people that are generating those data sets separate from the papers. And in fact, you can imagine that if a data set is collected and is the basis of a scientific paper, you can imagine authors A and B on the scientific paper. And authors A, C, D, and E as authors of the data set. And so the people getting credit for the data-related activity could be different or in different order than those for the scientific paper. And suddenly you have a real way to incentivize those kinds of activities. It's not just for the researchers. It's also for the general public, right? There are a lot of people that are more and more interested in finding out about health and disease and how can they find out what's going on. And the goal of this is not just for the researcher in the Ivory Tower or the MD, but also for the rest of the public as well.