I can give you an idea of some of the things that we're doing, but you've put your finger on one of the more difficult aspects of this.
The easier projects to measure in results achieved are the ones.... If we think about a scale between projects that are purely developmental—what I mean by that is projects supporting the audit function, ensuring that the audit office is capable of doing audits—through to opening up political space by working with dissidents, which is very political, our ability to measure progress over the short term of a project is much easier on that developmental side. We built the audit office. We helped them train the right people. We ensured that they had all the tools they needed, and they are now capable of doing audits of that particular department.
Supporting dissidents at the other end of the spectrum is even difficult to determine the proper measures. You provide support to them in, perhaps, a better understanding of how to use social media, and a better understanding of the options available that other countries have used to open up political space in a non-democratic country, and the results don't take place in a year. The results are maybe over a decade and you may not see those results for a long time.
Our ability to measure on that end has been quite difficult. This is why your previous question I think is important. Within that space, some recent studies have shown that drive to have measurable results has pushed a lot of spending down toward the development side and out of this space. It's called tame or non-tame democratic assistance.
By forcing us to work in purely developmental areas, you get results that are more easily explained, but there's a tendency now—talking obviously about the entire industry—to move out of the space that's more highly political and where it's more difficult to demonstrate true results.