The programs, I would say, that you primarily see in Bangladesh, because right now we do have some evolving humanitarian requests that are coming in from Bangladesh.... Before, CIDA primarily was a responsive organization, and it would respond to organizations, whether they're international, national, or more local, community-based organizations. Its partnership branch is set up to respond to Canadian organizations' requests, but even in the bilateral we would work with them, but a lot of the programs would be driven by individual organizations. I would say they were all very based with the best of intentions, good work, but they wouldn't be coordinated in a way that there was cohesiveness being undertaken.
In fact, even within organizations.... When I was in Colombia I saw two water projects undertaken by CARE International. For the one project, that team was doing everything for the local community. In the second water project, they took a little longer, but they set up a water project where it's now being locally managed and run. So even within the same organization you'd have two different kinds of water projects. You can see how, if you're being more responsive in indicating to the organizations what it is you want to accomplish in a country, they can come forward with their proposals.
As far as how the projects were selected before is concerned, they were selected according to the millennium development goals. If they were directly or even indirectly related to the reduction of poverty and one of the millennium development goals, they of course would be considered. CIDA itself formerly had some principle statements that they would support and areas where they would support, and of course those would again be supported. Consequently, what happened is you'd try to respond to any activity that would fit under those principles, without any focus in programming, and in actually articulating the outcomes you would like to see, the outcomes you were trying to achieve.
What I found at CIDA was that things were more measured by input than by output. So there was the reporting. Yes, they got the money, they administered it responsibly, etc., and then when you asked for output, what was the real result, some of the activities we were looking for, it was more how many people would be part of a program. The real measures--did it increase the productivity of a farm, did it prevent youth who were in the sex trade from returning to the sex trade--were measures that were never required of a project that we supported.
Those are some of the changes we're putting in place.