Another issue is that when you’re doing things at one lab, things are, at least relatively speaking, easy. I’ve got a person sitting next to me watching video tape. I just train them shoulder to shoulder. And they develop, and they develop the eye I have. If it’s how to interview I go with them or they go with somebody else, and we train them.

The minute you have a larger collaboration and multiple sites, things get almost industrial, and that’s been a real challenge for us because you have to write manuals of how you do things, which is actually good, because they’re easy to disseminate. But they take more time. You have to put in place really good systems for training everybody commonly, for checking reliability everybody commonly, and that kind of thing.

So don’t underestimate the infrastructure, costs, demands that it takes to do things with others in multi-site enterprises. I know in our project we had no appreciation of what was entailed with that. I think we’re probably better off for it because I’m not sure any of us would’ve gotten into it to begin with. Because you need bureaucracy, you need infrastructure, you need to make so many things much more explicit, put them down in writing so that everybody’s literally and figuratively reading from the same page.