There is obviously an element of Person as well as Task. That @Zach! proposed "non-technical docs" was a big factor in supporting it, because I knew he'd do great. Also, if a relative newcomer proposes something, they tend to have built up less of this knowing. So maybe it's not about the Person, but the Relationship?

Anyway, as it currently stands, the grant process is explicitly about Task, but with an defacto implicit aspect of person. If it was the other way around, a person oriented grant - I think task would still be a factor but become implicit.

on risk: I don't think that taking enough risks is the big challenge we have here. What we need to do is execute well. Adding more crazy ideas would be spreading our selves too thin! We need to make the crazy great ideas we've already have work really well. When you find the gold (risky fun) start digging it! (safe work). Searching for the gold is the fun part, but now we have to do the work of getting it out. It's the benefit of the results of the safe work that makes the risk worthwhile.

That said, if you find your own source of funding, you can administer your own grant process, you can make it as risky and person oriented as you like! I'm not against this in principle, I just think it's not what we need right now.