Part of why we've been discussing this so long and not made moves yet is that there are a lot of things that a foundation could do, like hiring devs. But we're unsure what we would want to commit to it doing. Hiring devs has some big advantages, but also a lot of disadvantages too. For now, we're focusing purely on the trademark ownership part, as a sort of MVP.
The Blender Foundation has been employing devs for, like, forever -- they're (usually) the ones who fix all the little things since everyone else wants to work on 'more cowbell'.
Maybe shoot Ton Roosendaal a friendly email or something as he has years of experience in this matter?
I'm curious on what or those from your point of few.
From the disadvantage point I can see: It won't be easy to figure out what they should work on. I think if you have a foundation which has sponsors that pay for dev-time, those might also want to see the issues and features prioritized that they are interested in. However devs might have different desires on what they want to work on. That's kind of true for every environment, but I guess it might even be more true in an environment like Rust which attracts a high amount of smart, creative and passionate people.
Now this can obviously be dealt with via project management. But I think that still might not exactly be what some of the Rust contributors hoped to get out of a foundation (getting paid for working on stuff they like).