Earlier quoted context omitted.
> "(b) being able to be free of the GPL" Wouldn't the easier path have been just for Google to contribute (or fork) Free/Open/NetBSD?
Seems like BSD/Linux seems fundamentally not compatible with what they want to achieve technologically. It is not a license issue.
Anyways, if you’re going to build an alternative or fork another stack —- you might as well hit two birds with one stone. Fuchsia’s relatively distinct capability-based, quasi-microkernel (it is not in fact a microkernel on a strict read) architecture is a chance to cleave off technical debt and start anew on security, and it’s relative modularity dovetails into the whole driver, kernel interface issue.