There are very specific use cases, where the concern is not about entropy in the series (because duh!), but from having a small seed value that is intrinsically not cryptographically secure but unpredictable (not quite the same thing) even in a controlled setup like a the start up of a VM image.

I would agree with the article that the right way to fix the problem is to seed each instance of the image on startup, but that will also avoid you having a problem with it blocking.

That's a special case though, where you aren't in the middle of a cryptographic handshake, you don't have real time constraints, and the fix to the real problem will also mean there is no problem. Don't use it for a network service's source of randomness.