This doesn't make any sense to me. You can't take a shortcut around understanding the systems as they are now.

Also, it's not clear to me that in this case they were even trying to find someone to be so bleeding-edge. They also need sysadmins to do other kinds of work -- fight fires, deal with adding servers, handle local IT, administer testing environments, and so on. If they are looking at people with the OP's background I'd hazard it was more about the latter.

Who are usually the same people.

More likely a subset. Just because you can read up on what other people have done and implement their ideas, doesn't mean you can come up with your own, better, ideas.

I don't know. Taking in the context of this discussion, it would mean someone with vigor and no understanding of basic CS concepts can come up with breakthrough in distributed computing. This is in spite of tens of thousands smart people who actually know the stuff and been researching it for years standing where they are now.

Just does not sound very probable.

EIDT: OK it's not quite what you've been arguing about. Yes, not everyone familiar with the field is going to ever push the envelope. My point was that the chance of an outsider there is quite small.

I think that was my point. Understanding the cutting edge within the field is necessary, but in no ways sufficient, to have what it takes to push the envelope. And if you need to hire someone to push the envelope then you need to look beyond simply what they know when hiring.