It's not a good question. It's both famous, so people already know how to game it, and it's overwhelming. I do not want to have to give an entire lecture, which essentially 3+ different university courses condensed into one, on the spot to show off my street cred.
Overwhelming is one way to look at it. Another is that it’s an opportunity to demonstrate both breadth by hitting the keywords and also depth by diving into whichever part you know best.
Any candidate who made it to an interview should be able to rattle off a bunch of stuff. An excellent candidate will have more depth than the interviewer on some portions.
The interviewer’s first job is to decide whether the candidate has a baseline-acceptable level of knowledge (harsh but true, good fakers make it this far). This question is fine but not great for that for the reasons you mentioned. However, it is best to go in assuming they do.
So then the interviewer’s next job is to decide if they are going to be a good teammate. Google is rife with nerds trying to out-nerd each other, finding out how that goes in the interview is an explicit goal. A great interview on this question will end up in a moderately deep technical discussion of some aspect where both the candidate and interviewer are knowledgeable, much like a discussion you would want to have with your teammate at work. And that is why this is a decent question, in my opinion.
