I've never had trouble viewing an article in incognito mode, or sometimes even hitting stopping a load before javascript runs, so I've been under the impression that it's mainly javascript trickery. My argument here was with that assumption, so perhaps it's moot.

> Alice tells the cashier that her name is Bob and shows an (obviously) fake ID. The cashier isn't paid enough to care

I'd say that the proper example is that the cashier doesn't check ID and isn't even expected to by corporate. Companies aren't even strictly against this type of soft-fraud word-of-mouth trick, as it just further helps their price discrimination and customer mindshare. Whether you get one free birthday desert per year or five, you're still buying meals.

The elephant in the room is that Google could easily setup a system whereby Googlebot got secure access to the articles. I think they haven't done this because of an idea that the same pages should be served to users. So who is defrauding whom?

Philosophically I'd assert that the Internet philosophy is directly opposed to the very concept of fraud - "identity" does not scale, especially across jurisdictions. We've finally got a way to formalize and mechanically execute contracts purely between private parties, so why cling to a heavyweight idea of post-facto enforcement based on nebulous ambient natural-language rules, especially as an overarching foundation? When you put a quarter into a "claw game" and it drops your prize before the chute you don't sue for "fraud" - you just stop putting quarters in.