I am very curious what the damages awarded to Facebook will be.

On one hand, none of the surveillance were possible without Facebook.

On the other hand, a lot of the people have public profiles, which is expressing interest in being discovered, and Facebook is acting against that. And the cost to Facebook for serving the scraping requests is very small.

Wasn't there a recent court decision that explicitly ruled scraping to be legal?

Isn't the case presented here slightly different? In LinkedIn's case the data was publicly available. In this case they were either accessing data that was only available to registered users (but otherwise "public"), or they only used fake accounts to bypass rate limits imposed on unregistered users.

Does the account requirement change whether the data is considered publicly available on sites with open signups?

It's hard to argue something isn't public if the only barrier is making a free account. But whether a court will be sensible about it is a different question, I agree.

It may be legal, but it can be against a (civil) contract saying you won't use their service to scrape other members. Why wouldn't you be able to sue for breach of contract?

I don't know how much weight courts currently place on click-through contracts, but given how many such contracts we "agree" to in modern life, how impenetrable their jargon is, and how often they change, we should be extremely cautious about giving them too much, or any, validity.

E.g. how would you feel if some no-name free website/service sneaked in a "you owe us $1000 if you use this service on Tuesday" term into the novel that is their ToS/EULA? Or if the grocery store put up a "by entering you agree not to disclose our prices to others" sign?

It is kind of ironic :)

People hate it when you try to move in on their grift.