Right; AR is significantly less useful without a massive data platform backing it up. But the real problem is that Snap doesn't really have too many paths for g

Right; AR is significantly less useful without a massive data platform backing it up. But the real problem is that Snap doesn't really have too many paths for growth that don't run into Facebook's existing network effects. Facebook is already a top aggregator -- and already has a lot of preference data from billions of users. Those are enormous network effects, and even if Snap is hugely innovative, Facebook will always be able to replicate Snap's products much faster and better than Snap will be able to replicate Facebook's network effects.