> I think that's an unfair comparison for two reasons: (1) People were already comfortable with the idea of paying for cabs, cable, games, books/electronics in the old world. Whereas the current norm is _not_ to pay for social media services.

Of course; this is because cabs, cable, games, books/electronics all provide value. Most social media services don't, at least not enough value for people to pull out their credit cards.

> (2) A transportation service, game, streaming video service, e-retailer can deliver value to their first customer. A social network is only valuable if others are using it. If Spark / Twitter 2.0 launches tomorrow, even if they can provide a bunch of product improvements over Twitter 1.0, the first users have no reason to pay b/c they're joining an empty / worthless conversation, so at launch it's almost necessary to allow people to join for free. And this creates the norms described in (1).

I disagree, I think the norms described in (1) exist because Twitter/Spark provide very little value to their users.

You're describing a first-to-market advantage, but that's not really relevant. If that were a signifiant deciding factor, Twitter could simply institute a subscription fee and be profitable by tomorrow. But they can't: if they did that, they'd lose all their users overnight.