Data storage is becoming cheaper at rapid rates. This is one reason why I don't ever expect a totally converged information superhighway, supplying our television, computer, music listening, etc., all in one service. Why obsess over your piping when you can have milk delivered cheaply at your doorstep? Netflix and Google's Gmail, rather than Verizon, may represent our cultural future. Data storage and delivery also tend to be less regulated than centralized piping, plus they limit natural monopoly problems. Under this alternative model, I might receive "cultural disks" in the mail, every month or week, and decide what on those disks I am willing to pay for. Yes there will be hackers but we will be rich, the discs will be cheap and convenient, and they will offer ancillary services of organization and presentation. I can hardly wait, except now I remember I don't even have time for the current menu of cultural offerings. Marginal Revolution: The economics of storage
This is similar to a discussion I had with Roland a little while back. His argument was that he wants a little broadband router sized device that takes care of all his services for him. My argument being that sure, I can do many of those things today (like my own gallery vs. Flickr), but that economics of scale will always see advances first happening in a centralized fashion then becoming decentralized as they become commoditized/processing power increases at the edge/etc.
So, I expect things like 3D images or other computationally or storage intensive applications to start at the center, and over time you'll be able to buy it in a small form factor that sits on your desk.
Oh yes, and in case you didn't get the memo, it looks like centralized telephony networks are doomed -- we can do all that stuff at the edge, all we need is someone to run a centralized directory.