Nicholas Negroponte said, in the context of the United States, that all that was carried on wireguides would shift to wireless (e.g., telephony) and all that was carried by wireless (e.g., television) would shift to wireguides. Wireless was better at connecting people who were inherently mobile; while wireguides made better sense for hauling large amounts of data needed to give people high-quality entertainment experiences. George Gilder called this the Negroponte Switch.
The US market, of course, was heavily wired to start with: twisted-pair copper from the phone company and co-ax from the cable company coming to most homes and offices. In this context, the Negroponte Switch made eminent sense. The refarming, for mobile uses, of 700 MHz frequencies that were inefficiently used for television, earlier in 2008 was a manifestation of the Negroponte Switch.
The difference in developing countries was that the incumbents had failed to roll out copper to most homes to start with. They were content with serving the government and business elites and themselves with the perks that come with monopoly. So when the entry barriers were removed, people obtained telephone services through wireless in the first instance, there weren’t much wireguides around to carry entertainment. And shortsighted incumbents neglected wireline maintenance when they saw the cost and other benefits of wireless connections.
Now it looks like at least one incumbent has seen the legacy copper network as an asset that needs to be developed and exploited a. This will not benefit those at the bottom of the pyramid, of course. And there is no guarantee that what worked in densely populated Hong Kong will necessarily work as well in the less dense urban areas of Sri Lanka.