It's easy to read too much into the news that Earthlink will pull the plug on its Philadelphia WiFi network, once heralded as the future of internet connectivity. Free marketeers will say that this failed experiment shows the folly of government efforts to pick winning technologies. And they'll be wrong---or at least they'll be overstating their case greatly. In a great many areas, the government provides (either free or on a subsidized basis) the infrastructure on top of which private enterprise runs: roads, sewers, sanitation, fire protection and many other services can be---and at various times in various places have been---privately provided, but national, state and local provision of such services have also been extraordinarily successful.

All technologies eventually become outdated. The Roman aqueducts were a marvel of public engineering. The fact that modern cities typically use underground pipes as a means of obtaining their water hardly shows that the Romans made an unwise investment in aqueducts. Likewise, if, in a future as envisioned by Gene Roddenberry, teleportation replaces ground transportation as the principal means of travel on Earth, that will not mean that the orphaned roads will have been a failure. It will only mean that their time has passed.

The problem with municipal WiFi, by contrast, seems to be that its time never was. Limitations of the technology made it suboptimal. But that need not remain true forever. In the long run, wireless communications---whether bouncing off satellites or terrestrial relay stations---has at least one enormous advantage over any technology that requires hard-wiring: mobility. Just ask anybody under the age of 30 for his or her "home telephone number" for confirmation of this fact: Although land lines still deliver much higher quality sound, Generations Y and Z rely almost exclusively on mobile phones.

The free marketeers have a point, of course: In periods of rapid technological change, neither the government nor any single individual or firm can be counted on to successfully pick winners and losers. The problem with municipal WiFi was not that the government acted but that it acted too soon, before it became clear what the best means for connecting people to the Net was. Ten years from now, however, municipal, statewide or even national wireless internet service provision may be a much better bet.

Posted by Mike Dorf