What if the Interstate Highway System Became Obsolete Every Five Years?

Tim Wu believes he has diagnosed the problems of public Wi-fi.  Public wi-ife is a great idea, he says, but the problem is that municipalities have not recognized they need to spend real money on it.

It's hard to dislike the idea of free municipal wireless Internet
access. Imagine your town as an oversized Internet cafe, with invisible
packets floating everywhere as free as the air we breathe....

Not quite. The basic idea of offering Internet access as a public
service is sound. The problem is that cities haven't thought of the
Internet as a form of public infrastructure that"”like subway lines,
sewers, or roads"”must be paid for.

It could be, however, there are a few tiny differences between public wi-fi and public roads:

  • Any wi-fi system you install today will be dated in three years and obsolete in five. In fact, given the long delay in public projects between design (and presumably technology selection) and deployment, the system may well be obsolete on the day it gets turned on.  Would we have made the same public highway investment we did if roads went obsolete every five years?
  • Roads don't tend to have private competitors.  And when roads are constructed by private entities, say in a new housing development, you can absolutely bet that the municipality doesn't feel the need to invest in "public" roads to run beside them.

Wu admits that both cable and DSL have a much lower cost to serve urban customers, which is why private efforts for urban wi-fi tend to fail.  Free municipal wi-fi will therefore be more expensive to build and operate than if you just provided direct public subsidy payments to poorer people to use existing private solutions.  Further, a huge part of the investment will go towards giving away free access to people who already have internet service from a private supplier and are willing and able to pay for it.

Note that Wu never actually names a goal for municipal wi-fi or a
problem it is solving, just this beautiful vision of a city-wide
internet cafe (are we going to provide municipal coffee too?)  This fascination with municipal wi-fi reminds me of nothing so much as a similar fascination with light rail.  You can see it in his opening comment about the "oversized internet cafe."  This is an aesthetic, not an economic, vision.  Our light rail project here in Phoenix is the same way.  It will haul passengers more expensively and at a far higher investment and with less flexibility than our bus and road system.   With the investment we are putting into the system we could have instead bought cars for every rider and had money left over.  It makes zero sense for the density and commuting patterns of this city, but still we are doing it, because there is a subset of people who love light rail as some sort of pleasing aesthetic vision.  Name any goal either one is trying to solve (e.g. access to transportation or internet) with public investments in light rail or municipal wi-fi and those goals could be solved more cheaply some other way. 

Postscript:  A while back, I wrote about another danger of municipal wi-fi:  That bureaucrats in charge of the system will try to protect their jobs by blocking new competitors:

[the municipal wi-fi authority] can use its government authority to block new entrants. ...  Take another large government network business: The Post
Office.  The USPS tried like hell to get the government to block Fedex,
and almost succeeded.  The government continues to block competition to
the USPS for first class local mail.  Heck, the USPS has tried at
various times to argue that it should have authority over email and the
Internet.  The government blocks new cigarette manufacturers to protect
the settlement money it gets from the old-line tobacco companies and it
blocks usage of Love Field in Dallas to protect D/FW airport.
Bureaucracies never, ever let themselves die, and there is no way a
municipal broadband business will ever let itself be killed by a
competitor - that competitor will be blocked, even if that likely means
that local broadband consumers have to stick with higher costs and
outdated technologies.

You see something very similar with municipal water systems trying to get the government to limit the growth of bottled water.  It happens all the time.  Already, examples exist of municipalities trying to shut down wi-fi competition from private companies.

Boston's Logan International Airport is attempting to pull the plug on
Continental Airlines' free Wi-Fi node, which competes with the airport's
$7.95-a-day pay service.

In an escalating series of threatening letters sent over the last few weeks,
airport officials have pledged to "take all necessary steps to have the (Wi-Fi)
antenna removed" from Continental's frequent flyer lounge....


  1. Sammy:

    “dated in three years and obsolete in five” is a red herring. The equipment could change – smaller, more reliable, less power consumption, stronger signal, but the actual Wi-Fi signals – format, frequency, etc. – can’t change else you block out current users.

    Can we say legacy systems?

  2. ErikTheRed:

    The comparison with the highway system isn't that great. Highways are important. Municipal WiFi is a convenience. Anyone who relies on it for "critical" tasks is, frankly, an idiot. If you need mobile Internet access, the newest 3G cellular cards are generally much faster and more reliable.

    @Sammy - Actually the formats and signals of WiFi do change over time, but they tend to retain backwards compatibility modes. The best example is the evolution of 802.11b to 802.11g to 802.11n. Each newer technology is backwards-compatible with the other. This is not necessarily a good thing! An 802.11b user (with a theoretical max transfer rate of 11Mbps at the MAC layer, which means about 6Mbps of actual user data) is streaming something at 1Mbps under perfect laboratory conditions, that means he's using about 18% of the capacity of the access point (AP). An 802.11g user streaming the same material would only be using about 6% of the AP's capacity, and an 802.11n user would be using about 0.5%. This is because WiFi networks only allow one device to broadcast at a time, and the older devices take much larger chunks of time to send and receive information. In the real world the disparity in impact is usually much higher than the idealized numbers I list here.

  3. NeoWayland:

    There is also the "small" problem of local politicos deciding what sites and services the public should and should not have access to on the "free" networks.


  4. Allen:

    The whole issue goes to show how politicians just latch onto something, even if it's not the right technology. WiFi isn't good at covering a wide area. It's not a good technology to use to try to cover a city with wireless coverage. Yet they close their eyes to that and plow ahead with it.