This story appeared on Network World at

This story appeared on Network World at
http://www.networkworld.com/columnists/2008/102708bradner.html

Are tax-payer-funded networks a good idea?

Mulling Palo Alto's attempt to offer broadband connectivity

'Net Insider By Scott Bradner , Network World , 10/27/2008

There are few things that annoy a telephone or cable company as much a town or city deciding to put up its own ISP infrastructure.

Every now and then, in spite of annoyed carriers, another municipality tries to figure how to make such an idea work. The latest to try -- Palo Alto, Calif. -- may fare as poorly as most previous U.S. cities and towns have, but if it manages to succeed, its approach could be a prototype for bringing high-speed Internet service to communities.

The goal of the Palo Alto plan is a new fiber network, with the city operating as a wholesaler and enabling 100Mbps connections for its homes and businesses. It's not a city-funded giveaway, however: Customers will have to pay (although I've seen no price estimates). And it's all at no cost to the taxpayers.

Palo Alto does not have a vendor signed up yet, and it may not find one willing to build the network. Even if it does, the venture may run into trouble, as other such endeavors have. Most big municipal-wireless plans have died because a business model was missing, and many previous municipal-fiber networks are in trouble or have been sold off.

For every failed implementation, however, numerous new efforts have been launched. In most cases, these are not just a case of unreasonable expectations but instead, an attempt by the municipality to survive when the local telco and cable companies refuse to provide good or cost-effective broadband services. Of course, as soon as a town announces a plan, it gets sued by the local telco to try to prevent competition for what the telco is not offering. Sometimes the announcement winds up pushing the telco into action, which can be a reasonable result if the price is rational. (See "City-owned fiber network a go as judge tosses telco lawsuit".)

The needs of, and problems with U.S. municipal fiber networks are explored in a good article on Telephony Online. In too many states, municipalities do not even have the option of seeing whether they could do better than the noncooperative telco because that same telco worked with the state legislature to outlaw such networks. (See "A warning about future telecom 'reform'?") There have been attempts in Congress to outlaw such restrictions, but none have succeeded yet. Still, I would not be surprised to see fresh attempts in the new Congress next year.

A great deal of the anti-network-neutrality rhetoric by incumbent carriers, both telephone and cable companies, is that they cannot afford to roll out new infrastructure if they cannot extort extra money from people to use the network (that's not quite the way they put it, but that is the gist of the argument). If that's the case, why not be open to letting someone else pay for the physical plant and just buy connectivity wholesale? A lot of the power system works this way -- one company owns the physical wires and another provides the power.

Under this model, a municipality might be an ideal provider of wholesale service -- a town or city can get low-cost loans to pay for it; and as long as the city is just providing dark fiber (or, as in some places in Europe, tubes to blow fiber through) rather than services, it will not get in the way of competition for higher-level services. Municipalities also can better understand their development priorities.

But don't hold your breath waiting for a logical outcome to this ongoing discussion.

Disclaimer: There is almost always logic behind what Harvard does, even if the distance behind varies. But I know of no university position on muni-owned infrastructure, so the above muttering is mine.