title:

Utopia, except for the phone companies

By Scott Bradner

Is there a viable business model for commercial Internet service providers in an end-to-end Internet? Telephone carriers looking forward to an Internet-based future do not seem to think so. Instead, they are looking for ways for them to be in-the-loop as content or services providers. But there may be other options.

A purist's Internet is an end-to-end service. You set up a service. I access and use that service. Maybe I pay you for my use of your service. As long as I use standard Internet protocols no Internet service provider (ISP) along the way between me and you knows that you are offering the service, that I'm using the service or that I'm paying you to use the service. No ISP gets part of whatever fee I'm paying you. I pay my ISP for Internet connectivity. You pay your ISP for Internet connectivity. If there are other ISPs between my ISP and yours they do not receive any specific chunk of the money we pay to our ISPs. Our ISPs might buy connectivity from them but the fee for that connectivity does change based on what our ISPs collect from us. So ISP service is basically a commodity and its hard, though not impossible, to make money selling a commodity.

It costs a lot of money to build an ISP, particularly an ISP that can offer very high-speed service to its customers. That takes fiber optic cable and fiber is expensive to buy and install. Customers, especially residential customers, are reluctant to pay enough to make installing such networks an economically viable proposition.

There is an alternative to the carrier installing the fiber themselves that is getting more popular and may get even more so depending on the outcome of an upcoming Supreme Court case.

If carriers can not afford to put in fiber or are unwilling to do so because they do not think people will be willing to pay for it then why not someone else install the fiber and lease it to ISPs? The November 17th edition of the New York Times carried a story about the plans by18 cities in Utah to install a fiber infrastructure, called "Utopia," that would reach 248,000 households and 34,500 businesses. A lot of other municipalities would like to do the same thing but this concept caused the phone companies to pull a nutty and get laws adopted in 10 states so far that block governmental entities, such as towns or cities, from competing with the local telephone companies. The legality of such restrictions is now in front of the US Supreme Court as Nixon v. Missouri Municipal League (No. 02-1238). Arguments in the case are scheduled to be heard on Jan 12th. Information on the case can be found at http://supreme.usatoday.findlaw.com/supreme_court/docket/2003/january.html and http://www.techlawjournal.com/topstories/2003/20030623.asp.

Another type of effort is represented by the California "One Gigabit or Bust Initiative." (http://www.cenic.org/GB/index.html) This is an effort to fiber up every educational institution, business and home in California by 2010. As a non-governmental project it does not have the same legal issues but would still create an infrastructure instead of waiting for the phone companies to do so.

The background studies for the California project foresee very large returns to society for installing very high-speed networks but not enough of these returns would flow to the carriers to make it possible for them to do the job on their own. A related question is what value the phone companies bring to the table in this case. You can guess my answer to that one.

disclaimer: At least some people think that Harvard has value (in addition to its endowment) but this guessing game is mine alone.