DIY Internet Infrastructure

An open source angle on muni Net infrastructure build-out

While Moore's law drives up compute power, network speeds languish. One reason I moved to Santa Barbara in 2001 was that Internet connectivity was much better here than in Silicon Valley, where I had lived for the preceding sixteen years. Six years later, however, my connection speeds in Santa Barbara are barely any better and the costs have gone up. Worse, the city appears to be a low-priority region for Verizon and Cox, our local telco/cableco duopoly. While Verizon is rolling out fiber to homes in hundreds of communities around the country, Santa Barbara isn't one of them. And while Cox is also rolling out higher speed services, Santa Barabara is smaller and farther away from Cox's Atlanta headquarters than any of its other markets.

And we're not alone. The same kind of story is happening in most communities across the U.S. This is why hundreds of those communities are doing what Linux and open source developers have done all along: taking matters into their own hands, doing for themselves as individuals and groups what proprietary businesses and industries can't or won't do. They're rolling their own infrastructure.

That's why I've been taking part in a local citizen effort to build infrastructure that the local duopoly can't or won't provide — to create a new open market for connectivity where the old closed markets for phone and cable service can't or won't do the job. This has involved lots of meetings and discussions -- not only amongst ourselves, but with local businesses, civic leaders, elected representatives, infrastructure (fiber, wireless) deployment contractors and other folks, including cable and phone company people. We want something that works for everybody: a new infrastructural tide that lifts all boats. Like Linux did, and still does.

This is why our emphasis is more on fiber before wireless. We want Net infrastructure that maximizes capacity for the longest possible time, and for the largest number of people and locations -- much as we might be doing if the challenge were building roads or reservoirs.

It's still early, and we're pioneering here -- as are many other communities. So, to gather useful thinking at this pioneering stage, I've put together an FAQ for myself — and for anybody else who wants the benefit of the same thinking. Since much of this thinking is informed by, and modeled on, the successful experiences of the of Linux, free software and open source movements, I thought I'd share my draft here.

Bear in mind that this is just thinking-out-loud at this point. It's provisional. I offer it in hope that it's useful to other folks in other towns that are trying to do the same kind of thing — and in faith that the smart readers of Linux Journal will help myself and others think and work our way to local Net infrastructure that favors everybody, and not just incumbent carriers.

So here goes.

Q: Why does (your town or region here) need Internet infrastructure?

A: For the same reason it needs roads, water, waste treatment and electricity. The Internet is becoming a "fifth utility" no less essential than those other four.

Q: Why not do wireless infrastructure instead of fiber?

A: Fiber optic cabling is the highest-capacity "pipe" for carrying the Net. Even cities opting for wireless build-outs require fiber "backbones." Fiber is also the most "green" and has enormous capacity. One fiber optic pair can carry thousands -- even millions -- of times more data than any form of copper wiring, or any form of wireless connection.

Q: Isn't the Internet just a service we already get from cable or phone companies?

A: No. The Internet is not itself a "service." It is a means for transporting data between devices anywhere -- without regard for the distance between them or the routes between them. The Net is a system for supporting all forms of connected activity, including everything that can take advantage of an open and networked marketplace. Again, it is pure infrastructure, like roads, electricity, water and waste treatment. Like those utilities, its primary and fundamental role is to serve as a "tide that lifts all boats". The Internet cares no more about how you use it than the ocean cares about what you float on it.

Remember that telephone and cable TV are optimized for single purposes. Even the new fiber build-outs by cable and phone companies are being designed and optimized for one-way distribution of TV programs. The Internet is still a side service for these companies.

Q: But aren't telephone and cable companies utilities as well?

A: Not in the same way roads, water, gas, waste treatment and electricity are utilities. They are businesses first. While the Net today is carried to our homes and businesses by these two businesses, the relationship in the long run is the other way around. Telephony and video used to be analog services. Now they are just two among many other forms of data carried on the Net's worldwide mesh of roads or pipes. They are two services on a transport system that can support an infinite variety of other services. This is why it is essential that we stop seeing the Net as gravy on top of telephone and cable service.

Q: Is Net infrastructure costly to deploy?

A: The Internet itself is not inherently scarce. In this way it is different than water, electricity and roads. Instead, the Internet was designed as way to remove distance between every connected device, and the people who use those devices. There is no "long distance" to the Internet Once a device is "on" the Net, it is zero distance from every other device in the world that's also on the Net. That's because the Net was designed so the functional distance between any two keyboards in the world is no greater than the distance between those keyboards and their screens.

Even fiber-optic cabling, which provides most high-capacity connections, is not expensive. Of course, the cost of digging trenches, pulling cable and installing and maintaining facilities is far above zero. But it is an extreme bargain compared to the cost of building roads, damming rivers and running power or waste treatment plants.

Q: Doesn't the Net cost money to maintain?

A: Yes. But the maintenance costs is more like roads than like phone or cable TV. Unless a town (or a county) wants to go into the phone and TV businesses, the engineering and maintenance costs do not have to be high. But capable engineers and maintenance workers are required, just as they are for other utilities.

Q: Isn't the Internet service we get from telephone and cable companies good enough? We have email. We can browse the Web. We can do instant messaging. We can search on Google and buy stuff from Amazon.

A: The phone and cable companies deserve credit for offering Internet services, and contributing to the expanded reach of the Net. But it is important to understand two things. First, the Net is far more than just an extra service provided by telephone and cable companies -- even though that's how those companies sell and bill for it. Second, the Internet is a third-priority offering for both telephone and cable companies.

Right now telcos and cablecos are working much harder at getting into each others' core businesses than they are at expanding Internet services. And the Net is about much more than email, searching and browsing. It is becoming an essential backbone for civilization itself. There are many public and private service needs that can only be served by reliable high-capacity Net infrastructure. Doctors can do diagnostics and even surgery over distances. Public safety services can communicate and share data rapidly. Businesses that deal in graphical imagery, high-quality audio and moving pictures can produce and share their work far more easily. Most importantly, businesses and civic activities of all kinds can be supported -- not just the two incumbent businesses that first brought the Net to our homes.

Q: Doesn't all this threaten telephone and cable companies?

A: It doesn't have to, because there is nothing keeping telephone and cable companies from also going into other businesses that are opened up by a growing Internet. In other words, there are benefits to incumbency other than leveraging original business models.

In fact the Net offers enormous opportunities to telephone and cable companies -- opportunities to provide countless services in addition to their traditional ones. That's because there is no limit to what you can do with the Net. Telephone and cable companies have a head start, both in existing facilities and existing relationships with thousands or millions of customers. Instead of being in one, two, or three businesses (the phone, cable and Net "triple play" businesses they're in now), these companies can offer an infinite variety of value-added services to individuals and companies and organizations that use the Net.

Q: Why do cities and counties need to build out Net infrastructure? Why not let the marketplace take care of it?

A: Unfortunately, the marketplace we have is not a free and open one. Telephone and cable companies are captive to long-standing regulatory environments that are changing very slowly. Also, neither side has shown much interest in putting Net build-out ahead of their core businesses. Both telephone and cable operators have powerful lobbying forces in Washington, and at the state level as well. In fact, "telecom" spends more than any other business category on lobbying in Washington. They even outspend the pharmaceutical companies. Right now they are working hard to protect their traditional businesses from the "threat" of free and open Internet growth. In fact, the threat is the other way around. By subordinating Net infrastructure to the TV and phone ambitions of two entrenched monopolies, Net growth in the U.S. is falling behind other countries which place Net infrastructure growth at top priority.

Q: Doesn't the government care?

A:. Depends on which levels of government we're talking about.

At the federal level, not much. While the Internet was born in the U.S., neither our federal government nor its protected communications duopoly has shown much interest in the Net's development, except as a side effect of telco and cablco business development. Nor has our government paid much attention to how well the Internet supports and sustains economic growth. Other countries -- Korea, Japan, Denmark, Netherlands and France, for example -- have encouraged Net build-out for most of the last decade. As a result, the U.S. is now 11th in "broadband" penetration. By deciding to leave Net build-out to the monopolists who feel threatened by it, our federal and state governments have assured limits on our economic growth. They have reduced our competitiveness in the world marketplace.

This is why so much initiative is happening at the local level, and with local governments either building out Net infrastructure or creating local building code and regulatory environments that encourage Net build-out at a grass roots level. They know how

At the state level, it varies. Some states — Utah, for example — prevent municipalities from retailing Net-based services while also encouraging cities and groups of cities to build out Net infrastructure. Telco and cableco lobbying in some states — Arkansas, Nebraska, Tennessee and Virginia, for example — have essentially outlawed Net build-out by local and regional government entities. In other states — Louisiana, for example), local and regional governments are fighting incumbent carriers in the courts. In nearly all these cases, the Net is defined legally as a "service" of "common carriers" such as phone and cable companies, which is an affort to the nature of the Net and a mistake akin to preventing automobiles from being made by anybody other than railroad companies.

Q: How have the carriers crippled our Internet service?

A: The Internet was designed originally as a symmetrical system. That means the "upstream" and "downstream" speeds should be the same — or at least not kept intentionally low in either direction. That's the kind of Internet connectivity we find in many universities, and large companies. But it's not what the telephone and cable companies provide to our homes and small businesses. What we get is asymmetrical: much higher downstream than upstream. The reasons are not necessarily bad ones. Most of us consume far more data than we produce. This is especially true when we download large graphical files, watch a YouTube video or listen to the live stream of a radio station over the Net. The carriers today see the largest opportunities for revenue production in the nation's persistent appetite for television, and growing demand for digital "high definition" TV. Providing capacity for those services is fine. But limiting upstream speeds intentionally is not good, and that's what the carriers have done since the beginning. Today if you want higher upstream speeds, or to run your own mail or web servers, you have to pay "business service" premiums that are so high that most small and mid-sized businesses simply don't opt for them — especially when they start with higher prices for lower downstream speeds than the carriers charge for plain home service.

So the asymmetrical systems maintained by the carriers continue to relegate each of us to a consumer role, and prevent us from becoming producers as well. This limitation is compounded by what are called "port blockages." This is where our phone or cable company prevents us from setting up our own Web server, or running our own mail server. Again, they have some good reasons for blocking the ports on our computers that those services could run on. Spammers, for example, can take advantage of open mail server ports on our computers. But these port blockages also prevent all kinds of uses, including the ability to set up home businesses of many kinds.

So, instead of, say, offering services that aid in the development of small and home businesses, the carriers just shut off possibilities to avoid hassles that might distract from focus their core phone and cable TV businesses.

Q: What is "Net Neutrality", and does it matter at the local level?

A: Net Neutrality is another name for the end-to-end nature of the Net as it was originally designed. It is also a big fight in Washington, with the telco and cableco carriers lined up against it. At the local level, however, there doesn't need to be a fight. If the infrastructure is installed and maintained as a primarily public utility, the principles of Net Neutrality and business opportunity are both well supported.

Q: Isn't local infrastructure build-out a case of government competing with private industry?

A: No. It's a case of citizens finding a way to do what a government-protected duopoly cannot. In this respect, municipal build-out is far more of a "market" activity than what we've seen from the carriers. Citizens want to open new Net infrastructure to use by anybody, including cable and phone companies. They have those companies' interests at heart too.

Q: Isn't this different than what we're seeing in other cities that roll out "triple play" fiber systems?

A: Yes, it is. By just installing the "transport", and leaving the services up to other parties (including the phone and cable companies), we are doing two things. First, we are saving money by not getting into businesses and facilities that actually cost a lot to start and maintain. Second, we are treating the Net as the simple utility it needs to be. As a city, we know how to maintain roads, water treatment and waste facilities, and so on. We also know how to support businesses rather than compete with them. That's what we are doing here.

Q: What will it cost to bring fiber optic cabling to homes and businesses?

A: We have estimates of up to $2500 per "drop" -- about the cost of a big flat-screen TV (without the immediate and certain depreciation). But the cost of fiber optic cabling itself is actually less in most cases than the labor cost of burying it in the ground or hanging it from poles. There is much we can do to drive costs down for the city, and to make it easy for citizens to connect to the network as they need it. For example, we can require that conduit and fiber be put in the ground everywhere a trench is dug. We can hang fiber on poles first and work out the individual drops later. We can create a regulatory and procedural environment that encourages citizens and local businesses to build their own neighborhood fiber networks, which we can then connect to our publicly or privately own owned "backbone" whenever they're ready. We would love to create an environment where local businesses do much of the build-out, and make money in the process.

Q: If you're not selling phone and cable TV services, how are you going to pay back the cost of installing the fiber cabling?

A: We'll have to approach this the same way we approach building out any new public infrastructure -- just as we did when we built our water systems, our waste treatment plants, our roads. We need to think creatively about this, but also take advantage of what we already know about building utility infrastructure.

Q: What do you expect will be the economic benefits of this?

A: We can begin to hint at those by looking back at what the Internet has done for our lives, and our economies, over the last ten years or more -- and multiplying it by countless citizens and businesses that are finally able to do what they want with a Net infrastructure that truly supports it.

Q: Are there any downsides to building this out?

No. And that's the real bottom line.