Universal Business Access to the Internet
by Kent M. Pitman (Monday, May 26, 2008)
Refining the notion of Universal Access to the Internet
A great deal of noise has been made about the the notion of Universal Access to the Internet but I'd like to see that terminology refined a bit.
Usually what is implied by this is the need of individual citizens to participate on the Internet in a basic number of ways, such as access to a web browser and access to email (sometimes through their web browser, sometimes through other means). Most people having that degree of capability can participate in a large number of online activities.
But there is another kind of access that I think has not been looked at quite so carefully, and that's establishing a baseline of access for small companies to participate in the marketplace.
The Issue of Verticality, and the Browser Wars
Let's look for a moment at how this issue played out on the desktop.
Microsoft made a computer operating system called Windows that was in competition with other operating systems as a platform for the writing of desktop applications. Through whatever means (some think it was more appropriate than others, but it's not relevant here), they achieved marketplace dominance (or some would say monopoly).
Then later there was a competition for products written on the Windows desktop, which we call a vertical market because it relies on the use and continued success of the underlying platform. Windows bundled its Internet Explorer in with Windows and it was alleged that it could control the browser market by manipulating the details of the support offered for browsers, both in terms of technical details and market positioning, such that its Internet Explorer had what was alleged to be an unfair advantage.
To understand this paradigm it is not necessary to agree that the claims made against Microsoft were valid, it's only necessary to understand the basic paradigm under which such a claim could have been valid so that such a paradigm can be recognized in another context.
For example, if there were competition for a set of highways and for whatever reason my company achieved dominance over those highways such that everyone pretty much always used my highways and no one else wanted to build or maintain highways, and then my company entered the trucking business and started to succeed better because my company's vehicles weren't stopped for tolls while everyone else's were, that would be a similar kind of issue. Clearly, a healthy market in trucking relies on the notion that no one's trucks get favored treatment over anyone else's for reasons that amount to nothing more than a bias toward success in some other market.
For more information on this, there are any number of write-ups by business analysts available on the web. One example would be:
Verticality of the ISP Market and the Web Hosting Market
It seems to me that a similar phenomenon has been happening in the Internet access space. In our zeal to have everyone have personal Internet access, companies offering such access have taken subtle steps to make it hard for businesses to grow up using that access in ways that might compete with them. In effect, as more and more people have gotten personal access, it has been harder and harder for businesses to compete.
Here are some examples of features that are needed by businesses that are not available in an open and uniform way. Each broadband service provider makes up its own rules:
Bandwidth Limitations. Broadband providers sell bandwidth to the end user, and yet then leave it to the end user, who is really not equipped to do so, to assure that the bandwidth limitation is not exceeded.
If the Internet were the telephone system, this would effectively say that you had to say how much usage you wanted but that if people called you more than you said, you would not just pay more but be in active violation of your agreement with the company.
Sharing Unused Bandwidth It would be easily possible for one user to rent bandwidth and then re-sell it others in the community nearby, except that it is forbidden under the terms of service of most providers. This holds prices artificially high.
If the Internet were Fenway Park, this would be as if to say that if you held a season ticket and could not use your seats for certain games, you would be obliged to just lose money and allow the park to resell those seats for their own profit.
Access to Ports. Some broadband providers block important ports, such as port 25 (used for e-mail transport via the SMTP protocol) or port 80 (used for transmitting web pages via the HTTP protocol).
Ports are like information doors at a commercial building. One might be the entrance for package deliveries and another for people to enter the store and yet another an entrance to business offices. On the net, email all goes to door #25, web traffic to door #80. Yet it is common for service providers to say “I'll put you on the internet, but you may not have a door #25.” because they don't want you receiving email directly. They want you to have email delivered to their door #25 and they want you to go to a different door (called POP, the post office protocol, at door #110) to pick up your email from them—possibly for a fee or subject to restrictions on the number of accounts you might have or how you use them.
Access to Static IP Addresses. Not all broadband services offer static IP addresses.
If the Internet were the telephone system, an IP address would be your phone number and a domain name would be your human name. When someone goes to lookup yourname.com on the internet, they look in a directory that tells them what IP address to use, effectively what phone number to call. If the address is static, it does not change, and the same IP address can be used repeatedly without looking it up. A dynamic address is the only thing many broadband service providers offer, which makes it basically impossible for you to list your phone number in the directory because it will change every day. So people who have no static addresses can call out to the Internet, but it's hard for anyone on the Internet to know reliably how to call them.
Privacy. Packet inspection, logging of URL use, and other questions of intrusion are also things it would be nice to understand rather than just assume would be handled properly.
Increasingly, the broadband hosting company is being tempted to “peek at” content in the packets (the bursts of “ones and zeros” that are bouncing around the net). This is like your mailman reading your mail and even possibly taking notes on what he sees or making annotations of his own in your as he carries it along. It's much harder to build a trust relationship with a customer about data you want to handle for them if the way you have to handle it is to hand it to a non-trusted carrier. There are digital techniques for safeguarding such information through encryption but it shouldn't be necessary to encrypt your personal mail sent by letter carrier to know that a mailman isn't going to read it.
Broadband providers have created subtle roadblocks to access to the information superhighway. The rationale, as far as I can tell, is that they themselves want to sell layered services (for example, web hosting and email), so they have each made certain key things hard to get. But the net result, if you'll pardon the pun, is that a business wanting a particular feature to be used for business cannot rely on obtaining it from multiple broadband service providers, hence the competitive forces of the market are not in play. And, if they want multiple such features, there may be no way to obtain that combination from any vendor, so in that case there is not any source of direct broadband access with the desired features at all.
What I Want (and What I Think is Fair)
In the real world, if I bought a piece of land, I could build my business on it as I please. Driving out onto the highway would not require me to sign terms of service or to only have a certain number of people visit that business. The burden of proof would not be for me to show that the things I wanted to do were really necessary to my business, but rather for the government or the maintainers of the roads to show why I should not be allowed to use my property as I felt appropriate. As a consequence of those freedoms, I would have the flexibility to experiment with what I want to make, even to make something new that others don't do with their business, in short, to innovate.
The writers of regulations need to acknowledge the quite obvious fact that the needs of new business entrants are not necessarily identical to the needs of end-user consumers nor of established businesses. Equal access to a wide variety of capabilities is necessary for the network to remain an incubator of new concepts and not just to become a streamlined way to deliver commodity items well-understood and tightly controlled by the same companies that already own the network backbone (the cables and switching systems that tie the core of the Internet together nationwide, and worldwide).
The requirements of internet access for businesses and for end-user consumers are not the same, and it is a mistake to assume that the regulations targeted at one are suitable to the other.
The requirements of established business are merely those required to consolidate market share and lock out entrants that might disrupt the shape of the market. The requirements of new businesses are often quite different than that.
Universal Business Access to the Internet is a concept that requires careful study independent of Universal Personal Access to the Internet. Achieving universal business access requires more than merely making sure the network reaches all businesses and requires more than making sure that all businesses are allowed to buy Internet services in packaged form from the same businesses that maintain the Internet backbone.
Universal Business Access should be a first-class priority of any serious policy regulating the Internet. It deserves explicit public discussion, and a formal commitment from politicians independent of the question of Universal Personal Access.