The SEACOM fiber-optic cable is one of the major data conduits connecting Africa's networks to the rest of the world. SEACOM has suffered a serious outage in the past 36 hours or so, and Internet users across a wide area are having problems as a result. Among those affected are the South African universities that use a SEACOM-reliant regional research networking service, TENET, which also means that I and about 800 or so other participants at a pair of international media conferences in South Africa are mostly disconnected from email and the other Internet services we normally find essential.
It's easy to grumble in such situations, and we're all doing that. And it's essential that we work on ways to increase network reliability -- and reduce our vulnerability to situations where a single point of failure can paralyze vital services.
Even as we realize how far we have to go before we can fully trust that when our online tools will be available when we need them, we need to remember how far we've come. A decade ago, on my first trip to Africa, dial-up phone access was the main way I connected to the Internet in my travels -- if I could connect at all. And just decade before that, hardly anyone in America, much less Africa, had even heard of the Internet.
The problem we're experiencing here will pass. Repairs to SEACOM are under way (though conferees may not be back online properly by the time we leave two days from now). And Africa's telecommunications system, while far behind the rest of the world, is steadily improving in a general sense.
Meanwhile, I'm sending this column via the somewhat iffy WiFi connection at the B&B where we're staying. The establishment subscribes to an Internet service provider (ISP) that either doesn't depend, down the pipe, on SEACOM or, unlike the university system, has found a way around the affected cables via other fiber-optic lines.
Yet the outage here reminds us that our interconnected world -- and the technology we've increasingly come to rely on -- remains all too vulnerable to areas of control that amount to outright choke points. Accidents or deliberate actions can achieve the same result: Network services go down or are censored; market-dominating software and Internet companies have critical bugs or use their power to make unilateral decisions.
Sometimes our problems are of our own making. Big Internet customers typically have backup plans for network outages. Home users typically don't. This is one reason why the epic consolidation of the ISP business in much of the world, including the United States, into duopolies or oligopoly is a dangerous trend. When Comcast or AT&T has a major outage, millions of people are stuck until service is restored. And the internet "backbone" system -- the lines that carry the longer-distance traffic -- is becoming more and more a fiefdom of a few big players.
Choke points are built into other parts of our culture and economy. We encourage single companies to control access to life-saving drugs. We have an energy ecosystem, especially the electrical grid, that's scarily vulnerable to disruption. We invite huge financial institutions to make wild bets with other people's money, grow too big to fail without bringing down the global economy, and then let them steal the rest of us blind when their insane bets go bad. A few thunderstorms in the wrong places create absolute havoc in airline traffic, in part due to an aging and nearly archaic Air Traffic Control system. Even a mild pandemic of a slightly more lethal virus than last year's H1N1 would overwhelm our already-stretched hospitals. And so on.
The best solutions are redundancy and competition, of course. But there seems little political will to make the kinds of decisions that would encourage more of either. A rare exception was last week's Obama administration proposal to expand wireless bandwidth; we'll see if the telecom industry's lobbyists find a way, as they always try, to scuttle any initiative that expands competition.
There's not much individuals can do about the mega choke points except tell politicians that they care, and vote. But we have more choices about the control points that affect our daily lives.
This is why you should keep in mind that data you post online -- in places like Flickr, YouTube, Google Mail, Yahoo, Facebook and so many others -- is much more in their control than yours. Can you get your information out as easily as you put it in? Rarely. And if you're using a small service that someday goes out of business, you might be entirely out of luck.
That's also why you should be wary of doing business with companies that sell you gadgets or other technology and then insist that they, not you, have ultimate control over the ways you can use them. Apple is most famous for this kind of control-freakery in its iPhone/iPad ecosystem, but is hardly alone. You may recall that Amazon (a company in which I own a small amount of stock) drew well-deserved fury when, for reasons it considered compelling, it removed books from customers' Kindles remotely. Apologies don't suffice; the way to avoid such gaffes is to be open and non-controlling.
And the way for the rest of us to avoid being on the receiving end is to have a Plan B for the essentials. Do you?
A longtime participant in the tech and media worlds, Dan Gillmor is director of the Knight Center for Digital Media Entrepreneurshipat Arizona State University's Walter Cronkite School of Journalism & Mass Communication. Follow Dan on Twitter. More about Dan here.