Is our wireless infrastructure a house of high-speed cards?
We've officially gotten ahead of ourselves. Millions of Americans -- and billions of people around the world -- have now entirely ditched landline service in favor of wireless, and for perfectly good reasons: wireless is more functional and entertaining, its uses stretch well beyond voice alone, and like the term "mobile phone" suggests, we can take it with us virtually anywhere we go. What we've failed to consider, though, is that wireless is still in its infancy -- so much so that it's still being wholly gutted and upgraded every few years to take advantage of new technologies and higher data rates, and with that constant churn comes unreliability. When's the last time your POTS provider waxed poetic about a next-generation network?
Granted, the comparison isn't entirely fair. Upgraded cellular networks have more to do with high-speed data than they do with voice quality -- which is pretty much the only thing POTS is responsible for these days, outside the occasional fax machine or rural modem user. But the fact remains that we rely on our cellphones day in and day out to place and receive calls, and for many of us, they've become only form of telecommunication readily available.
Therein lies the rub. AT&T's landline division, the Baby Bells, and old-school telecommunication infrastructure providers have long been famous for making insanely high availability a priority -- and that's a goal they've largely been able to achieve on account of the fact that the underlying technology has been maturing for well over a hundred years. As a rule of thumb, landlines are developed and operated under the policy that they need to be operational 99.999 percent of the time, which works out to less than nine-tenths of a second of downtime per day. As this weekend's AT&T outage reminded us, wireless carriers aren't there yet -- not by a country mile, in fact -- and they likely won't be for a long time to come.
Yes, we can levy blame on AT&T. We can pitch a fit that this never should've happened in the first place, and let's face it, it shouldn't have. But to an extent, carriers and their hardware partners are hogtied by the reality that they're operating in a fluid space where neither standards nor devices stand still for very long, and that leads to shortened development cycles, shortened testing phases, and a general lack of that "battle tested" label that we can comfortably slap on commodities like electric power, running water, landline telephone service, and taxes.
As of Monday, AT&T had announced that the massive Midwest outage had been fully restored to normal operation, and we've no doubt that they'll fully identify the failure mode and how to prevent it from happening again. Problem is, with an immature technology, there are ten unidentified failure modes for every identified one -- and that's a growing pain POTS more or less went through back when alcohol was illegal.
So give it forty, fifty years, folks. Until then, we'll be glancing at our phones merely hoping for a signal, not expecting one.
