Exploiting Boston

Predictably, the terrorist attack of the Boston Marathon brought the hucksters out of the woodwork more or less immediately. One of the first stories – and still the worst one, aside from Alex Jones’ “false flag” paranoid fantasy – was an attempt by self-styled “futurist” Anthony Townsend to use a false wire service story to argue for a set of network regulations that are frankly insane. Check the lede in “The Shame of Boston’s Wireless Woes:”

Almost immediately after Monday’s tragic bombings at the Boston Marathon, the city’s cellular networks collapsed. The Associated Press initially reported what many of us suspected, that law enforcement officials had requested a communications blackout to prevent the remote detonation of additional explosives. But the claim was soon redacted as the truth became clear. It didn’t take government fiat to shut down the cellular networks. They fell apart all on their own.

That’s right, the AP falsely reported that a government shutdown, but our futurist says network overload is “shame.” Gosh.

As we all know, communications networks of all types – plain old telephone service, cellular, broadband, and even the telegraph network – are not designed to carry information from all possible users all the time. They could be designed that way, but if they were most of their capacity would be unused most of the time, and the cost of providing service and the price of subscription would skyrocket to a point that would prevent any but the richest people from using them at all. In network design, the most important dimension is the “Erlang,” the factor of potential users expected to be active at any given time. Network capacity is assigned in accordance with the number of users expected to be active at a time, and the amount of load that the typical users is expected to offer; it’s then bumped up a little to accommodate surges in usage, but it’s never bumped up all the way to disaster proportions.

This is well understood by most of us, even if the tradeoffs behind the Erlang aren’t as well understood. If we wanted a cellular network to support a ton of phone calls during a disaster, one way we could do this would be to prevent users from running high bandwidth data apps when demand for calls is high; that would work a little bit, but would probably only double the number of calls that could be put through at the expense of all Internet access, so it wouldn’t be a good tradeoff. A more sensible course would be for people to learn some new etiquette for using networks in times of crisis.

This etiquette would consist of two things: Before you make call to a loved on in a place like Boston, ask yourself whether your call is really necessary. Chances are you’re simply going to ask the person if they’re OK, and then go on to explain that you were worried and you’re glad they’re safe, etc. If you’re really chatty, you might go on to express a stern condemnation for all acts of violence before condemning the hate object of your choice. This whole call does more for the caller than it does for the person in the midst of the crisis. If I’m a block from a bomb and running for cover, the last thing I need is a call from a random relative asking me if I’m OK. The sentiment is nice, but I take it for granted and I don’t want the distraction. Call me tomorrow and I’ll give you the whole rundown. If I don’t answer the call, don’t conclude that I’m dead, just hold your horses and wait until I can get back with you.

Most of these calls are only looking for one bit of information anyhow, a yes/no answer to the question “are you OK?” SMS is a better way to ask such questions. There’s a lot of SMS capacity in today’s networks, so if you’re a worry wart go ahead and use it. Even better, get a Twitter or Facebook account and look for my updates; that’s one thing social networks do really well.

Townsend exploits the situation in Boston by connecting it to a number of totally irrelevant issues, such as cell tower battery life and “hardened” networks:

But, as we learned in the aftermath of Hurricane Sandy, wireless carriers have also neglected to harden their networks against extended losses of electrical power. Thousands of towers were knocked offline in the New York region alone when backup batteries failed. Yet as a member of Governor Andrew Cuomo’s NYS Ready Commission this fall, I was stunned to learn that wireless carriers had never formally discussed plans with the region’s electric utilities to restore power to cell sites after a major disaster.

This is nonsense. Cellular networks outperformed the commercial power grid in the aftermath of Sandy, which was an event with a totally different set of challenges than the Boston Marathon Bombing in any case. The day after the bombing, the cellular network in Boston was running as it normally does, but they day after Sandy all of the New York infrastructure was in disarray – power, water, gas, highways, Internet service, and cellular. There’s no comparison. The cellular towers in New York and elsewhere are equipped with batteries and with diesel powered generators for long-term power outages, just as the POTS network is. The difference, of course, is that there are many more towers than telephone company Central Offices, so refills of diesel have to go to more places.

As to coordination, what actually happened during Sandy was that commercial power ended up borrowing diesel from the cellular operators because the cellular network was better prepared for the disaster than the power networks were.

In conclusion, Townsend goes totally off the rails:

The time to stop treating our cellular networks as an afterthought in preparedness, as expendable casualties during crises, is long overdue. In fact, they are the key to getting first responders to where they need to be, and an essential tool for resilient responses by citizens in the hours and days after a major disaster. The cellular industry has enjoyed the benefits (and profits) of access to public radio spectrum. With that access now comes enormous responsibility. We can’t afford a communications infrastructure that works only when we don’t really need it.

Actually, we can’t afford a network that can handle an infinite number of “are you OK?” calls when bad things are happening, so we need to learn how to use the one we can afford better. This means we need to do the following:

  1. Prepare networks for overload and disaster: Been done.
  2. Give dedicated capacity to first responders: In progress.
  3. Learn what to do and what not to do with our phones in a disaster: Not even started.

Aside from etiquette, we could do a lot better by giving priority to calls and text messages to 911 in times of crisis. This means discarding naive notions of “net neutrality” in favor of common sense. After the national etiquette lesson, let’s do that.