When the Levee Breaks

Hurricane Katrina has shown how vulnerable our communications systems are in the face of a natural disaster. Anybody got a pencil?

The early days of the hurricane Katrina tragedy were complicated by an almost total lack of communications among the various local, state, and federal agencies tasked with disaster relief, rescue, and aid. According to published reports, communications leading up to storm landfall on August 29th were continuous. After the initial damage, inter- and intra-agency communications broke down completely. Cell phone towers were down, land lines were disrupted, portable radios soon exhausted their batteries, and rechargeable devices were unable to be recharged due to lack of power. Previously staged rescue units from the wealthiest, most technologically advanced country in the world were unable to operate efficiently because they couldn't talk to each other. It makes me wonder if we have become victims of our own technology–so dependent upon fragile systems and with no effective backups.

Just a Few Years Ago

I was a naval officer during the cold war. Those of us who weren't actively involved in the war in Vietnam spent our time keeping track of Soviet military forces–just in case. We navigated by celestial navigation–shooting stars before dawn and the sun at noon. If we were within radar range of land, we could use radar to fix our position. After obtaining the star declinations using a sextant, it took about 45 minutes to an hour to do the calculations necessary to fix our position. We determined the time for star fixes by synchronizing a stopwatch with one of three chronometers that were carefully protected in a gimbaled housing. Each of these chronometers (which had been running for years, always wound at exactly the same time every day) was off by hours, but since we knew the rate at which they drifted, we were able to match accurately Greenwich mean time, upon which our star tables were matched. This primitive system was time consuming and had been around for about 200 years, but it worked…day in and day out.

As I was leaving the military service, my ship was outfitted with one of the first global positioning satellite receivers and computers. This first-generation machine was a huge ungainly beast with a really weird antenna. I can't even imagine what it cost in those days. But it was able to spit out our latitude and longitude on demand in a matter of minutes (yes, minutes–results were not instantaneous). It was accurate to something like 75 meters.

I can remember thinking at the time how we were starting down a scary path, a path that would lead to a day when naval officers no longer knew how to shoot stars and were entirely dependent on machines to do their navigation. Even in those days, we knew satellites were vulnerable to missile attack. We also knew about the EMP effect from nuclear weapons–an electromagnetic pulse is a side effect of a nuclear explosion. It essentially is a very powerful, intensely fluctuating magnetic field that can interact with electrical or electronic systems and cause internal current fluctuations that can damage computers, computer networks, and telecommunications equipment.

I have no idea how a modern navy would expect to operate in a real global war environment with technologically advanced enemies who can be expected to take out global positioning and communications satellites. I am pretty certain it would be difficult to dial in a firing solution for a cruise missile without an accurate fix on the launch location.

The Internet

In modern warfare, the fear always has been communications breakdown (see Led Zeppelin, 1969). It was 1962 when the RAND Corporation was commissioned by the Air Force to study how it could maintain command and control after a nuclear attack. The final proposal of that study was a packet-switched network: "Packet switching is the breaking down of data into datagrams or packets that are labeled to indicate the origin and the destination of the information and the forwarding of these packets from one computer to another computer until the information arrives at its final destination computer. This was crucial to the realization of a computer network. If packets are lost at any given point, the message can be resent by the originator." (Source: the RAND report.) That recommendation led directly to the creation of the ARPNET in 1969, which in turn evolved into the Internet.

As it turns out, the Internet has proved itself a reliable framework for communications following hurricane Katrina. The New Orleans Times-Picayune was not able to produce print editions but was up and running immediately on the Internet at http://www.nola.com/.

No Computer Is an Island

Over the years in this column, we often have discussed how much more valuable a computer is when connected to other computers. The exchange of data between computers both local and remote is the driving force behind 21st century business. The insurance business in particular is entirely dependent upon data and then upon interconnected computers to make use of that data. At least one major carrier is paperless–all of its customer, policy, and claims data is electronic. We have become entirely dependent upon digital information to maintain our businesses. So, what kind of disaster plans do we have in effect? And I don't mean just the disaster plan where the data center is compromised. I mean the nationwide disaster where there are not only data-center compromises but the inability to communicate between remote locations.

The Old Way

I was speaking with a retired salesman friend of mine the other day. He was saying how much easier his job would have been had he been able to use modern technology such as cell phones, laptop computers, and the Internet. He was on the road (literally) all the time. He would stop at diners and gas stations throughout the day to schedule appointments and phone in his orders to the home office. He kept all his customer records on index cards and his sales records in an "accounting notebook." He is convinced he would have been much more efficient had he been equipped with the present-day salesmen's tools. I am not so sure. I understand the efficiencies that can be created with technology–I am just not always sure technology solves all problems. Selling is about building relationships, and it is much easier to build a relationship over a phone call and a handshake than an endless series of BlackBerry messages. Technology does make some things easier, but it also creates more work–both for the end user and also for the large staffs that are required to support the technology. I am not even going to mention the nightmares road warriors can cause the IT help desk. "Hi. I just checked into the No- Tell Motel in Badlands, South Dakota. How do I…?"

The Weakest Link

It seems like the weakest link always is communications. I hate to return to my cold war tales, but I will. We were forever operating with various NATO allies. We all had secure encrypted UHF communications. Key codes were changed every day–these were very secure radios. But there was a problem. There always was one ship with a broken radio set. That meant we all had to revert to some other form of communications–and that meant lowering the lowest common denominator. If we could see each other in daylight, we could use signal flags, and flashing light at night. Both forms of communication were relatively secure–heck, if the Soviets were close enough to read our signal flags, we were close enough to shoot them. But they also were limited in the amount of information they could impart. It is very inconvenient to carry on a tactical conversation using signal flags.

So, we always ended up using unsecured single-side band VHF, which meant we were broadcasting all of our tactics in the clear over a frequency that could be picked up hundreds of miles away. The Soviets didn't need spies, they just needed a couple of decent VHF receivers. Advanced technology failed us. And it will continue to fail us if we rely on it too heavily. There is something very satisfying about a copper wire connecting point A to point B. It may not be the most sophisticated backbone, but it certainly is a sturdy one.

Back to the Internet

One way or another, we all use the Internet to connect our data-processing systems. You may (and probably do) have dedicated point-to-point lines between your major operation centers, but when you need to start communicating with your customers, individual agents, and agencies, you are going to rely at some point on the Internet. And the Internet is a fragile beast, not necessarily the packet-switching infrastructure. That has pretty much withstood the test and has proved to be a very reliable way to get information from point A to point B. You notice I say reliable–not necessarily efficient. A trace to my home 40 miles from this computer frequently takes 15 hops with side trips to cities 500 miles distant. But that doesn't matter. The backbone is solid.

But what about DNS? For most users, a DNS failure means something on the order of the inability to reach www.file_your_claim_here.com. Now, there are 13 root name servers–and a 17-year-old with the right connections can bring any one of them to its knees with a DDOS (Distributed Denial of Service) attack. It seems to me a reasonably affluent terrorist group could manage the same on all 13. I think it is inevitable terrorist attacks will increase on information systems. If Abby Hoffman and company were looking to disrupt the U.S. government today instead of the 1960s, I guarantee they would go after telecommunications.

Impudent Snobs?

There is a point to all this. I am not antitechnology. In fact, I love technology. As we become more technology advanced, we become more dependent upon those technologies. We will reach a point where "advanced" societies no longer will be able to function without technologies they will be unable to replicate. We are not that far from becoming a society where no one will be able to build the machines that build the machines upon which we rely.

I am not advocating a step backward in time–I truly do not long for party lines with operators and slide rules. But I do advocate businesses taking a close look at how they would function if the power were out for three months or if cell phones were down for six months and telecommunications severely degraded. We need at least a rudimentary business continuation plan–just enough to keep going, just enough to take care of our employees and generate enough services to ensure income when the "grid" comes back.

I don't think I want to rely on the government–at any level–for the survivability of my business. We experienced fairly widespread anxiety over the Year 2000 non-issues. Maybe we need to take a look at how we would perform in a real crisis. Could your business survive for three months using 1930s-vintage technology? Probably not. We actually have a few typewriters around here at National Underwriter, but alas, they are electric. Anyone have an inverter?

Special Reports /

When the Levee Breaks

Related Stories

Recommended For You