Schneier on Security
A blog covering security and security technology.
« Perceptions of Risk |
| "Cyberwar" in Estonia »
August 23, 2007
I live in Minneapolis, so the collapse of the Interstate 35W bridge over the Mississippi River earlier this month hit close to home, and was covered in both my local and national news.
Much of the initial coverage consisted of human interest stories, centered on the victims of the disaster and the incredible bravery shown by first responders: the policemen, firefighters, EMTs, divers, National Guard soldiers and even ordinary people, who all risked their lives to save others. (Just two weeks later, three rescue workers died in their almost-certainly futile attempt to save six miners in Utah.)
Perhaps the most amazing aspect of these stories is that there's nothing particularly amazing about it. No matter what the disaster -- hurricane, earthquake, terrorist attack -- the nation's first responders get to the scene soon after.
Which is why it's such a crime when these people can't communicate with each other.
Historically, police departments, fire departments and ambulance drivers have all had their own independent communications equipment, so when there's a disaster that involves them all, they can't communicate with each other. A 1996 government report said this about the first World Trade Center bombing in 1993: "Rescuing victims of the World Trade Center bombing, who were caught between floors, was hindered when police officers could not communicate with firefighters on the very next floor."
And we all know that police and firefighters had the same problem on 9/11. You can read details in firefighter Dennis Smith's book and 9/11 Commission testimony. The 9/11 Commission Report discusses this as well: Chapter 9 talks about the first responders' communications problems, and commission recommendations for improving emergency-response communications are included in Chapter 12 (pp. 396-397).
In some cities, this communication gap is beginning to close. Homeland Security money has flowed into communities around the country. And while some wasted it on measures like cameras, armed robots and things having nothing to do with terrorism, others spent it on interoperable communications capabilities. Minnesota did that in 2004.
It worked. Hennepin County Sheriff Rich Stanek told the St. Paul Pioneer-Press that lives were saved by disaster planning that had been fine-tuned and improved with lessons learned from 9/11:
"We have a unified command system now where everyone -- police, fire, the sheriff's office, doctors, coroners, local and state and federal officials -- operate under one voice,'' said Stanek, who is in charge of water recovery efforts at the collapse site.
"We all operate now under the 800 (megahertz radio frequency system), which was the biggest criticism after 9/11," Stanek said, "and to have 50 to 60 different agencies able to speak to each other was just fantastic.''
Others weren't so lucky. Louisiana's first responders had catastrophic communications problems in 2005, after Hurricane Katrina. According to National Defense Magazine:
Police could not talk to firefighters and emergency medical teams. Helicopter and boat rescuers had to wave signs and follow one another to survivors. Sometimes, police and other first responders were out of touch with comrades a few blocks away. National Guard relay runners scurried about with scribbled messages as they did during the Civil War.
A congressional report on preparedness and response to Katrina said much the same thing.
In 2004, the U.S. Conference of Mayors issued a report on communications interoperability. In 25 percent of the 192 cities surveyed, the police couldn't communicate with the fire department. In 80 percent of cities, municipal authorities couldn't communicate with the FBI, FEMA and other federal agencies.
The source of the problem is a basic economic one, called the collective action problem. A collective action is one that needs the coordinated effort of several entities in order to succeed. The problem arises when each individual entity's needs diverge from the collective needs, and there is no mechanism to ensure that those individual needs are sacrificed in favor of the collective need.
Jerry Brito of George Mason University shows how this applies to first-responder communications. Each of the nation's 50,000 or so emergency-response organizations -- local police department, local fire department, etc. -- buys its own communications equipment. As you'd expect, they buy equipment as closely suited to their needs as they can. Ensuring interoperability with other organizations' equipment benefits the common good, but sacrificing their unique needs for that compatibility may not be in the best immediate interest of any of those organizations. There's no central directive to ensure interoperability, so there ends up being none.
This is an area where the federal government can step in and do good. Too much of the money spent on terrorism defense has been overly specific: effective only if the terrorists attack a particular target or use a particular tactic. Money spent on emergency response is different: It's effective regardless of what the terrorists plan, and it's also effective in the wake of natural or infrastructure disasters.
No particular disaster, whether intentional or accidental, is common enough to justify spending a lot of money on preparedness for a specific emergency. But spending money on preparedness in general will pay off again and again.
This essay originally appeared on Wired.com.
EDITED TO ADD (7/13): More research.
Posted on August 23, 2007 at 3:23 AM
• 46 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
you cannot imagine what is to be uncommunicated in betwen different floors by the direct will of a millonaire madman which in that moment you don´t know who the heck he is
Does this having fewer points of failure not also increase the chance that some outage (spam or jam) would hinder more, as well?
One thing worth noting: in the report on Katrina, one of the few areas where communications did work was where volunteer amateur "ham" radio operators and their equipment were used. These operators were in many instances able to pick up and carry the load as the normal communications infrastructure (e.g. the phone systems and Internet) failed.
Which goes to show that it does *not* take pork-barrel spending to achieve the desired results. In some cases, one need only make use of resources already available.
This is really a shame when programmable radios with multiple channels are the norm and there are specific frequencies set aside (at least they are listed as such in scanner enthusiast guides) just for inter-agency comms.
The Feds are doing something: the National Incident Management System should be required of most first responders by now.
While it doesn't specify freqs, it does include things such as ICS and cleartext over the radio instead of code-speak like 10-4.
Interoperability has gotten technologically much more feasible in recent years. Now handheld radios are not only synthesized but can be programmed by the using organization to use whatever freq (within the operating band of the radio) they desire and have sufficient memory for many channels. So (as long as they do the corresponding FCC/NTIA paperwork) they can have the firefighters' radios with the ability to access a local police or ambulance frequency. They could even set up a "channel 19" where EVERYONE could play (of course someone away from the scene would need to be a net controller to keep it from sounding like the real channel 19).
My local community had a disaster in 1986 (ironically before I moved there) in which police from probably a dozen or more local communities participated. And they could not inter-communicate. So we had HAM volunteers riding in police cruisers and at medical facilities acting as comm relays since we could all be on the same frequency.
It also should be noted that Minneapolis's new 802.11x wireless network was being installed at the time and the area's around the bridge were the first to be put up. The wireless vendor temporarily turned off authentication and billing to allow local crews around the bridge quick access to the network.
Interoperability is good, and we've spent billions on it, but let's not deceive ourselves that it's a magic bullet. Once each of us can talk with any of us a number of challenges regarding discovery, routing and scalability that used to be masked by the bounds of technology come to the front of the queue.
Who is where and in what role? What information do they need? How much information can they handle? How can responders best manage their finite resources of attention when it seems like the whole world is calling?
The organizational paradigms of ICS and NIMS offer some broad guidance, but a lot of detail remains to be worked out if the current thrust toward interoperability is to be more than just a bonanza for the radio vendors.
Bruce has absolutely nailed it here: what we need is flexibility and adaptability in response to emergencies. Training for both foreseen and unforeseen events.
I teach and train martial arts and constantly stress in class that you WILL NOT be able to predict what type of conflict you will run into. If you expect hand-to-hand you will get a knife, expect a knife, get a gun, expect a gun get a car coming at you, etc, etc. The only thing you can do is train on core principles that keep you alive: Preparation, Awareness and Attitude.
Preparation - Not for a specific event but a multitude of different type of events. There are some simple, low tech things that all people can do to prepare for trouble. For example:
1) Have communication devices and know how to use them.
2) Know basic first aid.
3) Carry tools appropriate for a variety of problems and carry them all the time.
(i.e. Cell phones & swiss army knives)
Awareness - Learning how to learn what's going on. This is more difficult than it would appear. You have to want to see what's going on and then train to be able to see it. Also you have to work on seeing what's NOT happening, which can be even more important. e.g. - Sherlock Holmes and the dog that didn't bark - http://en.wikipedia.org/wiki/...
Attitude - Consists mainly of the following ideas:
A) Don't panic.
B) Stay Safe.
C) Fixed goal but flexible mind.
I worked for the public safety side of Motorola for several years, so I am well versed in radio systems and inter-agency communication. I agree with Bruce wholeheartedly that this is a major area we need to improve. The technology is readily available now (it wasn't really available even five years ago).
However, the problem is very complicated, for the reasons Bruce listed. This perfectly fits under the classic quote:
"The difference between Theory and Practice is that, in theory, there is no difference between Theory and Practice, but in practice there is."
This is a much more difficult problem to solve than to simply suggest we all use the same frequency. Trunked comm systems are pretty complicated and proprietary, trust me, and there is always agency-specific needs like Bruce pointed out.
We'll get there, but it's going to take a lot of work, and there won't be an overnight solution (even with a lot of money thrown at it)
Well spoken Bruce.
Interagency communications is a sure key to faster and more efficient response in Disaster situations.
NIMS and the NRP outline the chain of command structure, but fall short in dictating this key piece of the puzzle. And it needs to be there, nationally as well as regionally. It will save many lives, both victims and responders.
Additionally, as bzelbob pointed out, training and preparedness are also key.
Planning is great, but in practice we can hone our response, weed out problems ahead of time, and markedly increase the likelihood of saving lives.
Having standardized communications channels is a lesson we have learned too many times. We need to make that part of the plan, nationwide (heck...globally) and "make it so".
Nice essay. However, I disagree with one part: "No particular disaster, whether intentional or accidental, is common enough to justify spending a lot of money on preparedness for a specific emergency."
While this is true in most cases, I believe it's wrong for certain disasters which are both large and common. For example, people in Florida spend a lot of money on preparedness for hurricanes, and it's necessary and worthwhile because it's known that a hurricane is going to hit sometime pretty soon. Another example would be earthquake-prone areas. An unbelievable amount of money is spent in those areas building structures to a code which is meant to resist that one particular disaster, but it pays off big-time when an entire city doesn't burn to the ground after a big quake.
For most things, general preparedness is better but there are certain specific scenarios which deserve a great deal of money and effort.
I live in California, the home of the Incident Command System which the Feds are now making standard. Settling the issue of command and coordination between a number of different agencies is important -- as long as the Feds remind themselves that they are neither the "top" or the "command."
The problem at Katrina is that the @#!&! local law enforcement, National Guard and/or FEMA kept closing the damn disaster scene and pointing weapons at the people they were allegedly there to help.
That's not a radio problem. That's a command and control problem of enormous scope and magnitude.
Unless your community's disaster system is prepared to deal with both trained and emergent volunteers, your community is not ready.
Shout out to the amateur radio people. I have seen them drag coordination kicking and screaming out of a disaster ranging from a hospital losing its entire internal phone system up to and including a major regional earthquake.
Radio has made great strides over the last 5 years, but eveyone that needs to be in the loop does not have a radio, and many emergency issues can't be handled through simple talking. Data, enabled with GIS, and distributed through standard approaches like Common Alerting Protocol, doesn not have to be repeated, can be updated as required, and consumed by a variety of devices.
As IP clouds become more pervasive, and cell phones become multi-netowrk capable (using IP and cellular like hte new TMOBILE phones), and can store procedures and plans on an encrypted micro-sd card (1GB now, 4GB end of year), the utility of a hand held device becomes very strong. As interfaces bcome more prevelent, these phones can participate in radio networks via ROIP capabilities.
Oddly, I think it's exactly the kind of advanced equipment that's being bought lately in the interests of "improving" communications that hampers interoperability. Trunked digital radios are nice, but they require base stations to operate and won't talk to units using different protocols. Plain old FM short-wave radios, by contrast, pretty much don't care what brand the other end is and don't need base stations (repeaters are nice, but in a pinch anyone can relay messages for anyone else). It doesn't work as well in the best of circumstances (transmission discipline's needed, for example, to keep from stepping on other people), but it keeps working decently as things degrade.
A police office in the town that I live in, who's been through a slew of incident management training, and managed a tornado response or two, expressed to me a while back that allowing the various responding agencies eg) police, fire to all talk to each other frequently did more harm than good.
The different services do different things, with different lingo and communication patterns. What was needed was for unified command at the command level, not allowing Charlie Cop to speak to Frank Firefighter.
There is some really cool mesh technology that is slowly taking hold, which relays IP traffic over repeaters. A fireman, for example, that has an IP-enabled radio, can have five or six of these "repeaters" on their belt. As the signal begins to degrade, they just turn a repeater on and drop it, and the network builds the mesh itself.
Interesting blog post today on public safety communications.
You quote the FCC PSWAC report on interoperability.
There were other comments in that report which referred to an underlying organizational reason for lack of interoperability. Here are two such comments:
Interoperability (or the lack thereof) is often affected by non-technical factors
including reluctance to adopt new approaches and funding limitations.
Contending with the human factor is another critical element in achieving
interoperability. PSWAC p. 19.
There is a command and control issue that varies to some degree across the different jurisdictions and agencies, but is basically similar. Many commanders are willing to have personnel from other agencies join “their home��? system, but somewhat hesitant to allow the personnel within their own agency to join the radio system of another agency when it jeopardizes the commander’s ability to maintain communications with his own personnel. A user in the field who cannot be contacted is not available for assignment. PSWAC p. 313
The reference to "contending with the human factor" in the first quotation is, I believe, a euphemism for the fact that many fire and police chiefs don't want their people talking to people from other organizations. I believe that the basis for this is primarily a desire to keep the relevant staff focused on task—not bureaucratic empire building.
Jimmy Breslin wrote an article on the fire/police communications issues in 9/11 that gives a different view into the bureaucratic issues. A copy of it lives at
The failure of collective action insight is probably partially correct. But, there is a lot of history and technological evolution that is also involved—things that economists would call path dependency. As technology improved higher frequencies could be used; large agencies and growing agencies expanded to the higher frequencies; some smaller agencies stayed behind. The lower frequencies give better geographic coverage per dollar of infrastructure—the optimum frequencies for the Nevada Highway Patrol differ from those that are optimal in Los Angeles and probably even from those that are optimal in Las Vegas.
I think the best solution for interoperability is to build modern wireless systems with architectures more like modern commercial mobile radio systems—such systems have much more capacity per radio channel and permit easy interoperability. Look at the San Diego County regional radio system as a good example of such a system. http://www.rcs800mhz.org/web/ It provides efficient radio communications, good coverage, and support for interoperability is a no-extra-cost byproduct.
The bridge collapse and Katrina are rather different events. The bridge was a small, confined area with no collateral damage to communication networks or the power grid. New Orleans was without power, phones, or internet for weeks in some places. No matter how robust your community thinks its communications networks are, cell phones, radios, computers, and other devices must have power to operate and most devices also require some form of working communications grid. It wasn't compatibility that forced the civil war like communications, it was the destruction of the infrastructure.
So how much do you want to spend on a backup system? Do you want satellite phones for each officer, a portable generator, and some source of unlimited fuel for the generator? How will you supply the fuel if the roads are all washed out?
I think there are some possibilities for the future. Portable cell towers with satellite connections powered by solar, wind, and water and are capable of being dropped into place by helicopter would quickly redeploy a cell infrastructure. Portable power stations could provide the local responders with the needed juice for their equipment. There's no way a state or municipality would have the capability of developing or deploying such a system, though, so your comment about the feds is right on the mark.
As people have pointed out above, it's not just the radios. Any amateur involved in emergency services and coordination can tell you that different local agencies (different service, same town, same service, neighboring town) will have different meanings for the same codes and jargon. So often it can be useful to have trained people doing the relaying specifically to prevent the miscommunications that result from people talking directly to each other...
A UN-MOVIE-PLOT CHALLENGE:
How much of your thinking has been shunted into thinking only of First Responders instead of First Contacts?
Who is more likely to be exposed to a newly arrived contagion: an ER- paramedic, or someone who teaches in a school which is the first point of community participation of a newly arrived person who comes in (with or without documents)? I am talking about Adult English as a Second Language School Teachers.
Such as they are NOT on the list of those who can be vaccinated for smallpox, and yet they are center in an behavior environment that stretches contacts throughout every large community, includes all ages and service workers, does not listen to The Network News, and moves, lives, and works below the political radar.
Is security served by ignoring a large reality?
Over 20 million persons.
My Background: 20 year volunteer firefighter / EMT, with Instructor & Officer certifications. I don't know the exact total, my guess is 2,500+ incidents I've responded to over the years.
1) A police officer & firefighter do not need radios to talk to each other.
Maybe in small towns where informal lines of communication exist, fine. But there's some ridiculous expenditures out there, like Rhode Island's system of allowing a firefighter in Westerly to talk to a police officer in Woonsocket (I think that was proposed, not built).
For the most part on these large operations, each emergency service needs to act within it's chain of command. The police do not need to listen to peanut gallery opinions of firefighters of how they should do their job, the firefighters do not need to listen to the peanut gallery opinions of the police.
2) Commanders do. That's why you set up Unified Command Posts. So when a police officer reports something up his chain of command, the police commander can look at the fire commander and relay the information.
And then commanders do what commanders do: Issue orders in a coordinated manner based on the information.
Chaotic? Stressful? Certainly. But it's better then working level supervisors trying to parse information coming at them from other working level guys who may or may not have the proper training or big picture view.
3) "Management" particulary excutive level politicians often confuse their inability to talk to each other...to mean that causes difficulties for the ground pounders.
Senior management between agencies need to talk and have radios that can talk to one another when they're not present in person at a Unfied Command Post or Emergency Operations Center.
Far too much money is spent on overly fancy radio systems for the troops, in order to have features only needed by senior managers.
4) What is truly needed?
When you look at the response to a true catastrophe like Katrina, you see that human-based communication systems where overwhelmed by data.
The people couldn't process and match resources effectively.
That's a database problem. Match pledged resources with requests. Computers can handle huge amounts of chaotic data that humans can't.
So we need a Resource Management System, basically just a geographically distributed database.
And build a simple interface to it -- something using basic web standards that allow pre-registered departments to pledge what resources they have available, and be notified when and where to report...and to confirm they will accept the mission.
Keep the pages very simple, so we can run them on a dedicated, low-bandwidth network (not the general internet) that includes a self-establishing radio network. After the disaster, start opening up the boxes, plug in the batteries, raise the antenna and have the TCP/IP network start to discover it's routes.
Include in that network a Text Messaging system -- so you're not reliant on voice communications, and messages that are crucial can be flagged "K" to require a response. And will whine to a terminal user until they acknowledge receiving and if necessary acting on the message.
Such a system should arguably also be encrypted by default, with each terminal having it's own keys that are centrally revokable to control any malicious use of the net to issue false orders.
Such a system ain't that fancy -- it's a matter of integrating the software and hardware components, along with a process of registering agencies that will participate and resource typing their personnel and equipment.
If the big one hits, I could give a rat's ass about having a video conference with someone or having a radio that would allow me to talk to a Trooper 75 miles away.
But if handling an operational sector, I'd kill to have a text messaging system to keep an accurate track of requests to and from an emergency operations center.
Or if back at the EOC, to look at a computer screen and know:
1) My requests have been received;
2) My requests have been acknowledged;
3) Which of my resource requests have been matched, dispatched, mission accepted, and estimated time of arrival;
4) Which of my resource requests are pending acceptance of the mission;
5) Which of my resource requests can not be filled, and the anticipated time when (if ever) they will be filled.
A collapsed bridge is actually a pretty simple emergency -- long, but it's all right there, and we have plenty of resources to throw at it.
A situation like Katrina which present hundreds of sectors each as technically complex as the Minnesota bridge collapse is a different beast and absolutely needs much better solutions -- specifically resource management -- then we have in place today.
I was just reading the link to the Breslin article.
One point it didn't make well, the Fire Alarm Offices in cities were traditionally located in parks -- to keep them from becoming involved in conflagarations. Too often very old -- over a century old in this case -- lessons are forgotten. And NYC puts their EOC across the street from the biggest terrorist target.
The more important point is this though -- it isn't the place of a Fire Alarm Office(Communication Center) to talk to police resources like helicopters.
The failure was one of person-to-person.
On a simple scale, when we have an auto accident, we tell the Trooper when we're leaving (even if it's obvious, gee they're putting everything back on the truck) -- "Hey, we're all set. Need us here anymore?" And usually the reply is "Thanks, have a great night." But sometimes, "Oh geez, hey could you stay and keep your flood lights on while I finish up my investigation?"
Similiarily, the Troopers don't leave a scene without checking with us first, "Hey, I'm all set...I got another call pending, would you guys mind babysitting this till the wrecker is done?"
You shouldn't expect Fire Alarm to monitor or talk to the Police helicopter.
There should have been a Police official near the Fire Command post who could've looked over and said, "Shit, Chief -- our helicopter is reporting the building is shifting and we're pulling all our guys out."
Or less desirably, you'd have the Police Communications call the Fire Communications to tell them. Ideally they should be in the same building, and during an event like that have experienced dispatchers / supervisors in a joint center to relay messages in person from one "side" to the other as a fall back to what should be face-to-face interaction by the different agency commanders on the ground.
It is *not* a technical failure, and that part of the problem doesn't have a technical fix.
@Matt from CT
Your posts are interesting, thank you. What is especially interesting about them is that they conflict, in that you first assert that line officers do not need to talk to each other (only to their commanders), then talk about instances where it's desirable for the opposite to be the case.
According to the command and communications structure you describe in your first post, the state trooper who wanted your lights or the firefighter who was staying anyway would have had to talk to *his* commander, who would talk to the trooper's commander, who would relay the request/order (in his ignorance, by the way, because he doesn't have a good view of the situation as the line officer does).
I understand that you're talking about different situations (disaster response vs. routine incident response) but I don't see how a disaster makes the ability to use someone's lights or babysit someone's scene (when you can and it means they can go elsewhere) *less* useful.
The answer is to get the good bits of direct communication without the bad ones. For this, line officers need the technical ability to talk directly to each other, and a command structure that allows them to do this when necessary and not otherwise.
In other words, the direct responder-to-responder channels should be set up via the command structure you describe ("I need to talk to the fire crew on the floor above me now.") but the actual communication may be optimal if it's direct.
Not conflicting at all, David.
The Trooper on the scene is the top of his chain of command for the incident. He doesn't need a radio to talk to the fire officer in charge -- they just meet face to face.
If you had a State Police Sergeant on scene (usually reserved for incidents involving another Trooper, or some fatal accidents), then the investigating trooper would have his request for the FD to stay on scene to provide lighting go up through the Sergeant. In practicallity, the Trooper, Sergeant, and Fire Chief would all probably be standing next to each other when that request was made.
On a complex incident or catastrophe situation, it becomes broken down into a series of smaller "scenes" each being handled as it's own small incident. No need for radios, since the decision makers from all agencies can still discuss what's relevant to this particular sector face to face.
For the scenario of "I need to talk to the fire crew above me" :
Standard practice is for units on a scene to be operating on a common "Talk Around" channel. So if fire is getting by the hose team on Floor 4 and threatening to cut-off the stairwell to Floor 5, when they report that to command the officers & members on Floor 5 would also hear the report. The Commander's next major responsibility is to verify the crews on Floor 5 know they need to retreat -- something they should have already started to do based on standard practices.
Depending on the community, there may be a dedicated seperate channel for Command to Dispatch.
As the incident expands in size, different "sectors" should get their own channels; what constitutes a sector may vary. It may be putting Water Supply operations on one channel, and Interior Operations on another. It may be units in the West Wing on another channel from the East Wing.
At this point it's pretty standard for Command to Dispatch traffic to go to another channel, and Command to Command traffic (i.e. Incident Manager to Sector Commanders) to start operating on their own channel as well.
The guys in the East Wing need to know what other guys in the East Wing are doing; what's happening in the West Wing isn't that important as it won't immediately endanger them so that communication can feed up through to command and back out through the sector officers.
Even in my relatively rural area, you'll start getting other resources to help -- for example, an Incident Dispatcher will be assigned. That's a dispatcher whose called in on overtime and responds to the scene to handle all the communications from the scene to the dispatch center. The Incident Manager can tell the Incident Dispatcher what he wants, but all the actual talking to the dispatch center is done by the Incident Dispatcher, freeing the Manager to continue on to higher value work.
The Incident Dispatcher itself is just a modern update to the County Coordinator system developed in the 1950s to fill a similiar role.
How can different agencies learn to trust each other if they can't eavesdrop on each other?
What could be wrong with everyone hearing everyone's voice mail?
Is it just more movieplottophobia?
"'This is an area where the federal government can step in and do good.' Classic! Absolutely, the federal government should be put on the job! They did such a stellar job with Katrina, they've earned it! What more proof of their competence could we want?"
I'm not sure I understand your point. I believe that imposing interoperability standards from the top -- i.e., the Federal government -- is the best way to solve the interoperability.
You argue, if I have your point correctly, that the federal government's poor job w.r.t. Hurricane Katrina is evidence that imposing interoperability standards from the top isn't a good idea.
Honestly, I don't see how that follows. If anything, it was a lack of top-driven interoperability standards that exacorbated communications problems after Katrina. My original article makes that point.
While you make some good points, your post is likely to offend a significant number of first responders, and for reasons not relevant to the subject: the term "ambulance driver" is considered demeaning and insulting by paramedics and EMTs (Emergency Medical Technicians).
EMS (Emergency Medical Services) personnel tend to be highly qualified, sometimes with years of medical training. Driving the ambulance is a very small, and probably the least important, part of the job.
After all, police officers are not "police car drivers" and firefighters are not "firetruck drivers".
"You argue, if I have your point correctly, that the federal government's poor job w.r.t. Hurricane Katrina is evidence that imposing interoperability standards from the top isn't a good idea."
Well, what I meant to argue is not that imposing interoperability standards from the top isn't a good idea, but rather that invoking federal authorities to do the imposing, isn't a good idea.
The federal authorities have a poor track record with respect to disaster management. I don't think you'd argue against that point. But I would argue that the principal cause behind that poor track record is the same cause that would result in a poor planning process, a poor implementation, and poor maintenance of any such program. All areas of the "imposition" would be tainted by the relative incompetence that we saw firsthand with the handling of Katrina. That cause is the relative absence of proper incentive.
There is no reason that "interoperability standards" could not be agreed upon by parties involved. Merely because such standards aren't yet in place, is not a good enough reason to call in the Feds to force all of us children to straighten up and fly right. Resorting to the use of governmental solutions (which are actually "public" funds backed up with the legal use of force if necessary) should truly be a last resort.
Look at all the volutarily organized associations that exist in the U.S.A.--there are thousands. Government programs or agencies which require mandatory membership could have been created in place of any of these, but they weren't. People created these voluntary organizations because they chose to, to accomplish a common purpose.
Voluntary compliance with the standards of any group or organization that doesn't lord the power of the state over you will nearly always be less than you'd get as a result of threats of state-backed punishment. But people don't ALWAYS need to be threatened in order to incentivize them to do good. A grassroots solution has all the benefits of letting the best ideas rise to the top, rather than having a politically motivated solution from Camelot forced down from the top.
It's always tempting to demand that the government "do something" when a good solution to a pressing problem doesn't quickly present itself. It seems the marketplace of ideas has had its chance, and now it's time for the government to step in. But surely we've seen a sufficient number of examples where stifling the paternalist impulse to attempt to solve other people's problems with a "we're from the government, and we're here to help you" solution, has paid dividends in a fantastically creative approach to solving a problem brought forth from someone who was affected by the problem directly. (Would you like to know of some examples of these solutions? They're truly inspiring.)
Who are these people in this case? The First Responders. Do you honestly believe there are simply no leaders in that group? No leaders who will begin to address this problem without politicians nearby to hold their hand, or force their hand?
Surely you've seen the problems that come from relying on government agencies to come up with creative solutions (you comment daily on the very lack of such creativity in these agencies.) Give the creativity of these folks a chance, and if you have already, then give them another. I don't think you'll be disappointed.
Radio communications systems for first responders have supported interoperability for years. Most states and many counties have had "intersystem" frequencies since the 1970s, and almost all new trunked radio systems that have been constructed 1990s-present include talkgroups for interagency communication. Even New York City had a functional interagency emergency comm system in place on 9/11, the Department of Information Technology and Telecommunications (DOITT) trunked system, which supported the activities of the Office of Emergency Management (OEM). However, OEM was marginalized and refused a role by police commanders, and the 800 MHz interoperable radios reportedly sat unused all day.
Dedicated interagency frequencies are now available in New York City. I doubt they are in use. In San Francisco, all emergency agencies share the same trunked radio system and are capable of speaking to one another. I've never heard them do so (although, in fairness, police and fire commanders may talk with one another via cellphones).
We don't need to throw billions at tech fixes except in regions where there are real communications gaps. This only benefits large vendors such as Motorola and MA/COM, who are lobbying hard and fast for bigtime interoperability programs. We need cultural change to help implement real incident command. Unfortunately, it doesn't seem easy (especially for outsiders) to effect change in the way police and fire agencies work.
In response to Hilarious,
Your political arguments remind me of the various movies made about the signing of the Declaration of Independence and writing of the Constitution. Each party didn't want the power taken away from them and given to the "Feds". I wonder if you have had any dealings with various departments trying to communicate with other departments. In the county I live in we have various city public service chiefs who want their own empire and until 800 mhz came and was "strongly" encouraged by the State for all city's to join or loose funding they couldn't even talk to the adjoining town to let them know that they were in hot pursuit of a criminal. Fire departments are much the same way. Multiply that by every city/county and state and you have at least some idea of "my idea is better than yours", mentality. Everyone has a standard right now its called their frequency plan. Work on a wild land fire and see how many fire crews come in with different frequencies on their radios. In fact the My County Sheriffs department uses Ham radio operators in major disasters as they are able to communicate with all services through operators stationed at the different agencies throughout the county.
The Idea that I got from Bruce was that a standard from the Federal Government with money's allocated just for communications infrastructure and not some pork barrel that the individuals would spend for the wrong things is what is needed. Having each fire/police/EMS person using the same or communication device with similar capacities available to them makes more sense than different frequencies that can't communicate except to their own group. Most trunked systems can be set to allow different groups to communicate with other agencies, and if not needed the switch is set so the rank and file who don't need to can't.
The biggest problem with Fire/police/EMS personnel is that they lack training in using their radio's for emergencies like hurricanes,earthquakes,terrorist attacks. Even big fires can overload their system because they are trained to call for help NOW when they need it. The problems that result when 10 or 20 people are all calling for help at the same time on the same frequency is like putting them in the same room and each yelling what they need. No one can be heard,each thinks (rightly so) that their emergency is the most pressing and the dispatcher can't get in to sort out the mess. Amateur Radio is only "amateur" in that they don't get paid to do what they do. They spend hours practicing for the disasters that will happen and try to instill discipline in their operators to listen,listen,listen. Then talk if you need to, and make it brief because others need the frequency.
The unified Incident Command Structure makes sense because it trains everyone in the same goals. Managers manage, rank and file work in the trenches. Communications needs to be trained as an ongoing item to handle the times when overload will occur. It is every bit as important as a Fireman's turnout gear, a policeman's pistol, and an EMS IED/First response pack.
At first glance I was quite impressed that you specifically included EMS workers in your list of first responders. Often, as in the days after 9/11, EMS workers such as Paramedics and other Emergency Medical Technicians are either completely overlooked or subsumed under the blanket title "Rescue Workers."
I have to say, though, that I agree with A.M above. We are Emergency Medical Techinicans, some more advanced than others, but we're all EMTs. And we do more than drive ambulances and increase fire department revenue by artificially boosting call volume.
"Who are these people in this case? The First Responders. Do you honestly believe there are simply no leaders in that group? No leaders who will begin to address this problem without politicians nearby to hold their hand, or force their hand?"
I suggest you read the paper I linked to in my essay: "Sending Out an S.O.S.: PUblic Safety Communicaions Interoperability as a Collective Action Problem." It explains why leaders in the various first-responder groups can't solve the problem, why the collective-action problem results in sub-optimal results.
I know it's fun and fashionable and libertarian to poo-poo anything involving the government, but this is a counterexample; the basic economic mechanisms fo capitalism and self interest cannot solve it.
(I'm writing about another now; it'll appear in about four weeks. Really, it's always good to know the limitatations and boundaries of any system)
"While you make some good points, your post is likely to offend a significant number of first responders, and for reasons not relevant to the subject: the term 'ambulance driver' is considered demeaning and insulting by paramedics and EMTs (Emergency Medical Technicians)."
Apologies; I don't know the lingo.
I used "EMT" in one paragraph, and "ambulance driver" in another.
Hilarious has a little bit of a point: putting members of the current administration in charge of any large interoperability effort would likely result only in graft and confusion. Luckily no one in power will be listening to Bruce much before 2009 anyway...
"the basic economic mechanisms of capitalism and self interest cannot solve it."
Wow. Well, I'll give you this: you don't shy away from sticking your neck out.
Bruce, I'll bet you dollars to donuts that with less than 8 hours of reading of what I would call evidence to the contrary, you could convince yourself that your statement I quoted above is not true.
I'll go read the paper you suggested I read.
A parting thought: any time you find yourself ready to commit to thoughts like "[there needs to be a] mechanism to ensure that individual needs are sacrificed in favor of the collective need" and "[we need a] central directive to ensure interoperability", pausing to consider that you're taking a step down an extraordinarily well-travelled road, with quite literally thousands of examples of projects, plans, and proposals that turned out far differently than was imagined at their outset, not a single one of which lacked the very best of intentions, might inspire you to take a second look at a solution that seems intuitively to be a sure bet.
We already have the benefit of quite of bit of hindsight in the realm of economics. The temptation to think "this time will be different" is always strong, but it can be tempered by looking at this history.
>The biggest problem with
>Fire/police/EMS personnel is that they
>lack training in using their radio's for
>attacks. Even big fires can overload
>their system because they are trained
>to call for help NOW when they need it.
This is EXACTLY the trap senior managers fall into.
Hurricanes, Earthquakes, etc are disasters. The are *NOT* emergencies.
They are disasters and need a different communication system from the day-to-day tactical systems for use in coordinating wide area responses.
The tolerance for latency in strategic communications being relayed through Hams is probably an hour or better. It's minutes or less with tactical Fire / Police communications.
It doesn't mean you build the day to day radio network to handle it. Build a specialized network -- and in my opinion, with Resource Management System at it's core, to handle these strategic needs.
One of my petty pet peeves is referring to police/fire/EMT/... as "first responders"; they're not, they're the "second responders" (and they are essential and life-savers.) The first responders are those "ordinary people" who are responding to situations by making the calls to 911, and coping with the situation until help arrives. Sometimes they have training and are able to help even before the second responders arrive, sometimes they help without training, sometimes they mess things up.
And one of the things we need to do in this country is to train a lot more of those "ordinary people" to correctly help before the government-designated helpers arrive.
"First Responder" is a bit of lingo from the Public Safety field that started to become common in the mid-1990s.
It started off in the EMS side to describe a level of training above First Aid, but below that of an EMT. It's about 32 hours of training to be a First Responder v. 160 hours for an EMT-Basic today.
(There's also EMT-Intermediate who can start IVs and administer a very limited selection of drugs with 400 hours, and EMT-Paramedic with a much broader assortment of drugs and cardiac electrotherapy skills at 1,200 hours)
Since many police and firefighters are also either EMS First Responder or EMTs, the term seems to be familiar to all of them and kind of gained wide-spread use as a generic description for the front line of public safety agencies.
I don't necessarily disagree with you though -- just giving you the background of where the terminology comes from.
@Matt -- [sighs] You're correct about the term. What I dislike about its use is that with the public adoption and dilution of the technical term, a great many of the general public seem to have been willing to give up responding to situations at all -- they don't even call 911, preferring to remain totally uninvolved. I don't know if the term's use is the cause, or the effect, of the detachment; I'm just infuriated at the "let 911 do it, if someone else calls" attitude.
Somehow I doubt that you're one of those, though!
I agree with Bruce but I point out that these 800 MHz systems require a central message-tagging and frequency-assignment center that is involved every time you hit the "talk" button. In the Twin Cities, I believe ours is located in Shoreview. If this center fails or is out of service, the individual radios are now bricks, since they can't talk to each other without the central site switching and tagging their transmissions.
There is supposed to be a backup for this center, but given the way we've funded infrastructure in this state lately it probably doesn't exist. They're relying on the central site to never fail. [groan]
This particular problem is now taught as a class example in the Defense Acquisition University systems engineering courses. The entry condition is many independent comm systems. The goal is a fully interoperable system that works at local, state and Federal level. An on scene commander can talk to her people and to the President, if need be. The class walks through the phases of developing systems to go from one state to the next.
The general solution would involve each person having a component of an RF version of a LAN/WAN. It needs to be a terminal equipment for that user, a router/hub to communicate with others in the near area, and a bridge between higher level nets and on-site or to other departments on other channels. IT needs to be capable of being a relay, so that responders in RF isolated areas (deep basements) can still hear the general comms outside. It might need video and digital text capability. It almost certainly needs multiple voice channels to send and receive.
So how do you make hand portable 3G cell sites with long battery life, high ruggeddness, and incredibly easy to use? And cheap.
On First Responders:
(Edit: just noted someone has already spoken about hams, but these people are about the only highly skilled people called "amateurs!")
I've heard a lot about this problem of communication. Licensed amateur
radio operators, or hams, also play an important role when direct
communication is not possible because they are highly skilled and
trained in helping relay messages in emergency situations.
Recently there was a successful attempt to have the requirement of
Morse code removed from amateur radio licensing. This is a good thing,
and will probably get a lot more people licensed in ham radio. When
needed, the hams are listening in and assisting. If I recall correctly
they played a part in communications with 9/11. They certainly do in
situations when there's advanced warning like damaging storms and
It would be a good idea I think when bureaucracies move slowly to help
promote this highly satisfying hobby which can save a lot of lives.
It's interesting - kids can and have talked to astronauts! - you learn
about electronics and electrical safety, it helps promote interest in
science and technology, you learn how to assist when you hear a mayday
or pan pan, and lots more.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.