Craig Mundie: Phuket International ICT Conference

Remarks by Craig Mundie, Senior Vice President and Chief Technical Officer, Advanced Strategies and Policy, Microsoft Corporation
“Technology Transforming Disaster Response”
Phuket International ICT Conference
Phuket, Thailand
February 18, 2005

CRAIG MUNDIE: Good morning, everyone. Your Excellency, thank you for inviting us here this morning. It’s a pleasure to be back in Thailand, and I’m looking forward to my first visit, albeit brief, here in Phuket.

As we’ve looked at the question about the role of information technology and disaster response, I think there are a number of important things to realize as we talk about this. The role of technology continues to increase largely because the cost of technology continues to decline and the availability continues to go up. But it’s not just the case that tsunami-style disasters are impacted here or are important here but also the realization there are many types of events that create this type of disastrous result. Some of these are induced by people, being an example, some are a variety of natural disasters like earthquakes and other things, and some are medical disasters like AIDS and perhaps potentially the concern around things like the avian flu. Some are weather-induced. No matter what the causes of these disasters are, there are a great many issues that arise.

Microsoft has been involved increasingly through the efforts of its employees in addressing the opportunity to employ information technology in a number of these disasters. For example, it’s well known that in Sub Saharan Africa last year the AIDS virus caused the loss of life of 2.2 million people and that 25 million people live with this infection. Disease, drought, malnutrition, poor healthcare, poverty, all of these things are a disaster of a proportion that dramatically exceeds even that which Southeast Asia has seen in the recent tsunami. Many people die every day of diseases because they are poorly informed about the process of taking care of themselves or how to get clean water, for example.

In addition to national disasters, the developed world faces the challenge, for example, of global climate change. Last year in Europe there were a number of people who died (in fact, 35,000) simply because there was a 5-degree increase in the average temperature for a number of weeks, and people were not really prepared to deal with that.

In Kosovo, in 1999, it became clear that we had a disaster in the making in terms of the dislocation of people by the war there. Microsoft was involved with many other companies and it was one of the first times that you saw a novel employment of information technology. About 100 people who worked for Microsoft worked to develop a system that was a refugee registration system. There were actually half a million people in two centers who were disrupted by the war activities, and the ability to use new technology to issue identification cards and to keep track of the locations of these people turned out to be incredibly important. In fact, the UN high commissioner, who was involved at the time, said that the use of information technology had been a breakthrough for them in providing assistance to disadvantaged people in remote corners of the world and often in difficult situations. The Kosovo efforts led the IT community broadly to move to develop a thing that we now call FACT, which is the Food and Commodity Tracking system. And under the Mercy Corps leadership these agencies today, many agencies today, use this technology to track and distribute donations of food, medicine and other relief supplies.

Microsoft was also involved recently in developing a tool called Save the Children. What was done there was to distribute a number of these new technologies like these Pocket PC phones. This becomes an incredibly powerful tool in the hands of relief workers. It has the ability to communicate by voice and data, it has a built-in camera capability, and within a year you’ll see these devices not only able to communicate over the wide area cellular networks but even over wireless networks that would be installed locally. These kinds of technologies are actually quite inexpensive compared to the kind of things we used to have to do to provide first responders any kind of communication capability. From being able to put these in people’s hands, have them operate, in many cases, without local power and with emergency communications connectivity, these things have the potential to be transformational in the way that people address these problems in the field.

I also want to talk a little bit, and share with you some of the, I’ll say, lesser-known experiences that came about in the disaster recovery in the United States that followed the terrorist attack on Sept. 11. Clearly, there were many lost lives and that’s what mostly gets reported in the television shows on a global basis. Perhaps less well understood was that we also lost one of the major telephone switches in the New York area. It happened to be in the basement of the World Trade Center. When people designed telecommunications systems in the past they knew they can never design these systems to handle peak communications. In the United States the telephone engineers used to call this the Mother’s Day problem. The telephone networks would always come up short on capacity on the day that everybody honored mothers because everybody wanted to call Mom and it wasn’t possible to actually design economically a system that would allow every person to talk to their mother at the same time. We always depend on the statistical relief. But this also had other effects, and many people underestimated what the recovery period would be to get the economy back on track.

In fact the New York Stock Exchange was near the World Trade Center, but it was not physically affected by the attack. However, the New York Stock Exchange was not able to come back into service for over one week. And the whole reason was because the telecommunications system in the United States, while it lost only one switch in one building in one city, was so damaged in terms of its overall capacity that we were unable to provide enough data communications for the New York Stock Exchange to begin to coordinate trading with all of the people in the United States and around the world.

Another thing that was quite interesting from the point of view of command and control, the role of the government in trying to orchestrate a response to something like that terrorist attack, was that despite many efforts being put in place over the years in traditional telephone networks, when it was all said and done, the top people in the United States government found that the only reliable way to affect communication for command and control was using wireless, IP-based data devices. The messaging systems (not on their cell phones) were essentially over wireless data networks like the BlackBerry and some of the other ones that were out there.

Why was that the case? Because first, the loss of the switch had created great congestion in the United States. Everybody wanted to talk about this, figure out whether they had lost family members. Also, no engineering had really been done to provide any type of priority routing in the cell phone network, only in the wire line networks. And when the cell phone networks were immediately swamped throughout the entire eastern half of the United States, even government people found that they were unable to communicate. And so, even where there was the ability to coordinate response it was very difficult to affect.

The other thing that was interesting is that with this disruption the Internet, which many people may remember had its source in the Defense Department in the United States more than 30 years ago, it was designed to be able to withstand nuclear attacks and to automatically route around perhaps the loss of even entire cities. And so while that is something that’s lost as people just think today about surfing the World Wide Web or doing e-mail, it is these properties of these communication networks that turn out to be critically important in thinking about how you deal with large-scale disasters.

But these things actually are disasters on a more local level. Another thing that was little known is that the traditional radio devices, the walkie-talkie-type radios that were in the hands of all the first responders in New York City, those radios also failed. The structural impediment of being inside a collapsed structure the size of the World Trade Center turned out to be disruptive to the radio communication. And so the policemen and firemen who were risking their own lives to go in and evacuate people in those buildings found that they couldn’t talk to their commanders in the trucks and cars outside the World Trade Center. And many of their lives were ultimately lost after the collapse of the building or in the collapse of the building simply because they couldn’t talk either.

So as you begin to look at the role that individual people play and the role of cell phones, for example, there are some other fascinating stories. On the ground the cell phones turned out to be one way where family members and even first responders were able to identify the location of people who were trapped in the collapsed building. Other examples of this have been shown in other disaster relief environments.

Perhaps also notable was the role that cell phones played in the air even though it was not a design feature of cell phones to work from airplanes. The fourth plane that was hijacked is generally believed to have been targeted at either the White House or the Congress in the United States. That was its mission, to crash into those buildings. But that fourth plane never made it to , D.C. In fact it was stopped by the actions of the passengers on the plane who had been able to make a cell phone call despite the terrorist being in control of the aircraft, were told by their family members on the ground that they were in a plane that was likely to be used as a weapon and they decided to take action, take the matter into their own hands, and they stormed the cockpit and the terrorist and ultimately caused the crash of the plane and the loss of all their own lives. But it was in a field in the middle of instead of in the middle of , D.C. That would never have happened if people, individual people, did not have in their hands the ability to communicate with almost anybody in the world on an instantaneous basis.

So these technologies become more important not just in the hands of the government, not just in the hands of the professionals who deal with first response, but in the hands of the people who are involved and the families of those people.

In the 9/11 aftermath, Microsoft and a number of other people developed a family registration tool. One of the greatest causes of suffering was the unknown status of people and all their family members and how they could communicate. So we developed a database where people could go on the World Wide Web and register that the family had concern because they couldn’t find somebody, and the people who were taking care of the injured were able to register photographs of the people that they had removed from the World Trade Center. By using the Internet to match these things they were able to dramatically accelerate the identification of many of the people who were missing or lost.

One additional result of 9/11 was it forced a reassessment in the United States at the highest levels of government as to what the implications and planning needed to be in the development of a new network in order to deal with these realities. During the administration I was appointed to a permanent commission that had long been in place in the United States to advise the president on national security telecommunications. The focus of this is primarily around emergency preparedness, emergency management communications and other things that affect the national security. Recently that commission established a task force that is looking at what we called the next-generation network question. I think that this is something that each country is going to have to come to address.

What we recognize today is that the old model, you can say the Cold War model, of how we dealt with emergency communications (the idea that there was a designated telephone, the red phone, that you could pick up and affect some kind of emergency communication) is really no longer sufficient. It’s as important for computers to communicate with other computers and throughout the hierarchy of public and private communication as it is for the traditional command and control networks that we know in our military and governmental applications in the past. All of these things are going to require some significant changes.

Today, because this conference was developed on short notice, my boss, Bill Gates, was unable to join us in person. But he’s been personally quite involved both at Microsoft and through his foundation work.

[Note: Break in tape resulted in loss of a few seconds of transcript.]

Video Remarks by Bill Gates, chairman and chief software architect, Microsoft

Good afternoon. I would like to thank Prime Minister Thaksin, Minister of ICT Dr. Surapong and the organizers of today’s event for inviting me to speak. On behalf of everyone at Microsoft I would first like to say a special word of thanks to the community of Phuket, which took remarkable care of our employees from all over the world who found themselves in harm’s way on Dec. 26. Their kindness and heroic generosity were a dramatic demonstration of the Thai people’s spirit that is truly jai dii.

In thinking about the role of information technology in events such as this, it’s difficult to know where to begin. Let me offer a few thoughts. Years ago the world would have been left watching helplessly as this tragedy unfolded. But I’m encouraged by the way that companies and individuals everywhere responded through the Internet to channel help immediately to those most in need. Donations poured in, thanks to efforts of companies like Yahoo!, Amazon, Google and Microsoft. The Internet also became a critical communication vehicle as people attempted to connect with their families and friends. It enabled government agencies to begin collaboration in hours instead of days. It matched up volunteers to real-time needs in Sri Lanka, and it facilitated new mapping of the ravaged areas as old maps were suddenly obsolete.

One key lesson is that technology needs to be a core element of any national disaster plan. And we’re responding to numerous requests to develop solutions that will improve relief efforts. The dialogue you’re starting today is a welcome contribution as we all begin to think through the most effective ways to mobilize assistance as quickly as possible.

In terms of prevention, we will see the development of additional monitoring and early warning systems beyond just the Pacific. Here technology also plays a vital role. Beyond the collection and analysis of data, IT is critical in the coordination of emergency warnings and information which can flow immediately to government agencies and to the public. Safeguards need to be implemented to ensure that new systems can’t be compromised but building the right solutions can literally save lives.

While I’m sorry my schedule didn’t allow me to be with you in person, I look forward to hearing the results of this meeting. I’m sure Microsoft can help work with you to rebuild communities and individual livelihoods. It’s clear that we need to use every tool available to relieve suffering and prevent future losses. IT, when used wisely, is one tool I know can make a difference.

I wish you a productive discussion today, and thank you.

Continued Remarks of Craig Mundie, CTO, Microsoft

As Bill said, technology is a tool in disaster response. Hopefully in the future it will become even more of a tool in allowing us to predict what kind of problems might come from any of these medical, man-made or natural disasters. One of the things that’s clear is that many of the aspects of our daily life, communication, inventory management, identity management, warning systems, all of these things are essentially becoming digital, becoming interconnected and to a great degree, through the Internet’s capability, becoming global. There is a revolution under way within the IT community, with this general notion that the industry has now chosen to call XML-based Web services. This new architecture allows the design and deployment of applications with much less advanced knowledge and understanding of how they will have to interact in the future. The ability to have meta data that can be expressed in a computer lingua franca that allows people to continue to build an increasing array of applications and services, to do process automation, to develop new communications and search capability, all of these things represent an opportunity to transform the way in which we address these kind of problems. The technology is actually in its infancy for disaster mitigation. I mentioned one example, cell phones and their proliferation, and how that has changed disaster response. The ability to use SMS as a notification mechanism has already been demonstrated. The ability to use cell phones as a way to do geo-location on people has also been well established now. I think an increasingly important thing will be the emergence of new wireless radio technologies going beyond the type of things that we use in the traditional wide-area cellular communications, and frankly, even beyond the traditional notions of wireless broadcasting and things like television systems. The ability to deploy both Wi-Fi systems and ultimately, new radio technologies on a rapid basis without any advance knowledge of the geography or any central control will become increasingly important. The ability to blend that with other things like the GPRS or other data capabilities and the more recent cellular networks and the ability, for example, to deploy new things like WiMAX to be able to build a wireless, high-speed, wide-area Internet protocol-based communication system, these things all represent tremendous potential as we move forward in this area. And this Web services architecture, I think, gives us an architectural commonality that can be deployed to our great benefit in bringing together the applications because it’s very hard to predict exactly what services, in fact, exactly what organizations, will have to be brought together to affect any change. The ability to have machine-to-machine communication, not just people-to-machine or people-to-people, will be an increasingly important part in ensuring that no matter where the disaster occurs, no matter what language people speak, that it may be possible to interpose computer systems and their software to effect basic elements of the disaster recovery plan. They will all tend to lead to a more expeditious capability.

So what are the things that have already changed and made this possible? Cell phones are an obvious example along with this new Web services architecture. But there are even changes in things that we don’t usually think about, such as the opportunity to get electricity.

Today the historical technologies like solar panels and advanced battery techniques have resulted now in very small, miniaturized solar cell capabilities. Today people who go camping and who are avid users of digital photography, they have to recharge their batteries just to take pictures in the wilderness, and there’s no generators, there’s no place to plug in. So how do you do that? You unroll a little mat that essentially becomes a charger for all the batteries in your digital equipment. The ability to have this very inexpensive, personal, portable power system is something that, I think, will find greater application in these disaster relief environments because frequently, even if there was infrastructure to support communication and power, it’s disrupted in a large way by these natural disasters or military actions. So the ability to bring digital technology forward with its dependency on electricity requires us to bring some other technologies in to play. Connectivity is becoming cheaper and more powerful, we’ve got more reliable satellite communications, we’ve got new data capabilities in the cell phone networks, we have the ability to take both Wi-Fi and cell towers, mount them on the back of a truck and drive them into an area and turn them on in a matter of minutes. Many of these things, however, require a tremendous step forward in the software systems in order to support the provisioning of these networks. You can’t just turn them on and have them work. You have to be able to figure out how cell phones are made to talk to them, what their authorizations are, and perhaps, increasingly important in the future, how priority will be established in order to make sure that mission-critical communications continues to take place. And, of course, we’re getting cheaper and more powerful devices. The ability, as I said, to take these tiny computer systems that fit in your pocket and be able to take them out and recognize that they are, in many ways, a more capable computer system than the desktop machine of as little as six or seven years ago means that the power of these tools is incredibly great and the ability to combining, in a single device communication, video, voice, the ability to do Internet access, means that no matter what your role in the field is, the device can adapt to your requirement. So if you were to provision these things and make them available to first responders, to police, to fire, to medical workers to volunteer workers you could, in fact, ensure a level of communication an the ability to access information and plans that all could be prepared ahead of time and pushed out to them even though you didn’t know who was going to be important on the day the disaster actually occurred.

So what are some of the technologies that are on the horizon at Microsoft and other companies working on that I personally think could affect outcomes in the future? One of the things that I think is most promising is the ad hoc, mesh, wireless network. What is basically happening is that there will be a convergence in the way we physically manufacture radios. And, in fact, the same processes that we use to make microprocessors will be used to fabricate the capabilities for radio communication. The net affect of that is that it will become common place to think that any place that you can put a microprocessor and put some software you will also have the ability to have a built-in radio. The radio, in fact, in the future could be a software-defined radio, which means that, within reasonable bounds, it could migrate its ability to talk from one network to another. So if it found a Wi-Fi network it could configure itself to be a Wi-Fi radio. If there was no Wi-Fi, it could look for a GPRS capability on a cell tower. The ability to do this allows us to continue to lower the cost and power requirements to these devices and to make them able to adapt to novel communications facilities as they go forward. This adaptability through software and sophisticated hardware techniques, I think, will become increasingly important. Why? If we’re going to use this technology we’re going to have to figure out how we’re going to stockpile it and put it out there so that when we need it we can put it in people’s hands. That obviously requires a financial investment and one that has to be preserved over some period of time. So the ability to use software techniques to make these devices more malleable than have historically been the case with specialty devices to support medical activities or communications activities, I think, could dramatically improve outcomes and lower the overall cost. The mesh networks are particularly interesting in a world where there are pervasive radios because we have the ability to presume that there is no center to the network. These are organically formed radio communication systems that pass packets in a helpful way from one little radio to the next. So even though the radios are tiny and their ability to communicate may be over relatively modest distances, the ability to do that at large scale over hundreds or thousands or maybe even tens of thousands of devices, and to bring them together instantly into a network, is very important.

These technologies are already well under way in development and there are really two things that are driving it forward. Military organizations worldwide always have the equivalent of a disaster recovery problem in communications. When they essentially get deployed into a war zone, they can presume nothing about the existence of communications on the ground. And so increasingly they recognize that since that’s critical to their success they need an ability to essentially airdrop a lot of little radios around, put one on every car, every person, and let them all talk to one another. Perhaps, even more broadly though, I think this technology will play a role in providing low-cost communication in the villages of the emerging markets. In that environment it’s noneconomic for traditional carriers to build a wired or even wireless infrastructure. But perhaps the ability to have these things acquired by people as they just buy the things that they live with every day will allow the organic formation of low-cost radio networks. So I think that some of these things are quite interesting and hold great promise, and the challenge now is to adapt them to these disaster-recovery environments.

Other things that relate to identity, I think, are emerging quite quickly. We all know about identity cards and issuing smart cards and government ID cards. But in a disaster it’s very frequently the case that even if they had one it wasn’t on their person (very likely the case on the beaches in Phuket) and the question is, when the people arrive and you start to deal with the injured and the dead, how can you identify them? And how can you get other people to participate in that process? The ability to have RFID tags is something that could be deployed not dealing so much with the people but the logistics and the areas where problems exist. I think this has a strong potential.

And there are new technologies related to software and facial feature recognition that, I think, have applicability in the future here as well. Microsoft has developed a technology that we called Face Certs, which is the ability to use a digital camera and a personal computer and to issue a non-forgeable identity certificate that can be used both electronically and as a face-based recognition system. That could be very important. When you’re dealing with, for example, a refugee environment where you have a half a million people that are starving and you want to make sure that people get the food they’re entitled to and don’t get other people’s food, getting some type of non-spoofable, reliable, cheap, on-the-ground identity system is a very important capability. And these techniques are now emerging. And all we need to use is standard inkjet printers, a standard desktop computer and some of these new software techniques. Similarly this face-recognition software could be very valuable coupled with the World Wide Web, very small, local digital photography capability, the ability not just to depend on forensic specialists to do the identification work but to get families involved, to get, in fact, software involved to do matching between digital images that happen in the disaster site and what can be posted on the Web by people who think that their relatives have been lost. All of these represent interesting opportunities.

[Note. Break in tape resulted in the loss of several seconds of transcript.]

AOL and other companies are operating global networks that provide instant messaging using text, voice and video. And one of the things that’s intrinsic to the use of these things that many of you are probably familiar with is the buddy list, which is the ability to say your buddy just got online. The ability to use this kind of thing to promote a human network of communication in a disaster environment, the cascading effect, the word-of-mouth effect, being able to be operated through these global networks and instant messaging systems could be a powerful tool, and we need to find a way to harness that.

Clearly, medical issues are always a problem, and we now have an increasing ability to bring the world’s medical establishment to bear on some of these problem. And while that may be equally important to do that in some of the epidemiological problems or, for example, bio-terror problems that could emerge in the future, it’s also obvious that we could use a lot more help in dealing with the short-term emergency medical capability in a disaster area like here in the Southeast Asia area with the tsunami.

I think one of the things that also is interesting is that we’re now developing — not Microsoft but many people in the medical community — software that’s incredibly powerful at doing diagnosis. Many times in a disaster environment you just can’t possibly have enough trained medical people, and yet you have a lot of volunteers. The question is, how do you put the best tool in their hands to help the most people? And having tiny pocket computers that actually allow them to just click or check on a form the kind of symptoms they can observe and get someone to make an emergency diagnosis, not a person but a piece of software, that also looks to be available in the relatively near future and be quite powerful.

And finally the use of the meta data captured in XML coupled to very powerful search mechanisms allows us to do many things that we might not have anticipated in terms of finding and using information.

So, going forward, what do I think really has to happen? And how is that different than some of the things that have been involved in the past? First, I think there is a requirement for more of a public-private partnership not only in developing the technology but in preparing the plans for response. Microsoft and its people, who in many cases want to come forward and help in some of these situations, often find that it’s difficult, in fact, that their ability to help is often times delayed.

Why? Because there was never actually a plan put in place with the local authorities to say, “How would you deploy these specialty IT people and their knowledge and their tools?” And because there’s no planning that goes on that contemplates the important role of the large corporations in a particular country or city in response to these things it’s always historically been thought to be the role of the government to provide the emergency medical capability or the defense capability. But I think it’s increasingly true in every country I visit that people recognize that to be safe, to be secure and to deal with these things cannot be done strictly with the government acting on behalf of the citizens. You’re going to essentially have to have more of a partnership. And I would encourage a great deal of thought as to how that should be developed. I think we need to develop a problem taxonomy. One of my early slides shows that there is a broad array of disaster types. We have a common technology base but you can’t assume that you have exactly one tool for all disasters. So I think that we need to do more work to develop a taxonomy, and for each class of problem have a well established, perhaps, computer-automated checklist that you push the button and the thing starts to tick off what the actions are that somebody should take. Many times you have to anticipate that the people who are the experts might have been lost in the disaster and therefore you have to find some way to preserve that knowledge. Otherwise the situation only gets worse.

As I’ve highlighted in my own comments, we have a broad array of tools to bring to bear on these problems. But many of these things have not historically been deployed in the worldwide community for disaster recovery. So I think we need to inventory the technical solutions. We need to be sure about the ones that work today, and we need to be contemplative about these advancing technologies and what the right way is to introduce them into this environment.

And finally, I think we need to do trials. We need to do them at scale, and we need to do drills. And the drills have to employ the population, the private sector and the public sector. And as far as I know this kind of thing really just doesn’t happen in almost any country in the world. But as anybody knows, whether you’re training a football team or a military team, if you don’t drill them you can’t be sure that they’ll actually play the game right when the big day comes. But I think, given the scale of the disasters we know we face, whether through terrorism, medical problems, war or, in fact, natural disasters, the scale is going to large, it could be global, we’re going to have a lot more linkage between our countries in these problems through the transport of these problems through air travel. And so I think there’s going to have to be a lot more. And clearly governments are going to have to figure out how they’re going to partner with each other. But within every country, within every city, I think we’re going to have to do more to develop the trials and the drills.

There are a number of examples where people are becoming an important part by their own initiative in this. Here, the Southeast Asian Earthquake Weblog and other blogs. These things have only really existed in a popular sense on the Internet for less than two years. And yet this became one of the most highly used tools on the Internet as it related to recent tsunami efforts. It was established informally by a number of citizens on the Internet who used standard tools to build it. Last month there were 2 million people who went to that blog to either contribute information or to get information about what was happening to people in Southeast Asia.

There are some fascinating examples where, I think personal initiative shows the power of some of these techniques. I was told a story about some people in Sri Lanka who operate the cell phone system and they realized that one of the things that they always keep track of for billing purposes is who’s on a roaming cell phone and who’s on a local cell phone. They said, Wow, because we know that, we’ve got all these people visiting, and nobody knows who they are or whether they’re here, whether they were hurt, whether they have a problem. They’re not locals. So they decided to actually tell their cell phone system to just send an emergency SMS message just to the roaming cell phones under the presumption that that might be a class of people who would require special help. They sent out, I think it was 5,000 — not only 5,000 messages to roaming cell phones, and they got 3,000 messages back from people who said, Yes, I need help. Many of them didn’t even know where they were. They were disoriented, but their cell phone was there, and it worked, and they responded to it, and they used triangulation on the cell phones themselves to send emergency responders to those people. That wasn’t something that was planned ahead of time, it wasn’t something that was part of the emergency response plan, but it was the clever work of very smart technology people who recognized that these things can be used in many ways. And we need to encourage that kind of participation to deal with these problems.

So I think this is just one example of many that obviously occurred here and in many other disasters around the world that I think bode well for using these things to improve the situation.

So finally let me just emphasize that I think one of the things that seems to be at the heart of using technology in all of these things is getting communications right. If you can’t talk, you don’t have command and control, you don’t have access to this information. And so I think there’s a series of things that have to be re-examined. What is the role of emergency broadcasting? What is the new way to think about emergency broadcasting? I don’t think it’s just going to be television networks anymore or even traditional radio stations. They clearly have a role to play, but frankly they’re going to be displaced by some of these other more targeted capabilities. I think one of the things that every country should look at is adding priority routing for cell networks so the government officials and first responders can, in fact, use cell networks as a reliable command and control environment. But most countries have not implemented priority routing for that. And that requires not only that you have the mechanism but you have to identify who gets to use it and how they get identified, and that gets you back to the identity problem. I think as we’re starting to do in the United States, every government should probably form a task force — reporting perhaps even to the prime minister — that thinks about this question: What is going to replace their red phones, and what is going to be their next-generation, emergency preparedness and command and control network? I think that this will be a critical capability, it’s the only thing that will truly allow both redundancy and the deployment of all these technologies, and I think that it’s going to be a critical thing to do.

The last thing I think in getting it right is essentially going to be getting wireless communications right. Not just the traditional cellular wireless communication or broadcasting, but, in fact, the ability to make spectrum available in order to allow these new technologies to emerge. To allow the grassroots to participate, whether they’re the citizens themselves or whether it’s the people who are professional first responders. We’re going to need much more aggressive adoption and deployment of these things. The radios have to be stockpiled. It may be just as important in the future when you send that first wave of trucks out that has blankets and food and medicine, it maybe should have a whole bunch of wireless radios so that when you just drop them off the back of a truck every time you drop a blanket you drop a radio. And that may, in fact, do as much to enable people to communicate and manage the disaster as anything. But today, I would say that no country in the world has really opted to correct policies relative to spectrum and radio technologies in order to allow that kind of broad thing to happen. So I share those thought with you this morning.

I think we have just a couple of minutes if anybody, perhaps the ministers, has a specific question. I have to apologize. I’m traveling in Southeast Asia today with Dr. Kissinger. We’re meeting this afternoon back in Bangkok with a number of people in the Thai government, and so I won’t be able to stay for the rest of the conference. But I’ve enjoyed my time here and the dinner last night and the chance to talk with many of you privately. I thank you for organizing this conference and hopefully getting my colleagues in the IT business to come together with you and develop more plans.

Thank you very much.