U.S. Strategic Command

 

Speeches

AFEI Conference

By General James Cartwright | Washington, D.C. | February 21, 2007

Topic: STRATCOM and the Information Environment

General Cartwright: This idea of net centric and collaborative environments, it is a manifestation of culture. It is tools to some people. It oftentimes becomes tablets that came down off the mountain to others. So how do you bring this into warfighting? How do you make it a part of the way you do business and how do you allow it as a nation to become a venue for advantage and competition and the quality of life that all of us would like to continue to improve.

That's really the challenge that's out there on a global basis. You can't escape this, whether it's on the street, the idea of a flattening world or a number of other people that have made a lot of money telling us what's happening to us. But it's hard to escape, whether you're doing it with some of the capabilities that are provided by our space systems in addition to navigation and timing that actually affect how you do things and affect how you get to work and any number of things that are so invasive and so pervasive that we all just assume them away.

We were having this conversation at home the other day. And as always the technology centers around the VCR and the delayed programming VCR. My wife and I were working our way through that challenge and commented on the fact that we have a three year old grandson that lives here in the Washington area, and a daughter that lives in the UK. Our grandson has no problem pushing the right buttons at night to say good night to both families, any place we are in a VTC or a computer. We're still struggling with a VCR and he's doing VTC between Washington, Omaha and the UK every night just to say good night to grandpa and grandma. It is a different world that we're in, and there are some really incredible capabilities that are out there amongst us, enabling what we do in a very positive way. I can't imagine a retrenchment that would move back from that. This is with us. It's going to stay with us.

When you look at that world you have the glass half full, glass half empty. There is the threat side of the equation which is what at STRATCOM we get up every day worrying about. That threat doesn't necessarily have to be cyber in nature but it is certainly enabled by the access to information which is relatively effective which doesn't have any geographic boundaries. But freedom with which knowledge flows and information flows around this globe is a challenge. We also have the capabilities today on the negative side that are starting to proliferate out there. [Inaudible], weapons of mass destruction, and the ability to try to figure out how to build and use and deploy is becoming pervasive and it's a problem for all of us.

It's a huge problem on the kinetic side when you consider that almost any country can buy a ballistic missile, short to medium range, that can reach any of their neighbors in just a minute or two, that eventually they're going to be able to deploy on top of those things that have regret factors that are so significant that heretofore only large nation states willing to invest billions of dollars over many years could develop and they're going to be able to [inaudible]. That is a very difficult world enabled by these wonderful [inaudible] systems that we have.

So how do we start to organize ourselves to operate in minutes, not days, weeks or months? Change our capabilities at the speed of Moore's law, not the industrial complex?

Today we can talk about minutes for ballistic missile with huge regret factors, but if you're industry or you're the government and you live on the opposite side of the earth from where we are today, you go from that point out 23,000 miles in geosynchronous orbit and back to another place on the face of the earth, it's microseconds. With that we can steal intellectual capital, competitive advantage, the ability to govern ourselves based on property and borders of nation states. We can do that in microseconds. In the blink of an eye what it took you years to build up, intellectual capital, competitive edge, whether it's in business or as a nation, it can be [inaudible] to any place on the face of the earth in microseconds. That's the world that we live in. That's the decision cycle that we're going to deal with in the future.

They don't tolerate the Napoleon structures that we're used to where we come up this long chain of command that we describe, go back down, then up to the next headquarters, and by the end of that whole activity we've washed out all the ambiguity and we're totally irrelevant because the problem is over with. An adversary that would seek to disadvantage us in microseconds, use ballistic missiles for a few minutes to hold us off and then it's all over.

It's interesting that we can fly at twice the speed of sound. It's still 20 hours to the other side of the earth.

What are the skills, what are the capabilities that we need to start to bring to this environment? That's the challenge that we're trying to understand. That's the challenge that we're trying to address, and we're trying to do that in a timeline that is ahead of the problem, not behind it, and I can't tell you that we're always winning that battle. But that's the objective, that's the threat that we face out there that is coming with this information age.

Much of that threat is controlled by culture. We build as a culture power centers either by controlling resource or by controlling knowledge in the industry, in government, et cetera. We organize ourselves in such ways to ensure that those centers of power don't change because it gives us stability and we like stability, we like to know how things are going to work and who's going to do what. And the information age is terribly intolerant of those structures. It really pays no attention to them and it allows them to change at a speed that is very uncomfortable [culturally]. It's very uncomfortable to those who would seek to get advantage and then hold that advantage. It breaks us down.

The military, General John Jumper once had a quip that he used that I thought was very appropriate, that he would go out and visit forces in the Air Force as a Chief would do, and he would find what he would call tribes. These were organizations that were thrown together in order to solve some problem. My term is you take 10, 20, 30 people from a hot, sweaty pile, put them in a room, lock them up, feed pizza to them until they solve the problem and then you let them out. For about the first six months you get 90 percent output for 10 percent of the management [inaudible]. And in about six months it goes the other way. Ninety percent of the activity is making sure the organization stays together and 10 percent is output. And you can't ever make it go away because they spend most of their effort protecting themselves, making sure that the hammer they invented, everything is a nail and that hammer will solve all of the problems of the world. You can go look at almost any sector of the government and find that problem. And it's good and it's bad.

[Inaudible] say the [inaudible] an airplane. Cartwright grew up in the Rivet Joint and the Rivet Joint will solve all problems. That's the basis of my knowledge, it's what I understand. So when I'm put in the staff and asked to solve problems I do it with Rivet Joints. How do you start to broaden that construct out?

Part of the activity that's going on within the department is to start to understand, and we use the term stovepipe pretty loosely, and we make it very pejorative. The reality is vertical integration is a hugely good thing. It has advantaged this country for a lot of years. It is a great way to take a common activity, organize it in the most efficient, lean, quick, able to adapt. The difficult part is how do you make these stovepipes, or let's call them vertical cylinders of excellence instead of stovepipes so it's not so pejorative. [Laughter]. But how do you go cross corridor? How do you make two of them work together to get synergy? And when does it make sense to do that? Because you don't need to do that for everything. So what are the rules that tell you when it's appropriate to move cross corridor, so to speak, and what are the rules that say no, don't do that because you're really going to have greater regret in disaggregating or breaking down vertical integration. That's the piece that we're trying to understand today, and where are the synergies, where are the leverages to do that?

So you have really right now three what we're calling, some people call them global commands, some people call them functional commands, but when you look at the United States Special Operations Command, when you look at TRANSCOM and when you look at STRATCOM you're looking at three commanders and three commands who have global responsibilities to provide services for regional combatant commanders, functional services that they should be able to do, but are all over the globe all of the time. And whether it is transportation or special operations or integrated missile defense of cyber operations or ISR, all of those tend to be enablers for every regional combatant commander.

So the construct that we have tried to put together at STRATCOM where we tend to have, as the speaker said, five or six of these types of commodities so to speak, and functions, is how do you pull this together? How do you make it work on a global basis? How do you integrate it? How do you advantage your regional combatant commanders in a way that they would not have been advantaged otherwise?

We have put together at STRATCOM these joint functional component commands. The more letters I guess is a reflection of the contest that we had to name [inaudible]. [Laughter]. But componency has been around for a long time and functional componency has been around for a long time. That's really not a new invention. The thought process here is one, STRATCOM could have and may have paraded down a path to bring all these capabilities into Omaha, be the center of excellence for intelligence, surveillance and reconnaissance, integrated missile defense, information operations, the whole panoply of missions, and we would have grown to some behemoth size, violated almost every industrial or business process that you could have learned in college, and we would have gone on that path probably if we had not made a conscious decision to step away.

These components are far different, though, than probably most components that you'd find in a regional combatant commander. I could pick almost any one of them. But the construct here is that every one of these components is joint by nature. In other words we took a center of excellence and ensured that it represented a joint warfighter. Second, it represents the allied contribution. It has to. Third, it is proved in some way in business and academia. And for me that was important as setting up as a first principle because my sense, my reading of history is that we've started to move away both from the American public in defense and from the intellectual capital that's out there. When we go back and study say World War I and World War II, the integration of industry and defense, they were inseparable.

Industry has incredible intellectual capital to offer, incredible insight, incredible opportunity that if we're not tied together -- infrastructure is not part of the equation, we're going to miss the boat. So these functional components have all of those elements, one.

Two, these functional components have capabilities and authorities that no combatant commander will ever have. I don't have it. They have more power and more authorities than a headquarters. That's a reversal. That's very different than the way we've done business. They all have acquisition authority. STRATCOM does not. They all have requirements authority. I don't have any of those authorities. They all have resource authorities, don't have those. Then they have their unique authorities, whether they're in the intelligence world or whether they're in the missile defense world.

So the intent here was not to pull all the power up, it was to push it back down where it rightly needs to be and where the decision cycles can start to match the problem.

That's the organizational construct. That's the idea. The thought process behind this was that these are centers of excellence already in existence, hard to compare their value to anybody else that you would rebuild in the name of say ISR. For me, the DIA and the Department of Defense is clearly the center point for intelligence, surveillance and reconnaissance and therefore should be the commander and should be that [inaudible]. You don't need to reinvent that. For information operations you're not going to find in this country anything better than what you're going to find in the National Security Agency. So trying to rebuild that or have a clone is a tremendous waste of manpower and resources. It's not that we don't want to be competitive internally, but to go out and try to generate these capabilities and the manpower associated with them, the intellectual capital is really beyond the limits of our resources.

We put way too many people in headquarters as it is already. We need to get them down to the action level and we need to slim down the headquarters and get the decision cycles back in line with the operational realities we have to deal with.

So the ideal here is take these centers of excellence with all of the authorities that are inherent in those centers of excellence, put them into a construct where they can be collaborative, they can support regional combatant commanders, be global in their vision, understand what's going on globally, and move their resources to the problem, abrogate when necessary, and then when it's not take the excellence they provide and use that in a regional combatant command.

So if, let's just pick PACOM, what we used to think of a problem associated with missile defense, that they'd like to have some missile defense, they'll go to the components that offers that capability and create a missile defense. If when they get there they find out the real problem they had was missile defense but it needs to be enabled by ISR, it's already integrated. It's already moved to the problem very quickly and allow that cross-border, the integration of ISR and integrated missile defense. That's the whole point here. And we do it for the tailoring necessary for that regional combatant commander. So if they believe the problem in PACOM is an integrated missile defense problem and an ISR problem, they've got capabilities, we've already wired it together, the command relationships are there, it's not five months of staffing through the wonderful Joint Staff and OSD and move at the speed of glaciers. It's there. It's ready. It's available when the commander needs it.

Don't pick on the Joint Staff, it's not nice. [Laughter].

So that's the construct. It's got to be able to take these disaggregated centers of excellence, aggregate them on the fly when they need to be, tailored for the problem globally for all the regional combatant commanders.

So the first question I get is why not just have STRATCOM do that? Why not have STRATCOM solve that problem in the region? The experts are in the region. The experts are the people who know what they need and who know what will work. One size does not fit all. You've got to have that regional tailoring. You've got to have contextual information to know what will and what will not work. So having somebody from Washington or Omaha tell you how to do business is the wrong way to go to the problem. It's absolutely the wrong way. And we've got to be flexible enough that the mixing and matching between these vertical cylinders of excellence can be accomplished by understanding the problem better.

So ownership has got to get torn down and these things have to move at the speed of need, not the speed of staffing. And that's a major challenge.

Let me just hit three or four areas that I think are going to be challenges for us, and I have kind of avoided, and I see Charles [inaudible], so I know I've got to talk a little bit about the geek stuff and work my way through some of this stuff. [Laughter]. So I'll hit a couple of the challenges that I know will be debated and probably articulated far better than I can through the events in this conference.

But when you look at the problem of integrating either horizontally or vertically and being able to do it on a global basis and being able to do it at a scale that allows multiple transactions going on any place on the face of the earth at any given time, the organizational constructs, the information management constructs are huge challenges but are also the key enablers and they will differentiate you from your adversary or your competitor on any given day.

Right at the core foundation of all of this is data. And we have for I don't know how long had this notion at least inside of defense that if you could just format all of the data the same, then you would somehow enable all sorts of great activities. So that concept is the idea that there is no vertical integration, that everything will go horizontally, and that's just fundamentally flawed. The format for video does not do well as taking that format and exporting it to signals. It just doesn't make any sense. The technology is there today. It is not necessary to format all data the same. But what you've got to do is make that data discoverable and you've got to be able to when you discover it know what the format is so that you can take advantage and use it and let the formats grow with the technology that makes that functional let's say video better every time you cycle through. Don't kill -- One of my biggest pet peeves is those who would in the name of uniformity kill innovation. You can't do that. You just can't do that.

So it's not a question of standardizing the data format, it's a question of registering that data, making it discoverable, so that people can get to it and take advantage of it and you don't close off innovation. So that's a huge issue. How you manage that data is critical. This is where the culture piece kind of comes in for Cartwright growing up in a squadron, every week you've got the duty officer. You were the squadron duty officer for the day and you had this wonderful log and you wrote down everything that you thought was important, everything you thought the boss might want to know. Then in the morning the boss would come in and he would look at your duty log and say, tell me something about this, or tell me about that, or you idiot, why did you write that down. [Laughter]. So the decision cycle was 24 hours.

Obviously I was put there for 24 hours but nothing I was going to do was so important that we couldn't wait for the next morning. That's alive and well in almost everything we do today. Right now we get, at my level we get a power point briefing in the morning. A very interactive capability. You can talk to the slide all you want. [Laughter]. And it never talks back and never disagrees with you. Somebody obviously knows exactly what you want to know. You can see this in all sorts of media out in the regular world, so to speak. At some point I question whether or not what I need to know is on those slides. At some point I get tired of watching a rock star's head [inaudible] or the who is the daddy case. [Laughter].

How much of the broadcast is efficient and effective versus an interactive dialogue that goes on where you set the parameters, you decide what you want to know, you get it fed to you when you want it fed to you and you can engage and push back on whatever it is you want to know in real time. It's a very different thought process. It's not to exclusion, though. There is good in broadcast, there is good in having that capability, there is good in forms of communication that demand that there be someone at the other end of the lines -- chaperones, video conferences -- and there's good in things like e-mail and blogging that allow you to have a dialogue with any number of people at the time at which people want to engage and are available. There's a huge benefit in all these. It's not one or, it's all of these things. And where do they fit and how do they advantage you in the operational scheme for me? That's the key piece.

But it's all at the data level. Understanding how to discover that data.

One of the key technologies that I am pushing hard on within the department is this understanding of who are you. What are your certificates? How many cards do you have to carry? How many passports do you have to memorize to be able to have access to the data?

One of the inherent flaws that we have in the system today is that it is organized around the terminals. So everything is organized around those terminals. You build firewalls around those terminals, you do all sorts of things, and inside those terminals are whatever Cartwright wants to know or a function wants to know. But you can't get into my terminal because I've organized it such that you must ask my permission to know what it is that I've gained and can pull together. So we're back to that cultural thing of power centers. Instead of an architecture that says here's who I am and opens up the gates with one here's who I am.

We've got to find a way to figure out who we are so that we can move horizontally, because absent that all sorts of mischief can occur. And we don't have that yet. We're working towards it but we really don't have it yet.

The second level of activity that we are pushing on very hard, and let me go back to that one because there's one more piece. We have worked very hard to lay out the criteria for all data within the department by which you will register it and make it discoverable. We have exported that to our cousins in the intelligence community and our cousins in Homeland and our cousins in Justice and we are moving very quickly to put in place those standards. It doesn't drive you from what the standard has to be for your data, it just tells you how to register it so it can be discovered.

Now for some in this room that's a yeah, good idea. For others that's a horrible idea. It's really exciting to me to talk to a CEO who has two directors, one who's going this way and the other one who's crying because they don't like this idea. But we've got to start in the department, and let me make this pejorative in nature. We can no longer afford to pay the same guy three times for the same algorithm. It's just crazy. I can't make rules to make sure that that manufacturer, let's call him, can't even tell his own organization that he's doing it because of the security rules. We've got to break those walls down. We've got to build an architecture that allows us to get past that. Not because I'm after the competitive edge or the opportunity for profit, but because I can't stand the time delay. I can't stand that time delay of stepping across the line of departure or taking it to the end of the runway and lighting the afterburner and as soon as I'm airborne or as soon as I step across the line of departure, shit happens and you didn't plan for it. And I can't wait five years for the solution. I've got to have it faster. That's just absolutely critical.

The second piece is this idea of pilots that we are working very hard at STRATCOM on. But these are taking, geek speak, communities of interest. The groups of common functionality and pulling them together and saying okay, in this case, and let's take strike as an example. In this case I do want to have a common vocabulary. It actually will enable me to do a lot of things that I can't currently do. And we can do wrappers and we can make words in one database equal words in another database that's a completely different word. But at the end of the day you do have to come to some understanding of whether it is a wall or a bulkhead. You have to come to some understanding of whether or not taking a hill or taking a building is really a lease or whether you are attacking or whether you're defending. [Laughter].

You've all heard these stories but there is something to be said for the vocabulary and we've got to come to some common understanding in areas where we know that there will be advantage for very tight integration and high quality of service with very low error rates. We've got to come to some understanding.

So my Global Strike component has been working this one very hard and after nine months this week we are within staffing range of getting common agreement on eight words. [Laughter]. Eight words. It's killing me.

It is not the intent to tear down service ethos, but at some point that ethos is going to get in the way if you take it to the wrong extremes. If we take eight months to get to staffing, this is just to enter into staffing, for eight words -- we're missing the point here and there's not a sense of urgency.

If I gave this same task to some Marines in foxholes in Baghdad, it would be solved a lot faster. We've got to push through some of this cultural ethos to get to some of these common vocabularies, but where the need is obvious. We're trying to get these communities of interest to say here are the qualities of service that are going to be necessary. Let's just take integrated missile defense. We want to be able to work any place on the face of the earth, integrate the sensors in air and space and terrestrial, integrate them on the fly with footprints that equal at least a third the size of the earth in any one spot, and we want to be able to take that right down to the battery commander who is sitting there with his Patriot battery and avail him of that kind of knowledge, that kind of sensor cueing, that kind of optimization.

We need a common vocabulary. We need to set standards for the quality of service. We need to be able to set expectations on how the system will work. And those expectations have to be able to span independent of geographic boundaries. That's a huge challenge. You want to set that up in such a way that your colleague say in the UK can plug right into that at any time and with the right credentials have access to all of that information and actually, no kidding, be an ally, integrated into the fight instead of some other nice word that means you don't' really get the information. You're not advantaged by that kind of a [inaudible] network. That's where we've got to be.

So these pilots set those standards, set those expectations, challenge those cultural norms, and it's more about fighting through the cultural norms than it is about the technology at the end of the day.

Go all the way down to the platform level in this construct. The other piece that we really need to start to think about is given the industrial construct that we have today, the realities of that, it is without a doubt the best in the world, but I'm still paying and waiting three times longer and three times more expensive for any given spacecraft. We still tend to field platforms that can be declared legacy before IOC.

We've got to find some way to build into our systems the adaptability for a world that is changing at the speed of Moore's Law rather than waiting for tens of years to catch up with what changes on the battlefield or changes in business. We've got to find some way to integrate the information age better into the industrial complex.

The example that I use, particularly on the geek side of this thing, is aircraft and the buses for information distribution on aircraft. We're really patting ourselves on the back because we're almost to the point where we've got all those digital buses on aircraft. Of course they couldn't pass a reasonable amount of information even by today's standards and we're going to have that same bus at least for the next 25 years. Apple and Dell would go bankrupt on that construct. The mother boards, we'll change them out maybe once in the life of a 20 year platform. That's [inaudible].

My grandson, again, can plug a printer into any computer and expect the computer to know it's a printer. I can't do that on an airplane. I can't do that on a spacecraft. I can't add components based on emerging threats and emerging capabilities. I have to redesign, I have to wait three years for the whole software package to be opened up. The only thing that's slower probably is [inaudible]. But we've got to get a different construct and we've got to start to go after the next generation of buses and information distributions at the platform level. We have to do that now. We cannot --

If we start today we might be able to realize this by 2025 the way we work. Think about it. Think about what we just fielded in the F-22. Is this information system going to be viable in 2020? I can't imagine that it is. So we've got to change the rate at which we integrate and we've got to change the architecture that allows us to keep pace with the times.

It's going to change the profit centers, it's going to change the power centers, and everybody will say yeah, okay, but we've got to do it. This is going to be a huge challenge and there has to be an imperative to go that way.

The other problem that we are trying to struggle with associated with most information that we have today is one that I'll call templating the information or acknowledging or building for a known user but not for the unknown user. This really gets at the issue of being able to go horizontally.

If my F-18 was put together the way it is, the acknowledgement was that the information that I gather is to drop a bomb or launch a missile, and that information has no other use. Everything else that happens on that aircraft that is sensed is just thrown in the trash. For an aircraft, okay, got it, but you're flying over adversary territory, you're sending things that other people might want to know [inaudible]. So why not make the information available for the unknown user? The person who's willing to subscribe and manipulate that information for a completely different function. We don't do that.

There is a program out there that we have been working for about the last three years that started by trying to recruit space surveillance and then was picked up by missile defense. Essentially what it did was two things. One, it took sensors that were built for other functions, took the data and made that data available for unknown users -- missile defense space surveillance, et cetera.

The other thing that it did was something that we've got to figure out architecturally is it took what we'll call legacy platforms and turned them into viable producers for today and tomorrow's capability. And probably the biggest icon for this activity was a radar that we have up in Shimian, Alaska that was built in 1972 for part of the DEW line and the missile warning parameters, et cetera. By taking the information and by putting a reasonably modern motherboard in there so we could process the information and changing from the 1970s processors that would have been run by Atari to something more practical and useable today, number one. Number two, by taking all of the data and archiving it and tagging it in a way that was discoverable.

We took a radar that had one function that it wasn't doing very well and gave it three functions -- technical intelligence, space surveillance, and missile defense/warning. Able to do those missions simultaneously, run on business rules so that it moves from one mission to the other based on threat, based on threat factors, et cetera. It goes cross-agency because it's owned by and operated by the intelligence community but produces data across public [inaudible] only, distributes that data, and oh by the way, if you want to do something new with it it's all archived and so you can hand it to industry, you can hand it to academia and say what I'd really like to know is how many little bodies are up there around that big satellite? And they go off and work it, test it, and integrate it again in a timeline substantially shorter and without having to go back in and rebuild the hardware.

That's the kind of architecture that we're trying to drive. That's the known versus unknown user. We've got to move out of templating the information for one user and doing it right at the source so nobody else can use that sensor for anything else. We've got to get out of that architecture. It's just killing us right now to have to do business that way, particularly when you're dealing with very large terrestrial or space or air sensors. It just drives you crazy.

The last one, and then I'll turn over to Q&A. Hopefully I've irritated you enough that you've got a couple of questions. But the last piece is this distribution piece.

From a warfighter's standpoint, and I'll go with the uniform that I'm wearing. Marines think about everything in threes. We call it combat arms. We call it integrated arms. Whatever you want to call it. But the idea is that if you've got an air, a land and a sea capability that if they take air away it's not going to change the outcome. I'm just going to do it on land and sea. If they take one of the other mediums I'm going to find a way to do it. More than one way to skin the cat. Convince your adversary that no matter what they do the end state's going to remain the same. That's just [inaudible], but that's the warfighter approach to this thing.

If we distribute this information in only one domain and only one way we are incredibly vulnerable. We've got to make sure that the way we move information around, the way we get it to the warfighter has the resilience and survivability that all the rest of our defense [type of action] would have.

So today witness the ASAT test. Oh my gosh, they're going to completely take away our information capability because they're going to deny us space access, which has a huge enabling capability for us because it allows us to move places where we don't have infrastructure and distribute information to the forces that are on the move, et cetera.

We've gotten to the point in some cases where we lean too heavily on that way to communicate and that way to get information. We recently conducted some activities here in the States post-Katrina to take an architecture that exists in the States but was analog in nature that allows you to communicate from an airplane to the ground and then distribute that information once it gets into the ground fighter side of the equation, and update it to digital.

It's one of those things that's going to cost you billions of dollars to do it across the whole country. Well, we actually found out it only costs you about $5 or $6 million. And so if you're flying in an airplane and you want to get information from somebody you can go up to a satellite and then down, or you can go down into the fight.

There's no reason that that shouldn't be global in nature and that's the direction that we're heading. You've got to have more than one way to do that.

So that gives me two, and I have to think in threes. So I've got to have the same thing between my aircraft. I've got to be able to turn them into hubs that move information, don't care whether it goes up, down, laterally, but I can go three different ways to get to the same problem. And actually I'm not even going to tell you which way I'm thinking so that you cannot predict where I'm going to be.

We've got to get to that. If we can move to that direction you'd still end up with disadvantaged users that are out at the end. And we call them disadvantaged users, I really don't see them that way. But this is the construct that Cartwright is a sniper. And I really need to know where my target is. I need to know with a high degree of precision. And I need to know it now. I could send you a ten meg file with a picture from some airborne or space sensor and it would be kind of, you've seen the pictures on television, sitting there watching the thing load forever. It used to be a watched cake doesn't ever rise; now it's a watched file never loads. [Laughter]. That takes forever.

Or, and I'm not supposed to use commercial examples, but I'll hint at this and I think you'll figure it out. Or I could just manage the dots. I could actually build the capability to just put a dot on a map that's already on my PDA at the right place with the right information on it. Avoid the 10 meg file and just manage the dots. Don't screw with my dots. Dots are good. All that.

That's the kind of thought process that has to go into this because we can certainly have an appetite that far exceeds a capacity that's reasonable. So we've got to have an architecture that allows us to do that kind of work and a mindset that allows us to do that kind of work. And sure, everybody'd like to have the 10 meg file but is that really what you need to do your function? It's back to the known user and the unknown user. Is the known user an analyst who has to comb over it at a high degree of resolution and accuracy to understand what's there and read the license plate? Or do I just need to put a [point] between the dots? There's a difference in the quality of service and we've got to acknowledge that fact and manage that fact logically.

The last piece I'll talk to is national capabilities. I alluded to going between agencies and between organizations but today everybody talks about, particularly on the military side of the equation, about all elements of national power. You show me one of those elements of national power that's even connected to the same phone system. Now they're starting to recognize that, but how is the National Command Authority then to wield all of the elements of national power?

This is not an assault on checks and balances. This is an integration of the capability so they can be used in the way they were intended to be used and synchronized and planned and organized and strategized in a way that makes sense. We've got to get beyond that.

There's an awful lot of partnership going on to do this but there are some huge cultural walls in that environment also.

That's a little bit about the context of where the world is today and a lot of the great things that are going on, a little bit about the organizational strategy of the command, some of the challenges that I see. There's a heck of a lot more. I'm willing to go in any direction you want to go in a Q&A type format. Just beware, I will tell you what I believe. [Laughter]. Or I can go back to Omaha and shovel global warming. [Laughter].

Question: My question for you, sir, is one of the issues that you discussed particularly was functional component commands. If you get down to the smaller level of the Joint Task Force Commander, of course he has the [inaudible] and JFACC all working for him in different areas. Do you foresee a possibility that he'll need in the future, maybe now, a Joint Force [inaudible] Operations Commander who has an equal footing with those others or works in a different space? And how would that [inaudible] representative [inaudible], and how would PA, public affairs and public diplomacy fill in the gap?

General Cartwright: Let me make sure I know what you're asking. One is how does this scale down as you go down to the lower levels, and then how do you integrate it in such a way that it's useful and exercises the greater capabilities.

We have done some exercising below the combatant commander. In other words I kind of like to do things in steps, generally on six month centers and try to change. Two years sago we started exercising with other combatant commands, something that had not gone on in the past. Last year we exercised with multiple combatant commands. So a conflict in South America versus a conflict in the Pacific, running them simultaneously because, let's just take ISR. We're not going to have the luxury of just being able to look in one spot. We're going to have to move assets and reprioritize on a global basis. If you don't practice that you won't be able to do it when it's time.

This past year we also then in the middle of that activity in Pacific Command exercised a complete JTF simultaneously and provided the support down to that level. We learned a lot about doing that because the scaling is not the same, the expectations are a little different, and the level of expertise that's available to them is different. But how much of that can you bring to the problem and how much of it has to be at the problem? This is something that we're really trying to struggle with. But this is the idea of, Cartwright at one point in its life said that the way to succeed is give me the oversight and the authority. Let me own all of the functions and resources and I'll conduct the battle. A pretty interesting idea. Not terribly efficient. Good if there's only one problem, but it's just not realistic, number one. Number two, getting all of those functions to the fight in their entirety just doesn't make sense and takes too long.

So how do you start to get efficiencies in the program for the lower level users? I'm going to take it all the way back down. I gave you this Rivet Joint example earlier on, but this is the issue of moving everything to the point of transaction instead of -- people, equipment, et cetera -- instead of moving the transaction to a center where you can do business.

We'll take the linguist on the Rivet Joint. We launch out of Okinawa, go south to some unnamed country's border. We fly there for six hours. We spend two hours on station then we fly six hours back. So Marine math says that's six and six and two and that comes out somewhere in the neighborhood of 14. I did that without fingers. [Laughter]. Of which I had two hours worth of use out of that linguist.

Now I could have taken that same information and pumped it up through a satellite and back down to a work center in the States, in Okinawa, whatever, and let that individual linguist have an eight hour day to do translation versus a two hour day. I could have stared at that point instead of just patrolling it for a couple of hours using a different sensor.

So how do you start to find the right balance of not moving everything to the point of transaction, taking what's necessary and getting it back. The reachback, things like that? And when we looked at this, we sat down with the Air Force and said geez, the answer to this is we're going to pump this information back. My linguists, I have about a 25 percent retention rate. After the first tour I can only keep 25 percent of them. Why? I employ them for about 300 days a year every day that they're in the service. Great family life. Why can't we keep them?

So you say well, if I'm going to do this, I'm going to bring him back and we're going to make a nice place on a base where they have their family and they can go home at night and get eight hours work out of them a day. Everybody goes touchdown. Then you go try to put it into the culture and everybody rebels. You go, what's wrong with this problem? Well, what does the linguist get paid for? He gets paid for flying, not for translating. So if you put him in a base he loses 25 percent of his income because he just lost his flight [inaudible].

So the incentive structure on this is all [gooned] up. That's Marine speak. [Laughter].

But how do we start to make sure that that is our point of transaction for the lower echelon commands, that you have the right incentive structure, that you're getting them what they need in the quality of service that they need. We don't have that right, but it is kind of back to managing the dots, understanding the point of transactions of what you want to get accomplished, and then being able to operationalize that down at the lowest levels. That's the work to be done. It is good work and the services are going after it aggressively, but we're learning things as we do this, that we go geez, how could we have been so stupid? But that's just the way it is.

Question: General Cartwright, I'd like to go back to a point you made earlier about [inaudible] and so forth. [Inaudible]. About eight years ago we had the [good] Science Board Task Force [inaudible] open architectures for [inaudible] in combat support systems. And [creative] readiness came out of it. Number one, interoperability, [inaudible] measure. [Inaudible]. Second was the avoidance part of the technical obsolescence, we went to requisitions. We also [inaudible]. The third part was [inaudible] resistance. You can [inaudible] and hopefully somebody's [inaudible] that you can use When the briefing was taken up to the third floor for responses the comment came down, geez, that's great, but it's too hard, too expensive and nobody's ever going to change [inaudible]. Do you think the environment today is the same? Maybe we ought to do that now.

General Cartwright: I certainly believe that's the case. That's certainly where STRATCOM is heading in its role s an advocate for the warfighter. My [inaudible] is that absent that it would take a significant event, a Chapter 11 or another 9/11. Either a threat to the industry itself that it just can't do business this way and sustain itself, or the nation gets threatened because it's unable to do operations in an effective way. One of those two is the how close do we have to get to those before we compel the larger organization that it's got to step out of the [hole] that it's in. My sense is we're there. We're close enough that I'm willing to say that's something I'll fall on my sword for. I think that's critical.

Question: Sir, would you comment [inaudible] role of the warfighter, would you comment on that and [inaudible]?

General Cartwright: This has been -- advocacy for the warfighter and how does it fit with these functional commands.

To me the space that I want to be able to operate in as a functional combatant commander is two-fold. One is that for the centers of excellence, whether it's our ISR center of excellence, our missile defense, to help them advocate for those things and that vertical capability that will make them better. That's point one. They already have all the authorities they need, but having a warfighter standing with them, either in congressional testimony or in front of the budget committees inside the Pentagon, et cetera, has value. That's point one.

Point two, in those things that need to be integrated horizontally, then to stand up and be the advocate for those capabilities that have to go horizontal in nature, or enable horizontal integration. That to me is the second key advocacy piece for STRATCOM or any functional commander, really. Because the vertical integration will be handled fairly well by the services, fairly well by the functional centers of excellence like DIA or NSA or you pick it. But the horizontal really has no advocacy, so we run into the same problems that we were talking about here with the Air Force or with the airplane [inaudible]. Nobody's there to take care of the unanticipated user. Nobody sets the battlespace so that the unanticipated user can be [inaudible] quicker than his adversary. So to me that's where STRATCOM fits in. That's where our functional combatant commander fits in.

Question: General, I'd like to continue on that integration theme. We've had some initial dealings with your global innovation strategy center out at West Omaha and that seems to be a way that you are looking to use that integration [inaudible] as well as address some of the cultural issues. Can you talk a little bit about [NIC] and how you see that fitting in with integration?

General Cartwright: This is back to the principles, the first order principles that we used to set up the organizational construct and it's back to the issue of getting industry in, and in particular academia into the mix in a positive way, exposing them to our problems, particularly our hardest problems, and then exposing ourselves to a wide range of possible solutions rather than I'm from defense and I've done defense since I was ten years old and here's how you solve the problem. But start to look for innovation in ways that we have not enabled ourselves to do in the past. So that's what the innovation center is sort of set up to do.

The idea there is that we don't hire anybody. You've got to be compelled that this is really a problem and willing to give of your own time. That way I don't build General Jumper's tribes. We don't solve a problem and then have a group of people who just keep solving their problem the same way forever. So the idea here is to reach out to the smartest and brightest associated with any given problem, bring them together, facilitate for them, give them a deadline, a sunset clause, feed the pizzas under the door until they solve the problem, and then disband, take the range of solutions and see what we can do.

It's been interesting in that industry comes very quickly, the academic side has been a little slower to come but now really is beating down the doors for the opportunity. You allow yourself to get a cross-section not only of the vertical organizations, but you get a cross-section of industry and academia. You get a cross-section by, and I have to say this word right because I always get in trouble, but young to old. You do. You actually see problems looked at on very different perspectives and you can build the demographics in that allow you to do what this country does best which is take advantage of diversity rather than avoid it.

It allows you to do a lot of things and by putting a sunset clause on it, and by not paying them, you actually do attract the best and the brightest and you attract them for the reason that you want to attract them, because they want to contribute and they may bring a whole university system with them or they may bring a whole industry with them and you do not have to limit yourself to just the U.S.. You can reach out to where the best and the brightest are if they happen to be some place else.

So that's the idea behind it. Then you give them the problem and you say what's the range? When do I expect that it's realistic to solve this problem? Should I accelerate that? What would be the implications? Second and third order effects. Why does North Korea do what it does? Why does China do what it does? What's at the center of their calculus? Really interesting stuff.

One of the projects that they took on and offered a solution to that has really taken off was if either by man-made or by natural occurrence a bio-problem occurs in this country, whether it's a pandemic or whether it's introduced, how do you distribute [inaudible]? How do you know that it's coming? One of the indications that go on there, how does all that interplay?

And essentially we had not even the 30-somethings. We were dealing with the teens. These were a group of students that got put together and they came up and went to industry and got for one state a template of how to pull all of the industries and all the schools, without identity so to speak, so you didn't compromise identity, how to reach out beyond that to people like the Five Star Alliance and see their data, who showed up and who didn't who's in the hospitals all over the country and all over the world, pull that together for Center for Disease Control, give it to them, and then how to take industry and say you've got this part of the state, you use your trucks to do this, you use your trucks here. National Guard, you go to this area that's contested and [inaudible]. Put that whole thing together and now Homeland is using that as a template for all the states. This was done by a group of students.

We did have a group of older people that looked at this, but these are the ones that came up with it. They wrote all the programs, did all the software.

That's the kind of stuff that you're trying to incentivize. People do want to contribute, they really do. They find great value and great reward in being able to be part of something like that. As long as you can condense it, force it, drive it, and put a sunset clause on it it tends to add value very quickly. So that's a little bit about the kids.

Question: Two comments. There's no lack of innovation, but there's no government organization or customer who has the authority, the resources, and the technical expertise to do wartime integration given Title 10 and the requirements process. How are we going to connect the dots? It's more than drawing lightening bolts, it's doing the [inaudible] to do it and actually having someone in government pay for it.

General Cartwright: That's right. That goes back to the imperative issue. How close do we have to get to the edge of the cliff before people start to see the value in [inaudible], in one alternative versus another? And at what point does this net centric network enabled collaborative environment show sufficient promise that it's worth giving up something else that you're holding onto in order to do it?

Without using a book example, there is this tipping point. The reality is, at STRATCOM they call it the Disney Principle. We'll put something together on the [cheek] and if there's a line we'll find investors. If there's no line, you've got to be willing to shoot the baby. I'll be very brutal about shooting the failures, so to speak. But if there's a line I've really not had a lot of trouble getting the resource necessary to go do it.

You are right, though, that when we go do this, having been a former J8 is probably a great advantage for me. We know where most of the skeletons are. So we have been able to generate the resource necessary to go get capability. But the first principle is there's got to be a line at the door to show great value, and then it's really hard. But the process is designed not to advantage horizontal integration. You're exactly right. And so you've got to really fight hard to get capability that goes horizontally. And that shouldn't be -- The more it shows value the more that will change because people will align with success.

Question: Good morning, sir. I really resonate with your comments regarding incentivizing organizations. After many years in the intel community and also wearing the uniform, I always appreciated the fact that when I was in a producing organization I was judged by products. Not information, not services. I think one of the cultural barriers to making the transition to information sharing is just that -- organizations have to justify their budgets based on their support to the warfighter in terms of products.

I would submit to this group that maybe a different way to think about that is if you begin to appreciate that unexpected user as your primary consumer and we begin to look away from productizing our information sources, but rather recognizing that our product, be it a power point slide for your morning briefing or a cockpit display in your aircraft as you're flying over Baghdad, it only exists at your point of interface with that information.

As they think about that, maybe organizations need to be incentivized by advertising their services in a discoverable way rather than by identifying products in metric power points through budget controls.

General Cartwright: I think you're exactly right. That's right on the mark.

Question: I was wondering if you would talk a little bit about what sort of what you see as your biggest challenges right now, your command in STRATCOM. You mentioned some of your [inaudible] problems and I'm just curious what you see [inaudible].

General Cartwright: This is more along the lines of what keeps me awake at night? Okay.

It's never a simple answer and I'll try the elevator speech as best I can. But the span of the threat has gotten so much larger when a Hezbollah can act as a nation state with throw weight authority, or an al Qaeda can bring to bear weapons of mass destruction that heretofore only nation states could do, and there's no home address for who did this. How do you start to build those things that would deter an adversary that really doesn't have a home address, that you can't get attribution on? How do you start to move in that direction and not give up the fact that there are nation states that need to be deterred in their actions, that need to be [inaudible] or moved in a direction? That span is so broad.

So on one corner we're tugged with making sure that the nuclear arsenal is all it should be to deter the people that are appropriate to deter them [inaudible], but on the other extreme is the individual with no ties who can get up in the morning and go geez, I'm really irritable today. And oh, by the way, I've got access to weapons of mass destruction.

That huge span is to me the great problem for the country and the world. How do you start to understand that? It's not that an individual can't be deterred. That's what law enforcement is. That's the basis of law enforcement. It's not that we don't know how to do that, but how do you start to integrate this wide range of threat in a format that heretofore was really focused on the nation state construct? And to me, that's the piece.

That means you have to have effects that are well before [kinetic], that are setting the stage and allowing nation states to help themselves while preserving their sovereignty and yet be part of a larger collective defense and capability that is global that works against those things that would just be done on a whim, so to speak. So a lot more focus needs to be paid on that pre-kinetic activity and discouraging entering into the kinetic part of this equation. But to do that you've got to be credible in the kinetic part as well as being credible in this non-kinetic portion of the fight. That to me is the tough part.

Question: Sir, speaking of non-kinetic, information operations is evolving and maturing [inaudible]. Where do you see STRATCOM in terms of pushing that forward or advocating it to the services? Especially since the Air Force is moving out and has expanded the definition from IO to cyberspace?

General Cartwright: I guess it's where you sit. It's good that the Air Force has finally come on board. [Laughter]. And they're doing it as only the Air Force can by throwing everything they've got at it, so that I applaud. But I will tell you that [inaudible] Army and Navy have been at this for a substantial period of time and have probably the more robust capabilities today. That gap can be filled quickly but it's going to take a lot of work and it's going to take what you're seeing very publicly being done by the Air Force to get themselves up to speed, so to speak, or back in synch. That's not pejorative, it's just kind of a statement.

We have the mission. What we're trying to understand today in STRATCOM is one, this is a millisecond world so the decision processes, the organizational constructs have to be adaptable in a millisecond world to the [inaudible] construct is one that was put together ad hoc, really not even by industry but just by a whole bunch of interested people so that the architecture of this medium, unlike the sea or air or space, is [manning] so to speak. So that's a challenge. How do we make it and what are the pros and cons.

I talked a little bit about that architecture is really one of terminal organizations centered on the terminals. We've got to build a defense that is integrated offense, defense and battlespace preparation of the intelligence side of that equation in that medium, number one.

Number two, we've got to start to understand how to build a layered capability. We cannot just rely on firewalls. We have to be out there in the medium, just like we would be out there in the air or on the sea. We've got to be in that medium able to detect hostility in that medium and able to do something about that problem before it threatens us.

That means, just take a virus as an example just because it's relatively simple. Someone introduces a virus into the internet and they do it on the other side of the earth. You've got to detect it, process what it is, whether it's good or bad, decide what the appropriate action is, and then beat that virus which is moving at the speed of light back to the target. That's a pretty tall technical order but that's where we've got to go. We absolutely have to go there. We have to push out from the terminals. We have to get this to a global enterprise. We have to decide what are those things that are appropriate for the warfighter to be doing? What are those things that are appropriate for some other global commander to be doing? What's the taxonomy? How do you train and equip [inaudible] force associated with cyber operations? A lot of what you're seeing publicly being debated by the Air Force right now which is a wonderful debate to have.

And remember, that when you shoot a bullet in cyberspace it's the speed of light. It knows no boundaries geographically. The laws that are associated with that and the rules and the ROE are all based on geographic principles. So how do you translate them across?

These are the huge challenges that exist in cyber. The tools are getting ever better, as you say. The capabilities are getting ever better. But are the decision processes matching up with that? Is the architectural construct of how we defend matching up with that? That's the real question for the department. That's where we're putting our emphasis. And you should hear both from my network operations group which is really operate and defend, and my cyber network warfare group which is the battlespace preparation and the warfighting so to speak. Both of those will be here to talk, but you'll see how we're starting to organize.

We are coming to an understanding that the skills associated with operations and defense are pretty much similar. You can send them to the same school, so to speak, and train them up that way. They attack and the battlespace preparation skills are pretty much the same, but the two, while they can do each other's jobs, are different skills. Now how do you integrate them as an integrated offense and defense, as the critical thing that we're trying to understand. Organizationally and vertically through the chain of command. What's appropriate at the JTF level? What's appropriate at the nation state level to be done?

That's kind of a long answer but it's a huge problem. It's a huge challenge, let me put it that way. We have a way forward now that you're seeing by the services, exercising the debate relatively public, which is good, about how to think about this problem so that they can, back to the question up here, start to match the resource with the regret factors in this problem.

Thank you very much.

[Applause]