Obama administration aims for bottom-up approach to creating global standards for protection of IT and critical infrastructure.
It's Linton Wells II's little secret: The Defense Department has been paying him for his avocation, not work, for the past four decades.
"I remember telling someone in seventh grade that I wanted to go to the Naval Academy, and I had a chance to do that," Wells says. "I had the opportunity at 17 to take an oath to support and defend the Constitution of the United States, and I've had the privilege of doing that in uniform and out for the past 42 years."
For the last year and a half, that privilege has extended to running the systems shop at DOD as its acting CIO. Wells stepped into the role after John P. Stenbit left the job of assistant secretary of Defense for networks and information integration in March 2004. Wells will soon return full time to his role as principal deputy assistant secretary now that President Bush has tapped Raytheon vice president John G. Grimes to become the next Defense CIO.
Just as his stint as CIO was coming to an end, Wells talked with Fed Tech Managing Editor Vanessa Jo Roberts about the challenges of overseeing a systems operation as massive as DOD's and building out a network that can bring information technology's power to the edge—to the end user, whether that's Secretary Donald Rumsfeld in his Pentagon office or a soldier in a windswept tent in Iraq.
Fed Tech: You've been at the helm of Defense IT since March of last year. But long before that you began promoting the transformative nature of IT. How do you view the work the CIO does in meeting the mission of the department?
Wells: A charter signed by Defense leaders in early May lays out the role of the assistant secretary of Defense for networks and information integration and CIO.
It defines the role as both the principal staff assistant for the secretary and the key proponent for the secretary in areas like command and control, communications, spectrum management and that sort of thing. But it also includes being the strategist for the organization from an information, IT and information resources management perspective, and as the architect and the business executive. That charter had gone through several drafts since the late 1990s.
The role has now been defined after several years of evolution.
Fed Tech: Do you see yourself as the link to outside the Pentagon, to the IT world, bringing ideas in?
Wells: The Office of the Secretary's role is policy, oversight and governance. What we try to do is work within the major processes of the department to pull this together.
There are really five major processes: the requirements process; the resources process—planning, programming, budgeting and execution; the acquisition process; the operational processes for deployments; and the external processes for things like congressional relations, public affairs and interagency programs.
So it doesn't do us any good to stand alone and hurl Olympian thunderbolts into the system. We've got to leverage all the other things that people are doing.
If I'd like to take some wonderful, whiz-bang technology being developed in the private sector and make it useful for Defense, the way for me to have the maximum impact is to have the requirements process pick it up and create some kind of capabilities document that says, "When we build things we must have this capability, and this is why." But we need to learn to do this faster. It now takes us years to acquire things while the commercial product process proceeds in months. We're working on this.
Fed Tech: One of your personal buzz phrases is "power to the edge." Could you talk about how that's fundamental to the network-centric goals of the department?
Wells: The basic premise of net-centricity is the existence of a network that allows for increased information flow, which allows people to develop and share awareness of the situation around them.
Based on their understanding of the command intent, they can take that situational awareness and self-synchronize their actions rather than having to wait for hierarchical orders from the command structure. This dramatically improves the speed of mission accomplishment and effectiveness.
To have the net-centric piece, you need to have the network. Our three goals here basically are to build the network, populate the network and protect the network.
The edge of the network is where the action is happening. For the soldier in the foxhole, the edge of the network begins with the secretary of Defense. For the secretary of Defense, the edge begins with the guy in the foxhole.
What we want to do is move away from what used to be a smart-push approach to sharing information to a smart pull.
For a smart push, think of a telephone system: I need to know where you are and what you need to know in order to convey the information.
A smart-pull approach says I probably won't know exactly where you are and I certainly won't know what you need to know. But if the network acts as the repository for the information and that information is discoverable, accessible and understandable, you can define for yourself what you need.
For a squad leader, say, that might mean having access to everything within 10 kilometers of his position that's happened in the last 24 hours. That leader can pull the information, rather than having somebody else trying to figure out what's needed.
So what we're seeking with power to the edge is that people at the edge of the network, where the action is happening, are empowered by the information that they receive from the network and also rewarded for the information they put back into it. It's got to be a two-way street.
Fed Tech: So the network part of that is the Global Information Grid Â
Wells: The Global Information Grid is not just a network. It's also the people and the processes that make the whole thing work. It's the entity. It's the vision. It's an architecture.
Fed Tech: With the Net-Centric Enterprise Services, you're creating sort of alleys of information and services for people on the grid, correct?
Wells: We want to make the GIG more like the Internet. You have data, and you have applications. The enterprise services are basically what tie the data to the applications. Enterprise services are things like discovery and messaging, and security and mediation. It's where you would tell your discovery agent, "This is the information I would like to have pulled to me. Go find it."
Fed Tech: It's essential then, isn't it, for power to the edge, to have those services work in tandem, to be able to control them and then to add the security factor?
Wells: Security is utterly critical. The network is rapidly becoming our center of gravity. We can't let it be our Achilles heel. And security needs to be designed in from the start. It can't be appliqued on afterwards.
Fed Tech: Where would you say you are in achieving this power to the edge? It's clearly a never-ending sort of a goal because it's dynamic, but where do you think you are in terms of the network being really useful?
Wells: I think we have pockets of real excellence, and we have a migration path to the future.
There are six major programs that constitute the GIG.
Four deal with transport of information: the GIG-Bandwidth Expansion (GIG-BE), the terrestrial fiber piece; the Joint Tactical Radio System (JTRS), which is not just a flexible radio replacement but also the router on a mobile Internet Protocol network; the Transformational Satellite, providing global laser communications on the move; and the Teleports that link the terrestrial to the on-orbit segments.
Then there are the Net-Centric Enterprise Services, which tie the data and the applications together, and there's Information Assurance, which underpins it.
Plus, there are three major concepts that go with this. We have to be able to do "net ops" and network resources management on this very heterogeneous network that runs 10 gigabits per second across terrestrial stable fiber to maybe 10 kilobits a second on dynamic networks on the battlefield.
So how do you apportion bandwidth? How do you apportion new numbers of radios? How do you apportion information?
We also have to manage the radio frequency spectrum—that's key to mobile operations.
Finally, there is the absolutely critical third piece: the underlying data strategy so that all data is tagged to make it discoverable, accessible and understandable.
Where we are:
In summary, I think we have the key pieces in place. We are now shifting the emphasis from the big architecture piece to implementation at the edge of the network. We are going to the tactical edge and working our way back in.
Fed Tech: What's your trick as CIO to have a quick idea of what's going on in all the programs that you oversee?
Wells: I think the answer is that there's no closed-form solution. You just need to keep your trap lines out there and try to build an atmosphere where people are willing to come tell you the information rather than having to go drag it out of everybody.
The hardest thing I found in command was to get bad news early enough to still act on it to avoid problems, not to learn it after something fell off the cliff.