The Future of Information Management in the U.S. Intelligence Community
Frederick Thomas MARTIN <email@example.com>
This paper will describe the future of information management within the various organizations and agencies that collectively are known as the US intelligence community, including the CIA, NSA, DIA, and the now declassified NRO. The intelligence community of the US government recently confirmed that its budget -- kept secret as classified information in all 50 previous years since its inception -- totaled $26.6 billion dollars last year. Because most of these funds are spent on providing information, the central focus of this paper will address what the US intelligence community believes to be the "information revolution" of the third millennium. This paper will provide an explanation of the possible role and impact that the Information Technology Management Reform Act (ITMRA), passed by Congress in August 1996, will have on the future of information within the intelligence community and how that relates to Internet and intranet working professionals. It will describe the transition to Web-centric electronic publishing of our nation's intelligence reports, known as "finished intelligence," into an integrated information space. Describing the future, this paper will explore the concept of a more "agile" intelligence enterprise, giving us insight into how the US intelligence community plans to achieve its goal of an electronically networked environment for the production and exchange of intelligence, a goal deemed absolutely essential to national security in the 21st century.
Perhaps the greatest paradigm shift in modern intelligence production is the transition to Web-centric, electronic publishing of our nation's intelligence reports. As a result, there is much effort now being devoted within the government to addressing a number of electronic publishing concerns. Sharing the results of several successful projects, this paper will take a look at a number of these concerns, including the issue of implementing "push" technology and the debate in many enterprises today over the appropriate roles for standards such as SGML and the role of new emerging standards such as XML. Examples will include specific applications taken from the National Security Agency, the Office of Naval Intelligence, the National Imagery and Mapping Agency, and the Joint Intelligence Center, Pacific. These success stories have direct application to the organization or business today faced with the production and distribution of large volumes of documents, which describes virtually all businesses today!
The paper will also provide us with a look into the future: how does the US intelligence community plan to implement all of the information management improvements that it is working so hard on? We can begin to answer this with an examination of the underlying carriage to all of the information management improvements of the US intelligence community and a glimpse at the real future of the world of intelligence: the concept of "virtual intelligence." What will the world of virtual intelligence really look like?
To answer this, we will examine the idea of a more "agile" intelligence enterprise, as envisioned by Dr. Ruth David, the current CIA deputy director for science and technology. This paper will explain that vision, including the perceived problems, security issues, and management challenges. The "agile" intelligence enterprise concept, combined with architectural, security, and other standards applied to the existing and planned telecommunications infrastructure, represents the primary components of one of the US intelligence community's most important goals.
In the United States, the production of intelligence is a multi-billion-dollar effort that is vital to the country's security and prosperity, and arguably, that of the free world as well. The various organizations that are collectively known as the US Intelligence Community are in the process of tremendous cultural and technological change as the third millennium approaches. With technology changes, declining resources, an explosion of information, and a completely different set of geopolitical priorities, the US Intelligence Community must find smarter ways to manage its most critical resource: information. This paper addresses the future of information management within the Intelligence Community and describes how they plan to become flexible, adaptive, and more "agile" as an enterprise. It describes their development and implementation of "Intelink," an intranet that provides electronic publishing and dissemination of their finished intelligence as well as a multi-security-level collaborative environment. Two organizations with operational prototypes are cited to demonstrate the progress that is being made to become more "agile." Lastly, the paper describes the transition to an "agile intelligence enterprise," where "virtual teams" collaborate online and all data becomes a shared asset as part of an enterprise-wide "information space."
The leaders of civil and military organizations have always desired to have information that would forewarn them of danger, reduce uncertainty in the decision-making process, and otherwise prepare them for whatever future critical decisions they may face. The practice of intelligence can be defined as the sum of efforts required to achieve the fundamental needs of these leaders and decision-makers. The practice of intelligence involves tedious collection, interpretation, analysis, and reporting of information; and like so many business processes today, it also must grapple simultaneously with removing or at least reducing the various problems and barriers encountered. Most people think of intelligence activities as occurring among foreign nations, powers, or other organizations (although "industrial espionage" involves the practice of intelligence among commercial, nongovernment entities) to determine various activities, plans, and even capabilities of those involved. But regardless of the "flavor" of intelligence, much of it comes down to the same set of activities that are so important to businesses today: making sense out of a deluge of information.
In today's world, the practice of intelligence is a key activity of the US Government; indeed, it is vital to all governments, large and small, with an actual or perceived role on the international stage. Every single day, the various components of the US Intelligence Community provide crucial information to those who manage our nation's strategic interests. Generally, the "requirements" of intelligence include the following:
Collectively, the business of intelligence is really about information: gathering, processing, analysis, reporting, and archiving for potential future analysis or reporting. The amount of information collected and processed is staggering and is measured in trillions of bytes.
The US Intelligence Community is a coordinated network of thirteen primary agencies whose customers are a number of civilian and military organizations that work together to support efforts to ensure and maintain the security of the United States and its allies. This task is accomplished through the collection, analysis, and reporting of huge volumes of information that keep the President, Cabinet members, members of Congress, and other senior government decision makers informed. Intelligence has become a vital element in every important activity of the US Government. The thirteen primary components are
In addition to the thirteen primary components above, many people broaden the definition of the "intelligence community" to include the federation of US Government organizations and users that have access to the primary intranet that serves the official Intelligence Community. This federation includes other Department of Defense civilian and military organizations, the Department of Justice, the Drug Enforcement Agency (DEA), the National Aeronautics and Space Administration (NASA), and many others.
In recent years, the lines between various intelligence activities have become less distinct, but the goal has remained unchanged: to support key government decision-makers with the best possible information. In his report to the National Performance Review in 1996, then Director of Central Intelligence John Deutch stated
For members of the Intelligence Community, these "missions" refer to ensuring that policymakers and military leaders are provided with the information necessary to make decisions on such matters as
Special "select" committees control congressional oversight for the US intelligence efforts. In the US Senate, this is accomplished by the Senate Select Committee on Intelligence (SSCI), chaired by Richard C. Shelby of Alabama and in the House of Representatives by the House Permanent Select Committee on Intelligence (HPSCI), chaired by Representative Porter J. Goss, a former CIA officer. The overall cost of intelligence production to the American public has remained secret as classified information since the inception of the Central Intelligence Agency over 50 years ago. However, the Intelligence Community of the US Government confirmed in the summer of 1997 that its total budget for the previous year was $26.6 billion dollars. This total includes the budgets of the Central Intelligence Agency; the Pentagon-run agencies such as the National Reconnaissance Office, the National Security Agency, the National Imagery and Mapping Agency, and the Defense Intelligence Agency (DIA); and other operations and organizations at the Pentagon, State Department, and the Federal Bureau of Investigation.
It is interesting to note that behind the primary missions of the National Security Agency (specifically the "signals intelligence" [SIGINT or "code breaking" mission] and the information systems security [INFOSEC or "code making" mission]) is the fascinating story of the evolution of automation and information management. From the early code breaking days through the advent of the digital computer, the National Security Agency and the Intelligence Community played a critical role in shaping early automation, including mainframe and networking developments. For example, early cryptanalytical research was instrumental in the development of the first large-scale computer and the first solid-state computer, leading to mainframe computers and today's modern systems.
As the ultimate information producers, spending billions of dollars annually, the Intelligence Community has always faced enormous information management challenges in the collection, processing, and dissemination of intelligence. In this ever-changing environment, they needed several things: an architectural framework that allowed for sharing of intelligence data, products, and information across the various intelligence organizations; a methodology for collaboration to facilitate their production; and an improved mechanism for their dissemination electronically. Several important factors shaped and intensified these needs, including
Steven T. Schanzer was the first director of a senior intelligence department established to promote the interoperability of all automated information systems supporting intelligence operations. He decided that an intercommunity electronic information service for intelligence information should be the principal mechanism for implementing the goals, operating efficiencies, and cost savings envisioned by his new organization. Steve is generally considered as the "father" of Intelink: his vision of a secure communications and processing environment for intelligence dissemination and collaboration. Intelink was destined to become part of the prototype of the "agile intelligence enterprise" described in Section 5.
In short, the Intelink is a secure, private collection of networks implemented on existing government communications networks. These networks employ Web-based technology, use well-established networking protocols, and are protected by firewalls to prevent external use. Intelink encompasses one of the world's largest data repositories and addresses demanding data management issues that are at the very extreme of what an enterprise normally encounters.
Within the Intelligence Community today, Intelink is commonly perceived as simply the intranet/extranet that interconnects various intelligence organizations and operations, i.e., connects the various providers and users of intelligence information. It is this and more: Intelink is actually a full suite of information services that have been put in place by various government organizations for their authorized users. It uses advanced technology such as commercial Internet applications and protocols, all operating on existing US Government as well as commercial telecommunications resources. The ultimate purpose of Intelink is to provide robust and timely access to all available intelligence information, regardless of location, medium, or format, for all interested users that have authorized access to the various available security levels of Intelink service.
By using standard commercial World Wide Web browsers, such as Netscape Navigator or Microsoft Internet Explorer, users can access intelligence products over the Intelink family of networks. They can exchange e-mail, access on-screen "video-on-demand," conduct video teleconferences, and use other tools that allow users to collaborate or work together in real-time. As such, Intelink captures the essence of current advanced network technology and applies it to the production, use, and dissemination of classified and unclassified multimedia data among this nation's intelligence resources. It has been declared as the strategic direction by the US Deputy Secretary of Defense and the Director of Central Intelligence for all Intelligence Community "finished intelligence" dissemination systems. ("Finished intelligence" or "end-product" refers to the various reports and studies that are disseminated to the customer or end-user and collectively represent the output of the intelligence process. A threat projection, technology summary, capability study, scientific handbook, a detailed national intelligence estimate on some subject, or even an individual translation, are all examples of finished intelligence.)
Will an intranet such as Intelink help the Intelligence Community achieve its information-sharing and collaboration needs as shaped by the four factors cited above? Timing is everything: intranet and Web-based technology development have come on the scene and are maturing at precisely the right time to play a key role in managing information for the Intelligence Community, and indeed, the entire US Government. Several reasons point to an intranet as the right solution:
With these advantages in mind, an intranet like Intelink is clearly the right answer to the problems that the Intelligence Community faces.
Intelligence Community leaders are not the only ones who have been concerned with these issues. The US Congress has also taken a strong interest in the information management needs of the government. In fact, Congress recently passed important legislation to improve the way federal agencies manage information technology investments and streamline the acquisition process. Known as the Information Technology Management Reform Act (ITMRA) and signed into law in August 1996, it requires each federal agency (including the intelligence agencies, of course) to implement a process to optimize the value of information technology acquisitions. It also established the position of corporate or Chief Information Officer (CIO) in every federal agency to serve as the senior information technology manager or official. This legislation, also known as the Clinger-Cohen Act, has special significance to many members of the Intelligence Community -- namely those in the Department of Defense -- because the current Secretary of Defense (former Representative and Senator from Maine from 1979-1997) William S. Cohen was the primary author. The new legislation has led to the formation of a number of forums that have facilitated exchange and cooperation of mutual information technology concerns. This in turn has helped to focus on the information technology needs of the Intelligence Community as an enterprise and the examination of the value of a collaborative, "agile" environment using intranet technology.
The bottom line is clear: when one examines the functions that are delivered by Intelink (i.e., what Intelink actually does) and its content (i.e., the information available on Intelink), a compelling business case can be made which shows that Intelink improves intelligence production and distribution. As a prototype of the future world of intelligence, Intelink has become the strategic tool for open information management within the US Intelligence Community.
Many enterprises today are information-based. The true assets of the enterprise are the data and the intellectual resources that transform that data into information. For the Intelligence Community, information is its most critical resource, and managing that resource is its number one priority, as the following examples demonstrate. In both cases, Intelink provided the catalyst for their success. In addition, both organizations are taking advantage of SGML to improve the production and management of their information.
The Office of Naval Intelligence (ONI) is the national production center for global maritime intelligence for the United States. As such, it has become the nucleus of their expertise for all important maritime issues, including sea-related weapons systems, foreign naval sensor systems, submarine platforms, and ocean surveillance systems. ONI supplies a large variety of maritime products to Naval and Joint Forces, ranging from large scientific and technical (S&T) handbooks distributed to a wide customer base to short threat summaries for limited distribution to decision-maker audiences. For ONI and other intelligence centers, the products are extremely perishable. Intelligence that is delivered either late or in the wrong format has lost both its impact and usefulness to the customer.
In the past, ONI's production unit was staffed and oriented exclusively toward printed media production. The rapid build-up and entry into combat during the Gulf War outstripped the long production times for traditional publishing, which could not keep up with the demand for printed intelligence in large volumes. Indeed, the delivery of some products as large pallets of paper documents was inconsistent with the rapid deployment and movement of those military forces. The sheer weight and volume alone made it extremely difficult to deliver paper documents to mobile theater units and it was difficult for the units to transport the products with them.
Using SGML, ONI developed an automated document management system that provided a number of options for information delivery: print media, CD-ROM, and electronic dissemination over Intelink. ONI was then able to use this system to put intelligence in the hands of naval and joint forces quickly and in a form that was compatible with their mobility. The CD-ROM format was especially important as it allowed ONI to reach customers where Intelink was not available.
This effort provided a significant, documented manpower and real cost savings. As a result, ONI is able to produce sixty percent of its former product volume with only about five percent of the previously required resources and in less than 1 percent of the time. More specifically, the traditional paper version of the Naval Ship and Submarine Handbook series (11 regional volumes of about 1,000 pages each to 900 customers) cost approximately $250,000 to print, while the cost to produce the CD is about $20,000. Moreover, the CD contains more information, is fully searchable, and offers more ways to view the information. In addition, the production labor has been cut from approximately 204 weeks to only 1 week of labor.
Even more important than the manpower and cost savings, the quality of the end-product was vastly improved. For example, it is now easy to tailor a unique product for a specific audience. The customer gains the ability to process more information in a medium that has virtually no space or weight restrictions.
The second example of the usefulness of Intelink comes from the Joint Intelligence Center, Pacific (JICPAC). JICPAC is a US military intelligence production center responsible for producing general military intelligence, targeting, and operational intelligence information for over forty countries in the Pacific Rim and Indian Ocean areas. JICPAC is moving from an almost exclusively hardcopy environment to a Web-oriented environment with Intelink as its basis. As such, they represent the true prototype of the future intelligence production world, where information is managed as an enterprise, in a true "agile" environment.
To accomplish this, they have implemented an SGML-based information management, electronic publishing, and distribution system. This new system uses SGML to create, manage, and maintain the necessary databases for their own local "information space" and to produce and deliver intelligence documents over Intelink. The new system consists of four system environments:
Their implementation of the "information space" concept, an interoperable information infrastructure that supports collaboration and knowledge sharing among intelligence users and producers, exemplifies the evolution toward an "agile" intelligence enterprise. As implemented on Intelink, this concept provides the critical tools and capabilities needed by the intelligence analyst of the 21st century, including
all within a secure environment protected from unauthorized access or tampering.
With the implementation of their new system in February 1998, JICPAC has become the herald of the future world of "virtual intelligence" within the US Intelligence Community.
Intelink, with all of its advantages, is actually only
the beginning. One of the primary goals of the US Intelligence
Community is an evolution towards "Virtual Intelligence."
The plan is to create Virtual (or unlimited) Intelligence as part
of implementing an "agile intelligence enterprise,"
a fully integrated, distributed "information space."
The concept of an agile intelligence enterprise, which has been
deemed as absolutely critical to the nation's security in the
21st century, is the vision of the current Deputy Director of
CIA for Science and Technology, Dr. Ruth A. David.
I believe that the Intelligence Community today is at a crossroads of progress: Our choices are evolution or revolution. . . . (Dr. Ruth A. David, October 1997)
Like many businesses today, the Intelligence Community is at a crossroads of progress. The issue is how to tackle a long-range strategic problem: meeting the expectations and needs of intelligence customers that require greater speed, flexibility, and capacity. In addition, these expectations and needs must be met in within an environment that is constantly changing to respond to technology changes, an information explosion, an unstable political environment, and certainly an era of declining resources. The answer: create an "agile intelligence enterprise."
The concept of agility in an organization is not new. Manufacturing enterprises developed the concept of a leaner, flexible organization to meet changing market requirements in a global competitive environment. Early experts such as Rick Dove, Steven L. Goldman, Roger N. Nagel, and Kenneth Preiss at the Iacocca Institute of Lehigh University in Bethlehem, PA authored a number of papers defining the concept. One notable paper was the "21st Century Manufacturing Enterprise Strategy," written by Dove and Nagel in 1991 for an industry-led consortium partially funded by the Department of Defense. The "Agility Forum" has held a number of annual conferences to promote strategies for implementing this concept.
To begin our discussion of applying the concept of agility to the US Intelligence Community, we need to examine the composition of the various organizations that constitute the intelligence producers and the community of intelligence users. Like many businesses, the Intelligence Community consists of a number of individual organizations that perform distinct missions and yet support overlapping sets of customers. These so-called "stovepipes" contain the separate operations that collect, process, and exploit the various types of intelligence data. The very existence of some agencies is "classified information," i.e., a state secret. For example, the predecessor organizations of the National Reconnaissance Office (NRO) date back to the Cuban Missile Crisis in the early 1960s, yet the existence of NRO was classified information until only a couple of years ago. Other agencies, such as the Central Intelligence Agency and the National Security Agency, have existed openly for decades. There are some benefits to having stovepipes: the ability to train and maintain individual specializations, better accountability, and improved security. In spite of these advantages, however, the Intelligence Community has developed separate cultures and artificial barriers that tend to separate the various agencies. For example, the longtime practice of reducing security risks by disseminating information on a "need-to-know" basis -- that is, only to those who have been both approved and deemed to actually need the information to perform their job -- further exacerbates the problems associated with the stovepipe concept. By contrast, much of the private sector has a person or governing board at the top that essentially gives orders and allocates the necessary resources to back them up. This is not always so among agencies in the US Government, and this fact can have a negative impact when agencies must interact in a given scenario.
It is now recognized that these separate stovepipe cultures do not always serve the intelligence users well. At least one of their customers, the US Congress, as well as other "watchdog" organizations within the intelligence hierarchy, may view certain activities as waste or redundancy. While studies have shown that, in fact, little duplication exists, the Intelligence Community must minimize any waste in this new era of fiscally restrained resources.
There are three primary components in the concept of managing information in an agile "Virtual Intelligence" enterprise: an electronic network infrastructure, multidisciplinary teams, and shared data.
Many of the basic assumptions of intelligence have drastically changed since the fall of the Berlin Wall in November 1989. Prior to that time, the free world was faced with a Soviet nuclear threat masked behind the Iron Curtain of Warsaw Pact allies. The US strategy was one of certainty -- total and complete risk avoidance -- with respect to Soviet capabilities and intentions. Today, rather than a single and primary intelligence goal, i.e., the containment of Communism, there is a proliferation of potential intelligence areas of interest. The US must deal with a diverse set of national security interests and focus on risk management over risk avoidance. In this post-Cold War era, the primary concerns are topics such as international terrorism, the proliferation of nuclear and other weapons of mass destruction, economic intelligence, and drug trafficking.
In addition to these new interests, the US Intelligence Community finds a much more complex and ever-changing set of customers or authorized users of intelligence within the government. As the mission and functions of the various customer agencies expand, so does their need for intelligence to support those functions. The challenge, then, is to provide intelligence users the benefit of the totality of the distributed knowledge of the entire Intelligence Community, i.e., to leverage all of the information available when providing a response to a customer. That means they must take advantage of all of the information in all of the individual stovepipes; and they must do this cheaper and faster. As shown in this paper, an intranet -- namely Intelink -- which services the Intelligence Community and its customers is the first step towards providing this needed functionality. Yet, they must not be content with their past successes such as those exemplified by ONI and JICPAC. Rather they must continue to evolve into a fully agile intelligence enterprise if the United States is to be capable of dealing with the intelligence problems it will face in the next millennium.