Intelligence Center Mines Open Sources

March 2006
By Robert K. Ackerman
E-mail About the Author

The information is at hand; now the community must get its arms around it.

The wealth of information available worldwide from open sources has impelled the U.S. intelligence community to establish a new center dedicated exclusively to exploitation and dissemination of valuable unclassified products. This center will scour the world’s environment of readily available information for snippets of data that could complete a vital intelligence picture as well as for messages among enemies that travel in the open through the global village.

Mundane open sources of information, such as mainstream news media outlets, can provide vital information for intelligence products on a daily basis. Because these products are unclassified, they can be shared with many customers ranging from foreign coalition partners to U.S. law enforcement agencies.

To exploit open-source information fully, the Office of the Director of National Intelligence (DNI) has created the Open Source Center, or OSC, headquartered at the Central Intelligence Agency (CIA). Douglas J. Naquin, director of the OSC, offers that the center has a community role in providing open-source products as well as brokering roles throughout the government.

This center has assumed several missions, some of which are new. As the central clearinghouse for open-source intelligence, it must provide good pull access to its data while simultaneously pushing vital information to customers. It also is striving to set standards for open-source exploitation. And, it seeks to train a generation of open-source experts in exploiting that vast source of information.

“The community in general has a very small, isolated, open-source effort,” says Eliot A. Jardines, assistant deputy director of national intelligence for open source. “There are no common standards for training or dissemination. We are looking to create, in essence, a guild of experts who can lead the way.”

The center is structuring its activities so that it can tailor its support to the characteristics of individual government organizations. It can offer different levels of service ranging from providing full-service information to helping an organization set up its own capability.

“We’re going to consolidate the process of collection requirements and the dissemination process,” Jardines says. The center is not going to “obliterate” elements of open-source excellence throughout the intelligence community, he emphasizes. Instead, it is facilitating open-source work while it also disseminates vital information. The common denominator among the OSC’s services is that it has an open-source product that it can provide throughout the community.

“Any product that we produce is only good if someone has seen it,” Naquin states.

This new center has absorbed the old Foreign Broadcast Information Service (FBIS). Naquin had been director of that organization, but the OSC exceeds it in both scope and function. Naquin relates that the FBIS began to assume more of a community orientation after the September 11, 2001, terrorist attacks.

Although based at the CIA, the DNI OSC serves the entire intelligence community along with the U.S. Defense Department and the U.S. Department of Homeland Security. Its assigned personnel totals have not been made public, but they run in the hundreds. The DNI is looking to double the size of the center in terms of personnel over the next five years, and these new positions will be coming from throughout the intelligence community. Many of those people will be “forward deployed” among other agencies, Naquin offers.

In addition to being the source for open-source information, the new OSC strives to create a center of excellence for building a communitywide infrastructure for the open-source arena. Efforts along these lines traditionally have been initiated by small groups of isolated individuals, Naquin relates. Instead, the OSC seeks to develop an infrastructure both for tasking the community at large and for disseminating vital information.

This touches upon one facet defining the OSC. “We don’t see this center as a monolithic big circle where everybody sends their requirements,” Naquin says. “We see this center more as a hub.

“There are some things that make sense to centralize,” he continues. “You buy data once as opposed to 65 different times. You train people for tradecraft.

“We want to be able to export across the community expertise in open-source exploitation—whether it’s in how to use technology, how to do advanced Internet exploitation or how to do media analysis,” Naquin declares.

The center includes an open-source academy offering about two dozen courses focusing on open-source exploitation and analysis. Students have come from as far away as Japan to take these courses, Naquin says.

Jardines explains that the center is trying to expand the long track record of professionalism at the FBIS throughout the community. “We are trying to create a gold standard of what open-source exploitation done correctly looks like,” he says.

In addition to training, the OSC endeavors to serve as a center of excellence for related information technologies and for data acquisition as well as for the actual production of open-source information. The goal is to export skills and expertise along with information.

Every piece of information posted on the unclassified side is pushed to the classified arena. There, the customers include the military, which uses the secret Internet protocol router network (SIPRNET) and the Joint Worldwide Intelligence Communications System (JWICS), and the other intelligence agencies.

This unclassified information includes both raw and processed data. For raw information, the OSC buys commercial databases on behalf of the community. Each month, the center produces up to 30 analytical products, and its filtering service translates about 30 million words. These types of information are placed in large repositories organized along topical and geographic categories so that users can conduct searches as needed.

All told, more than 98 percent of the center’s products are unclassified, Naquin estimates. The less than two percent that are classified usually are in response to classified requirements. Classification also may be applied when the OSC must protect a classified source. Generally, the use of classified information will remain the purview of the all-source analyst, Naquin maintains.

One advantage to the unclassified nature of the OSC’s information is that the center can support customers ranging from the president to local law enforcement. It already has lined up some police departments as its customers, Naquin notes. The OSC will move more people from outside organizations into the center as well as move its own personnel into these other organizations to serve as links between the two groups.

The center hosts an unclassified, password-protected, open-Internet Web site that serves as a resource for the community. Authorized users can find data organized in what Naquin describes as “AOL-type fashion” to pull the information they need.

Customers also can have information pushed to their e-mail inboxes by establishing user profiles. For example, if a user is interested only in a specific country or geographic region, that user can set up a query preference so that generated information will be sent to that person’s in-box around the clock.

Ensuring that the center is not overwhelmed by the available mass of open-source information is its critical role, Jardines offers. As more agencies bring Internet to the desktop, more individual analysts are doing their own open-source research. However, these analysts may not have the background or language skills. So, the OSC provides value-added products that feature more background on the sources.

“No matter what resources the community has, it will never keep pace with the amount of data that is available,” Naquin adds. “We need to focus as much on asking the right question as on just trying to collect everything.” The center trains personnel to seek out information that may be needed by policy makers. “We want to make sure that whatever they’re interested in is what we’re on top of, and we also want to feed them things that they don’t know they’re interested in yet,” he says.

Jardines points out that getting these customers the information they need also is pegged to ensuring that they receive it in their work environment. If analysts must move away from their classified computers and go to an Internet kiosk, then they are not as likely to use the material regardless of its quality. Conversely, if they must switch from one system to another at their desktops, then having the open-source product in electronic form aids in its utility.

Naquin reflects on how the OSC’s predecessor, the FBIS, often generated good products that were not seen by customers. For one reason or another, the right people did not have access to that information. The OSC cannot allow that to happen, he adds.

The OSC does not offer itself as the last word in open-source intelligence. Naquin allows that the center will be tapping external expertise and capabilities. “There are a lot of people who do open source in some form or fashion very well,” he says. “We are much more interested in bringing—whether government or outside government—those capabilities to bear in serving the entire community.” Over the past few months, the OSC has focused on Defense Department commands and on the Department of Homeland Security. That department has been engaged actively with the OSC in developing open-source capabilities and support, Naquin relates.

Jardines states that the OSC is aiming to create a robust public/private partnership to leverage resources and capabilities from the private sector. Many firms—publishers, academia, the news media—engage in private exploitation of open sources. These firms may be producing resources of interest to the OSC, he says.

And, the center will be working with the private sector to develop new techniques and technologies for open-source collections. The vast majority of the OSC’s information technology work is outsourced to commercial companies, Naquin attests. These include integrators and companies with niche technologies such as large-scale data exploitation.

The OSC uses the CIA’s venture capital firm, In-Q-Tel (SIGNAL Magazine, April 2001), for cutting-edge technologies that may not even be commercially available. These may include video exploitation and filtering, Naquin says.

And, this approach opens the door to adopting and adapting more commercial off-the-shelf technologies. Adapting a commercial technology for the unclassified arena may take as long—or longer—than developing a system in house, which creates stovepipes and expensive legacy systems.

The center has a long wish list of commercial technologies that it needs to accomplish its mission. “When you’re dealing with the potential universe of data, it’s pretty daunting,” Naquin allows. “So one of the things that we look at is anything that will help us filter that information.” This includes filtering tools such as machine translation, video filtering and data mining, for example.

On the production side, Naquin looks for the ability to integrate geospatial products with text and video products within a year or two. This would permit the center to provide podcast-type products from its Web site. There might be other production capabilities that would help merge or integrate various types of media.

The center’s business model includes workers stationed around the world in facilities or even homes. The ability to perform virtual collaboration and production also is key, and new tools may help facilitate that vital capability.

The OSC also can use large-scale Internet exploitation, which is a combination of elements ranging from visualization tools to link analysis—a combination of data mining machine translation and other capabilities. “The integration of all of these different capabilities will help us both filter and portray our products in a much more usable fashion than traditional searches with 16 million hits,” Naquin warrants.

Jardines observes that the DNI OSC can leverage other resources, such as the associate director for national intelligence and intelligence community chief information officer, Maj. Gen. Dale Meyerrose, USAF (Ret.). Gen. Meyerrose has authority over many information technology issues in the community, and he can exercise that authority for the OSC if necessary. Another resource is the associate director for national intelligence for science and technology, Dr. Eric C. Hazeltine. He is focusing on supporting the center’s efforts to target technologies needed for development.

The biggest challenge facing the new center is scaling up to community scale, Naquin offers. The capability that was developed in the FBIS serves as the foundation for the OSC, but it is much smaller than the goal envisioned for the new center. As it grows, it must avoid stretching itself too thin too fast, Naquin points out.

With its mandate to scour open sources for information collection, the OSC operates under the authority that it can collect information in the United States and overseas. However, Naquin emphasizes, it still cannot target U.S. citizens. The center can collect information from a newspaper, for example, but it cannot do so for the purposes of tracking U.S. citizens. That is the purview of law enforcement organizations.

So, while the center can facilitate information collection for a law enforcement matter, it must turn over information to the FBI or to local law enforcement organizations. That applies even to information gleaned from a foreign Web site if that data comprises activities involving U.S. citizens, Naquin explains.

After the OSC turns over to a U.S. agency information about an illegal activity on a foreign Web site, the center can have one of its employees detailed to that agency for a period of time if that organization wishes. Those circumstances are rare, but processes exist for that contingency.

The center is striving to flesh out this issue in a national open-source strategy, Jardines notes. Now that domestic organizations such as the Department of Homeland Security are part of the intelligence community, the handoff of information to these domestic government entities must work smoothly. Simply throwing the information over the wall is no longer an accepted option, he adds.

Issues to be resolved concern roles and authorities of personnel—including those from other agencies and local governments—inside the OSC. The same issues apply to OSC personnel who may be stationed at other agencies’ facilities, Jardines notes. “The bottom line is that because it is unclassified, there is far more collaboration and facilitation than you might find on the classified side,” he concludes.


Web Resources
Office of the Director of National Intelligence (DNI):
DNI Open Source Center (limited access):

Cultural Changes Weigh Heavily on Center Success

Eliot A. Jardines, assistant deputy director of national intelligence for open source, states that greater use of unclassified information raises a cultural issue in the intelligence community that he is trying to change. Generally, the community “develops high”—develops systems that are tailored at great effort and over long periods of time for higher classification levels. Instead, he is challenging the community to develop low—develop technologies and systems for the unclassified level, and then move them high. “Rather than spending a great deal of time and energy trying to get software products and technologies approved for use on the classified system, first we should move to deploy those systems at our unclassified FOUO [for official use only] level,” he declares.

“The Open Source Center has led the way in that regard—deploying the technology at the lowest possible classification so that it obviates the need for highly cleared programmers and likewise a complicated review process to get the software back down to the lowest level,” he adds.

Douglas J. Naquin, director of the Open Source Center (OSC), points out that the center can do more with integrated multimedia, such as using video collaboration tools, because of the unclassified nature of the OSC’s information. Many existing technologies are much more difficult to integrate on the classified side than in the unclassified arena. The center already has begun operationally demonstrating some of these technologies, he adds.

“When you don’t have to deal as much with concepts such as compartmentation or need-to-know or data ownership, the possibilities of being able to share that data are much greater,” Naquin offers. “And, information sharing is one of the key DNI goals.”

Naquin continues that open source has not competed well in the intelligence community because many people have felt that unclassified information is not really intelligence. However, a change in attitude is taking place, and this bodes well for the long-term success of the OSC.

Jardines allows that he is challenged to change a deep-rooted culture that says classification is the imprimatur for the validity of intelligence. The community no longer can afford the mindset of the higher the classification, the more valid the information. Instead, the community must move to a mindset of, “Given the operational tempo and environment, classification is not a good thing but a necessary evil.” This is even more evident in sharing information with coalition partners, he points out.

Enjoyed this article? SUBSCRIBE NOW to keep the content flowing.