Omaha, Nebraska, police investigate an early morning warehouse fire that points to arson. Included in the incident report are the names of witnesses, including one Joe Downs. The case goes unsolved.
A year later, fire investigators in Chicago, Illinois, respond to a warehouse fire in which two homeless vagrants die. Witnesses include a Joe Downs. This case also goes cold.
While both fires are fictitious, the investigative dead-ends are starkly true, because there is currently no way to correlate both the Omaha and Chicago police reports.
That's about to change. The FBI is getting ready to launch the Law Enforcement National Data Exchange (N-DEx), a means by which local incident, arrest and event reports can be shared on a nationwide basis.
Now, once N-DEx goes live, detectives in Omaha and Chicago can be notified
that the same witness has shown up in multiple arson investigations. Police on
the street also can learn if the person stopped for a traffic violation is
wanted as a person of interest in open investigations elsewhere in the
country.
Information gap
Following 9/11, the FBI took a close look at the database systems housed in its Criminal Justice Information Services division (CJIS), among them the National Crime Information Center (NCIC), Integrated Automated Fingerprint Identification System (IAFIS) and the Interstate Identification Index (III) of criminal history record information (the gun background check system).
"What we noticed was an information gap in national information sharing," says the FBI's David Cuthbertson, the deputy assistant director of CJIS. What was missing were local police incident reports — the basic form of law enforcement record management.
"Those were not being shared on a nationwide basis," Cuthbertson adds.
N-DEx will attempt to orchestrate this integrated justice discord by
harmonizing data entered into local law enforcement systems for common search by
agencies anywhere in the United States. The system is a counterpart to R-DEx, a
regional data exchange system that provides pointers to the location of specific
information held by other law enforcement agencies.
Wish list
An exhaustive national database has been high on police wish lists for decades.
Any citizen can access their bank account through any ATM from anywhere in the country and beyond. Yet when police officers pull someone over for even routine traffic stops, they still cannot get a complete profile of that person's criminal history from mobile technology provided in patrol cars or even from dispatchers.
Other than the FBI's NCIC open warrant database (which by some estimates lists only about 20 percent of the nation's felony fugitives), local law enforcement typically has access to only local criminal databases. Probably the most egregious example of what can happen from this failure to share information through some database network is the oversight of Mohamed Atta, one of the masterminds of the 9/11 attack.
Atta was a fugitive in Florida when he and his fellow hijackers steered jetliners into the World Trade Center towers. Atta, who had an outstanding warrant in Broward County, Florida, for failing to appear in June 2001 for driving without a license, was actually stopped for speeding one month later in another Florida county, but his warrant did not appear in national, state or county databases. Because the Delray Beach, Florida, police officer who stopped Atta in July did not know about the outstanding warrant, he was cited for speeding and sent on his way.
The law enforcement community no longer considers it acceptable that offenders routinely evade apprehension because local police don't have access to a national criminal database. It's particularly appalling because data sharing technology exists.
It's the non-technical challenges of establishing a national database that until now have impeded adoption.
"Security concerns in a computer-savvy world, lack of a widely adopted set of standard protocols, political boundaries regarding ownership of data and funding issues complicate wider adoption of information exchange," comments Gary Bunyard, president and CEO of Tiburon, a Pleasanton, California, provider of law enforcement management solutions.
The intent of N-DEx is to change that.
"When we approached our stakeholders at the federal, state and local law enforcement levels with the question of whether they want to share their police incident reports, we heard a resounding 'Yes,' " Cuthbertson says. "There is a need and a desire to be able to search and share those basic records."
The multi-year N-DEx project will eventually spawn a set of standards and best practices for agencies and vendors to follow in support of a true national criminal database.
Raytheon Co. of Waltham, Massachusetts, was awarded the $78 million contract to build N-DEx earlier this year. Increment One is due to go on the air with live data early in 2008.
"Agencies will then be able to connect the dots between data and link
information across jurisdictions that, at first, appear unrelated," Bunyard
says.
All aboard
There is virtual unanimity of support from the law enforcement community.
"With respect to using technology to solve crimes, N-DEx is the biggest revolution in policing since NCIC was originally created," says Paul Wormeli, executive director of the Integrated Justice Information Systems (IJIS) Institute. "Investigators in the field will be able to correlate suspects with crimes, find suspects based on M.O. and detect criminal patterns."
National law enforcement organizations, such as the National Sheriffs' Association (NSA) and the International Association of Chiefs of Police (IACP), have signed off on the N-DEx idea, provided local agency input is included in design planning and adequate federal funding is provided to ensure local participation.
"The IACP supports N-DEx, but it's a complex issue," explains Harlan McEwen, chair of the IACP's Communications and Technology Committee. "There will need to be extensive federal funding so local agencies can access the database."
In a position statement issued early in 2007, the IACP and NSA, along with
the Major Cities Chiefs Association and Major County Sheriffs' Association,
stated they believe strongly the financial cost of N-DEx must be addressed.
Without adequate federal funds to support local, tribal and state participation,
"the funds being spent on developing the core FBI system will be wasted."
Bottoms up
Local funding is critical because the information used to populate the huge N-DEx database comes from the bottom up — from local and state agencies.
"The most prolific way we will get data is to have data pushed from state and local major metropolitan area record management systems (RMSs) into N-DEx," says Kevin Reid, N-DEx Program Manager. "The majority of the data will be pushed to us into one big central database hosted by the FBI."
Centralization of data is a favorite federal tactic. N-DEx is essentially the next generation of the National Incident Based Reporting System (NIBRS), an early federal attempt to centralize certain types of information useful in analyzing crime trends.
"NIBRS wasn't as successful as anticipated, in part because it was extremely tedious and time-consuming to complete NIBRS-compliant police incident reports," says Chicago attorney Wil Nagel, a former policy analyst for the Illinois Integrated Justice Information System.
N-DEx, however, differs from NIBRS in that no new data is to be created, meaning there are no forms to complete. "N-DEx will collect data that already exists," Reid explains. "We're not expecting agencies to create new data — this data already exists in their RMSs."
Nevertheless, the N-DEx centralized architecture is the opposite of
successful data sharing models using distributed data sharing schemes rather
than one centralized database.
Central v. distributed models
The model for a successful distributed approach to regional crime databases already exists, pioneered by the Portland, Oregon, company Thinkstream. This model, which some consider a technically appealing way to avoid political squabbles related to ownership and logistics of data, does not require local agencies to surrender their data to a global federal bucket. Instead, the distributed network interrogates the data where it's created.
"This is a distributed system that provides real-time access to data at the site where the data is located without having to aggregate or move the data to a national data center," says Thinkstream president and CEO Barry Bellue.
Thinkstream's software has been in place in the state of Louisiana since 2004, where all state law enforcement agencies are connected by a system called LACCIE (Louisiana Civil and Criminal Information Network). County jurisdictions in California, Texas and Florida also are implementing the scheme.
"The patrol officer through any wireless device with access to the Internet, such as a laptop or Blackberry, can run record checks on individuals, check local warrants and misdemeanor warrants, or check stolen property records in any system in the state from any location in the state," says Sheriff Richard Edwards of Jefferson Davis Parish, Louisiana.
When the officer enters a name, race, sex and date-of-birth inquiry, the system now searches more than 300 different databases with one click. Results are typically returned in 20 to 40 seconds.
"Prior to LACCIE, we would have had to do 8 to 10 separate searches to obtain information that is now provided with just one query," says Capt. Bobby Baker, Ouachita Parish (Louisiana) Sheriff's Office.
Edwards advises agencies to move forward to sharing information. "We have not
in any way created a new database, nor are we mining data," he says. "We are
reading other systems and sharing that information via LACCIE."
Growing privacy concerns
Regardless of system architecture, one of the most serious snags in any national crime database relates to the issue of privacy invasion. The larger the database, the greater the concern the safety is off and the system will be neither accurate nor secure.
Many of the data protection laws passed in the United States that limit the government's collection, use and dissemination of data about its citizens evolved because well-meaning, but over-zealous, government officials abused the information.
Data accuracy is one issue. "How do we insure that the information in the N-DEx database is accurate?" asks Christopher Calabrese, counsel for the American Civil Liberties Union's Technology and Liberty section.
It's not like conviction information where someone has gone through a trial or plead guilty to a crime and law enforcement can have some confidence the information being reported is accurate, he notes.
"Here, you are talking about accusations, and some of them will be unfounded," Calabrese says.
For instance, suppose police respond to a domestic abuse complaint. They come to the house and take a report. The report goes into the local RMS. Later, the investigation determines the accusation to be unfounded.
"It's not clear that N-DEx would require a second report saying the incident was investigated and found to be unfounded," he explains. "The police have no incentive to keep the system updated, and initial reports could be left hanging in the system — potentially damaging to someone's career or reputation."
Security is another issue. Crime data is of potential interest to parties outside law enforcement.
"Everyone from landlords to insurance companies to criminals are going to want to see what's in this database," Calabrese says. Criminals, for instance, might want the identities of witnesses against them or the status of investigations.
Cuthbertson says N-DEx complies with all existing CJIS division security policies. But while the FBI may have a strong interest in keeping the database secure, Calabrese says in many ways N-DEx is a more powerful system than they have probably ever compiled before.
"Whether they are up to the challenge of keeping it secure I think is a fair question," he notes.
If N-DEx can meet the challenges placed before it — data accuracy, security and a sound architecture — it may just be the information sharing silver bullet the law enforcement industry has been looking for.
Douglas Page ([email protected]) is a science and technology writer living in Pine Mountain, California.
