Editor’s note: This article originally appeared in Police1’s digital edition, “Empowering law enforcement through data sharing.” Click here to download this free publication.
The concept of “data sharing” may sound ultra-modern, but it’s really a very old practice. Before there was the internet and instant access to information of all types, there were wanted posters displayed in post offices and other public buildings. That may have been primitive when compared to present-day methods, but it worked to put thousands of criminal fugitives in jail. People were eager to put a desperado in jail, especially if there was a reward for turning them in.
Now that we have the ability to reference nearly any topic on a whim and receive results instantly, data sharing is much more efficient. Law enforcement agencies assist one another by making their data available to other departments and keeping the citizens they serve informed, making them partners in the effort to suppress crime.
The trouble with silos
Law enforcement databases tend to be “siloed.” Information such as lists of identifiable stolen goods, addresses of chronic trouble spots, the names and affiliations of street gang members, and the names of people with active arrest warrants all reside in indexes that don’t interact with one another. Further, these information siloes tend to be accessible only within the agency that creates and stores the data.
A detective who has identified Joe Sixpack as his prime suspect in a string of residential burglaries may be unaware that Joe has an active arrest warrant in the next town over. Joe would be much easier to interrogate in jail, rather than wandering around selecting his next target.
National Crime Information Center (NCIC)
The first U.S. effort to share data between law enforcement organizations was in 1967, with the implementation of the National Crime Information Center (NCIC) by the Federal Bureau of Investigation.
Before NCIC, a police officer making a traffic stop had no way of knowing that his violator was driving a stolen car or that the car was used in a violent armed robbery unless that information arose from his own employer’s investigations. NCIC housed databases of stolen vehicles and other property, wanted persons, missing persons, and the first national database of people with criminal records. The data came from the same agencies that used it.
Making inquiries of NCIC required using technology we regard as primitive now. Operators had to first type the inquiry into a teletype terminal that recorded the keypresses onto a paper tape by punching holes in the tape. The tape was then loaded onto a reader, and the terminal connected to it dialed up the closest access line over a regular telephone circuit. Responses to inquiries were printed onto a sheet of continuous-form newsprint, which was torn off and given to the operator who initiated the query. Often, the response was that there was a typo in the inquiry message and that the inquiry had to be composed and sent again. It would be years before the interface was updated with a conventional keyboard and video display, and even longer to put this technology into patrol cars and handheld devices.
Most of the data contained within NCIC was contributed by the police agencies that used the system, with the FBI supplying the infrastructure. Local agencies, many of which were still running radio rooms with time-stamped incident cards to record traffic stops and calls for service, didn’t need to provide any computer gear except the teletype terminal and a dedicated telephone line.
From paper records to CAD/RMS
Most law enforcement agencies maintained some or all of their records on paper stored in file cabinets. Periodically, older records were optically scanned and archived onto microfilm or microfiche to save space and keep the records from being destroyed by mold, water, or rodents.
When computer systems matured enough to occupy a desk-size enclosure instead of a large room, local police acquired the first computer-aided dispatch/records management systems (CAD/RMS) and started storing much of their data digitally. It still took another 20-30 years for police to move the bulk of their initial data collection from paper forms to data fields on an electronic display.
That transition made data sharing among police agencies much easier, especially if those organizations used similar or mutually compatible CAD/RMS. Those with dissimilar CAD/RMS operated with a handicap, sometimes remedied in part by separate software solutions designed solely with data sharing in mind. These were used extensively by small organizations operating in proximity to one another, as with rural areas served by small police and sheriff’s departments.
Now that nearly everyone is using some form of CAD/RMS, many of these small agencies have seen the wisdom in buying compatible CAD/RMS so they can communicate and share data more easily.
Robust data networks
Faster computers and multi-channel, more robust data networks have also improved the delivery of information to officers in the field.
NCIC originally delivered information over plain vanilla twisted-pair telephone lines. As more efficient and more accessible transmission methods became available, law enforcement agencies slowly added those conduits to their data centers. Wireless data, ported over cellular telephone networks and multiplexed public safety radio channels, now delivered directly to the officer in the field. That officer, deputy, or trooper no longer needs the assistance of a dispatcher or to be next to his patrol car radio to get the information he needs.
The generation of officers now entering police service for the first time has never known a world where they can’t access information almost instantly and from multiple sources via their desktop or laptop computer or smartphone. They get frustrated quickly when they don’t have whatever answers they need immediately available. Data sharing helps satisfy that desire for answers on the go and improves retention of officers who were recruited and trained at considerable expense to their police employers.
Crossing the boundaries
Law enforcement agencies necessarily observe operational boundaries. A city police department rarely responds to calls for service outside the city limits, leaving those calls to the sheriff’s department. The criminal element does not feel the need to constrain their activities to one jurisdiction, striking wherever it’s convenient and opportunity presents itself. When crime data isn’t shared, two adjacent jurisdictions can be working to find the same offender without knowing of one another’s efforts, or even know that their crook is capering over a wide area. Pooling data helps reduce duplication of effort and running down unproductive investigative leads.
Benefits are realized from the sharing of data between industries and services helping to achieve the overall goals of public safety and allied players on the landscape. A study conducted in Cardiff, Wales in the United Kingdom tracked police and hospital activities regarding woundings and less critical assaults over a 51-month period. When compared with 14 other cities deemed “most similar” by the British Home Office, hospital admissions dropped from seven to five per 100,000 population in Cardiff, as opposed to an increase from five to eight in the comparison cities. Police responses to calls for service in Cardiff involving wounding increased from 54 to 82 per month per 100,000 population, but the comparison cities experienced a rise from 54 to 114 calls per month.
Citizen reaction
When a non-police organization shares data with law enforcement, there can be a political backlash. Many people recoil at the thought of any of their personal information being available to the police. This can be offset when the sharing produces a desirable outcome. After the investigative details of the Golden State Killer case were made public, a survey indicated that 80% of people supported the use of genetic testing businesses by police investigators when the objective was to resolve an especially heinous case.
This public support of data sharing with police is far from universal. Many users of social media such as Facebook and Twitter assume access to their accounts is heavily restricted and delivered to law enforcement only on receipt of a warrant signed by a judge. They would be even more annoyed on learning that Google received over 39,000 requests for user data between July and December 2020, and responded to over 80% of those requests without requiring a warrant or subpoena. Those requests included data from the accounts of more than 89,000 users, most of whom were and are unaware their accounts were made available to law enforcement.
The world becomes more complex with each sunrise and information stores are only growing larger. With the right access, nearly any aspect of someone’s personal life can be revealed to an investigator. Some of those investigators are cops, and others are people who have criminal ends in mind. Data sharing by and between police agencies is needed more than ever, but that process requires sensitivity and finesse to keep from being labeled as Big Brother. When it comes time to buy new CAD/RMS platforms and other data storage solutions, consider the value of being able to share data with your partners in other agencies and communities. A properly vetted selection can make your operations more efficient and effective.
Data sharing beyond public safety
Effective data sharing is not restricted to public safety and other organizational players. California law enforcement agencies obtained the assistance of genetic profile company 23andMe to identify new leads in the Golden State Killer case that included murders and rapes over 44 years old.
The customers of 23andMe submit a saliva specimen that necessarily contains buccal cells. The DNA in the cells is mapped and compared against other 23andMe customers, identifying family members the customer often does not know.
A separate company, GEDMatch, consolidates genetic data collected by 23andMe and other genetic mapping services.
By comparing a surreptitiously obtained DNA specimen from Joseph DeAngelo against GEDmatch, detectives from the Sacramento County Sheriff’s Office obtained an arrest warrant for DeAngelo. He eventually pleaded guilty to thirteen counts of first-degree murder and kidnapping and was sentenced to life in prison without the possibility of parole. His guilty plea was in exchange for an agreement by the prosecutor not to seek the death penalty.
