Richard Stallman is famous for creating the GNU operating system, and founding the free software movement, which changed how we develop software. Now he's published an essay in Technology Review about how police body cams should switch on automatically any time an officer pulls out a weapon.
Stallman's proposals are somewhat similar to what we've seen previously from groups like the ACLU, which crafted a model bill for regulating policy body cams. In that model bill, the ACLU suggests that police should turn body cams on whenever they respond to a call or interact with a member of the public — except when it would be dangerous for the officer to turn the camera on. Here's the exact language:
Both the video and audio recording functions of the body camera shall be activated whenever a law enforcement officer is responding to a call for service or at the initiation of any other law enforcement or investigative encounter between a law enforcement officer and a member of the public, except that when an immediate threat to the officer's life or safety makes activating the camera impossible or dangerous, the officer shall activate the camera at the first reasonable opportunity to do so. The body camera shall not be deactivated until the encounter has fully concluded and the law enforcement officer leaves the scene.
The problem with this rule, however, is that it relies on the police to determine when to switch on their cams — and it leads to blackout periods during violent confrontations. That's why Stallman suggests a technological fix.
He proposes that body cams should automatically go on during certain "significant events" within a distance of "perhaps 50 meters." That would mean, essentially, that police couldn't argue that they were in too much danger to activate their cams, because they would automatically activate at these moments. Stallman suggests these significant events should include any time the officer wields a weapon. Here is his proposal:
1. Whenever the agent removes a gun from its holster.
2. Whenever the agent takes a weapon in hand to use it. Weapons would include guns, tasers, sticks, and others.
3. Whenever the agent pushes a button to declare an event. Agents should be trained and required to do this when they see a violent attack or an injury, and then to aim their cameras at least briefly toward whatever they saw.
4. Whenever the system's microphone detects a gunshot.
Stallman goes on to argue that footage should be uploaded promptly, and retained until "citizens can verify that they are not being watched without grounds." He also suggests that footage should not be released to the public or officers unless "a judge rules that they cover part of an act of violence." He does not explain what kinds of tech would be involved in building this automated event detection system, but one has to assume it would involve a set of algorithms that would be constantly evaluating sensor data coming from microphones and the officers' movements.
It's interesting to get this perspective from someone who has worked for so long to create transparency and freedom in software development. Essentially, he's asking that we use software to verify that there's transparency in law enforcement, too.
[via Technology Review]