A Test of Police Body Cameras Defies Expectations • Hacker News Board

Re: A Test of Police Body Cameras Defies Expectations

#251

Post by icebraining » Sat, Oct 21, 2017, 8:32 AM UTC

BurningFrog wrote: post #242

rconti wrote: post #166

Earlier quoted context omitted.

I couldn't disagree more. The police are there to gather evidence. They're effectively tainting evidence if they mistreat people or ask leading questions or the like. You'd better believe it matters when they come after you for (say) making false statements to the police, and it's their word against yours, and the court always sides with the cops.

This discussion about what police tactics work best very likely contain zero actual cops, but several people with certain knowledge.

On that note, I found the book 400 Things Cops Know - written by an actual Sergeant - to be a pretty good read, reasonably informative and occasionally surprisingly candid.

Re: A Test of Police Body Cameras Defies Expectations

#252

Post by dpark » Sat, Oct 21, 2017, 8:39 AM UTC

FabHK wrote: post #248

dpark wrote: post #51

Earlier quoted context omitted.

> This is literally impossible. Researchers tracked reports of use-of-force incidents, which is factually rather different. This is unfair. You're complaining that they didn't do something that you admit is impossible. They measured essentially everything they could.

I've been wondering about that. What if they gave a third group a prominently visible dummy body cam that did not record, and the police officer knew it, but to others it would look like there was a body cam? And possibly a fourth group a working body cam that was barely visible to others? With variations like these, one might be able to tease out more information about the causal relationships.

Yeah. I agree they could have done more to capture the factors here better.

I wonder about the "dummy" body cam, though. I feel like that's an obvious group to include for the sake of teasing out contributing factors, but I also wonder if that would have gotten them sued. "You were supposed to be filming but somehow the footage is missing and now you claim it wasn't even a real camera?!"

Re: A Test of Police Body Cameras Defies Expectations

#253

Post by matt4077 » Sat, Oct 21, 2017, 9:17 AM UTC

throwaway000025 wrote: post #209

cryptonector wrote: post #87

Earlier quoted context omitted.

Fixing this will have to be the next step. The first step was getting the body cams. Also, you could probably file a suit to get the unredacted video, and you could make a big stink about this. That's how you get to the next step.

The body cams need to serialize their stream to a special satellite or cell provider, and a chain of auditable custody needs to be preserved at multiple data centers. Police cannot be the masters of their own footage. It needs to exist beyond bribery, lust and international intimidation. It needs to be out of their hands and into a web of trust, as soon as the photons hit the sensors. Then it comes down to jamming, t…

> serialize their stream special satellite cell provider chain auditable custody preserved multiple data centers bribery lust international intimidation web of trust photons sensors jamming tampering malfunction sabotage misbehavior recordings wearables turned off recorded

You're trying to solve a legal problem with technology. That's dangerous, because it's hard upper limit is the technology's long-term security. Even worse, it's real upper limit is the min(technology, law)

Re: A Test of Police Body Cameras Defies Expectations

#254

Post by matt4077 » Sat, Oct 21, 2017, 9:21 AM UTC

a3n wrote: post #164

Digory wrote: post #136

Earlier quoted context omitted.

In many (most?) US jurisdictions, bodycam video is cloud-hosted by Taser.[0] I don't have direct experience, but I'd bet Taser doesn't let podunk police departments really "erase" video. And given the volume, you'd think it's got to be built on AWS or another big cloud provider, with pretty good forensic tools. [0] https://www.nytimes.com/2016/07/13/business/taser-internatio...

> I don't have direct experience, but I'd bet Taser doesn't let podunk police departments really "erase" video. They will if they want to keep getting contracts.

Maybe once. But if that's a repeating problem that the legal system can't solve, it'd be a symptom of something so fundamentally broken it really makes police behaviour the least of our problems.

Re: A Test of Police Body Cameras Defies Expectations

#255

Post by mulmen » Sat, Oct 21, 2017, 9:26 AM UTC

kbutler wrote: post #236

mulmen wrote: post #193

Earlier quoted context omitted.

If airline pilots crashed half of their flights we would be right to criticize them. Having a hard job is no excuse for doing it poorly.

And if police catastrophically fail half their interactions, we should criticize them, too. We only end up discussing the tiny fraction of police cases where things go badly.

What percentage of airline crashes is acceptable? What percentage of police abuses of power is acceptable? Half was a made up number. The point is we are right to hold the police to a high standard and demand solutions when they fail in their duties.

Re: A Test of Police Body Cameras Defies Expectations

#256

Post by Banthum » Sat, Oct 21, 2017, 9:26 AM UTC

dsr_ wrote: post #45

jandrese wrote: post #40

Earlier quoted context omitted.

Why is the lifetime so short? Are the batteries non-replaceable? I'd definitely be looking into replacing worn out batteries on $1000 devices.

Oh, it will probably go down over time, but remember that police departments aren't going to go out and buy $399 Go-Pros. No, they'll be buying tactical harnesses and military-tough specially cased relabeled $299 Go-Pros for $999 for each complete system, from a certified minority-owned business in partnership with Halliburton.

The way you chained that together was almost musical.

Re: A Test of Police Body Cameras Defies Expectations

#257

Post by matt4077 » Sat, Oct 21, 2017, 9:29 AM UTC

hedora wrote: post #191

djrogers wrote: post #128

Earlier quoted context omitted.

You would have to prove a) that it was evidence of a crime, and b) the police had reason to believe or know it was such. In the absence of laws requiring full transparency and access to body cam footage, the body cams are just likely to be abused.

That doesn’t sound right. My neighbor is accused of a crime, and upon hearing that, I destroy all the evidence that would exonerate them. That doesn’t count as obstruction of justice / destruction of evidence?

Your neighbour is suspected of brutally murdering their parents in his backyard. One Sunday afternoon, you maw their lawn along with yours because they're really old and it really isn't that much work once you're at it anyway.

In the process, you destroy the evidence exonerating them, i. e. footprints of shoes much smaller than theirs.

Obstruction of justice? Nope.

Now imagine the same thing happened. Except that this time, you include their lawn not just because you always do, but to get them thrown in jail. You're really tired of doing all these chores for them, and it's time for new neighbours.

Obstruction of justice? Yep.

Intend matters.

Re: A Test of Police Body Cameras Defies Expectations

#258

Post by matt4077 » Sat, Oct 21, 2017, 9:31 AM UTC

QAPereo wrote: post #171

katastic wrote: post #156

Earlier quoted context omitted.

>You would have to prove a) that it was evidence of a crime, and b) the police had reason to believe or know it was such. Missing audio and a big "redacted" sticker is a pretty damning evidence that "Something" was removed. IF there was nothing to hide, why did they remove it?

That’s logic, not Law.

Please send nudes. Thx.

Re: A Test of Police Body Cameras Defies Expectations

#259

Post by dmurray » Sat, Oct 21, 2017, 9:37 AM UTC

dpark wrote: post #252

FabHK wrote: post #248

Earlier quoted context omitted.

I've been wondering about that. What if they gave a third group a prominently visible dummy body cam that did not record, and the police officer knew it, but to others it would look like there was a body cam? And possibly a fourth group a working body cam that was barely visible to others? With variations like these, one might be able to tease out more information about the causal relationships.

Yeah. I agree they could have done more to capture the factors here better. I wonder about the "dummy" body cam, though. I feel like that's an obvious group to include for the sake of teasing out contributing factors, but I also wonder if that would have gotten them sued. "You were supposed to be filming but somehow the footage is missing and now you claim it wasn't even a real camera?!"

They don't get sued (successfully) for having body cams and actually destroying evidence when it suits them, so they should be pretty safe with not collecting it at all.

Re: A Test of Police Body Cameras Defies Expectations

#260

Post by matt4077 » Sat, Oct 21, 2017, 9:40 AM UTC

zaroth wrote: post #221

rconti wrote: post #166

Earlier quoted context omitted.

I couldn't disagree more. The police are there to gather evidence. They're effectively tainting evidence if they mistreat people or ask leading questions or the like. You'd better believe it matters when they come after you for (say) making false statements to the police, and it's their word against yours, and the court always sides with the cops.

It turns out that the best way to gather evidence isn't actually polite deference to every wild story you hear! I don't understand your point at all about "tainting evidence" and how that could possibly follow from brusque questioning.

"Brusque questioning" will obviously be applied completely subjective.

Once you start deferring to police officers' immediate judgement, the legal process is tainted to amplify every single stereotype police officers hold.

White victims will obviously be treated with more deference than blacks, Spanish-speaking people will all be assumed to be illegal immigrants etc.

So the young man accused of something will be "brusquely questioned" within the first minutes and confess to the crime. While the old woman is treated with deference, has a chance to talk to a lawyer, and never gets convicted.

And the worst thing is the spiral, self-amplifying nature: start like that, and the sort of people already suspect will be the only ones actually being charged with crimes and convicted, which will reinforce all the ideas the process started out with.