Why do police have to turn their body cams on before an incident? Why not record the entire duty shift? I know the tech exists to allow it. Wouldn’t it make more sense to record everything an officer does?
Put a button on that flags the next 5 minutes of recording to be skipped if using the bathroom. The reviewer could then skip it, unless something suspicious happens.
If you want, put a second (easier to reach) button on for flagging an “incident”.
So why don’t they record all the time, even the boring bits like shooting people out of their window?
They do record all the time, they just don’t save the records. When a recording is started, the camera then saves the record of the time 30 seconds to 2 minutes before the recording was started.
Not constant recording mainly because of the lack of recording space. Remember that wearable cameras are very small, they do not have a very long recording time. Adding more storage space would make them bigger & heavier.
Presumably, a system could be devised to have a ‘base station’ in the car, with the wearable camera downloading to it periodically. But that would involve adding a transmitting capability to the camera, and no doubt other complications.
Also, there is the problem of storing & cataloging all the recordings from all the cops & dashcams for all the police force 24 hours per day. And then filling all the requests for copies from every lawyer & reporter in town.
Here’s a body cam I found (specs here).
It has 64g of storage. That should be enough for a full shift of video, and if it’s an SD card it could possibly be swapped out for a larger one.
It also has wifi, so it could, in theory, make a direct connection to their laptop to dump off video as it needs to, right?
Analogous would be the cockpit voice recorders on all commercial airline flights. IIRC, they are only accessed by the NTSB and only if there is a problem with air traffic or an accident. They aren’t there for assorted disciplinary fishing expeditions… “Joe’s a problem, he’s too strong a union guy, let’s go find some dirt on him…”
IMHO organizations get the union they deserve. If police unions are really strong, that’s indicative that office politics and vindictive bosses are common. I wouldn’t want my boss “looking over my shoulder” every minute of every working day. I’m sure police are no different. What happens outside that time, what you say to your partner outside of an investigation time, whether you forget to check something that one time, etc. - that should all be unavailable to the boss. Being able to turn off the camera is about as close as they come to privacy.
Also there should be a penalty to the officer who forgot to turn his on when he had the knowledge that he should and the time to do it.
Penalty standards related to seriousness of the event.
Police used to be the good guys. Unfortunately, in today’s world view they are the bad guys who abuse their power unless they can prove they did not.
Wonder how that came to be? A few bad apples that were known but even the good guys would lie under oath due to the blue line or some such.
If you are not willing to be held to a higher standard, don’t be police. If you are good police, keep your house very clean. Don’t and you will be condemned as bad police.
I believe the storage issue of the camera will be a moot point soon if it isn’t already. The storage and management of all that video over the long term is no small problem. AIUI, most of the cost associated with these cameras isn’t the device but the storage of the data. Also, I believe that there is a balance being struck between the officers expectation of privacy and the public’s need to know what is happening in police contacts. I doubt any of us would want to have everything we did and said at work recorded and subject to review. The technology is still evolving and it wouldn’t surprise me if, soon, the cameras are activated by dispatch when a call is put out. There is already a system that turns the camera on when a gun is drawn in testing. As officers get more acclimated to the BCs there will be fewer incidents of “I forgot to turn it on”. It will still happen (sometimes) in sudden, traumatic events.
If the transmission were encrypted no information would be compromised. But as noted, it’s a moot point since eight hours or more of video can be stored in a chip a fraction of an inch across.
I don’t follow this; many people currently (and for decades) do work in environments where they were either constantly filmed, voice recorded, or both… it doesn’t seem to be an issue for them. And many of the recordings of these thousands of convenience store cashiers, casino workers, 911 dispatchers, etc seem to be kept less secure than the recordings of police during their shift. And for the most part, nobody actually watches live or ever does review 99% of the footage anyway. I’m very sure cameras catch cashiers every day scratching their butts, picking their noses, or goofing around on company time; nobody cares. But occasionally when money starts disappearing from the till the footage is reviewed and the cashier can be seen slipping it into their pocket, and the situation is dealt with. I don’t see why the workers in one specific public service occupation need special privacy considerations that most everyone else doesn’t, or why that one group should be exempt from scrutiny.
But when you get to a couple of thousand officers, for 30 days - the storage needs get to be quite large. Not impossible, but complex. We’re not just talking a few body cams. there’s the central servers, the people to run them, the software to delete old video and archive items requested for trial, track who creates what video (after all, an officer with a broken camera could take a different one - need to track that…)
Cost, basically…and policy. While recording everything in a shift using only the onboard storage is simple enough, basically, if they record it they have to transfer it to storage, so that means infrastructure (if they are doing this wirelessly or don’t have the little upload banks) and storage (and encryption), all of which can be quite expensive, especially if you don’t have a really good IT department supporting you. There are training issues as well, plus you’d have to figure out what the policy actually is…do you keep all video for 30 days? 60? 90? A year? 5 years? Forever? Who gets the video and who has the authority to delete video or copy it? If you have a department that has 30 officers in it all of who have cameras, you are talking about 30 GB a day per person, so it can quickly become an issue, especially if specific policies don’t exist…which, in most cases, they don’t. There are usually no state wide policies, there aren’t any federal policies (yet) on this (which means it’s not a mandated requirement, meaning it’s often not funded), and many police departments don’t have a good IT division or the infrastructure needed to support all of this…and offsite storage is pretty expensive and necessitates some basic infrastructure as well.
Basically, the devil is in the detail on this, and most people (even most officers) don’t think through the ramifications of how to make this work. There are also untested legal hurdles (mostly wrt how long and how much to keep) that need to be hashed out. If it seems like something should be a no brainer then you should dig in a bit more to find out why it’s not being done. In this case, there are definitely reasons.
It isn’t just the officers privacy – that’s minor when they are at work; many people are on camera at work.
More important is the privacy of the public who are interacting with the police. Police spend a lot of time talking to crime victims, often inside their homes, often only partially dressed. The first interview with a rape victim, crying & trying to hold her torn-up clothes around herself – that should not be available to the public. Justin Bieber was arrested for drunk driving and strip searched – bet the paparazzi would pay a lot for that footage!
So it can’t be just an automated system to catalog & archive the body camera footage – some person has to go through it and identify it as non-public, and blackout of pixellate the private parts. That takes time & effort, and costs taxpayer money.
And think how much any kind of criminal organization would want body cam footage of a meeting between an officer and a confidential informant. And for many criminal organizations, if they really want it and it actually exists, they’ll find a way to get it. The next time the police meets with the CI is after digging him out of the expressway with a jackhammer and a backhoe.
The post I was responding to, by t-bonham@scc.net, proposed transmitting from the body cam to a base station in the officer’s car, not to a central server. A central server receiving transmissions from the field would be problematic for some of the reasons you mention as well as others, such as what happens when the officer is in an area without cell coverage. However, storage in the central server would not be a serious issue. Two thousand officers sending SD video 24 hours a day for 30 days would use less than 1 TB of disk space.
Why does anyone need to classify it as “non-public”?
This footage is not going to be available to the public. This will only be available to the police department, and to any prosecution or defense lawyers who need it for evidence. Any footage that gets released to the public would have already been vetted through several hands, many of which would have great interest in protecting the privacy of any victims.
I do not see this as an issue in any way, shape or form.
It’s only a serious issue in the real world of budgets and where we use realistic video storage amounts. 30 GB (which is not unusual for a single officer with reasonable quality video on a single shift) times 2000 is a hell of a lot more than 1TB. 10 officers doing 30 GB a day is 300 GB a day…that’s what we are averaging on our system for an officers 8 hour shift. And you can’t just put that storage on a USB drive, once you have it you need to maintain it. That means SAN storage systems, backups and encryption. And this is a storage amount that grows over time, as you don’t just toss out all the video…hell, it’s moving towards having to keep ALL the video for 30-60 days now, with rumblings about 6 months or 5 years. Or forever. You wouldn’t use cell phone connections for this as you wouldn’t be able to keep up, so that means some method to get it from the camera to the server, whether that be wireless (which is how we do it) or base stations connected via trunked Ethernet (which is how we used to do it).
People try and translate what they are used to wrt personal video on their phone to police, but it doesn’t work that way. Once you implement such a system there is then a public expectation of quality and availability. There are factors such as you don’t want this to overly impact any officer wrt administration or overhead of them to use the system. You have to worry about watermarking the data (so there aren’t any calls of foul wrt the chain of evidence and the like). You have to be able to search this data, to review it, which isn’t a big deal but it’s time and software. You have to be able to access it and catalogue it. None of this is rocket science, but it’s also not as blithely easy as you are trying to make it out, and all of it adds up in cost. If you don’t have access to a decent IT department or the infrastructure to do this stuff then it becomes even more difficult…and outside of the really large departments (where it’s going to cost even more for something like this) they can’t really afford it.
Our system has pushed the envelop of storage every year, far beyond projection, as requirements and policy have changed…and the costs continue to go up. I know counties that don’t do this, or do it in such a haphazard way as to make it pretty much useless. And I’ve seen decisions made because of fiscal reasons for officers to only turn on their video if there is an incident, or only have squad car video on if the sirens and lights are turned on.
I didn’t say anything about the difficulty of managing the data, which I’m sure is quite as complex as you describe. I was specifically talking about the idea of the body cam transmitting the video to a car unit which stores the data. There’s no difference in data management between the current system where the video is stored in the body cam and then transferred to your server, vs. the hypothetical system where the video is stored in the car and then transferred to your server. It’s the same amount of data.
You’re right, my calculation was off. I calculated the storage for ONE officer for a month rather than 2000 officers. If you’re getting 30 GB per day per officer, your video quality is higher than I expected, or your camera’s compression is pretty crappy. Based on your numbers, 2000 officers at 30 GB per day is 1800 TB per month, which is indeed a lot of data. Nevertheless, you need to handle the same amount of data on the server whether the data is initially stored on the camera, in the car, or transmitted directly to the server.