Markdown Extra syntax via https://michelf.ca/projects/php-markdown/extra/

Problems:  You would only be able to do this once, unless you had a trustable method of generating this pseudorandom stream again – entering the generation seed would allow an attack on the system?  I’m probably getting this wrong.
This way, the original ‘signer’ of the video – presumably the operator of the signing device – would be able to validate the stream’s integrity for a court or other examination by providing the key to generate the same pseudorandom stream again – which would be then compared to the pseudorandom stream extracted from the playback.  If they match, then the video has not been edited since it has been signed.
@Mike Sherwood
… if there is continued resistance after being tazered …
Trust me, there is no continued resistance after being tasered. You may have to BBQ a PCP twice, but that’s it.
However, where I failed was the production of the pseudorandom stream seed – that’s required to validate the stream integrity (as opposed to individual frame integrity).  You can’t give that out to anyone, because then they could generate the same stream – you have to assume that they could reverse engineer the pseudorandom algorithm.  But I can’t think of a way to validate the stream that doesn’t also require generating the same stream – which also requires that same key, not a paired key.
As for the camera, I am not opposed to its use.  I think that it would provide an effective tool to ward off false accusations, reports and complaints against honest officers.  Everyone seems to think that the police dislike having the cameras in their cars, and the idea of cameras on their Tasers.  Its just the opposite.  We are thankful to have them because of the enormous amount of BS complaints that are filed against us.
“The camera would only capture relevent events when turned on and pointed in the target’s general direction. The events leading up to the officer’s decision to acquire the target would not be recorded. To me, it seems the camera would be of limited use – particularly when the decision was made in haste.”
I swear there was an article on Slashdot.org about some guy going into places with camcorders strapped on him, to film places taking video footage of anyone, to drive a point, something like: ‘where is our OWN record(s) of the footage?’*
If the video signal were relayed in real-time to a manned monitoring station, then the operator, on seeing an unusable signal, should be able to remotely disarm the weapon.
While I’m all for increased police oversight, I don’t think that the sort of actions you’re suggesting are realistic.  After all, there’s no system (that I’m aware of) that sends video from cruisers’ dash cameras to a central location for real-time viewing, nor do we hear about cops disabling their cameras in order to harrass citizens during traffic stops.
Anyone thought of just storing the image in read only memory? All these ideas sound great but they are a long way from being possible.
I am sorry to hear about your nephew that died.  If you have a COD that shows the TASER was at fault, I would like to see it, or you can send it to TASER International.  As an instructor, I have taken the ride on multiple occasions, and although it doesn’t feel any better each time, I get right back up.  As of the last course I taught, there have still been no deaths that have been linked to TASER use.  I just want to set the record straight for fairness.
That would be easily accomplished by recording the video wirelessly into the cruiser then automatically sending this to the central data center.  Prevents loss of data and or later tampering.
The Taser Cam is an audio and video recorder that attaches to the butt of the gun and starts taping when the weapon is turned on. It continues recording until the weapon is turned off. The Taser doesn’t have to be fired to use the camera.
Now, to help better examine how Tasers are used, manufacturer Taser International Inc. has developed a Taser Cam, which company executives hope will illuminate why Tasers are needed—and add another layer of accountability for any officer who would abuse the weapon.
Ride the lightning! Related but not directly connected to this thread, before the cops can be allowed to use pepper spray and tasers, they get hit themselves. This is to let the cops know exactly what they are putting the perp through. Already the tasers have a shots used counter on them and each shot has to be reported. I think adding a camera is a good idea. The events leading up to using it would be covered by the dash cam, and the taser cam would give a better close up view of what happened. Sometimes the dash cam can’t capture the important stuff because of its fixed location.
Of course it has limited use. All security has limited use.  But it seems like a really cheap, really useful, auditing tool — a good security trade-off.
On-the-fly real-time generation of cryptographic hashes of every frame of the video followed by a gpg signature? Would this be possible?
The good part is that if there is continued resistance after being tazered, or the subject is not an immediate threat, the video could help provide justification.  The down side is that the officer has to have the weapon trained on the subject already for the video to provide evidence of the attackers agression.  A video that is just the blur of the tazer being drawn from its holster and fired is of no use.  Expecting the tazercam to provide justification requires that police officers escalate to the threat of less-lethal force earlier in their encounters.  In threat escalation terms, this puts the officer in the position of making the situation more hostile rather than trying to diffuse the situation.
(*)If anyone finds the article(s) I was talking about above on Slashdot or perhaps Wired, please post the URLs here as they did drive an interesting point.
To put this in perspective, in training, people are taught that a knife wielding assailant at 7 yards has an advantage over a person with a gun in a holster.  A tazer is a short range weapon, meaning that a failure of the tazer to operate correctly puts the user in greater danger because they thought they were armed, but were not.
If the signal were also relayed in real-time to a volunteer group of civil rights lawyers, the cops would know it might be hard to ‘disappear’ the evidence. The threat of prison terms would tend to curb cheating by the police.
If you used the GPS pseudorandom stream, you’d have a ready-made way to encode location/time of the video.  If you encrypted each frame’s GPS location/time pseudorandom using the private key, then during the recovery, you could use the public key.  The proper GPS pseudorandom stream for a particular time/location is a matter of public record.  Thus you would have a means of encoding time, place, and integrity with a public/private key verification.  Does this hold up?
Of course, this still wouldn’t be perfect, but you would have an advantage of knowing which officers ought to be at desk jobs and which ones handle the stressful situation of a physical confrontation with a subject well.
On the whole, I agree with Bruce in that this is a relatively cheap auditing tool that can help prevent misuse, or at least enforce consequences of misuse.
why can’t the power be lowered to half of the 50,000,bringing down the # of deaths,which deaths are going to increase more and more,because of more users! the large big cop lasering a young woman,several times should have to take his chances by recieving the same amount lasers shots? I believe,if you held this steady for a time,you could actually electricute a person! laser deaths have doubled in the last few years!
Join thousands of FFLs who, since 2010, have trusted FastBound Electronic A&D and Electronic Form 4473  with a guaranteed legal defense backed by attorneys.
Eventually, We should have a much more robust monitoring solution.  Imagine not only a visual record and an audio record, but bio monitoring on the officer as well (think the combat monitoring system from “Aliens”).
The camera would only capture relevent events when turned on and pointed in the target’s general direction.  The events leading up to the officer’s decision to acquire the target would not be recorded.  To me, it seems the camera would be of limited use – particularly when the decision was made in haste.
Multiple Tazer hits are not uncommon (that is more than 4). While current is being applied the target is not able to move, but once the current ends recovery is very fast. If no pain is felt the threat returns.
In general, though, I’m a little leery of cameras as an auditing tool, since they still leave a lot open to interpretation.  Cameras in patrol cars have the same failure problem -> they can make a visual record of a traffic stop (which is great when someone shoots a cop and you now have a record of a license plate), but they can’t record audio, so you’re still going to be relying quite a bit on people’s interpretations of an event.
This is where cameras do good: when they lessen a power imbalance.  Imagine if they were continuously recording the actions of elected officials—when they were acting in their official capacity, that is.
Having a tattletale built in will encourage the police to tamper with the recording apparatus. As an example, simply break the tape before using. Or insert a cassette that’s at the end of its reel. Or smear a glob of mayonnaise on the lens. All of those can be explained away as ‘accidental’ disabling.
(What’s interesting is that the taser manufacturer is supporting this, in an attempt to defend its product against allegations of misuse.)
The Taser Cam records in black and white but is equipped with infrared technology to record images in very low light. The camera will have at least one hour of recording time, the company said, and the video can be downloaded to a computer over a USB cable.
Being able to remotely disable a weapon is a terrible idea.  It endangers the person expecting the weapon to work and shortens the amount of time available to discard the nonfunctional weapon and employ another tactic.  Less-lethal weapons are employed more readily than firearms because they are not expected to cause permanent damage or death.
I think the cameras are a good start. As for tampering, anybody could tamper with it.
Like the old saying goes, “A lock keeps an honest man honest, locks don;t keep thieves out.”
My nephew was killed by a Taser gun and if it had a camera on it, we might know the truth today about his death.
http://www.citizens4taser-reform.com
This year, like in the past, there have been numerous reported stories of people dying in police custody or being arrested.  Camera’s shed light on what happened, when, where and how.  Leave the why to the judicial system to figure out.  And actually there have been many police brutality videos revealed.  Sadly many more people have died then will ever be known.  The facts just won’t be revealed.
I am a public-interest technologist, working at the intersection of security, technology, and people. I've been writing about security issues on my blog since 2004, and in my monthly newsletter since 1998. I'm a fellow and lecturer at Harvard's Kennedy School, a board member of EFF, and the Chief of Security Architecture at Inrupt, Inc. This personal website expresses the opinions of none of those organizations.
The audio may be a better tool than the video in many cases. Things move fast in a use of control situation but the audio will tell you a lot about the emotional state of all involved.
The camera adds information that can be used in both positive and negative ways. As long as the limitations of what is being shown is understood then that is a good thing.
Ideally, you’d be able to generate two streams, which were the ‘private and public’ key of each other, but I don’t know if that’s mathematically possible, I’m too stupid. 🙂  If you could, then you could validate the stream integrity without providing the key.
What I meant to say was that if you use a private/public key system, there should be a way to implement this such that the operator can provide their public key to allow anyone to validate the videostream without being able to reproduce it.
A TASER didnt kill your nephew JP and you know it. And the camera does not use a tape Mr. Owens as you said. Nor would the police smear mayonaise on the lens. Where do you people get your thoughts???????? The problem is not the police it is the morons who need to break the law and fight with the police. WE dont get paid to fight with idiots. DONT BREAK THE LAW AND YOU WONT DEAL WITH THE POLICE IN A NEGATIVE WAY. GROW UP AND STOP THE PHONY FALSE UNTRUE MADE UP STORIES ONCE AND FOR ALL…..
In the event of a “questionable” shooting, for example, you’d have not only a much more authoritative record of the victim (with the audio and video), you’d have a record of the cop’s pulse, blood pressure, etc. which gives some indications as to the officer’s physiological state during the event.
This brings to mind something I was noodling over earlier…if you have a video record of the taser in use, it makes sense for the video record to incorporate anti-tampering.  I had asked on my blog (at http://www.sharp-tools.net/archives/000245.html) if in fact it was viable to build into a video device some manner of tamperproofing – perhaps using pseudorandom streams and keys embedded into the video – such that the ‘continuity’ of the video could be later verified for a court.  I’m not enough of a mathematician to know.   It seemed to me, though, especially if people were going to be taking cameras to public places to record events (protests, etc. – the NYC RNC protests and associated camera footage use were what triggered this thought) then the ability to have a device which embedded this antitampering track into the video would be good.  IIRC, the police footage from that trial had turned out to be edited, and the court had not been told.
FFL eZ Check is a tool provided by the Bureau of Alcohol, Tobacco, Firearms, and Explosives (ATF) for

                            easily checking whether or not an FFL license is valid and verifying the shipping information. This is

                            vital for an FFL dealer or any person performing a firearm transfer of any kind.
Of course, cameras are only as useful as their data.  If critical recordings are “lost,” then there’s no accountability.  The system is pretty kludgy:
Anyhow, it struck me as something that could be implemented as part of a camera or as a video processing pass-through unit.  Even if performance was an issue, at worst case, you could have a separate ‘lower framerate’ version of the video which was thus encoded to compare to the original.
It would be nice to know if the data is protected against tampering. I wonder if other data is recorded such as durration of current applicaiton?
There’s a few problems that I can see from my uninformed perspective…you’re trying to validate the identity of the camera (and user?), the integrity of the frame, and the integrity of the stream.  There would absolutely need to be a GPG-type key system implemented, but what gets done with that key is the hard part.  I think what I suggested doing (in an off-the-cuff musing) was hashing each frame with the public key and with a pseudorandom stream generated from a secret seed – probably the private key.  The results of the hash would be encoded into the video stream, and some low-rez version of the hash could be placed into the stream as a visual artifact.  That way, even if the video stream was imaged and then re-edited, you’d know that something was wrong, even if you couldn’t fully validate the original datastream, when the visual artifact symbols indicated a break in flow.