Dog searches should require consent or a warrant, IMO. Too many cases where the dog is "alerting" based on cues from the handler and not an actual presence of any substances.
The phrase is more commonly translated as: "Who will watch the watchers"
https://en.wikipedia.org/wiki/Quis_custodiet_ipsos_custodes%...
Dog sniff tests like lie detectors should just be forbidden from being used in any capacity like every other pseudo scientific instrument.
The problem is that dogs have evolved to please humans and humans are very biased.
But the only thing they had to suspect Wendy Farris of was being a responsible motorist in stopping when too tired to drive safely. Maybe a breathalyser test could be warranted just to be safe, and if passed apologise for wasting her time.
It sounds like the officers were either running a scam or were just looking for people to harass as make-work. “Field sobriety test”, what is that even? Sounds like a total scam.
IMO the bigger scam is the number of baseless infractions that cops can use to pull you over in the first place. If a cop wants to pull you over, they can find a reason if they follow you for a few minutes.
If there is anything that sounds reasonable about law enforcement in America cops all over America will nearly without a doubt find a way to make it unreasonable.
For example they'll tell you to walk the white line. You'll do it easily. Then they'll tell you to do something else. Then they'll write in the report you failed to remain on the white line for the second task even though it was obvious from the context that wasn't the intention of the officer. The prosecutor will ask the cop on the stand "did you ever tell the defendant to get off the white line" and they'll be able to truthfully say no.
I could easily do all of those things and still be too impaired to drive safely. So it's useless even if the officer has the best of intentions.
This seems like an overreaction. Surely we'd want an explosives sniffing dog walking around at the Boston Marathon, for example. There is abuse when bad cops use $TOOL, but that doesn't mean that there are no legitimate uses for $TOOL. The proper fix is to create a legal and social culture in which government officials of any kind are held to a high moral standard.
It's not like dogs are machines, even small amount of unstructured positive and negative reinforcement can distract a dog from what it was trained to do.
If the Americans manage to clear up this category error, they won’t want to use these defective dogs any more than they’d want to use a dim torch or a cracked magnifying glass.
I mean, "yes, I consent to an easily corrupted dog search?"
People dislike black-box ML and "algorithms", but the approach seems miles ahead of Woof Woof I Smell Drugs as a Service.
This is absolutely true if the only relevant threat in your model is law enforcement action.
In cases where that isn’t the right threat model, the issue is more complex.
The simplest case is when you wee the obvious target of the past crime the police are investigating and it might recur or escalate, hence why people who don’t have a particular expectation of police animus tend to report crimes and freely cooperate with police in investigating them.
But this can be true even where you aren’t the obvious past victim.
If police already are intent on searching you... then you are a subject of investigation.
In the simple case where you know that you are the obvious past target of crime...actually, they still might ask for consent to search beyond what you have initially offered up.
> If police already are intent on searching you... then you are a subject of investigation.
That’s probably technically true, but only because you used a word that you probably didn’t mean; basically, anyone whose conduct is on any way relevant to an investigation is a “subject”, you seem to intend to refer to a “target”.
But even if you are actually a target, unless you are maliciously targeted (which for sure happens far more than it should), inhibiting the investigation may just make you a target longer rather than getting LE to revise its theory of the crime. And the most likely people to be mistakenly-but-not-maliciously targeted (and also the nob-targets most likely to be subjects) are people with some other nexus to the crime, including people who might be in danger from it.
That's all I need to know to feel strongly that dogs should not be used to establish probable cause.
I consider 95% to be highly accurate and reasonably acceptable, which is what I'd want in a drug dog. But I have no knowledge of what it should be, I'm not a dog expert. That's who I'd ask when making the determination of accuracy %. If it's anything less than 95%, though, I'd remove dogs completely.
95% is not highly accurate, it is arbitrary.
The point of my comment is that there should be a standard at which the dogs have to meet.