Exactly. Some people seem to have the opinion that a thing is good if it generally proves effective a capturing criminals. Others have the opinion that capturing criminals is only good if there's a very low risk of innocents being caught up in the process.

I'm in the latter group.

In general, though, doesn’t more (and more reliable) information permit more accurate conclusions? Wouldn’t this type of data reduce the risk of innocents being caught up, compared to asking cops or neighbors to hazard their best guess at who might have been around at the time of the crime?

> Wouldn’t this type of data reduce the risk of innocents being caught up

This really depend on how you use the data, doesn't it?

If you have a presumption any suspect is innocent, and investigate until you have overwhelming evidence, yes, it does.

If you have a presumption whoever the algorithm selects is guilty, and investigate to prove it, this gives you exceptional capabilities to persecute whoever you catch.

I definitely agree that it depends on how you use the data, although that seems to bleed into a judicial rather than an investigatory role.

In the case of this geofence kind of data request, though, it seems to self-enforce that a little bit, though, right? If you cast too broad a net, then you “catch” hundreds of people, and the jury rolls their eyes at you pointing the finger at any arbitrary one of them. If you construe the request narrowly, then you get a small number of leads, but those leads are in actual fact that much more incriminating by dint of how precise they are.

In some sense it reminds me of the way the squishier sciences dealt with p-hacking by normalizing preregistration: you kind of have to set the “power” of your request ahead of time, or the black box that spits out the results becomes less useful to you (and less convincing to the people you have to make agree with you).

Yes. And law enforcement seems to have consistent problems with p-hacking any new kind of evidence they are allowed to have. That's why the use of things like this tends to be denied.

That said, if some country manages to create a law enforcement organization with the right culture, it does indeed become much less of a problem. But it needs to guarantee the culture won't change either, and that the data won't become available for different organizations.

What makes you think the government cares about accuracy? I’m pretty sure they only care about the size of the net, not what it catches.

I think cops (or at least their bosses) care about their arrests resulting in successful prosecutions.

A request ambiguous enough to net hundreds of suspects seems unlikely to net the investigator a successful prosecution. Not without corroborating evidence, which might be the fruit of a thorough investigation.

But if the initial request leads to an investigation that develops enough evidence to prosecute somebody—that is, if the person really did do the bad thing, and this was one of the ways the government figured that out—what is it that’s so abhorrent about this technique that makes it right to overlook the bad deed?

If the status quo is throwing around nonsense like bite marks and sneaker prints to try and associate somebody with a crime scene, geofenced mobile data requests seem like a smaller rather than a bigger net compared to “everybody with teeth who I have a hunch about” or “everybody who owns Sketchers.”

> I think cops (or at least their bosses) care about their arrests resulting in successful prosecutions.

A successful prosecution is measured by a guilty plea or a finding of guilt. It is not the same thing as justice or convicting the right person. There are mountains of cases where it is clear that prosecutors are playing with dirty tricks for a conviction of anyone, rather than seeking to convict the right person.

I’m ok with TVs and jewelry getting stolen occasionally if it means my government is not constantly tracking my every move

Isn’t the whole geofence request paradigm an elegant compromise to address that, though? The requestors don’t get to track everyone’s every move. Instead, they have to specify exactly where and when based on a thing that actually happened, and the private corporation controlling all that location data decides whether or not the request is narrow enough to answer.

A mechanism like this allows them to realize the social benefits of that kind of a data trove existing, while providing some kind of a check on the way they use that data. However flimsy that check may be, it still seems really different from the government itself collecting and controlling all that data itself. And if anything, it takes the wind out of the argument that government properly should be the custodians of that kind of data: they can use it in the rare cases where they can describe a clear purpose, but they can’t just go frolicking through the movements of every person in their jurisdiction for funsies [0].

Is it necessarily bad that, in the US system, it turned out to be just the tacky companies slinging ads who control that utterly comprehensive archive of spy data, and it’s the guys with guns who have to ask nicely to access it?

[0] at least not through mobile geolocation data. Contrast with Flock Safety’s nationwide mutual data exchange compacts: https://www.newsobserver.com/news/state/north-carolina/artic...

That happens all the time during normal, legal place investigations. What's wrong with that? The concept of an investigation almost necessarily requires there to be innocent people involved, to differentiate the guilty from the innocent.

Eh. There is a big difference between an investigation ( as one sees in most American cinema hence why this particular phrasing is likely chosen ) and a fishing expedition, where LEOs go through data they in likelihood should not ( and to make it more annoying -- are highly unlikely to delete once done ).

I am saying this as a person who got LEO request for JUAN no last name and no other identifiers. As you can imagine, some of us are not amused by such requests.

edit: Location is much, much more telling than basic name and address.