The program is terrible. Collecting data on persons not under active investigation is simply wrong and contrary to our laws.
That said, the program seems more focused on high crime low income areas than the “highly racialized” description warrants.
I agree with your first point. While there isn’t a legal expectation of privacy out in public, I think there is a basic human expectation of being left alone if you’re just existing.
For the second part, that may be true–it’s hard to say without some quantitative measure–but in the southeast part of town there’s a visible lack of license reads in a low income area with significant crime. It’s such a start drop off that I wonder if some data was lost. The “ALPR Data vs. Crime Data” map at the EFF shows this.
I’d also like to see a normalization for population density and/or traffic density.
The difference is not immediately obvious to me. Discrimination does not need to be intentional to be real.
you can pretty easily see they don’t go into the hills neighborhoods. just up and down broadway terrace…
I’m surprised that drivers haven’t started covering up their plates when their cars are parked so that their data can’t be recorded. Assuming it’s not illegal, it seems like a valid form of protest against an intrusive program. Of course license plates would be recorded if they’re uncovered while driving and you encounter a patrol car. Maybe someone should come up with an automated license plate cover that activates when your vehicle is turned off.
That would be illegal.
Would it? If you park your car in your garage, you’re protected from the gathering of data. Another form of unintended discrimination (against people who can’t afford garages). What if you propped a folding chair against your car to cover up the license plate? Is this illegal?
As an oakland resident and ashamed former supporter of the EFF this bothers me not at all. Nobody has the right to operate a vehicle upon a public road in secret. Cars are a menace and a nuisance and the biggest threat to life and property a city faces. Having their plates photographed is a minimally intrusive yet highly effective check. If these do nothing other than solve oakland’s rash of hit and run crimes that will justify the entire program.
Nobody is talking about operating the vehicle in secret. The plates would be covered only when the vehicle is not being operated. The main use case I can think of for bulk scanning plates of vehicles, particularly those not in active operation, is to track the movements of people not under investigation (alternately: to place everyone under investigation at all times).
You assert the effectiveness and minimal intrusiveness of the program despite having no possible evidence for the assertion - the cops themselves are trying to be secretive about it, so nobody can know whether it’s effective, or if it is, toward what ends it is effective.
The automatic readers can’t tell the difference between a license plate and a plate-esque page of a different size pasted to a telegraph pole, a computer monitor in a house window, a tablet on a mailbox, a sign in your yard. They scan everything that fits the profile.
The database can be salted if enough people decide to protest.
Another option might be to use some tech which obscures your license plate to the readers, but not to people. How this might work depends upon what tech the use. For example, strong infrared LEDs near the plate. Yet another approach is to attack the license cams directly on the police cars. A dozen or so pissed off residents with lasers could fry them faster than they can be replaced.
I recall the Burbank, CA police ticketing a guy who would put a car cover on his car when he parked it – he had to cut holes to expose the license plate to make them happy.
It’s easy enough to put that down to looking at only 8 days of data. That area may have been heavily ALPR’d the day before or the day after the data set for all we know. We also don’t know if the police cars assigned to that area are even equipped with ALPR.
Again, I condemn the practice but, until I can see more data, calls of racialized policing seem more of a biased based conclusion than a fact based one. Besides, there are so many real concrete examples of racialized policing out there that applying that label to this data seems a waste and possibly even a distraction.
We saw how politicians in NJ shamed people based on their EZ Pass history, and it seems likely that some unscrupulous people with access to ALPR information will abuse it similarly. Then there’s the stalking aspect to it. And the overreach.
This is why I donate to the EFF.
Somehow I don’t think this program is targeting reckless drivers.
As someone who has used similar data in another city I was very ambivalent about it. On the plus: “holy crap, the suspect in a very violent crime’s car was parked outside the crime scene! Awesome. Yeah technology.” (Actual example FWIW) On the minus: “we keep every single record of every single car for how long? Hmm…”
Edited to add that from what I have seen this is used to find hot cars or cars that need a visit from the boot fairy, but yes it can also be used to run a car for investigative purposes.
In my experience they get used mostly to find hot cars or cars that need a visit from the boot fairy (at least where I am the operators are informed of such things in near real time). More practical/money driven than nefarious, but yes the massive amount of data can be abused.
It could easily be implemented in a non-privacy abusing manner: send a list of plates of interest out to the squad cars, and compare the plates scanned to the local list, immediately discarding any that don’t match.
Instead they opted to send a stream of license plates and their locations to HQ, compare those to a central list of plates of interest, and for no good reason retain indefinitely the time/place/plate records that didn’t raise alarms.
That design choice and data retention choice speak volumes about the actual project goals.