Comment by jdross
10 hours ago
I realize how unpopular flock is, and I will first say that I have literally never personally looked into the privacy concerns. But one city you don’t see named here is SF, which has cited Flock as a primary driver of its 10x reduction in car break-ins, and 30% reduction in burglaries. Those were a quality of life plague while I lived there
Crime's been descending from the COVID blip for a while, everywhere, Flock or otherwise. My city saw zero murders in Q1; 2021 saw ~15 by now.
In other words: https://www.youtube.com/watch?v=xSVqLHghLpw
it's clearly not a covid effect https://www.sfchronicle.com/projects/sf-car-breakins/
The data is open, and so we don't have to do the visual reasoning off an imperfect graph. SF Chronicle has done a pretty rare (but I think good journalistic practice) of specifying the source of the data: https://data.sfgov.org/Public-Safety/Police-Department-Incid...
First to match the graph you make sure you pick 'Larceny - From Vehicle' only (there are some others one might argue matter) and ensure you're only counting incidents once (many rows reference the same incident). That lets us recreate the original graph.
When looking at many things I like to look at seasonal effects just to see, and it doesn't look like they are significant here (but you can see the Mar 2020 drop to the next year quite easily which I like): https://wiki.roshangeorge.dev/w/images/2/2e/SFPD_Vehicle_Bre...
I also tried overlaying various line charts but that's useless for visually identifying the break.
One thing I thought would be fun is to run a changepoint algorithm blindly https://wiki.roshangeorge.dev/w/File:SFPD_Vehicle_Break-Ins_...
I like PELT because it appeals to my sensibilities (you don't say ahead of time how many changepoints you want to find - you set an energy/cost param and let it roll) and it finds that one changepoint. You can have some fun with the other algos and changing the amount of breakpoints or changing the PELT cost function. And then you can have even more fun by excluding 2020 or excluding Mar 2020 onwards or replacing it by estimates from the previous years (quite suspect considering what we're trying to do but hey we're having fun - a bunch of algos all flag Nov 2023 as some moment of truth)
Anyway, anyone curious should download the data. It's pretty straightforward to use and if I goofed up with off-by-one or whatever, you can go see for yourself.
1 reply →
The spike in your link's chart clearly starts in early 2020.
And "While our data extends only to 2018" is... important, yeah?
8 replies →
I could believe that perma-cameraing every inch of public space is more akin to chemo than to vitamin gummies, that SF had the city equivalent of bone cancer, and that this doesn’t mean healthy midwestern towns need Flock in any way.
The byproduct of habitually coddling criminals with zero consequences.
Do you have empirical evidence that we "habitually coddle criminals"? The united states locks up more of their people than pretty much any other nation...
3 replies →
It's a much tougher problem then "idk just throw them in jail lol" and anyone who claims otherwise is dishonest.
[flagged]
Any evidence that the reduction is actually due to the cameras?
Don't people tend to behave if they know the are being watched?
yes, people tend to act differently. not the people they're trying to afect, just random people just minding their business. but it is not an effective deterrent to things like "violent crime".
• Meta-analyses (studies that average the results of multiple studies) in the UK show that video surveillance has no statistically significant impact on crime.
• Preliminary studies on video surveillance systems in the US show little to no positive impact on crime.
https://www.aclu.org/sites/default/files/images/asset_upload...
I thought he asked for evidence?
[flagged]
There is very solid evidence it wasn't the cameras.
https://www.sfchronicle.com/projects/sf-car-breakins/ has a chart of the car breakins.
It shows the drop starting in September of 2023.
https://www.sanfranciscopolice.org/your-sfpd/policies/depart...
> Starting on March 19, 2024, Flock Safety began installing ALPR cameras in various strategic locations across San Francisco. This rollout is expected to take place over the next 90 days.
In other words, the cameras were added where I've annotated the chart with a black rectangle here: https://imgur.com/a/i00Gna0
To my knowledge, Flock doesn't have a time machine offering.
2 replies →
Sorry, this is russell's teapot falacy. "the burden of proof lies with the person making an unfalsifiable claim, rather than on others to disprove it"
If there is evidence this is related to cameras, then the onus is on companies making these cameras and claims to provide the data. Not on others to prove that they don't stop crime.
There's a reason you always start with the null hypothesis.
> which has cited Flock as a primary driver of its 10x reduction in car break-ins, and 30% reduction in burglaries
Are there reports or studies released which explains how the flock system influenced these reductions?
ALPR does help with some things but stationary burglaries are largely not among them.
Unfortunately, Flock really has been doing some shady stuff and the alliance of 1) people with legitimate concerns about Flock operations, and 2) the much larger population of people who are accustomed to getting away with petty crimes is, together, politically successful.
It would be easy to create a camera network that is locally owned and operated by public agencies, and if any place in America could so that it should be SF.
The crime did not happen because of a lack of technological capability or resources availability at a given price point. It happened because of politics and priorities. The 1984 camera dragnet vendor is no more responsible for the change in politics and priorities and subsequent crime reduction than whatever vendor sold the tires for the cop cars.