YHK - yet another hacker news clone

Edit: added scare quotes because it's difficult to be sarcastic on the internet these days.

So we are now distracted while they reshape the world with hunger and death

"He said 103 arrests has been made so far and there was one false alert which did not result in an arrest.

Supt Luke Dillon said overall crime figures in the Fairfield ward, where the pilot has been held, fell by 12% per cent.

"We're seeing reductions in almost all crime types, certainly in shoplifting and robbery," he said. "

https://www.bbc.co.uk/news/articles/c2056r07rjlo

"Last month BTP announced it was extending the trial for a further four months and expanding the deployments to include London Underground stations. "

Pretty much all I ever read about these systems is heinous false positives. I've currently set up automation to pull the local data that includes police activity. One of the towns over from me has, at least, 6 Flock cameras (probably more) for a locale of less than 2000 people. I've not once seen anything related to ALPR or any camera system providing any influence on assisting a crime. You'd think Mr. Langley would be begging police chiefs around the US to put in a positive spin to public information around these digital watchers.

But the morons buying these systems are also the morons taking state tax dollars to buy SWAT vehicles for 100s of thousands of dollars because they want to cosplay GI Joe. Yet... You want a park? Or something that's actually useful in your local community? Fund raise it yourself is the response.

The article doesn't say, and I haven't seen the cameras, but it's not like you're getting every single person parsing through those stations.

But, yes, the numbers seems low.

Some would call it discrimination but the majority of people who commit crimes in my area are people who can walk. I rarely see the wheelchair crowd pulling off B&Es.

Oh, or people who wear clothes. Criminals seem to like wearing clothes. Point proven: you never see nudists shoplifting Old Navy.

I've spent a whole of 15 seconds thinking about this and I don't foresee a single drawback here, lets just try it and see if we can prevent future crimes 100%, because then the future will have been saved.

The strongest predictor (by far) for committing a crime is having a prior criminal record.

It's "fell swoop".

It would probably pay for itself.

Since governments are run by people, this safetyism has become enshrined in law, typified by nonsense like criminalizing free-range parenting and mass surveillance.

its one of the value function that makes it almost impossible to evaluate.

Lets imagine that catching someone in the Interpol list of wanted terrorists is a huge catch. You would expect the avg catch rate to be almost 0 but once one is caught the avg value of the system becomes very easy to justify.

Its the same idea as having photos of known terrorists in airports and train stations to see if guards recognise someone. But with the added benefit that the facial recognition should not miss them.

If the evaluation function is it costs X to run each day and we caught Y number of criminals who robbed phones worth Z > X then the system will come off pretty badly in terms of ROI.