Paved with good intentions

[unixronin]

Via cracked.com, Bitmines, and paulesyllabic¹, the five most popular safety laws that don’t actually work.

Highlights:

• “Traffic calming” costs more lives than it saves.  “One report from Boulder, Colorado suggests that for every life saved by traffic calming, as many as 85 people may die because emergency vehicles are delayed.  It found response times are typically extended by 14% by speed-reduction measures.  Another study conducted by the fire department in Austin, Texas showed an increase in the travel time of ambulances when transporting victims of up to 100%.”
• [1997 article]  A 1997 study found that in 1994-1995, crime (both violent and overall) decreased by three times as much in states without Three Strikes laws as in states with them.  In eight of the thirteen states that had three-strikes laws, crime actually increased during that period.
• Fully half of the 233 Amber Alerts issued in 2004 were for children who were in no danger.  48 of the 233 alerts — more than 20% — were for children who hadn’t even been abducted.
• Sex offender registries don’t distinguish between serial rapists and somebody who took a leak in public while drunk.  Besides, 95% of sexual assault victims, child or adult, already know their attacker anyway.

And I’m sure I don’t have to point out the utter stupidity behind zero-tolerance policies to anyone here... not to mention the rampant abuses.

[1]  Not necessarily in that order.  Or any order.

Comments

[jmkelly.livejournal.com]

The last three (or four, if you count zero-tolerance policies) are all species of removing discretion, making action automatic so as to avoid negligence or bad decisions -- the assumption is that most of the time the automatic action will be better than action taking after real thought, or better than nothing. Sometimes that's harmless, as in the Amber Alerts -- if there are a lot of false alerts, what's the harm? (unless, of course, there are so many that people stop attending). Other times -- three strikes laws, "protect the children" sex-offender laws, etc. -- there's real injustice.

Removing discretion isn't such a bad thing, in moderation. California voters have discretion to amend their Constitution by a simple majority vote, and the result is horrible, one of the United States' worst state constitutions. If the requirement were raised to a two-thirds supermajority, we'd retain the ability to amend our constitution while being protected against the passions of nutjobs. Similarly, three strikes and Megan's Laws might not be so bad if they let judges retain power to short-circuit them when they were clearly abusive. Unfortunately, drafting them to do so takes actual thought, and risks incurring the wrath of the tough-on-crime and protect-our-children crowds.

[unixronin.livejournal.com]

Sometimes that's harmless, as in the Amber Alerts -- if there are a lot of false alerts, what's the harm?

As was pointed out in the articles that section linked to, all the false alarms waste resources that are already in short supply; meanwhile, the kids that HAVE been abducted by psychopaths are usually already dead by the time the alert goes out anyway.
What we need is hard numbers on how many kids who really are in danger are actually saved by "Amber Alerts". Those are probably hard numbers to get.

Similarly, three strikes and Megan's Laws might not be so bad if they let judges retain power to short-circuit them when they were clearly abusive. Unfortunately, drafting them to do so takes actual thought, and risks incurring the wrath of the tough-on-crime and protect-our-children crowds.

Exactly. And these days we're all about quick, headline-friendly fixes. Actually thinking about the problem? Not so much...

[perspicuity.livejournal.com]

they had 3 amber alert phone bank calls in the last couple years in my town.

in all cases, the people they were looking for "wandered off" and were found at a friend's house. absolute over-reaction. plus bad parenting.

iirc, they did not call the amber alert OFF after the fact for at least 1 of them.

they also had some kind of private alert thing going on where some lady lost her prize show dog, and somehow managed to get an entire community on patrol looking for it. we had a lot of crazy people driving around here, at speed, in a private community, for hours, despite being told to slow down (EFF EWE!), and that nobody saw their stupid dog, and SLOW DOWN THIS IS PRIVATE PROPERTY YOU ALMOST HIT A KID (EFF EWE!) - fine, we're calling the cops - SCREECH! see ya!

the highway amber alerts i've seen, involved a "legal abduction" of a father asserting his rights, a girl on a joy ride with her boyfriend and freaked parents (and the girl was a legal adult...).

apparently the old 24-48 hour rule is thrown out the window to test these new systems.

#

[ithildae.livejournal.com]

Trying to nail down the cost of resource utilization is a futile exercise. If the emergency workers and law enforcement personnel are already on the job, what is the incremental cost of having them look for something specific? Do we count it against all their time, figuring that if they were not looking, they would not be getting paid and consuming resources? Do we only consider extra costs, like the amount of gasoline that was consumed over what would have been used on the shift? Not and easy answer, and one that is ripe for gaming.

As a business example: When we lost power to our datacenter, the CFO was trying to get into the server room to tell us that we were losing $5000 per 15 minutes we were down. (An average outage lasted about five hours.) When we introduced the cost of the UPS's or generator to protect the datacenter, the income was only deferred, and we only actually lost about $4000 per outage. We finally got the UPS's. But the numbers were never the same from the same guy. It all depended on whether we needed it, or he wanted it.