Hacker News new | past | comments | ask | show | jobs | submit login

Automation of police enforcement allows a precision and completeness in ticketing that was never envisioned while writing the laws or fines. It would only start with blocking bike paths or bus stops. Once the technology is there, there is incredible incentive by the city to expand the scope to generate more revenue, as we've seen with red light camera installations.



I'm all for allowing that particular incentive to run rampant. Running red lights is unsafe and kills innocent people. Are you suggesting running red lights shouldn't be enforced 100% of the time??

Everyone has the option to obey the law!

Also not envisioned while writing the laws was a wish for some people to break the law and get away with it. They weren't hoping for partial enforcement, laws are written to communicate the standards under the hope that all people will abide. Enforcement was never intended to be less than perfect, it has always been to do as much as is practical. New technology is making it more practical, but isn't changing anyone's intent. There is no sacred amount of precision or completeness in ticketing that ought to be preserved in the face of improved ability to enforce laws.

I'm also hopeful that self-driving cars, and more bike support, and other changes to transportation, make this issue (and it's enforcement) disappear.


Numerous studies have shown increased accidents with the use of red light cameras (others have shown decrease in collisions but also increase in severity). In Florida for instance, fatalities and collisions went up. [1]

Speed limits are also based in large part on the 85th percentile speed of actual traffic, which presumes 15% of drivers are breaking the law.

So no...I don't believe absolute traffic enforcement was ever envisioned, and I don't recall ever hearing of a politician running on that platform.

Lest you believe I'm a law breaker looking for leniency, in over 20 years of driving I've never received a moving violation.

[1] http://www.sun-sentinel.com/news/florida/fl-nsf-red-light-ca...


Okay, I read that article, and it said traffic went up over 8.32% overall, and then showed some increases in accidents near 8%. It looks like they reported absolute numbers without controlling for the increase in traffic. If traffic was up 8.32%, and angle crashes were up 6.72%, doesn't that imply that angle crashes were actually reduced by around 1.6% per capita?

What about this one:

"Insurance Institute for Highway Safety study stated that fatal crashes from red light runners increased by 30 percent per capita after cities turned off their red-light camera programs."

> Speed limits are also based in large part on the 85th percentile speed of actual traffic, which presumes 15% of drivers are breaking the law.

It's a misleading assumption to suggest that the way speed limits are determined actually implies a design intent to retain a 15% minority of law breakers. The stated and explicit intent of posting the speed limit is to set the maximum, and for nobody to exceed the maximum.

Also: "The speed limit is commonly set at or below the 85th percentile operating speed (being the speed which no more than 15% of traffic is exceeding) and in the US is frequently set 4 to 8 mph (6 to 13 km/h) below that speed." (https://en.wikipedia.org/wiki/Speed_limit#Maximum_speed_limi...)

> So no...I don't believe absolute traffic enforcement was ever envisioned

It doesn't matter what was envisioned. We don't need to preserve what was envisioned.

> in over 20 years of driving I've never received a moving violation.

That's great! I would be more impressed if you hadn't done any speeding or run any red lights in 20 years. (I can't claim that, though I also haven't had any tickets in 20 years.)


A better solution would be to do regular speed studies and actually update the speed limits posted. Also, prevent municipalities from passing laws on their own speed limits and creating speed trap situations.

I see a lot of speed limits that are woefully out of date and/or very different from the 85th percentile speed on any one motorway.

> It doesn't matter what was envisioned. We don't need to preserve what was envisioned.

Then we should just ditch the whole speed limit concept. You should be punished for variance from 85th percentile speed on a motorway in either direction. Too slow or too fast are both dangerous


> Are you suggesting running red lights shouldn't be enforced 100% of the time??

The majority of red light tickets are for people not coming to a complete stop before turning right on red. Quite often, the non-turn lanes aren't even instrumented because it's not profitable.

There's also the issue that red light camera vendors encourage shortening the yellow light, which increases revenue and harms safety.


Do you have any sources on these claims you can share?

Adding in the problems of private vendors commercializing law enforcement wasn't on my mind, but is a very good point. I would hope that, despite any pressure to shorten the yellow light, laws and safety advocates would prevent that from getting out of hand.

What is currently most out of hand between unsafe driving and unreasonable enforcement revenue, is clearly unsafe driving. So even if enforcement isn't perfect, I still think a correction towards enforcement is a good thing on the whole. I hate driving because of how crazy unsafe large numbers of people will be for really marginal benefits. The wish more more enforcement crosses my mind almost every day.

And, again, we can easily defeat any evil government revenue enforcement schemes by simply following the laws we should have been following all along.


Chicago is a good example, they reduced yellow lights below federal minimums to maximize revenue, and the contract was rife with corruption.

"A Tribune-sponsored study of the red-light program in 2014 found that nearly 40 percent of the intersections equipped with the cameras are likely making the streets more dangerous. The study found that the cameras caused a 22 percent increase in rear-end crashes, yet provided no safety benefit at intersections that never had a problem with right-angle crashes in the first place."

http://www.chicagotribune.com/news/local/politics/ct-rahm-em...

https://www.justice.gov/usao-ndil/pr/former-redflex-ceo-sent...


I don't really want to spin here on red light cameras only; there are many more ways to enforce traffic laws than that, and if we could use people for 100% enforcement I'd be in favor of that.

I want self-driving cars and other mass transport to eliminate the need for enforcement completely.

I have to be honest though, the stuff you're throwing at me is tingling my spidey sense just a little. These quotes are stated in a way that sounds convincing and bad, but fails under math scrutiny.

The quote you picked left out the 15% improvement in angle crashes they measured (* cited in the bigthink article below).

You can have 40% of intersections with a 1% increase in crashes, and also have 60% of intersections with a 1% decrease in crashes, and you have an overall slight reduction.

These resources seem level headed:

https://www.fhwa.dot.gov/publications/research/safety/05049/

https://www.cdc.gov/motorvehiclesafety/calculator/factsheet/...

https://www.ncbi.nlm.nih.gov/pubmedhealth/PMH0012902/

And these feel like fluff, and even propaganda:

http://bigthink.com/ideafeed/study-red-light-cameras-ineffec...

https://www.motorists.org/issues/red-light-cameras/studies/


The Tribune is a pretty factual paper, but studies have been mixed over the years. Confusing the data is the implementation of RLCs at sites that never had an issue with right-angle crashes, that’s a key reason the Chicago data showed an increase in collisions.

I don’t think we’ll agree on this issue, but I understand your arguments and they are reasonable.


> I don’t think we’ll agree on this issue, but I understand your arguments and they are reasonable.

Hey thanks for saying that. I can return the favor, you’ve been very reasonable too. I don’t disagree that red light camers specifically might have some tradeoffs, and maybe it’s true they’re a net negative right now, I don’t know.

As far as the safety goes, have we actually given it enough time? Would it change anything for you if we fast forward into the future, and imagine that all intersections have cameras and everyone expects them? Today’s trade off between T-bones and rear-ends when there are cameras is presumably caused by people not expecting a camera and then realizing fairly late that they might get a ticket. Is it reasonable to assume that if cameras were everywhere, the rear-enders would eventually settle at a lower rate?


If we fast forward, I do think we’ll see at least a marginal improvement in safety. The cameras would never go away though, and the laws would have to be rewritten in a way to maintain revenue through minor infractions. That cost would land hardest on the poor and working class.

I’m a careful driver, wealthy enough to pay the fines without a care, so it wouldn’t really affect me. Neither would congestion tolls, or the numerous other solutions involving regressive taxes or fines. Less traffic would be worth the cost...I find myself often wishing they’d just triple tolls and gas taxes so my commute would be shorter. Not proud of those thoughts.

TBH, I would have agreed with you a few years ago, but lately my perspective has shifted. I don’t wish to muddle the safety argument with inequality, but it is part of my aversion.


A possible solution... remove revenue. Hell, have the fines be earmarked as a direct cash back tax cut or something. Redistribute the money from the guilty to the innocent, and keep the motivations clean.


Rolling right turns on a completely empty intersection late at night done at 5mph is treated the same as running through a red light with traffic with a $550 ticket.

Because of financial incentives for municipalities, the laws tend to stick that way, even when law makers want to make distinctions in red light camera types. And it hurts the poor the most.

See more red light camera bullshit here: http://www.highwayrobbery.net/


> Rolling right turns on a completely empty intersection late at night done at 5mph is treated the same as running through a red light with traffic with a $550 ticket.

And it takes all of 2 seconds to circumvent... by not rolling through. Get a ticket only once, and you'd probably learn your lesson and never do it again. You always have the power to avoid that ticket.

> See more red light camera bullshit here: http://www.highwayrobbery.net/

No, thank you. I sense bias and propaganda in the URL name.

This feels the same as the people who claim there's a conspiracy and that bicycle helmets don't increase safety, or that requiring motorcycle helmets is abuse of government power.

As a citizen who wants peace and safety for myself and my family, I am in favor of more speed limit and red light enforcement, among the many other apparently controversial things that I'm happy to pay a little bit of tax for, and/or happy that the government can make some revenue on.


Does data matter to you? If increased cameras let to increased accidents would you still be in favor? Do you think that there are unintended consequences to ultra strict enforcement?


Yes data matters to me. If cameras or other types of enforcement truly made things less safe overall, I wouldn't want them. I don't think that's the case, but I'm certain there are unintended consequences to certain kinds of enforcement.

I can ask you the same. Do you care about data? Your question about data follows a link to highwayrobbery.net, which is using FUD and inflammatory language, and I would bet some misleading characterizations of "data" to rile people up over this issue.

If the "data" is being politicized and falsely presented, if people are actually mad about a perceived tax or government control, and not about safety, do you consider that data that matters?




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: