Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'm a bit surprised that .08 BAC corresponds to 70% ability.

That means that in many cases, a good-but-legally-drunk driver will be vastly out-performing a terrible-but-sober driver.

I'm just glad my car will drive itself soon.



>>That means that in many cases, a good-but-legally-drunk driver will be vastly out-performing a terrible-but-sober driver.

Not really. Drunk-driving is illegal due to increased reaction times, not decreased driving ability. You can be drunk and properly follow the traffic laws, but if you run into a situation that requires quick reflexes you're much more likely to not react quickly enough.

Whereas terrible drivers are generally terrible in other ways.


Someone who has a low reaction time overall might have a better reaction time drunk than someone with a reaction time of a dead cow. My friend claims 0.1 reaction time on a game where you have to click your mouse (I believe him because he's a professional gamer so it makes sense), while I score 0.2 at best (which is average) and older people score worse and worse


Hence why some people are fighting the politically unpopular fight to get older people off the road. No one wants to be seen as the one that "stole grandma's freedom".


I bet the reason is unawareness of poor reaction times. I have found sluggish video game vehicles pretty easy to drive.


Good driving doesn't have that much to do with your reaction time. 40% slower to respond, with a well planned drive, is a lot safer than 40% faster driving too close.


Sure, but one aspect of "terrible" driving would be: following too close, leaving oneself insufficient space to brake in response to leading vehicles. So, someone who is terrible in that respect could be more dangerous sober than a better driver is when drunk.


Of course, the real reason the limit is there is because the greatest danger is the terrible driver, drunk. When you take 30% worse of "terrible", you get "really terrible".


That's the reasoning, but to me it just proves that drunk driving shouldn't be a crime separate from poor driving. (I guess it would be OK as an aggravating factor for other traffic crimes.) Someone who drives acceptably while drunk drives acceptably, period.


I figure the reason it's set up that way is because BAC is way easier to determine on the side of the road during a traffic stop than a comprehensive study of someone's driving, so we have to lean to the less permissive side.

(We opt not to lean to the more permissive side because drivers who are drunk are indeed already responsible for a great deal of injury, loss of life, and damage)


I don't know. You can't, unfortunately, stop everyone who drives poorly. Too many people do - and even if they just got off with points on their license, quickly enough they'd be banned and then out of work.

Seriously, I can drive for fifteen minutes and see dozens of people who are driving too close together or committing some other sin. All those people who, just going into a corner, you see lighting off their brakes one after another - when all you, a little further back, have to do is roll your toes back a little to take some of the speed out.

Considering random breath tests don't seem to be that big of a thing, (I think - https://www.askthe.police.uk/content/Q723.htm - that they can't do truly random testing,) they're probably picking the people liable to be the worst examples anyway. Which I suspect is the most we can practically hope for at the moment.


not yet... now, if only we could use the video record of our driving for the trip prior to the traffic stop to determine if we were driving reasonably... I'll just leave this here: http://www.extremetech.com/extreme/146909-darpa-shows-off-1-...

for better or worse, it's not far off. imagine we could cross reference some street level CCTV (license plate, maybe gate and facial recognition) and "select driver_dl from trips where driving_ability < 0.7"; revoke licenses from those who can't drive well, consistently... to pay for it, maybe, we could just issue some DUI tickets where the car was parked in-front of a bar for a while and the driver appeared to walk unusually (dozens of violations per habitual drinker, ez). cross reference some phone/text and CC metadata to determine sociability and spending, maybe more, and you can get a pretty good certainty who's probably driving under the infuence and/or a poor driver, and take them off the road.

barely fiction (maybe not at all?), but it won't be hard forever and they won't need to "stop" many people in the sense that is hard to do, for much longer.


Too many people do - and even if they just got off with points on their license, quickly enough they'd be banned and then out of work.

In the absence of usable public transit and a populace capable of bicycling, I guess we'll just have to wait for the self-driving cars...


I couldn't help but think of this http://www.youtube.com/watch?v=URupBZbfbJg


Well... I'd say good driving works to minimize the importance of your reaction time, but that doesn't make reaction time unimportant.

I do agree with your general point though.


>That means that in many cases, a good-but-legally-drunk driver will be vastly out-performing a terrible-but-sober driver.

How do you know what percentage of ability corresponds to bad driving?


>How do you know what percentage of ability corresponds to bad driving?

It is based on reaction time. So if you approach a stop light that turns red and you have a 70% score, you are taking approximately 30% longer to hit the car brakes.


Good driving is not about reaction time, it's about anticipation. That's why traffic lights have an amber stage.




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

Search: