Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Traffic Hacks: How One Driver Can Vastly Improve Traffic

trafficwaves.org

Post reply on HN
Print view View original

31–40 of 149 posts

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#31

Post by amalcon »

joezydeco wrote: post #16
mrshoe wrote: post #8

This man is a saint. For years I've wished more people understood these principles. Instead, I get drivers who are angry at me for leaving a big space in front of my car. If only I pulled up right behind the car in front of me, they'd be 60 feet closer to their destination!! I wish more people would check this site out. Traffic is a real problem that wastes billions of taxpayer dollars and hours on the road every yea…

These ideas work, but only if you can get everyone else to play along, knowingly or unknowingly. In Chicago at rush hour, if you leave more than a car's worth of space ahead or behind you, it will be filled instantly.

People think this about 128 in Massachusetts, too, but generally what I observe is that people leave my "empty space" about as often as they enter it. I leave a big empty space for safety reasons, and also because I actually understand how inefficient constantly accelerating and braking is. I'm usually able to maintain it without too much trouble.

Whether or not it actually helps traffic, I couldn't say.

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#32

Post by randomwalker »

kyro wrote: post #18
zck wrote: post #4

Some Japanese scientists did a study to show how traffic jams can appear of of nowhere. They told 22 drivers to drive in a circular path at 30 km/h. A traffic jam appeared, and moved backward. The video is here: http://www.youtube.com/watch?v=Suugn-p5C1M

It's not 'out of nowhere.' It's because we're humans. We're paranoid and reactionary. You wouldn't see that with robots.

Cynical much? How about it's because we have bounded cognitive abilities, limited perceptive accuracy and slow reaction times?

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#33

Post by RyanMcGreal »

PieSquared wrote: post #27
joezydeco wrote: post #16

Earlier quoted context omitted.

These ideas work, but only if you can get everyone else to play along, knowingly or unknowingly. In Chicago at rush hour, if you leave more than a car's worth of space ahead or behind you, it will be filled instantly.

Is that an assumption or something you've tried? If it's the former, he addresses that concern on his site.

In my experience these ideas work even if you're the only one following them.

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#34

Post by joezydeco »

PieSquared wrote: post #27
joezydeco wrote: post #16

Earlier quoted context omitted.

These ideas work, but only if you can get everyone else to play along, knowingly or unknowingly. In Chicago at rush hour, if you leave more than a car's worth of space ahead or behind you, it will be filled instantly.

Is that an assumption or something you've tried? If it's the former, he addresses that concern on his site.

It's from 25 years of experience. Noodle from Atlanta has a similar experience in the comments below.

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#35

Post by zngtk4 »

_cbsz wrote: post #7
zngtk4 wrote: post #6

Interesting, my instinct has always been to try to "punish" the cheaters by preventing them from merging in, thinking that this would discourage this behavior. But I can't really stop this, and the article provides a good explanation for why you might as well just let people merge in front of you.

The problem is that allowing merges incentivizes zooming ahead in the merge lane, which should result in there still being drivers who can't merge, creating the jam. I don't have a good solution for this.

To an extent I think you're right and some people will do it regardless. On the other hand, smoothly flowing traffic creates an incentive to not zoom ahead in the merge lane. There's less of a feeling of a need to cut ahead and people don't seem to do it as much when there isn't already a jam. (That is, the jam provides an incentive to cut in front of everyone)

And if people were to implement this strategy, the provided space gives an incentive for those zooming ahead to merge a little sooner rather than risk waiting half a minute for someone to let them in right at the exit. It's been my experience that most of the "cheaters" will merge a little sooner if there's space rather than zoom up right to the end. They'll still be passing you, which is IMO aggravating, but at least things will move a little sooner.

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#36

Post by kragen »

kyro wrote: post #18
zck wrote: post #4

Some Japanese scientists did a study to show how traffic jams can appear of of nowhere. They told 22 drivers to drive in a circular path at 30 km/h. A traffic jam appeared, and moved backward. The video is here: http://www.youtube.com/watch?v=Suugn-p5C1M

It's not 'out of nowhere.' It's because we're humans. We're paranoid and reactionary. You wouldn't see that with robots.

You can certainly program robots to do that. The question is whether you can program them not to without losing something important (robustness against unexpected conditions or malicious robots, for example).

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#37

Post by iigs »

I respect this guy's patience, but the primary reason that one car can make such a big difference in this case is because 520 (one of the most important commuter roads in the Seattle area) is just two lanes in each direction. One car on a three lane road will just be passed.

The counter-point to this, in my opinion, is that at some point psychology comes in to play and you have people starting to get aggressive and angry because of a few people forming a "rolling-road-block". It's not rational but it definitely happens.

I would support 3PM-7PM dynamic speed limits on 520W toward the bridge. If the speed was brought from 60 (just past 405) to 30 as you approach the bridge it could probably keep things moving smoothly through the traffic burst.

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#38

Post by erlanger »

If you want to help traffic, stop hitting your damn brakes so much. Brakes are for strong deceleration, and most deceleration in traffic can be handled by letting off the gas. Once you get good at this, follow as closely as possible in a traffic jam without applying your brakes.* There's nothing better a skilled driver can do to alleviate a traffic jam. The jam is physically shorter and people behind you are less likely to apply their own brakes as a result of these measures.

* This has the added benefit of making for a smoother ride, which also encourages those behind you to ease off the brakes.

Re: Traffic Hacks: How One Driver Can Vastly Improve Traffic

#39

Post by zngtk4 »

noodle wrote: post #25

i think its interesting that i used to drive like this site describes for the exact reasons it talks about. and now i don't/can't. the difference? i moved to atlanta. rush hour drivers here operate under a greedy algorithm. they see an opening that will allow them to speed up or gain extra distance immediately, and they'll take it, for no other reason than to be going faster right now, to be farther down the road rig…

That (in addition to the fact that everyone lives in the suburbs) perhaps provides some of the explanation for why Atlanta rush hour traffic seems to be some of the worst in the country, despite having more lanes and a smaller population than cities that have better traffic.
Post reply on HN