3 ms·

At the risk of sounding cold hearted: I get that it doesn't seem fair, but logically, it's the right call to drive to the end of the lane. 1) It keeps the rule

At the risk of sounding cold hearted: I get that it doesn't seem fair, but logically, it's the right call to drive to the end of the lane. 1) It keeps the rules simple. If we should merge at some arbitrary point before the lane is up, it will always be arbitrary, and different drivers will have different opinions. 2) By getting to the end and slowing down, the driver now has matched the speed of traffic. It's a self correcting problem. 3) Not going to the end of the lane is wasting capacity. If it's 4 lanes reducing to 3, how far back is it fair to expect that ending lane to go unused? 4) For on ramps, it gives more cars the chance to merge at once, which will either help the flow of traffic, or quickly reach equilibrium with the through lane. 5) And finally, if everyone always used the full lane, it wouldn't give "cheaters" the opportunity to pass everyone else and cheat. >causing the traffic column in the thru-lane to brake dramatically Theoretically, where the car merges doesn't have any bearing on this occurring or not. It shouldn't be any different at the beginning of the lane or the end. If traffic isn't moving, there's no breaking to be had. If it is, it should only give better opportunity to match speed and merge without disruption. Practically, the problem has as much to do with the aggression of the through lane drivers as it does the merging one. They certainly have the right of way. I'm just saying that where the merging driver comes over doesn't really change any of that, unless the through traffic is behaving differently in response.

FPGAhacker 7y ago
>Theoretically, where the car merges doesn't have any bearing on this occurring or not. It shouldn't be any different at the beginning of the lane or the end. It’s a scheduling problem where people that join the lane early are starved by the people joining late. In cases where there is roughly an equal exchange of vehicles between two lanes it doesn’t matter much, and maybe that’s the norm where you are. Where I am, congestion is more localized to chokepoints where too many people need to use an exit or on ramp. In this case if time matters it’s usually better to be aggressive and merge at the last minute rather than be passive and get in line early. People may think you are a jerk. For people where it’s important not to be thought of badly, they get in line first. For people who don’t care (or occasionally for other reasons) they skip the line and merge late.
wwweston 7y ago
> At the risk of sounding cold hearted: I get that it doesn't seem fair, but logically, it's the right call to drive to the end of the lane. There's no dispute on this point. It's the right call to merge at the last workable moment. One can also make another wrong call in failing to match speed and negotiate tooth-gap arrangement (still in separate lanes) before doing so. If drivers don't do that, it isn't a zipper merge, even if everyone waits.
asdff 7y ago
On paper, maybe the zipper is excellent. In practice, with prideful LA drivers, its a shit show. I know the maneuver the previous commenter described, it only leads to pain. - everyone effectively stopped in the exit lane has to stop for even longer - everyone in the thru traffic lanes has to deal with this person merging all the way from the left to the right with no blinker, then coming to a stop in the right most thru lane desperately trying to force themselves into the exit lane, usually driving over the gut in the process - the end result is the entire freeway becomes even more backed up because this one person couldn't think more than 1 exit ahead, or worse, they convinced themselves that risking a brutal death like this is worth the 20 seconds it saved them
- beerandt 7y ago
  What you're describing is a problem, but a different one. Def not zipper merge. For merging (on ramps), waiting to the last moment is collectively efficient. For diverging traffic (exits) it isn't, especially if there's a backup.

FPGAhacker 7y ago

>Theoretically, where the car merges doesn't have any bearing on this occurring or not. It shouldn't be any different at the beginning of the lane or the end. It’s a scheduling problem where people that join the lane early are starved by the people joining late. In cases where there is roughly an equal exchange of vehicles between two lanes it doesn’t matter much, and maybe that’s the norm where you are. Where I am, congestion is more localized to chokepoints where too many people need to use an exit or on ramp. In this case if time matters it’s usually better to be aggressive and merge at the last minute rather than be passive and get in line early. People may think you are a jerk. For people where it’s important not to be thought of badly, they get in line first. For people who don’t care (or occasionally for other reasons) they skip the line and merge late.

wwweston 7y ago

> At the risk of sounding cold hearted: I get that it doesn't seem fair, but logically, it's the right call to drive to the end of the lane. There's no dispute on this point. It's the right call to merge at the last workable moment. One can also make another wrong call in failing to match speed and negotiate tooth-gap arrangement (still in separate lanes) before doing so. If drivers don't do that, it isn't a zipper merge, even if everyone waits.

asdff 7y ago

On paper, maybe the zipper is excellent. In practice, with prideful LA drivers, its a shit show. I know the maneuver the previous commenter described, it only leads to pain. - everyone effectively stopped in the exit lane has to stop for even longer - everyone in the thru traffic lanes has to deal with this person merging all the way from the left to the right with no blinker, then coming to a stop in the right most thru lane desperately trying to force themselves into the exit lane, usually driving over the gut in the process - the end result is the entire freeway becomes even more backed up because this one person couldn't think more than 1 exit ahead, or worse, they convinced themselves that risking a brutal death like this is worth the 20 seconds it saved them

beerandt 7y ago
What you're describing is a problem, but a different one. Def not zipper merge. For merging (on ramps), waiting to the last moment is collectively efficient. For diverging traffic (exits) it isn't, especially if there's a backup.