Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 



Forgot your password?
typodupeerror

Comment Re:Turn signals are a good thing (Score 1) 469

For that, there's brakes. It has the disadvantage that it allows the lane-merger to "win", but at anything above parking-lot speeds it clears the hazard without creating any new ones (assuming that the guy behind you maintains a safe following distance -- but then the hazard was not newly created, it was already there), and it is always an option, whereas changing to the other lane might not be an available choice. For a robotically controlled car, that would certainly be the algorithm, because it ensures that the robot will not be at fault if a crash occurs following the avoiding maneuver. Impatient+savvy human drivers might learn to aggressively cut off robot-cars, but this would also be true even if the reaction was a swerve, even an optional swerve.

Braking has the secondary advantage of reducing speed in any crash that might subsequently result, which generally reduces the severity and scope of the crash.

Comment Re:Turn signals are a good thing (Score 2) 469

It would, but people manage to simultaneously hold in their heads both "I am a better than average driver" and "the anti-idiot squad might come after MY driver's license". There's a huge diversity of opinion about exactly what "good driving" is, too -- I try to be very conservative with following distance (not slow, just plenty of space, and not in the fast lane, either) , and sooner or later someone who ends up behind me will get upset that I am doing this, because look at all that empty space we could be driving in. As far as they are concerned, I'm one of those idiots, even though I am taking deliberate steps to avoid known-dangerous driving. Obviously, I think THEY are the idiots. At least one of us is wrong.

Comment Re:Other people controlling your car? (Score 1) 469

Forest for the trees. We tolerate far worse already, and think that it's wonderful.

Do you have any idea how bad it is for us, that so many of us use cars when we could instead be walking or riding bicycles? There was a large, multi-year study in Denmark, and (as they phrased it) non-bicycle commuters had a 38% higher mortality rate (from the US POV, 29% lower mortality rate if you get serious amounts of regular exercise (*)). Follow-on to that study found that bicycle-commuting increased lifespan by 2-5 years.

(*) 29% lower = 38% higher; study found this associated with bicycle commuting, not "exercise", presumably because commuting is daily and non-trivial, unlike some forms of "exercise".

Comment Re:US-sized in that you get more car for the dolla (Score 1) 469

I've never owned an American car, and I've happily driven my various dinky little cars the full length of I-10 and most of I-5. I really don't get this love of the giant cars. A car is just transportation, for those trips in the range where bicycles and airplanes don't work so well.

Comment Re:`why not stop the car? (Score 1) 469

Such technology exists in various forms already. There's bicycles, mass transit, scooters, and small motorcycles. If we had these hypothetical collision-avoiding self-driving cars, they wouldn't need to weigh so damn much "for [my] safety".

Given that someone driving a car can kill people and entirely escape any criminal penalty, I think we've got a long way to go.

Comment Re:My Prius (Score 1) 469

How much extra weight are we talking about? Consider how the driver could probably extend his/her life by losing an amount of weight that is probably comparable, it's unlikely to matter. Add to that, all the people who buy grossly overweight cars because they think they are safer.

One other problem with the "driving off the road" death metric is that some fraction of those are cars traveling at speeds (or on wet/snowy/icy surfaces) where nobody, human or computer, will be able to keep the car on the road.

Comment Re:Turn signals are a good thing (Score 5, Insightful) 469

"Some emergency situations" occur how often? No doubt, for every safety feature on a car there is some fringe case that it makes worse, but the net is (usually) better. What if you *needed* to lock up your brakes and slam the car into a skid, and the ABS prevented it? But overall, ABS is a good thing. We (humans) seem easily distracted by "fault", "intent", and "blame", when it would make a lot more sense to just try to minimize the body count.

Comment Re:What about Google driverless car? (Score 1) 603

PS - I'm going to bet that they are also taking a bonus risk because of the hill. I've been riding bikes for over 40 years, some racing, commuting, some touring. So on the one hand, I can sort of visualize the recipe for intersection X and how it might play out -- but on the other hand, I've finally gotten to the point where I notice that I've been tricked into doing stupid shit. Hills are one of those places -- you get going so fast, the momentum is so lovely, why would you want to slow down and waste it? Cars do this too (not stopping all the way), but it's a big deal on a bike, since that momentum came from your own personal muscles.

Stupid shit #2 is variations on "the awareness test". There are some intersections where you get distracted by (say) the left-turning traffic, and next thing you know, you're running a red light. There is one light that I have zero intention of ever running, where at least twice, I have found myself smack in the middle of it with the light red, without ever noticing the yellow. My attention is clearly drawn to something else, but what? I suspect that this may be a basic car-vs-bike difference, because drivers are usually focussed on the signal (I once watched a driver, focussed on the signal, start her car forward right into a jay-walking pedestrian), where on the bike, you're assuming that anything might happen, because if it does, it hurts, so you are looking at all the other stuff around you. And sometimes, there's enough "other stuff" going on, that you don't notice the light turning yellow.

This might have some relevance to the original topic of this article, way up at the top :-).

(Sorry for the late reply -- I started this in the morning, it's been a busy day.)

Comment Re:What about Google driverless car? (Score 1) 603

A-ha. Their safety "depends". If they are coming downhill in the same direction that you are turning left, and if there are two travel lanes, OR if there is a good place for them to bail (a sidewalk, a shoulder, parking spots), then they're safe. If it's two lanes, they're counting somewhat on your actually executing a clean left turn, though the more experienced (i.e., scarred) cyclists aren't going to depend on that.

By-the-way, at night your headlights give you away, and unless you are driving a hybrid, probably also your engine noise, especially once you start to move. Some cyclists are tricky that way. Others wear earbuds.

The question is, are there often any major swervy altercations, where they have to "correct" very gracelessly? You braking hard is not necessarily an indication that a collision is imminent (they may have a plan that you don't know about, so your braking is redundant safety, not that this is a bad thing), but them engaging in all sorts of nonsense is an indication that they don't have a plan.

If they're coming downhill and cutting across your left turn (i.e., traveling opposite your intended direction), that's just bone stupid, especially since they're rolling down a hill, but that doesn't sound like the case here.

Comment Re:What about Google driverless car? (Score 1) 603

So on the one hand, what you describe is phenomenally dumb, but on the other hand, are you sure it's really blind, and are you sure they're really not looking, and that they're really riding straight across? It's possible they're doing the R-U-R (stands for Right-U-Right, also Rossum's Universal Robots) stunt for getting across an intersection. Enter with a right turn, and as soon as you are IN the intersection and can see what's going on, do a (flattened) U followed by a right turn. This works even better if the first "R" happens to be uphill.

I assume it has to be something like this, because if they were really blasting straight across without looking, there would be bad crashes. What you see, if it's done right, is a lot of "lucky" cyclists who look like they're blasting straight across, and then a number who seem to turn right for a bit, possibly noodle around, and then go.

Slashdot Top Deals

You do not have mail.

Working...