Right. And then there will be the kids who, like all kids, start out as shitty drivers. So, since the kid is a shitty driver, like all beginning drivers, they use their parent’s driverless car. And when they get old enough to need their own vehicle, they get themselves a driverless car because that’s what they’ve always used. And that kid never really learns to drive even as an adult, because they never had the time behind the wheel you need to become a decent driver. So they get locked into driverless vehicles.
In a generation or two, driving your own car is seen as an affectation.
What I’d love to see is driverless cars that all communicate with one another. If a car has to suddenly slow and can instantly tell all the cars behind it what it’s doing, they can all adjust accordingly and avoid the ripple effect that slows everything down. Traffic could also be intelligently directed to alternate routes if the original is particularly busy or a robot car has gone on a rampage and blocked the road with broken bodies.
Exactly. My great grand-dad rode/drove a horse for work. The only skill I have relevant to horses is to recognize one by the roadside and say “Hey everybody! Look at the horsey.”
Car driving as a general population skill will be gone in 50 years. And just like horsemanship today, some folks, e.g. SpeedwayRyan, will still be manually driving cars for hobby or competition.
Well, maybe not like riding a horse to work. More like riding a motorcycle every day. It’s not weird, exactly. But it’s too much hassle for most people. I never learned to ride a motorcycle. I’m sure I could do so pretty easily, but I never have, and so buying myself a motorcycle is something I’m never going to do, just because it sounds like too much work and not much fun.
It’s going to take a lot more than 50 years.
Google cars might begin to be sold in about 10 years…but there are hints of serious delays, and huge debates about very basic issues. These issues will have to be decided by new laws, and by market forces. It will take years to reach a consensus over what people want, and what the legal system will require of them.
For example…up till now, the Google car has always been required to have a licensed driver in charge, who could grab the steering wheel at any time.
But recently Google says they"saw some things that made us nervous", so they are trying to eliminate the steering wheel altogether.
Will the law allow it? Will a self-driven “pod” car with no brake pedal and no steering wheel be allowed to pick up a 16 year old kid after school? how about a 12 year old? Will there be an emergency kill switch, and who is responsible for the accidents which , yes, will occur?
And will the public buy it?
It’s nice to advertise a car which allows you to sleep on the way to work—, but will people buy a car that took you to work but then made you go hungry, because the car prevented you from stopping for breakfast at the McDonald’s drive thru? (say, you hit the button for “go straight to work”, and didn’t realize that your spouse re-set the defaults to use the fastest lanes, so the car continued in the far left lane when you wanted to turn right into the McDonalds.) Not everybody wants to give up control .
It will take many years for the market forces to work and for the manufacturers to correct their mistaken assumptions about what consumers want.
When the first google cars begin to appear on the road, they willl be only 1 per cent of the traffic. Cars are expensive, require loans and lots of paperwork. People won’t rush out to buy one the way they buy the latest iPhone.
And the other 99 per cent of the traffic will be regular cars, which have a lifespan of 15 or 20 years.
I predict that 50 years from now, maybe half the cars on the road will be driverless…but there will still be plenty of us old-fashioned Luddites out there on the road, too. Driving is fun!
Great example. Just like there are horse trails and barns today in the future there will be driver trails and storage for your drivable auto, since you wouldn’t want to have to drive it to the trail.
Driverless cars are made of a few sensors and a pile of computing power and software. Compare computing power today with that available 50 years ago. (I’ve programmed on machines from back then.) Add the fact that driving algorithms will incrementally improve. Think of setting up a new peripheral 20 years ago versus now. Then you had to load the floppy and maybe throw some dip switches - now it is done for you. Think of how Netscape was at a disadvantage because many people were nervous about downloading and installing software. Now it happens automatically.
In 50 years cars are going to be such better drivers than humans that driving yourself, except as a hobby in an area where you can only hurt fellow nuts, will be considered anti-social.
Markets respond pretty quickly to what people want. Look at the now rapid growth of hybrids and electric cars.
That’s like saying that you can’t listen to the radio because your wife set the buttons to stations you don’t like. The car won’t read your mind, but I can’t conceive of a model where you won’t be able to go to McDonald’s when you tell it to go there before leaving. We already can find McDonald’s thanks to Google Maps.
People want control of where and when they go. I don’t think many of us want control of which lane we go into to get there. Especially when the car by communicating with the road can make better choices than we can.
I live in the Bay Area. 10 years or so ago hybrids were rare and expensive. Now they are the most popular car around. The number of plug-in hybrids have been exploding in the past year or so. One big incentive is that they save the buyer time by allowing him or her into the carpool lane. Think of the incentive of saving even more time by letting you do other stuff while driving. Once they hit the right price point, the numbers will explode.
The biggest issue could have been who will spend the money to develop the prototype - but Google has solved this for us. Google is excellent at monetizing ideas - I’m damn sure they have a way of monetizing this one.
One thing that’s going to be a bitch because it already is one (and I’m seriously hoping someone at Google reads this): multilingual areas. Setting your navigator for those is a pain in the ass because the databases aren’t set to have multiple names for a spot; I imagine the autopilot will be based on the same databases.
I have seen only one “map” database which can deal with multilingual addresses, and it’s Bilbao’s public transportation website. For any of its locations it has names in the two local languages plus several foreign ones, plus nicknames, plus a “guessing” algorithm; if you leave out the tilde in España, no problem, it still figures out you’re looking for Plaza Moyúa (Plaza España is its old name), aka la Elíptica (because it’s elliptical). The rest, including google maps, split streets with multiple names into several bits (some parts are called Avenida Loyola, others Loiola Etorbidea), give meshups of several of the names (Avenida Loyiola Etorbidea), or limit themselves to a single language (generally a political decision, not a problem with the database per se).
I’ve had to list my address in the Basque version everywhere, because if I use the Spanish name I get a call from whomever needs to find me saying “uh… I can’t find it…” as the maps have my house under the Basque name, and only under that one. The village, I need to use Spanishname/Basquename because again, otherwise depending on which software they’re using it won’t be found; by giving both, so far the human operators have always been able to find it.
[QUOTE=kanicbird]
I could see some roads, particularly highways, at certain times such as rush hour be autonomous only - want to manually drive, take another route. Perhaps starting with special lanes comparable to HOV lanes, but once this foothold has been established I do expect it to spread to a point if traffic conditions warrant you must switch to auto drive mode or pull over and wait or get off that road with a escape route provided.
[/QUOTE]
[QUOTE=Voyager]
In 50 years cars are going to be such better drivers than humans that driving yourself, except as a hobby in an area where you can only hurt fellow nuts, will be considered anti-social.
[/QUOTE]
Bye bye cyclists, whether the bi- or motor- kind. :mad: It’s [del]to keep the robocars flowing smoothly[/del] for your own safety, of course. :dubious:
Note my bolded words…People dont always think ahead.
When you have a 45 minute drive, how often do you suddenly change your mind after you’ve left?
You see a sign that there’s a low price on gas today, so you decide to pull into the gas station.
You suddenly remember you need to pick up a carton of milk, or you want a cold drink so you stop at the mini-mart on the way home. Or this morning, you want a latte, so , unlike yesterday,you stop at Starbucks, because today is sunny, not raining like yesterday, when you didn’t have an umbrella in the car, etc.
People like to be in control.
The car will have to flexible enough to allow that. It’s possible to do…maybe you will need to swipe a card and type in a PIN number before using the GPS keypad. But it will take time and years of marketing for the public to decide what features they want , and for the legal system to allow it.
For Me…You can take my steering wheel when you pry it out of my cold, dead fingers.
Actually it will make the roads better for us (sometimes) cyclists. The cars, using V2V communication, will be clumped tight together leaving more road space for me on me on my bike. Human drivers often do not percieve me on my bike despite my day-glo windgreaker and lights … they are looking for cars; the computer will see me and give me space. Maybe the day will occur when bikes will have emitters stating where they are to the cars too.
I agree generally, but I was responding to the suggestion that roads (in general, or specific roads) would eventually go “robocars only” because non-automated users would be seen as grit in the machine. If that’s not the case, you’re right that robocars would make things safer for cyclists because they eliminate clueless “didn’t see you there” :dubious: and aggressive “teaching cyclists a lesson” :mad: stunts.
I’d bet that there would be a market for older cars that can’t drive themselves. I’d bet new ones, even when a person is driving, would have all sorts of checks which would be annoying to someone going off the standard roads. Also, no reason to soup up a driverless car.
So, you’ve never changed the destination on your GPS in the middle of a trip? If the GPS can handle it, why not the car?
The one difference is that when you decide you want to get gas at the last minute, your car won’t cut through three lanes of traffic to get to the gas station they way the idiots around here do all the time. These things are trivial to implement.
I think what you want is called a key. If you’ve got authority to start the car, why need anything else? However you might have personalized menus of destinations - then you could either log in or have a personalized key.
In software, which this basically is, you don’t wait to implement until you have user requirements perfect. You go with a good set and then refine with later releases. And driverless cars are already legal in Nevada I think, and clearly Google is not breaking the law in California. Liability is far more interesting, but I suspect a good compromise will be worked out. After all, the car is to blame for some accidents already.
You might change your mind when your work colleagues do an hour more of work a day, while being driven, and get the raises and promotions.
