I believe Mark Snoodlegrass said something very similar about episodic television:
"I've heard a lot of insane talk about episodic content. Very little of it makes any actual sense. It's a broken business.
Customers are supposed to watch half the story, then wait a week for the next episode? When I watch a show, I want to watch a new one. Episodic shows will inevitably be using a lot of recycled content, walking through the same sets and filming the same characters with the same actors.
They're competing against massive movie budgets. Distribution without advertising is worthless. You can't buy weekly advertising for a show people can watch for free." He added, "Movies have a cohesive start, middle and end."
Yes, this is sarcastic but it is also to illustrate a point. If television production companies tried to film every episode the way movie companies make movies then it would be practically impossible to have TV series. We have TV series because people modified the process. Some of the change was more or less a no brainer. You didn't sign your star to a new contract before each episode or tear down your sets at the end of each one. Other parts of it required more significant changes. The process a movie goes through for the creation of a script (one or two people working on it at a time before handing it on to someone else) and the process a TV episode goes through for a script (teams of writers working together) are fairly different.
Of course if you assume that game companies have to produce episodic content the same way they've been producing stand along content then it will look like it can't work, but I suspect people will figure out how to modify the process to make it more manageable.
Say when a patrol gets ambushed, they engage the firing system that puts 120 bullets in the area (any area) from which the system detected gunfire.
What happens if the shooter is on the opposite side of the patrol from the robot? Suddenly they are not only being shot at be the original shooter but the robot is firing 120 rounds through them at the sound of the gunfire.
The idea that an autonomous system can be set up by giving it simple parameters doesn't work in situations where a failure of the system can lead to death or severe injury. Systems based on simple parameters are far too likely to have accidental failures due to unforeseen circumstances. Sure, the parameters that drive a Roomba are simple, but then a Roomba isn't likely to kill you if it fails to detect you in its way.
Making this even worse is the fact that for systems used by the military the enemy will be trying to actively defeat it. Try to make your system safer by issuing all soldiers with radio emitters so the robot won't target them and the enemy can just jam the transmitters. Program the robot not to fire at a certain color or pattern and the enemy can wear that same thing.
That doesn't mean that it's impossible to make an autonomous gun. Considering a robot was able to drive-off road along a programmed route for over 100 miles, passing another car along the way I'd say that the capabilities of designing robots to handle real world conditions is pretty high. However that robot had a lot more parameters to follow than to simply drive along a GPS route from point to point. All I'm really saying is that to design such a system takes a lot of work and can't be done through making the robot follow a few simple parameters.
According to the article they are suing because EchoStar is using the technology TiVo patented to let you watch one show while recording another, not the simple fact that they can do so.
If that's the case and an ill informed writer isn't making a mistake or misusing the English language then TiVo's case is a lot more valid. They would not be saying that no one is allowed to record one show while letting you watch another. Instead they would be saying that you can't do it through this particular method which they developed.
Of course that assumes the writer has his facts correct, isn't making a grammatical mistake, and the technology involved isn't extremely broad in definition.
Is this guy for real? Anyone who pauses for 30 seconds to think will realize that the idea of the games coalescing is completely unfeasible.
The idea of an avatar that goes from one game to another, taking with them all their skill and items, makes as much sense as allowing a football player onto the basketball court in full pads and say it's ok for him to tackle whoever has the ball.
Certainly a football player can remove his padding and play basketball with the same rules as everyone else, but at that point he's not taking his rules (which would equate to an avatar's skills) or equipment from one game into another.
The examples he gives to support his theory that the world is going to move this way are laughable. The software worlds of the Matrix and Snowcrash are fictional. Just because someone has envisioned them doesn't mean they will occur as soon as technology permits. People have also been envisioning and building amphibious cars for decades but the last time I drove my Honda into a lake it didn't do very well. What's more, even in their fiction they are really a communication method, not a game.
Combat occurs in the Matrix not because the computers wanted to make the virtual world fun and competitive but because they were trying to fool people into thinking it was real. Combat in Snowcrash on the other hand only existed because of the implausibility of the world. Imagine the chaos if I could go to any email or web post anywhere and click on a button and 'kill' the original poster, kicking them off the Internet for a period of time. It would cut down on flame wars drastically but you'd have kids abusing it just for fun all the time.
Of course there is some precedent on the author's side. Just look at table top RPGs. Originally they were all separate systems. When GURPS came out all those other systems went away since there was now a single system to handle everything. I'll tell you, it sure made it a lot easier for me to find RP groups since I know my fusion cannon wielding cyborg is compatible with what everyone else is playing, even if they were doing a medieval campaign.
Re:Great... just what space entrepreneurs need...
on
Golf in Space
·
· Score: 1
Doh! My bad. I forgot that gr is the abbreviation for grains. Obviously the bullet doesn't weigh 45 grams unless you are talking about a.50 machine gun round or something similar.
Using grains instead of grams produces a golf ball which weighs a bit over one pound which is only off by one order of magnitude instead of two.
Re:Great... just what space entrepreneurs need...
on
Golf in Space
·
· Score: 1
...The golf ball I expect weighs 200x the 220's 45 gr slug....
Man, I'd love to see the impact a 19 pound golfball (45g x 200 = 9kg) would make, assuming someone could hit it as far as a regular golf ball.
Forget worrying about divots. That thing would leave craters.:)
Using that logic nearly every MMO ever made is a failure. The only ones I can think of where they may currently be at their highest subscriber base ever are games like Eve Online, and the only reason for that is because they have always had fairly low subscriber bases but are slowly gaining ground.
Of course eventually they are going to have a declining subscriber base. That's sort of an inevitability. Does that mean that when it finally happens they will be a failure?
How do you figure the naysayers will have been proven right? Even if WoW drops a million subscribers it will have around 4 million subscribers, enough to be considered a wild success by pretty much everyone.
You know, you would think that at least the reporters who wrote the article would use the proper verb, the ability to use English being related to their occupation and all.
'Scientist to implant' means that the scientist is actually doing the implanting. One would assume what the author meant was 'Scientist to have [electrode] implanted'.
I mean, actually implanting anything into your own brain, unless done at velocities exceeding 300 feet per second, would be a pretty neat trick.
You complain that the writer for Wired (along with the editor and fact checker who looked over this) should have caught something like this. I, for one, am comfortable with giving them the benefit of the doubt. I mean, they aren't opthamologists or anything.
What floors me is how Blum, who is an opthamologist, couldn't explain it.
Sony didn't release WoW or Matrix Online. Those were released by other people (Blizzard and Monolith, respectively).
As for trying to follow the theory that if it complicates matters there should only ever be a single MMO you are using the false logic of Post Hoc, Ergo Propter Hoc. Just because there are multiple code bases that exist doesn't mean it doesn't take increased effort to maintain those extra code bases. Obviously it does, otherwise the number of people employed in the manufacturing and maintenance of MMOs would be roughly the same as it was when Everquest first shipped.
This increased effort, however, is taken on by other companies rather than the original 'parent' company and does not result in a direct increase of the amount of effort the parent company must spend, though there may be indirect increases due to the new competition. Why do the other companies take on this burden? Possibly because they feel that the parent company is unlikely to give them a portion of the market revenue just because they ask nicely. This is the nature of competition.
So really your argument should be that if it were true that it takes an increase of effort to maintain two code bases that there should only be one MMO per company. Unfortunately that argument fails because it neglects to take into account diversification. Companies supporting only a single product are vulnerable should that product no longer become desirable. Additionally they are limited in market to only those people interested in that single product. Making multiple products (or code bases) allows the company to draw from a large pool of potential customers.
So following this logic Sony should have kept the old game and made an NGE game and the additional expense of maintaining two code bases could be written off to diversity, right? In theory, sure, but the problem is that if you have two products that draw from the same market pool you will end up with both of them 'cannibalizing' each other. Just because one code base will support and audience of X players doesn't mean two similar code bases will support an audience of 2X. Yes, the combined total will probably be larger than the single code base but it may not be large enough to justify the additional effort of the second code base, possible client frustration because of the confusion of a second code base (oh, I wanted to play SWG Classic) or the lowering of 'player density', something which is important for MMOs, because they are spread across two code bases.
You say that Sony should only cut a code base after it is 'dead'. Using that same logic you should probably drive your car until you wear holes in your tires. After all, until that moment you can still squeeze a few more miles out of them, can't you? Sure, there's some extra headaches with the fact that you have to be more careful driving on the bald tires, and when they finally do quit it will probably be at an inconvenient moment, but you're still saving money because you're not putting wear and tear on a new set of tires just yet, so they will be able to carry you that much further.
Well, I suppose we can live with the ignomy that it was the British who first broke Enigma. England and America has had a special relationship for a long time.
I'm going to put my reply here rather that lower simply because there are so many 'shooting it with a gun won't work' threads down below and I don't feel like cutting and pasting the reply to each of them.
Lots of people are trying to argue how guns will poke small holes in the structure and with low pressure and cellular designs those small holes are manageable. There's even examples given with the Canadian and British trying to shoot down a weather balloon. All of these seem to forget a basic premise of military technology, that if current weapons are inadequate you design new ones.
I would imagine that it probably wouldn't be too terribly difficult to design a new shoulder mounted missile that was designed to penetrate the side of the blimp and then detonate. Payload could be in the form of some incendiary (with its own oxidizer since the helium would smother it otherwise), cluster munitions, or perhaps just thousands of objects designed to rip big jagged holes in the cells as they are converted to shrapnel.
That doesn't mean that the idea is completely unworkable. Modern troop carrying airplanes are fairly vulnerable to shoulder mounted missile fire as well. They deal with the vulnerability by either landing at friendly airfields and off loading troops there or by having the troops jump out at high altitudes. I would imagine the deployment of these blimps would be very similar with 'airfield' being more subjective since you can use a field, but it would still need to be under control.
The premise is this, some radical elements of a religion, really pisssed off over [an opera], [send death threats] in protest for depicting their religion as being [violent, intolerant, and full of hypocrisy]. People seeing these folks reacting like this ([violent, intolerant, and full of hypocrisy]) now see this particular religion as being [violent, intolerant, full of hypocrisy,] and everything that the protesters say they're not. So what happens? People become really afraid of this religion because their actions prove the [opera is] correct. And when people become afraid , they start to do some radical things.
The fact is that just about any religion is going to have people who take such extreme views that they can find argument to kill people (though I will admit I haven't heard about any Buddhist car bombings recently). Islamic extremists embody the beliefs and practices of the majority of Muslims just as much as the Klu Klux Klan embodied the beliefs and practices on main stream Christianity. I don't invoke the name of the KKK in any attempt to be inflammatory but to use them to illustrate a very real point. Just as Islamic extremists center their rhetoric around the beliefs of Islam the KKK wrapped its own rhetoric in the beliefs of Christianity. Even today we have Christian extremists who use bombs against people who do not support their views or who call for the assassination of foreign heads of state.
Does this mean that the religion itself is to blame simply because a few individuals attempt to co-opt bits and pieces of that religion to justify their own skewed world view? Do we assume that because one preacher claims that the stroke suffered by Ariel Sharon is divine retribution all Christians must believe such a thing is true? I should certainly hope that we don't.
Blaming Christianity for the actions of a few extremists is no less fair than blaming Islam (as your post seems to suggest) for the actions of a few of its extremists.
Unfortunately this complicates the system because now the cars need to be intelligent enough to deal with the random and unpredictable events of human drivers. As you have defined it (slow down if there's something ahead of you) the system has no way to compensate for a person weaving back and forth between lanes. An obstacle moving cross-ways to traffic, which a human would spot and slow down for, would be ignored until it was too late since it wasn't in the lane ahead of the car.
Something else you don't seem to have considered, how is the car able to tell it's driving on a safe road? Is the road covered with some special substance that the camera can identify? What happens if a can of paint falls off the back of another car? How can the computer tell the difference between black paint on the road and a break in the roadway?
The rules you've laid out work as long as nothing unexpected happens. Unfortunately you can't engineer a system like this and assume nothing unexpected will happen. You have to try and consider all the events that have any reasonable chance of occurring (meaning you can leave out things like meteor showers). Each new rule you add increases the complexity of what the computer has to do. I'm pretty sure that by the time you have a set of rules comprehensive enough to drive safely your system will be far to complex for a modern affordable computer to handle.
I assumed a human drove the car until they reached the roadway that was exclusively for computer controlled cars. Since it is being driven by a human at this point there is no problem with it driving around cars that do not have the ability to be computer controlled.
Let me ask you this, what is the purpose of having a roadway that is exclusive to computer controlled cars? Is it to prevent the computers from having to deal with human drivers? If so then how can you turn control of some of the cars over to humans? It seems to defeat the purpose. If that isn't the reason for the segregated roadway then what is? Perhaps I am misunderstanding the reasoning behind having it.
As for having no idea where I came up with my ideas, I know exactly where I came up with them. I made assumptions about the reason you proposed certain things. Obviously you seem to be indicating there is some other reason for this segregation of cars so I will now ask you to explain it.
They have a human engineer controlling the train.:)
Ignoring the problems of asking citizens to leave the safety of their vehicles so to remove the obstruction what happens if the obstruction is too large to be moved? One car breaks down and everyone is trapped.
No, I'm not being stubborn. You were the one who proposed to segregate computer controlled cars from human controlled cars, I assume so the computer controlled cars wouldn't have to deal with the unpredictable behavior of humans. You even agreed earlier on to the necessity of limiting access into the system so that a human driver can't drive in, whether accidentally or maliciously. Now you are proposing turning control back over to a human who is still in the system.
If all it took was to have humans to drive carefully then there would be no need to segregate the two. As for the fact that we have humans driving around humans the system works because humans are better at dealing with unexpected events than computers.
I'm not saying that cars will never be driven by computer. They will be able to, just as soon as people can make a computer that handles unexpected situations as well as a human can. Unfortunately what I don't think is workable is the idea of having computers that can't drive as well as humans but that's ok because they all drive in this special lane where nothing unexpected happens.
A nice idea except that a single cardboard box causes major problems. Being restricted to rails you can't simply drive around it and because you aren't on the special section of combined roadway-rails you can't switch to tires.
A robot car with no human driver can still have people in it. They would be passengers, which should answer most of the other questions you ask about the point to a robot car. When I refer to 'no human driver' I am taking the view that once on the roadway with the computer in control the human simply becomes another passenger (albeit one in position to retake control of the vehicle when it exits the roadway or if something should happen).
Anyway, we're really just splitting hairs here. Your design assumes that the cars on the controlled roadway are all being driven by computer, talking with one another and behaving in a predictable pattern, which is why they don't need to be as 'smart' as a normal human driver. As soon as you turn control over to a human being you break that system. Yes, the car can still talk to the cars around it, but now instead of saying 'Get ready, I'm about to change lanes' it is restricted to 'Watch out, I am currently changing lanes'. You will still need to slow all traffic down to the speed of the human controlled vehicle or risk collisions as he changes lanes with your simplified system, with all the problems that I detailed in my previous post.
Even after the advent of QBASIC the Altair never got beyond being a hobbyist's toy. Would it have been as successful without QBASIC? Probably not, but I still think it's a stretch to credit QBASIC with revolutionizing the home desktop computer industry.
If you are really looking for the 'killer app' that revolutionized the home computer industry I would probably go with VisiCalc by Dan Bricklin in 1979. It took the home computer from being a toy and gave it a function that many people had a use for.
Defaulting to a single action like stopping is not always safe. What if the hazard is actually moving towards you? (Perhaps something that fell from a vehicle moving the opposite direction)
Giving control over to the driver is also not an option because of the extreme simplicity of the program behavior. Your whole design assumed there were no human drivers on the road. As soon as you turn control over to a person you have the very real danger of someone making a lane change without alerting the cars around. If the driver, who is now travelling slower than surrounding traffic, changes lanes without adequate room another car will run into him, even if there is a free lane it could swerve into. Forcing all the cars on the road to slow because of the obstruction isn't workable because unless all cars slow to the speed of the human controlled car (which may become stopped) he may cut over without enough room, causing an accident. Stopping all traffic until the human can drive around the obstruction isn't workable because then you will have to stop the traffic for the next car as well. One broken down car would practically shut down the entire freeway.
The 'all or nothing' aspect of the sonar is also problematic. You can't simply decide to slam on the brakes because an object that is not the road or another car crosses the vehicles field of vision. Doing that would force the car (and all the other cars on the road) to stop because of a plastic bag blowing in the wind.
The activity of driving simply has too many quick and subjective decisions that need to be made. Attempting to simplify the experience so that a computer can handle it may be practical in limited scale (which is what car companies do right now when they have robot cars drive thousands of miles around and around to determine wear and tear on new designs) but to implement so that it is practical (sure, the freeway stopping six times a day is safe, but no-one can get anywhere) on a large scale simply isn't possible.
Slashdot Mirror
Yes, this is sarcastic but it is also to illustrate a point. If television production companies tried to film every episode the way movie companies make movies then it would be practically impossible to have TV series. We have TV series because people modified the process. Some of the change was more or less a no brainer. You didn't sign your star to a new contract before each episode or tear down your sets at the end of each one. Other parts of it required more significant changes. The process a movie goes through for the creation of a script (one or two people working on it at a time before handing it on to someone else) and the process a TV episode goes through for a script (teams of writers working together) are fairly different.
Of course if you assume that game companies have to produce episodic content the same way they've been producing stand along content then it will look like it can't work, but I suspect people will figure out how to modify the process to make it more manageable.
The idea that an autonomous system can be set up by giving it simple parameters doesn't work in situations where a failure of the system can lead to death or severe injury. Systems based on simple parameters are far too likely to have accidental failures due to unforeseen circumstances. Sure, the parameters that drive a Roomba are simple, but then a Roomba isn't likely to kill you if it fails to detect you in its way.
Making this even worse is the fact that for systems used by the military the enemy will be trying to actively defeat it. Try to make your system safer by issuing all soldiers with radio emitters so the robot won't target them and the enemy can just jam the transmitters. Program the robot not to fire at a certain color or pattern and the enemy can wear that same thing.
That doesn't mean that it's impossible to make an autonomous gun. Considering a robot was able to drive-off road along a programmed route for over 100 miles, passing another car along the way I'd say that the capabilities of designing robots to handle real world conditions is pretty high. However that robot had a lot more parameters to follow than to simply drive along a GPS route from point to point. All I'm really saying is that to design such a system takes a lot of work and can't be done through making the robot follow a few simple parameters.
If that's the case and an ill informed writer isn't making a mistake or misusing the English language then TiVo's case is a lot more valid. They would not be saying that no one is allowed to record one show while letting you watch another. Instead they would be saying that you can't do it through this particular method which they developed.
Of course that assumes the writer has his facts correct, isn't making a grammatical mistake, and the technology involved isn't extremely broad in definition.
This iPod can go up to 11.
The idea of an avatar that goes from one game to another, taking with them all their skill and items, makes as much sense as allowing a football player onto the basketball court in full pads and say it's ok for him to tackle whoever has the ball.
Certainly a football player can remove his padding and play basketball with the same rules as everyone else, but at that point he's not taking his rules (which would equate to an avatar's skills) or equipment from one game into another.
The examples he gives to support his theory that the world is going to move this way are laughable. The software worlds of the Matrix and Snowcrash are fictional. Just because someone has envisioned them doesn't mean they will occur as soon as technology permits. People have also been envisioning and building amphibious cars for decades but the last time I drove my Honda into a lake it didn't do very well. What's more, even in their fiction they are really a communication method, not a game.
Combat occurs in the Matrix not because the computers wanted to make the virtual world fun and competitive but because they were trying to fool people into thinking it was real. Combat in Snowcrash on the other hand only existed because of the implausibility of the world. Imagine the chaos if I could go to any email or web post anywhere and click on a button and 'kill' the original poster, kicking them off the Internet for a period of time. It would cut down on flame wars drastically but you'd have kids abusing it just for fun all the time.
Of course there is some precedent on the author's side. Just look at table top RPGs. Originally they were all separate systems. When GURPS came out all those other systems went away since there was now a single system to handle everything. I'll tell you, it sure made it a lot easier for me to find RP groups since I know my fusion cannon wielding cyborg is compatible with what everyone else is playing, even if they were doing a medieval campaign.
Using grains instead of grams produces a golf ball which weighs a bit over one pound which is only off by one order of magnitude instead of two.
Man, I'd love to see the impact a 19 pound golfball (45g x 200 = 9kg) would make, assuming someone could hit it as far as a regular golf ball.
Forget worrying about divots. That thing would leave craters. :)
Of course eventually they are going to have a declining subscriber base. That's sort of an inevitability. Does that mean that when it finally happens they will be a failure?
How do you figure the naysayers will have been proven right? Even if WoW drops a million subscribers it will have around 4 million subscribers, enough to be considered a wild success by pretty much everyone.
'Scientist to implant' means that the scientist is actually doing the implanting. One would assume what the author meant was 'Scientist to have [electrode] implanted'.
I mean, actually implanting anything into your own brain, unless done at velocities exceeding 300 feet per second, would be a pretty neat trick.
What floors me is how Blum, who is an opthamologist, couldn't explain it.
Worst.Opthamologist.Ever.
As for trying to follow the theory that if it complicates matters there should only ever be a single MMO you are using the false logic of Post Hoc, Ergo Propter Hoc. Just because there are multiple code bases that exist doesn't mean it doesn't take increased effort to maintain those extra code bases. Obviously it does, otherwise the number of people employed in the manufacturing and maintenance of MMOs would be roughly the same as it was when Everquest first shipped.
This increased effort, however, is taken on by other companies rather than the original 'parent' company and does not result in a direct increase of the amount of effort the parent company must spend, though there may be indirect increases due to the new competition. Why do the other companies take on this burden? Possibly because they feel that the parent company is unlikely to give them a portion of the market revenue just because they ask nicely. This is the nature of competition.
So really your argument should be that if it were true that it takes an increase of effort to maintain two code bases that there should only be one MMO per company. Unfortunately that argument fails because it neglects to take into account diversification. Companies supporting only a single product are vulnerable should that product no longer become desirable. Additionally they are limited in market to only those people interested in that single product. Making multiple products (or code bases) allows the company to draw from a large pool of potential customers.
So following this logic Sony should have kept the old game and made an NGE game and the additional expense of maintaining two code bases could be written off to diversity, right? In theory, sure, but the problem is that if you have two products that draw from the same market pool you will end up with both of them 'cannibalizing' each other. Just because one code base will support and audience of X players doesn't mean two similar code bases will support an audience of 2X. Yes, the combined total will probably be larger than the single code base but it may not be large enough to justify the additional effort of the second code base, possible client frustration because of the confusion of a second code base (oh, I wanted to play SWG Classic) or the lowering of 'player density', something which is important for MMOs, because they are spread across two code bases.
You say that Sony should only cut a code base after it is 'dead'. Using that same logic you should probably drive your car until you wear holes in your tires. After all, until that moment you can still squeeze a few more miles out of them, can't you? Sure, there's some extra headaches with the fact that you have to be more careful driving on the bald tires, and when they finally do quit it will probably be at an inconvenient moment, but you're still saving money because you're not putting wear and tear on a new set of tires just yet, so they will be able to carry you that much further.
We can screw up the other planets later.
At least it isn't as if someone like the Poles broke it first.(http://www.codesandciphers.org.uk/virtualbp /poles/poles.htm)
Lots of people are trying to argue how guns will poke small holes in the structure and with low pressure and cellular designs those small holes are manageable. There's even examples given with the Canadian and British trying to shoot down a weather balloon. All of these seem to forget a basic premise of military technology, that if current weapons are inadequate you design new ones.
I would imagine that it probably wouldn't be too terribly difficult to design a new shoulder mounted missile that was designed to penetrate the side of the blimp and then detonate. Payload could be in the form of some incendiary (with its own oxidizer since the helium would smother it otherwise), cluster munitions, or perhaps just thousands of objects designed to rip big jagged holes in the cells as they are converted to shrapnel.
That doesn't mean that the idea is completely unworkable. Modern troop carrying airplanes are fairly vulnerable to shoulder mounted missile fire as well. They deal with the vulnerability by either landing at friendly airfields and off loading troops there or by having the troops jump out at high altitudes. I would imagine the deployment of these blimps would be very similar with 'airfield' being more subjective since you can use a field, but it would still need to be under control.
http://entertainment.timesonline.co.uk/article/0,, 14929-2013698_1,00.html
The fact is that just about any religion is going to have people who take such extreme views that they can find argument to kill people (though I will admit I haven't heard about any Buddhist car bombings recently). Islamic extremists embody the beliefs and practices of the majority of Muslims just as much as the Klu Klux Klan embodied the beliefs and practices on main stream Christianity. I don't invoke the name of the KKK in any attempt to be inflammatory but to use them to illustrate a very real point. Just as Islamic extremists center their rhetoric around the beliefs of Islam the KKK wrapped its own rhetoric in the beliefs of Christianity. Even today we have Christian extremists who use bombs against people who do not support their views or who call for the assassination of foreign heads of state.
Does this mean that the religion itself is to blame simply because a few individuals attempt to co-opt bits and pieces of that religion to justify their own skewed world view? Do we assume that because one preacher claims that the stroke suffered by Ariel Sharon is divine retribution all Christians must believe such a thing is true? I should certainly hope that we don't.
Blaming Christianity for the actions of a few extremists is no less fair than blaming Islam (as your post seems to suggest) for the actions of a few of its extremists.
Unfortunately this complicates the system because now the cars need to be intelligent enough to deal with the random and unpredictable events of human drivers. As you have defined it (slow down if there's something ahead of you) the system has no way to compensate for a person weaving back and forth between lanes. An obstacle moving cross-ways to traffic, which a human would spot and slow down for, would be ignored until it was too late since it wasn't in the lane ahead of the car.
Something else you don't seem to have considered, how is the car able to tell it's driving on a safe road? Is the road covered with some special substance that the camera can identify? What happens if a can of paint falls off the back of another car? How can the computer tell the difference between black paint on the road and a break in the roadway?
The rules you've laid out work as long as nothing unexpected happens. Unfortunately you can't engineer a system like this and assume nothing unexpected will happen. You have to try and consider all the events that have any reasonable chance of occurring (meaning you can leave out things like meteor showers). Each new rule you add increases the complexity of what the computer has to do. I'm pretty sure that by the time you have a set of rules comprehensive enough to drive safely your system will be far to complex for a modern affordable computer to handle.
Let me ask you this, what is the purpose of having a roadway that is exclusive to computer controlled cars? Is it to prevent the computers from having to deal with human drivers? If so then how can you turn control of some of the cars over to humans? It seems to defeat the purpose. If that isn't the reason for the segregated roadway then what is? Perhaps I am misunderstanding the reasoning behind having it.
As for having no idea where I came up with my ideas, I know exactly where I came up with them. I made assumptions about the reason you proposed certain things. Obviously you seem to be indicating there is some other reason for this segregation of cars so I will now ask you to explain it.
Yeah. I'll definitely agree with that.
Ignoring the problems of asking citizens to leave the safety of their vehicles so to remove the obstruction what happens if the obstruction is too large to be moved? One car breaks down and everyone is trapped.
If all it took was to have humans to drive carefully then there would be no need to segregate the two. As for the fact that we have humans driving around humans the system works because humans are better at dealing with unexpected events than computers.
I'm not saying that cars will never be driven by computer. They will be able to, just as soon as people can make a computer that handles unexpected situations as well as a human can. Unfortunately what I don't think is workable is the idea of having computers that can't drive as well as humans but that's ok because they all drive in this special lane where nothing unexpected happens.
Anyway, we're really just splitting hairs here. Your design assumes that the cars on the controlled roadway are all being driven by computer, talking with one another and behaving in a predictable pattern, which is why they don't need to be as 'smart' as a normal human driver. As soon as you turn control over to a human being you break that system. Yes, the car can still talk to the cars around it, but now instead of saying 'Get ready, I'm about to change lanes' it is restricted to 'Watch out, I am currently changing lanes'. You will still need to slow all traffic down to the speed of the human controlled vehicle or risk collisions as he changes lanes with your simplified system, with all the problems that I detailed in my previous post.
If you are really looking for the 'killer app' that revolutionized the home computer industry I would probably go with VisiCalc by Dan Bricklin in 1979. It took the home computer from being a toy and gave it a function that many people had a use for.
Giving control over to the driver is also not an option because of the extreme simplicity of the program behavior. Your whole design assumed there were no human drivers on the road. As soon as you turn control over to a person you have the very real danger of someone making a lane change without alerting the cars around. If the driver, who is now travelling slower than surrounding traffic, changes lanes without adequate room another car will run into him, even if there is a free lane it could swerve into. Forcing all the cars on the road to slow because of the obstruction isn't workable because unless all cars slow to the speed of the human controlled car (which may become stopped) he may cut over without enough room, causing an accident. Stopping all traffic until the human can drive around the obstruction isn't workable because then you will have to stop the traffic for the next car as well. One broken down car would practically shut down the entire freeway.
The 'all or nothing' aspect of the sonar is also problematic. You can't simply decide to slam on the brakes because an object that is not the road or another car crosses the vehicles field of vision. Doing that would force the car (and all the other cars on the road) to stop because of a plastic bag blowing in the wind.
The activity of driving simply has too many quick and subjective decisions that need to be made. Attempting to simplify the experience so that a computer can handle it may be practical in limited scale (which is what car companies do right now when they have robot cars drive thousands of miles around and around to determine wear and tear on new designs) but to implement so that it is practical (sure, the freeway stopping six times a day is safe, but no-one can get anywhere) on a large scale simply isn't possible.