First of all, I think it's cool that they finally finished this. I hope they'll put it out on DVD.
But what really got me was this translation of a letter from a kid at boarding school, 4400 years ago, writing to his wealthy mother:
quote:"From year to year, the clothes of the (young) gentlemen here become better, but you let my clothes get worse.
"The son of Adad-iddinam, whose father is only an assistant to my father, (has) two new sets of clothes . . . while you fuss even about a single set of clothes for me. In spite of the fact that you bore me and his mother only adopted him, his mother loves him, while you, you do not love me!"
posted
I wish I could remember the exact place I found this (I think in a severely outdated history book my husband got second hand), but I stumbled upon a reference to the children of the nobility who traveled with the first crusade continually asking if they had reached Jerusalem yet when they were still only in Italy.
I love the idea of a bunch of bored medieval kids going "Are we in Jerusalem yet?"
Posts: 4089 | Registered: Apr 2003
| IP: Logged |
posted
Me as well, but here in Phoenix we dont call these people white, we just refer to them as tourists or Scottsdale types.
Posts: 2302 | Registered: Aug 2008
| IP: Logged |
quote:Originally posted by sarcasticmuppet: I wish I could remember the exact place I found this (I think in a severely outdated history book my husband got second hand), but I stumbled upon a reference to the children of the nobility who traveled with the first crusade continually asking if they had reached Jerusalem yet when they were still only in Italy.
I love the idea of a bunch of bored medieval kids going "Are we in Jerusalem yet?"
Hah, yea, that's the same thing as going on a four hour car journey.
Posts: 3564 | Registered: Sep 2001
| IP: Logged |
posted
Man, that really puts things in perspective, eh? Very interesting.
Maybe one day, thousands of years from, someone will look back and see some of the stuff we've written and think "wow, look at that, these people were just like us". Except they'll live in floating cities and their video games will essentially be legalized drug trips thanks to their Matrix-esque technology.
Posts: 1324 | Registered: Feb 2011
| IP: Logged |
quote:Maybe one day, thousands of years from, someone will look back and see some of the stuff we've written and think "wow, look at that, these people were just like us". Except they'll live in floating cities and their video games will essentially be legalized drug trips thanks to their Matrix-esque technology.
There had better be floating cities, dang it. And flying cars.
Posts: 407 | Registered: Jul 2003
| IP: Logged |
posted
Flying cars are an awful idea. Do you really want your teenage son flying around office buildings?
Posts: 4136 | Registered: Aug 2008
| IP: Logged |
posted
Five years ago the newspaper I work for was celebrating its 150th anniversary. I had the opportunity to do some research and write weekly articles about the paper and the area. In 1901, they had some articles about what the area would be like in 1999, and flying cars was part of it. In fact, there was a whole article about a hobo being caught sleeping in a barn, escaping in his flying car and being caught by the pursuing policeman near the Isle of Man. I though it was hilarious that they continued the idea of hobos sleeping in barns but thought the hobos would take off in flying cars. I think I have a copy of the article around here somewhere ...
Posts: 2034 | Registered: Apr 2004
| IP: Logged |
posted
Flying cars are the ultimate "you can't have nice things" — sorry, it's not happening. People are just too dumb.
Posts: 15421 | Registered: Aug 2005
| IP: Logged |
Anyway, flying cars wouldn't be too practical anyway. Now robot maids and jetpacks are a different story.
Posts: 1324 | Registered: Feb 2011
| IP: Logged |
posted
"Times are bad. Children no longer obey their parents, and everyone is writing a book." — Marcus Tullius Cicero, 106 BC – 43 BC
Also from Cicero:
"Six mistakes mankind keeps making century after century: Believing that personal gain is made by crushing others; Worrying about things that cannot be changed or corrected; Insisting that a thing is impossible because we cannot accomplish it; Refusing to set aside trivial preferences; Neglecting development and refinement of the mind; Attempting to compel others to believe and live as we do."
Posts: 7790 | Registered: Aug 2000
| IP: Logged |
quote:Originally posted by Chris Bridges: "Times are bad. Children no longer obey their parents, and everyone is writing a book." — Marcus Tullius Cicero, 106 BC – 43 BC
How did you know I was writing a book?
quote:Also from Cicero:
"Six mistakes mankind keeps making century after century: Believing that personal gain is made by crushing others;
But our society kind of demands that we do this, and the result is usually a better job, a hotter girlfriend, more money, or some other personal gain...so I don't think it has much to do with believing as it does with the fact that it actually works.
quote:Worrying about things that cannot be changed or corrected;
Wait...according to Cicero here, nothing is impossible, which means everything can be changed, so worrying about something that we can't change doesn't make sense. Dammit Cicero! You and your circular logic!
quote:Insisting that a thing is impossible because we cannot accomplish it;
Right, because 1+1 can totally equal 7, as long as we believe.
Also, by saying that nothing is impossible, Cicero is also saying that finding something impossible IS impossible, which means this is a paradox! OMG!
quote:Refusing to set aside trivial preferences;
Says you, Cicero. I happen to enjoy my X-Box 360 and Baked Ruffles potato chips, thank you very much!
quote:Neglecting development and refinement of the mind;
So the atom bomb is a good thing, then? Huzzahs!
quote:Attempting to compel others to believe and live as we do."
Henceforth every Friday shall be known as Happy Pants Day, and we all shall have root beer and tacos for dinner! All those who oppose my decree shall suffer the wrath of the whiffle bat!
Lol, by the way, I'm only kidding
[ June 08, 2011, 01:02 PM: Message edited by: Jeff C. ]
Posts: 1324 | Registered: Feb 2011
| IP: Logged |
quote:Originally posted by Samprimary: Flying cars are the ultimate "you can't have nice things" — sorry, it's not happening. People are just too dumb.
Well human piloted flying cars are a no go but with sufficiently advanced computers and navigation automated flying cars are possible.
Really taking humans out of driving any cars would probably save countless lives.
Posts: 305 | Registered: Jan 2008
| IP: Logged |
quote:Believing that personal gain is made by crushing others;
But our society kind of demands that we do this, and the result is usually a better job, a hotter girlfriend, more money, or some other personal gain...so I don't think it has much to do with believing as it does with the fact that it actually works.
No, the result of crushing others is usually going to jail, or to live in a society so destroyed that no gain is made.
quote:
quote:Insisting that a thing is impossible because we cannot accomplish it;
Right, because 1+1 can totally equal 7, as long as we believe. [Razz]
Also, by saying that nothing is impossible,
He didn't say "everything is possible as long as we believe", he said "you can't call something impossible, just because *you* can't do it."
The second part is important. The fallacy isn't in calling things impossible, the fallacy is in calling them impossible because of your personal (or even collective) inability to do them.
quote:Well human piloted flying cars are a no go but with sufficiently advanced computers and navigation automated flying cars are possible.
Aren't we calling these flying cars "helicopters" and "private small planes"?
Posts: 676 | Registered: Feb 2003
| IP: Logged |
quote:Well human piloted flying cars are a no go but with sufficiently advanced computers and navigation automated flying cars are possible.
Really taking humans out of driving any cars would probably save countless lives.
You dramatically overestimate the current state of artificial intelligence. The day when computers will be able to drive cars more safely than humans is still a very long way off.
Posts: 12591 | Registered: Jan 2000
| IP: Logged |
quote:Believing that personal gain is made by crushing others;
But our society kind of demands that we do this, and the result is usually a better job, a hotter girlfriend, more money, or some other personal gain...so I don't think it has much to do with believing as it does with the fact that it actually works.
No, the result of crushing others is usually going to jail, or to live in a society so destroyed that no gain is made.
quote:
quote:Insisting that a thing is impossible because we cannot accomplish it;
Right, because 1+1 can totally equal 7, as long as we believe. [Razz]
Also, by saying that nothing is impossible,
He didn't say "everything is possible as long as we believe", he said "you can't call something impossible, just because *you* can't do it."
The second part is important. The fallacy isn't in calling things impossible, the fallacy is in calling them impossible because of your personal (or even collective) inability to do them.
quote:Well human piloted flying cars are a no go but with sufficiently advanced computers and navigation automated flying cars are possible.
Aren't we calling these flying cars "helicopters" and "private small planes"?
No. I was thinking the traditions SCI-FI "flitter"
Posts: 305 | Registered: Jan 2008
| IP: Logged |
quote:Well human piloted flying cars are a no go but with sufficiently advanced computers and navigation automated flying cars are possible.
Really taking humans out of driving any cars would probably save countless lives.
You dramatically overestimate the current state of artificial intelligence. The day when computers will be able to drive cars more safely than humans is still a very long way off.
No it's not. Googles's experiments show how surprisingly easy setting up such a system could be, The real issue is liability concerns and getting the public on board. Both of those problems are harder than the technical aspects.
Posts: 305 | Registered: Jan 2008
| IP: Logged |
quote:No, the result of crushing others is usually going to jail, or to live in a society so destroyed that no gain is made.
But social darwinism says it works, and so it must be true.
Now if you'll excuse me, I'm going to put my hands over my ears and yell "lalalalala" really loud so I can't hear your rebuttle. Muhaha!
Posts: 1324 | Registered: Feb 2011
| IP: Logged |
posted
aeolusdallas, I've read quite a few research articles about AI driving systems and you're wrong.
Posts: 12591 | Registered: Jan 2000
| IP: Logged |
quote:aeolusdallas, I've read quite a few research articles about AI driving systems and you're wrong.
No, he's not. While current driving AI is by no means able to deal with all scenarios (for instance, a major thunderstorm at night is right out), it is nearly at the point of approximately comparable accident rates in normal driving scenarios; certainly no more than five or ten years away. What's more, the current good driving record of google's AI cars (~140k miles of almost human free driving and ~1k miles of trips that had absolutely no human intervention, last they updated anyone on the numbers, which was late 2010, so I'd expect double that), where the only accident they've been in was being rear ended, is without enhancement of highways and without there being other self driving cars on the road to communicate with. Add one or both of those, and self driving cars suddenly become exponentially more capable.
That's the state of the technology for self driving cars: nearing good enough to beat humans in 90% of the driving we do (and smart enough to be able to pull over and tell the human to take over). The state of social and legal acceptance of self driving cars is probably further off (of course, the process won't really start until they're more common), though I'm optimistic that, following experiments in some of the less trafficked states, they'll see rapid adoption due to saving lives (and time, and so many other things).
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
quote:aeolusdallas, I've read quite a few research articles about AI driving systems and you're wrong.
No, he's not. While current driving AI is by no means able to deal with all scenarios (for instance, a major thunderstorm at night is right out), it is nearly at the point of approximately comparable accident rates in normal driving scenarios; certainly no more than five or ten years away. What's more, the current good driving record of google's AI cars (~140k miles of almost human free driving and ~1k miles of trips that had absolutely no human intervention, last they updated anyone on the numbers, which was late 2010, so I'd expect double that), where the only accident they've been in was being rear ended, is without enhancement of highways and without there being other self driving cars on the road to communicate with. Add one or both of those, and self driving cars suddenly become exponentially more capable.
That's the state of the technology for self driving cars: nearing good enough to beat humans in 90% of the driving we do (and smart enough to be able to pull over and tell the human to take over).
I based my assessment on what I've learned from reading the details and not just the press releases. AI systems are very good at freeway driving, where they fall short is complex mixed use traffic systems. In the Google AI tests, for example, the human took control of the vehicle in situations where there were pedestrians, bicycles, cars backing out of parking spaces and so forth. The critical thing is that these are the situations where most accidents occur.
As a human driver, you see a kid playing with a ball or a car sitting in the parking lot exit and you make an assessment about what they are likely to do. If the ball starts heading for the road, you know the kid is likely to follow and you slow down. Current AIs are no where near sophisticated enough to make those kinds of assessments, which is why the human drivers take over in those situations.
And sure, the human driver can take over in those situations, but there is a huge difference between how a human AI developer testing a radical new system is going to act and react, and the way your typical Joe is going to act and react when he's "driving" a car that's on auto pilot. Once people become comfortable with autopilot, they won't ride along studying the street alertly for potential traffic hazards that might require their intervention. They barely do that when they are 100% in control.
Posts: 12591 | Registered: Jan 2000
| IP: Logged |
Blayne Bradley
unregistered
posted
I'm waiting for my holographic AI singing idols.
The day I can install Hatsune Miku on my computer and fully interact with her as I would a person!? Omg omg omg.
IP: Logged |
quote:I based my assessment on what I've learned from reading the details and not just the press releases. AI systems are very good at freeway driving, where they fall short is complex mixed use traffic systems. In the Google AI tests, for example, the human took control of the vehicle in situations where there were pedestrians, bicycles, cars backing out of parking spaces and so forth. The critical thing is that these are the situations where most accidents occur.
And if you read closely, you'll see that the person riding along noted that the car was already avoiding the situation when the human took over, and did not seem like it was going to crash. The human took over because the human is liable in the off chance and Google desperately wants to avoid bad publicity, not because the AI wasn't capable of handling the situation. Unfortunately, Google isn't releasing information on their simulations of unexpected driving conditions on courses they control (and can thus take the risks of crashes on), but given they're testing the cars on lots of normal city streets (including the steepest one in SF), they don't seem to be acting like they haven't got in a lot of progress on the problem.
Keep in mind no one is promising perfection. When it comes down to it, humans aren't very good at handling those sorts of situations either, and when an object does enter the roadway unexpectedly, computers are much, much faster at reacting, and able to react in a more appropriate way from the start.
As you say,
quote:They barely do that when they are 100% in control.
That's why the goal of self driving cars is closer than you think it is: because they only need to reduce accidents (especially major accidents) relative to human drivers to be a great idea. Given how many accidents are caused by, for instance, serious judgment errors by sleepy drivers, that goal is astonishingly realizable.
To take your example:
quote:As a human driver, you see a kid playing with a ball or a car sitting in the parking lot exit and you make an assessment about what they are likely to do. If the ball starts heading for the road, you know the kid is likely to follow and you slow down. Current AIs are no where near sophisticated enough to make those kinds of assessments, which is why the human drivers take over in those situations.
And a human who is over focused on, say, looking for the street they want to turn on is often incapable of noticing a small child wandering out into the street in front of them, but an AI car is right now capable of noticing that and going into active avoidance, avoiding the child before a human could even if they were looking straight at the child as it wandered out from behind a parked car.
Posts: 15770 | Registered: Dec 2001
| IP: Logged |
quote:And a human who is over focused on, say, looking for the street they want to turn on is often incapable of noticing a small child wandering out into the street in front of them, but an AI car is right now capable of noticing that and going into active avoidance, avoiding the child before a human could even if they were looking straight at the child as it wandered out from behind a parked car.
Can you give me a source on this because it's contrary to everything I've read.
Posts: 12591 | Registered: Jan 2000
| IP: Logged |
posted
Not that i wouldn't totally want a remote controlled car, but some studies have been done that suggest the economics of it are just downright unpleasant. I'll explain...
So, in order to get an AI controlled vehicle to perform adequately, there would have to be sensors on the roads. These sensors would probably need to be placed every couple dozen yards away from the last. Imagine doing this all over the United States and how much it would cost (it would be a lot). If you wanted to have these cars fly (which I, for one, would love), there would still need to be sensors on the ground and on buildings.
Next, imagine the lawsuits. That's right, the lawsuits. Everytime an AI messed up, the car would stop, crash, or something else, and people would sue. And if it wasn't the AI, it would be the sensors. Granted, wrecks happen all the time, but usually it's due to human error rather than the car itself (although there are certainly times where the brakes will go out, to be certain, but if you're hovering 300 feet in the air, you're going to die). Ultimately, it would prove too risky of an investment, which means it would need to be privately funded (which could happen) and then be shown to be truly safe, which is unlikely considering the current glitchiness of computer technology (I know, I fix them everyday).
Gas would also be a problem. People run out of gas all the time, and that's not going to change. If it happens in a flying car, you aren't just stranded, you're dead. And no one wants to be dead (unless they're depresed or married).
But yes, those are the three main reasons why cars are not automated or don't fly. At least according to the History Channel.
Posts: 1324 | Registered: Feb 2011
| IP: Logged |
quote:So, in order to get an AI controlled vehicle to perform adequately, there would have to be sensors on the roads. These sensors would probably need to be placed every couple dozen yards away from the last.
This is not true with current AI technology. The google AI car is able to drive on regular existing roads. There are AI car competitions, where cars drive off road over desert terrain.
The big limitation on AI cars (as I understand it) is the ability to anticipate. They respond very rapidly when something happens but have no ability to anticipate.
Posts: 12591 | Registered: Jan 2000
| IP: Logged |
quote:So, in order to get an AI controlled vehicle to perform adequately, there would have to be sensors on the roads. These sensors would probably need to be placed every couple dozen yards away from the last.
This is not true with current AI technology. The google AI car is able to drive on regular existing roads. There are AI car competitions, where cars drive off road over desert terrain.
The big limitation on AI cars (as I understand it) is the ability to anticipate. They respond very rapidly when something happens but have no ability to anticipate.
That's cool. Still, lawsuits and cost would be a problem. Of course, cost can always go down and safety can go up, but it will be quite some time before that happens.
I think the biggest issue plaguing the auto-industry is fuel. We really need to make a transition to something other than fossil, but that would probably destroy the world's economy.
Posts: 1324 | Registered: Feb 2011
| IP: Logged |
quote:I think the biggest issue plaguing the auto-industry is fuel. We really need to make a transition to something other than fossil, but that would probably destroy the world's economy.
The adjustment will happen quite naturally as prices adjust.
Rabbit: there are several companies that make the detection parts of such systems, able to accurately and instantly identify obstacles, including relative location of the obstacle, including unexpected obstacles. It's a pretty much solved problem, at least under non-extreme weather conditions. Here's one such company: http://www.mobileye.com/
Human reaction time, however, is about 1.5 seconds before they even begin to react to something, and that's if they're paying attention. Someone focused on something else will easily take two to five seconds to react to a child running into the road, by which time a self driving car will have easily been able to divert around it/brake (avoiding known obstacles isn't hard, nor is predicting their motion -- something the google cars already do, including the motion of things not yet on the roadway).
Printer-friendly view of this topic
![[Big Grin]](biggrin.gif)
![[Wink]](wink.gif)
![[Razz]](tongue.gif)
![[Wall Bash]](graemlins/wallbash.gif)
![[ROFL]](graemlins/rofl.gif)
