I don't think this story is really very much about the record industry starting to recognize consumer frustration and so on. They simply want to distribute digital music through channels other than iTMS and still maintain access to the iPod market, which is enormous. This is the sensible explanation put forth by the WSJ (although they speculate consumer demand is a driver as well): In a Turnabout, Record Industry Releases MP3s
From the article: Blue Note and other music companies are beginning to think they will have to sell some MP3-formatted music both to satisfy customer demand and to provide access to Apple Computer Inc.'s iPod for songs that are sold by online stores other than Apple's iTunes Store.
It's just the way Western societies have been largely static for centuries now... Rapid technological progress and the careers associated with it is a new phenomenon, and the Western mind hasn't really adjusted...
China may be advancing on the US, but only economically, and not because it is somehow more willing to embrace science and technology. Even if science and technology were the issue, how exactly can one look at the last few centuries of Western scientific progress and changes in daily life and say that it has been "largely static"?!
Chinese people are very much in tune with what is practical for getting ahead, both as a country and individually... A society which respects litigation and playing the stock market more than science and technology won't stay ahead too long.
I agree the Chinese may be more practical about getting ahead and that this could be the reason China is advancing its economy as it is. But Americans have their reasons for being impractical. We so deeply cherish our legal system and stock markets because these institutions preserve and promote individual rights and freedoms, which we see as more important than anything else.
The Chinese conception of individuality - rooted partly in Confucianism, as you noted - does not place the same premium on personal rights and freedoms as the Cartesian conception. An interesting argument could even be made that the Confucian conception may prove in a global economy to be the more practical of the two. And an even more interesting argument could be made that the Confucian conception is inherently more gender-neutral than the Cartesian.
These issues have nothing to with decade-long trends in science and technology, though. They are fundamental values that will differentiate our great societies for many centuries to come.
That said, I think Apple has grander plans than you give them credit for... Don't be shocked to see an Apple prosumer-grade digital camera for around $500, and Apple solid-state camcorder, and Apple PVR...
Okay, I hate to tell you there's nothing really grand about that. Maybe to you, the consumer, it seems grand, but these cheap consumer devices are a means to an end for Apple and everyone else: get people hooked on your cheap stuff today so they'll buy your expensive stuff tomorrow.
That was the entire purpose of the iPod. First, Apple tried to use the iPod as a Mac-only value-add. That didn't increase Mac sales. Then they tried to use the iPod as a try-it-on-Windows "gateway" to the Mac. That didn't work either. Hate to say it, but Mac sales haven't even come CLOSE to matching the success of the original iMac in 1998.
The iPod came out 4 years ago. Thats an eternity in the tech industry. Apple hasn't been able to leverage iPod sales like a lot of other companies would have been able to. And new consumer devices? They'd have done it by now.
So if the iPod is so successful, but Apple can't use it to get you to buy a Mac, the problem is obviously with the Mac. And if the product itself isn't the problem, the problem is obviously with the business of selling Macintoshes.
The only grand plans Apple has are new business plans that sell tons of Macintoshes in creative, new ways. Perhaps the move from IBM to Intel has major technical implications. Big deal... what it really was was ditching one Macintosh partner for a new one - a new business plan. The first of many to come.
Maybe in the early days, no one at Apple would admit there were problems with the business of selling Macintoshes. But Jobs is older now, and Apple doesn't interview the same college drop-out idealists they used to. Now, they're willing to recognize that the business plan is at least as important as the technology plan.
If Intel were to acquire Apple and begin licensing OS X to new Macintosh partners like Dell, it would just be a bold, new business plan. I happen to think it makes a lot of sense, given Cringley's perspective.
The innocent days of the drop-out CEOs making brash decisions and reading poetry at shareholder meetings are over. Its time to get down to business. The tech industry holds the future of the US Economy in its hands and $170B companies don't intend to be left out of the game.
I just finished reading Baghdad Year Zero. Thanks for passing it along. It was tremendously informative and insightful. A must-read for anyone who thinks like a journalist and just follows the money.
The mark of a free society is that you can get lawyers, go to court and fight things like this.
I agree this may be the mark of a free society, but that's exactly why I and so many others are concerned that detainees at camp X-ray are often denied due process.
I suppose a point could be made that these are not American citizens and therefor aren't afforded similair rights, but then again, where do you draw the line?
Well, it has to be challenged and make it into the courts before the courts could do anything about it.
This is an extremely important point. It is not unusual for the Congress to pass unconstitutional laws. But the courts can't do anything about them until until they hear a case concerned with them. Some of this has already ocurred.
So this is why Supreme Court nominations are even more important than these individual Acts in the long run.
I'm not going to disagree that the fans and cooling zones in the G5 tower are efficient and very well engineered. In fact, the point is that they had to be well engineered because the G5 is such a hot chip.
Just because the fans don't run at full speed *most* of the time doesn't mean they don't run full speed *some* of the time, as when one renders a scene in Final Cut Pro and the temperature of a G5 approaches 200 degrees Farenheit. http://www.mac360.com/index.php/mac360/comments/12 7/
I agree its amazing what was accomplished with HyperCard. I didn't know the original Myst was just a HyperCard stack. I should point out though that HyperCard stacks aren't necessarily authored with AppleScript. In fact, its more likely that they're authored in HyperTalk, the native language of HyperCard and in many ways as readable as AppleScript.
As a long-time Mac user, I know that the processor in the Mac mini is totally fine. In fact, the only reason to step up to a higher-end Mac, as far as processor is concerned, is if you were going to jump to duals, which OS X definitely takes advantage of. If you're not even considering duals, then the Mini is perfect.
Replace the internal HD with Hitachi's TravelStar 7K60, which is 7200 RPM w/ 8MB cache and put in 1 GB of ram. The total cost with the 1.25 GHz Mini, and prices off of pricewatch, is about $730. Still the cheapest Mac with great real-world performance.
Let me just say I'm relieved we obviously both understand each other's positions with clarity. For a bit there, I was worried we weren't actually making any progress. Your last post is simply phenomenal, though, and quite thought provoking indeed. Ultimately, however, it is flawed.
First of all, let me point out exactly what it is about morality that I believe is incompatible with functionalism. When we encounter other people, we are actually faced with the same choice of whether to "give [him] full rights and moral concern," as we would be if "faced with a full human-level AI." In neither situation is there actually conclusive evidence the entity before us is sentient. In other words, it is possible that you are the only sentient, experiencing being and that all other people and machines are zombies. This seems absurd, of course, because, despite the lack of evidence, you consistently choose to believe other people are sentient, just as you are. From my standpoint, this is precisely what constitutes morality, or at least its foundation: your belief that you aren't the only sentient being.
Why is this incompatible with functionalism? According to American Heritage, functionalism is "the doctrine in the philosophy of mind according to which mental states are defined by their causes and effects." In our arguments so far, I have labeled you a functionalist because you see in A.I. the possibility for a causal environment (a machine of some sort) that is robust enough to facilitate all of our mental states. I, on the other hand, believe there are mental states outside any causal chain that merely supervene on the daily cause-and-effect activities of our bodies. I've used two examples: appreciation and experience.
Now, getting back to morality, or your belief that you aren't the only sentient being, if all possible mental states did fit inside a causal environment, as you the functionalist would argue, then surely there would be some kind of evidence that the mental states of appreciation and experience are present in other people and machines. After all, why merely believe something if there is already a causal chain that can lead you to the same conclusion? The fact of the matter is that there is no conclusive evidence of these mental states and we only have morality to posit them. Since there is no conclusive evidence of these mental states, they must exist outside the causal environment, which is forbidden by functionalism.
As for your example of the person who piece-by-piece has his brain turned into a toaster, I would have to say I would still "give him full rights and moral concern," even as toast popped out of his head. The reason for this is that his brain was replaced one piece at a time, which allowed "the torch to be passed" so-to-speak. Think of a company of 20 people. If one employee leaves and another replaces him, it's still the same company because the body of knowledge and experience is preserved in the other 19 people and he adapts to it while they adapt to him. If on the other hand, the company was completely replaced by 20 brand new people all at once, I would say it is a different company altogether, because no one was left behind to "pass the torch" to the new people. Similarly, if our crash-prone loved-one had his entire brain replaced all at once, any indication of him "still existing" would merely be functional, and his moral entitlement wouldn't exceed that of the toast popping out of his head.
I think we're talking past each other here. Let me see if I understand your point.
Instead of approaching the task of creating A.I. by programming all the different "features" of our own intelligence, another aproach is to simply program one feature: learning.
Once a learning system has been developed, developing A.I. will be as easy as raising a child (or perhaps as hard as it:-)
In order to "get the ball rolling" in this neural network, which presumably would otherwise be nothing more than a pile of innactive nodes, we'll need to program a handful of "straightforward low level processes." You're suggesting again here that these processes need go no further than enabling learning capacity.
My concern is that not all aspects of human intelligence are learned. As my previous example had indicated, there is strong evidence that our ability to rephrase sentances in two voices ("I bought the shirt" and "The shirt was bought by me") is an inate skill, present at birth.
The key thing is that it is conceivable one might be able to teach only one of these two voices to a learning machine, and it would be completely unaware an alternative was available. This would be a striking difference from humans, whose pre-built brain structures require that both voices are available to the speaker, regardless of what they have or haven't been taught.
No doubt this pre-built brain structure evolved over generations.
So maybe the point that needs to be made is an extension of yours: Once a learning system that contains inate skills developed through the course of (simulated) human evolution has been developed, developing A.I. will be as easy as raising a child.
You cannot program anything into a mind, tou have to teach it things like language, the same way a baby learns.
Yeah, but there's a lot of evidence that babys are indeed born with a certain amount of pre-programming. Some would argue that even language, to use your example, is pre-programmed at very basic levels (e.g. syntax structures). This would be why, for example, every human language has both the "active" voice and "passive" voice, even though many of them developed completely isolated from each other.
So then the question becomes, "well, how much and what should we pre-program?"
And that, of course, just opens up a whole 'nother ancient can-o-worms.
I think what I'm trying to do is argue against functionalism, which is indeed quite popular these days and seems to be what you are in favor of.
What I am trying to show here and in other threads is that just because two things are functionally identical does not mean they are completely identical. The "experience" and "enjoyment" I speak of may only supervene on our functionality (e.g. not actually having any causal impact) but I think moral concern for other beings is rooted in our understanding of this supervenience and belief that it exists in others, even though they may *seem* on the outside to be the mere functions of neurons, as you may put it.
So, functionalism "misses" something about us (the supervenience of experience and enjoyment) and it is precisely that something that morality is entirely concerned with.
I guess the upshot of all this is that I would argue a staunch functionalist couldn't be a moralist. Functionalism and morality, from my point of view, are incompatible.
I don't expect everybody to buy this: it's philosophically sound but still many people regard it as counterintuitive.
The argument may be philosophically valid, but not necessarily sound. The argument, if I understand it correctly is something like:
i = me
m = my memories
b = my old "hardware" (i.e. my body)
n = my new hardware (i.e. a machine or another body)
Cxy = x contains y
Txyz = x is transferred from z to y
Cbm (my old hardware contains my memories)
Tmbn (my memories are transferred from old hardware to new hardware)
IF Tmbn THEN Cnm (if my memories are transferred from old hardware to new hardware, the new hardware contains my memories)
Cnm (the new hardware does indeed contain my memories, by modes ponens on 2 & 3)
i = m (me equals my memories)
Cni (new hardware contains me, by transitivity of identicals on 4 & 5)
Valid logic shown here. However, steps 3 and 5 are semantically debatable. Until proven that transferring memory from one thing to another is fundamentally possible and that "I am my memories," the argument remains valid but not sound.
My point was that just because an organism or a machine desires to live by instinct doesn't mean we should really be very morally concerned about them.
I think what we should spend our moral energies on is protecting the lives of organisms that don't just desire to live by instinct but desire to live because of their enjoyment of it.
If enjoyment of life is some sort of philisophical "pie-in-the-sky" concept, then I suppose you may have a point that we're nothing more than our instincts and a machine could be made like us. However, I believe enjoyment of life is a simple, straight-forward concept that we all know about and is precisely what would separate us from any kind of machine made to imitate our instincts.
I think you and I are on the same page with regard to believing that a mind can't be built. I agree that building an experiencing and appreciating mind is beyond our abilities; necessarily and permanantly.
However, I don't believe the reason for this has anything to do with not understanding our own minds. I think if it were possible to build a mind, we could do so long before understanding our own minds. Much of science is about inventing something that's not really understood and then inventing the science to explain it.
To build an artifical intelligence, we first would have to understand how human intelligence worked, and if you know anything about psychology, you know we're pretty far from that.
I had thought this for a long time as well. However, Prof. Weinberg at Indiana U. quite correctly pointed out that AI may emmerge from a collaborative effort between hundreds of scientists who don't understand "the big picture."
In this case, each scientist or small team would "understand" the functioning of one small part of human intelligence (like visual pattern recognition), but no one would really understand the whole thing, if it emmerged.
You couldn't be more right, Brian. I think people greatly underestimate how fundamentally different are experience and appreciation from mere reaction to stimuli.
While we may indeed get computers to the point where they react to stimuli in much the same way humans do, getting them to actually experience a beautiful sunset and appreciate it, for example, is necessarily beyond our inventive capacity.
In fact, a computer might actually be so well "trained" that it would indeed beg for its life. However, such begging would only be a triggered response and not due to any sort of experience and appreciation of its life. So, to hell with the damn thing and lets spend on energies on debating the morality of pulling plugs on those things that do experience and appreciate their lives!
Slashdot Mirror
I don't think this story is really very much about the record industry starting to recognize consumer frustration and so on. They simply want to distribute digital music through channels other than iTMS and still maintain access to the iPod market, which is enormous. This is the sensible explanation put forth by the WSJ (although they speculate consumer demand is a driver as well): In a Turnabout, Record Industry Releases MP3s
From the article: Blue Note and other music companies are beginning to think they will have to sell some MP3-formatted music both to satisfy customer demand and to provide access to Apple Computer Inc.'s iPod for songs that are sold by online stores other than Apple's iTunes Store.
It's just the way Western societies have been largely static for centuries now... Rapid technological progress and the careers associated with it is a new phenomenon, and the Western mind hasn't really adjusted...
China may be advancing on the US, but only economically, and not because it is somehow more willing to embrace science and technology. Even if science and technology were the issue, how exactly can one look at the last few centuries of Western scientific progress and changes in daily life and say that it has been "largely static"?!
Chinese people are very much in tune with what is practical for getting ahead, both as a country and individually... A society which respects litigation and playing the stock market more than science and technology won't stay ahead too long.
I agree the Chinese may be more practical about getting ahead and that this could be the reason China is advancing its economy as it is. But Americans have their reasons for being impractical. We so deeply cherish our legal system and stock markets because these institutions preserve and promote individual rights and freedoms, which we see as more important than anything else.
The Chinese conception of individuality - rooted partly in Confucianism, as you noted - does not place the same premium on personal rights and freedoms as the Cartesian conception. An interesting argument could even be made that the Confucian conception may prove in a global economy to be the more practical of the two. And an even more interesting argument could be made that the Confucian conception is inherently more gender-neutral than the Cartesian.
These issues have nothing to with decade-long trends in science and technology, though. They are fundamental values that will differentiate our great societies for many centuries to come.
That said, I think Apple has grander plans than you give them credit for... Don't be shocked to see an Apple prosumer-grade digital camera for around $500, and Apple solid-state camcorder, and Apple PVR...
Okay, I hate to tell you there's nothing really grand about that. Maybe to you, the consumer, it seems grand, but these cheap consumer devices are a means to an end for Apple and everyone else: get people hooked on your cheap stuff today so they'll buy your expensive stuff tomorrow.
That was the entire purpose of the iPod. First, Apple tried to use the iPod as a Mac-only value-add. That didn't increase Mac sales. Then they tried to use the iPod as a try-it-on-Windows "gateway" to the Mac. That didn't work either. Hate to say it, but Mac sales haven't even come CLOSE to matching the success of the original iMac in 1998.
The iPod came out 4 years ago. Thats an eternity in the tech industry. Apple hasn't been able to leverage iPod sales like a lot of other companies would have been able to. And new consumer devices? They'd have done it by now.
So if the iPod is so successful, but Apple can't use it to get you to buy a Mac, the problem is obviously with the Mac. And if the product itself isn't the problem, the problem is obviously with the business of selling Macintoshes.
The only grand plans Apple has are new business plans that sell tons of Macintoshes in creative, new ways. Perhaps the move from IBM to Intel has major technical implications. Big deal... what it really was was ditching one Macintosh partner for a new one - a new business plan. The first of many to come.
Maybe in the early days, no one at Apple would admit there were problems with the business of selling Macintoshes. But Jobs is older now, and Apple doesn't interview the same college drop-out idealists they used to. Now, they're willing to recognize that the business plan is at least as important as the technology plan.
If Intel were to acquire Apple and begin licensing OS X to new Macintosh partners like Dell, it would just be a bold, new business plan. I happen to think it makes a lot of sense, given Cringley's perspective.
The innocent days of the drop-out CEOs making brash decisions and reading poetry at shareholder meetings are over. Its time to get down to business. The tech industry holds the future of the US Economy in its hands and $170B companies don't intend to be left out of the game.
I just finished reading Baghdad Year Zero. Thanks for passing it along. It was tremendously informative and insightful. A must-read for anyone who thinks like a journalist and just follows the money.
The mark of a free society is that you can get lawyers, go to court and fight things like this.
I agree this may be the mark of a free society, but that's exactly why I and so many others are concerned that detainees at camp X-ray are often denied due process.
I suppose a point could be made that these are not American citizens and therefor aren't afforded similair rights, but then again, where do you draw the line?
Well, it has to be challenged and make it into the courts before the courts could do anything about it.
This is an extremely important point. It is not unusual for the Congress to pass unconstitutional laws. But the courts can't do anything about them until until they hear a case concerned with them. Some of this has already ocurred.
So this is why Supreme Court nominations are even more important than these individual Acts in the long run.
Legislators haven't picked up on PICS, so why should they support an even worse system?
Because its easier for them to understand.
$Episode[0]... He can still get away with it... but no more after this.
I'm not going to disagree that the fans and cooling zones in the G5 tower are efficient and very well engineered. In fact, the point is that they had to be well engineered because the G5 is such a hot chip.
Just because the fans don't run at full speed *most* of the time doesn't mean they don't run full speed *some* of the time, as when one renders a scene in Final Cut Pro and the temperature of a G5 approaches 200 degrees Farenheit. http://www.mac360.com/index.php/mac360/comments/12 7/
Yeah, getting those "fairly cool and power efficient PPC chips" into a laptop has proven so easy to do, Apple would be foolish to leave the G5 behind.
o sure06082004.jpg
I mean, honestly, what about this graphic says "fairly cool" to you?
http://images.apple.com/powermac/images/indexencl
I agree its amazing what was accomplished with HyperCard. I didn't know the original Myst was just a HyperCard stack. I should point out though that HyperCard stacks aren't necessarily authored with AppleScript. In fact, its more likely that they're authored in HyperTalk, the native language of HyperCard and in many ways as readable as AppleScript.
As a long-time Mac user, I know that the processor in the Mac mini is totally fine. In fact, the only reason to step up to a higher-end Mac, as far as processor is concerned, is if you were going to jump to duals, which OS X definitely takes advantage of. If you're not even considering duals, then the Mini is perfect. Replace the internal HD with Hitachi's TravelStar 7K60, which is 7200 RPM w/ 8MB cache and put in 1 GB of ram. The total cost with the 1.25 GHz Mini, and prices off of pricewatch, is about $730. Still the cheapest Mac with great real-world performance.
Let me just say I'm relieved we obviously both understand each other's positions with clarity. For a bit there, I was worried we weren't actually making any progress. Your last post is simply phenomenal, though, and quite thought provoking indeed. Ultimately, however, it is flawed.
First of all, let me point out exactly what it is about morality that I believe is incompatible with functionalism. When we encounter other people, we are actually faced with the same choice of whether to "give [him] full rights and moral concern," as we would be if "faced with a full human-level AI." In neither situation is there actually conclusive evidence the entity before us is sentient. In other words, it is possible that you are the only sentient, experiencing being and that all other people and machines are zombies. This seems absurd, of course, because, despite the lack of evidence, you consistently choose to believe other people are sentient, just as you are. From my standpoint, this is precisely what constitutes morality, or at least its foundation: your belief that you aren't the only sentient being.
Why is this incompatible with functionalism? According to American Heritage, functionalism is "the doctrine in the philosophy of mind according to which mental states are defined by their causes and effects." In our arguments so far, I have labeled you a functionalist because you see in A.I. the possibility for a causal environment (a machine of some sort) that is robust enough to facilitate all of our mental states. I, on the other hand, believe there are mental states outside any causal chain that merely supervene on the daily cause-and-effect activities of our bodies. I've used two examples: appreciation and experience.
Now, getting back to morality, or your belief that you aren't the only sentient being, if all possible mental states did fit inside a causal environment, as you the functionalist would argue, then surely there would be some kind of evidence that the mental states of appreciation and experience are present in other people and machines. After all, why merely believe something if there is already a causal chain that can lead you to the same conclusion? The fact of the matter is that there is no conclusive evidence of these mental states and we only have morality to posit them. Since there is no conclusive evidence of these mental states, they must exist outside the causal environment, which is forbidden by functionalism.
As for your example of the person who piece-by-piece has his brain turned into a toaster, I would have to say I would still "give him full rights and moral concern," even as toast popped out of his head. The reason for this is that his brain was replaced one piece at a time, which allowed "the torch to be passed" so-to-speak. Think of a company of 20 people. If one employee leaves and another replaces him, it's still the same company because the body of knowledge and experience is preserved in the other 19 people and he adapts to it while they adapt to him. If on the other hand, the company was completely replaced by 20 brand new people all at once, I would say it is a different company altogether, because no one was left behind to "pass the torch" to the new people. Similarly, if our crash-prone loved-one had his entire brain replaced all at once, any indication of him "still existing" would merely be functional, and his moral entitlement wouldn't exceed that of the toast popping out of his head.
I think we're talking past each other here. Let me see if I understand your point.
My concern is that not all aspects of human intelligence are learned. As my previous example had indicated, there is strong evidence that our ability to rephrase sentances in two voices ("I bought the shirt" and "The shirt was bought by me") is an inate skill, present at birth.
The key thing is that it is conceivable one might be able to teach only one of these two voices to a learning machine, and it would be completely unaware an alternative was available. This would be a striking difference from humans, whose pre-built brain structures require that both voices are available to the speaker, regardless of what they have or haven't been taught.
No doubt this pre-built brain structure evolved over generations.
So maybe the point that needs to be made is an extension of yours: Once a learning system that contains inate skills developed through the course of (simulated) human evolution has been developed, developing A.I. will be as easy as raising a child.
You cannot program anything into a mind, tou have to teach it things like language, the same way a baby learns.
Yeah, but there's a lot of evidence that babys are indeed born with a certain amount of pre-programming. Some would argue that even language, to use your example, is pre-programmed at very basic levels (e.g. syntax structures). This would be why, for example, every human language has both the "active" voice and "passive" voice, even though many of them developed completely isolated from each other.
So then the question becomes, "well, how much and what should we pre-program?"
And that, of course, just opens up a whole 'nother ancient can-o-worms.
I think what I'm trying to do is argue against functionalism, which is indeed quite popular these days and seems to be what you are in favor of.
What I am trying to show here and in other threads is that just because two things are functionally identical does not mean they are completely identical. The "experience" and "enjoyment" I speak of may only supervene on our functionality (e.g. not actually having any causal impact) but I think moral concern for other beings is rooted in our understanding of this supervenience and belief that it exists in others, even though they may *seem* on the outside to be the mere functions of neurons, as you may put it.
So, functionalism "misses" something about us (the supervenience of experience and enjoyment) and it is precisely that something that morality is entirely concerned with.
I guess the upshot of all this is that I would argue a staunch functionalist couldn't be a moralist. Functionalism and morality, from my point of view, are incompatible.
I don't expect everybody to buy this: it's philosophically sound but still many people regard it as counterintuitive.
The argument may be philosophically valid, but not necessarily sound. The argument, if I understand it correctly is something like:
Valid logic shown here. However, steps 3 and 5 are semantically debatable. Until proven that transferring memory from one thing to another is fundamentally possible and that "I am my memories," the argument remains valid but not sound.
My point was that just because an organism or a machine desires to live by instinct doesn't mean we should really be very morally concerned about them.
I think what we should spend our moral energies on is protecting the lives of organisms that don't just desire to live by instinct but desire to live because of their enjoyment of it.
If enjoyment of life is some sort of philisophical "pie-in-the-sky" concept, then I suppose you may have a point that we're nothing more than our instincts and a machine could be made like us. However, I believe enjoyment of life is a simple, straight-forward concept that we all know about and is precisely what would separate us from any kind of machine made to imitate our instincts.
I think you and I are on the same page with regard to believing that a mind can't be built. I agree that building an experiencing and appreciating mind is beyond our abilities; necessarily and permanantly.
However, I don't believe the reason for this has anything to do with not understanding our own minds. I think if it were possible to build a mind, we could do so long before understanding our own minds. Much of science is about inventing something that's not really understood and then inventing the science to explain it.
To build an artifical intelligence, we first would have to understand how human intelligence worked, and if you know anything about psychology, you know we're pretty far from that.
I had thought this for a long time as well. However, Prof. Weinberg at Indiana U. quite correctly pointed out that AI may emmerge from a collaborative effort between hundreds of scientists who don't understand "the big picture."
In this case, each scientist or small team would "understand" the functioning of one small part of human intelligence (like visual pattern recognition), but no one would really understand the whole thing, if it emmerged.
You couldn't be more right, Brian. I think people greatly underestimate how fundamentally different are experience and appreciation from mere reaction to stimuli.
While we may indeed get computers to the point where they react to stimuli in much the same way humans do, getting them to actually experience a beautiful sunset and appreciate it, for example, is necessarily beyond our inventive capacity.
In fact, a computer might actually be so well "trained" that it would indeed beg for its life. However, such begging would only be a triggered response and not due to any sort of experience and appreciation of its life. So, to hell with the damn thing and lets spend on energies on debating the morality of pulling plugs on those things that do experience and appreciate their lives!
Do you want to sell sugar water for the rest of your life or do you want to come with me and sell sugar water bundled with free music?