I think there are some important distinctions that can be made between "True Communism" as espoused by Karl Marx, Communism as practiced by China and the Countries Formerly Known as the Soviet Union, and simple property sharing.
I don't believe that Marx's initial conception of Communism was correct. While state ownership of resources can be a useful economic tool, he missed the boat when he decided that labor was the only thing that had economic value, and his belief that Communism was an inevitable force that drove history turns his conception into something more akin to religion than social science. Finally, if you read his work, you'll realize the whole thing is drawn from some really hare-brained speculations on physics.
Still, I think Marx would have been horrified to see the sort of brutality and supression of thought that marked our modern Communist regimes. Whatever his flaws, he spent his life formulating dialectic materialism because he believed it would bring freedom and happiness to an oppressed working class, and therefore he couldn't be satisfied with the way things turned out.
I'm still of the opinion that small, voluntary communes could work well for many people, as long as it was approached in a pragmatic rather than an ideological manner. That would be communism with a little 'c'.
Just as a tangent, I really doubt the U.S. Government would have valued the life of the dog over the winning of the space race. Therefore, I don't see how that example is useful as a comparator.
While you cannot disprove the possibility, it's impossible to do science unless you assume that conjecture is false.
No point in even doing science, if there's a meddlesome, omnipotent creator who has demonstrated a perfect willingness to mess with your test tubes. You can never be sure that the results of a given experiment are the result of natural laws, if you leave open the possibility that God is breaking in and introducing systematic errors to get you to draw wrong conclusions.
"Theory" is not speculation. In science, you start with a speculation, which is used to frame an initial hypothesis. Once this hypothesis has passed numerous attempts at disproof, then scientists start treating it as though it were likely to be correct. Only then does it become a theory in the sense that scientists use the word. No amount of further confirmation will elevate the idea any higher than "theory".
"Empirical observations" are where scientific laws come in. A law describes an apparent systematic, repeatable observation. For example, "The Law of Gravity" says that masses consistently attract each other. A Theory of Gravity might try to explain the law as the action of gravitons, or depressions in the fabric of space-time.
You say "empirical observation" means "speculative assignment of causational relationship in the absence of direct observation", which I would attempt to refute if only the statement didn't leave me confused and doubtful of your rationality.
Trying to claim equivalence between creationism and evolution is silly. Evolution has millions of observations across billions of years, while Creationism has a dusty old book and a few tired arguments based on misconceptions of evolution. Men wrote the Bible, God wrote the rocks. You choose.
Genes are "mutable" beyond the sort of mixing that comes from sexual reproduction. Viruses can implant their own genetic material. Radiation and other mutagens can change our genetic code.
Once you accept this, all that is required is to accept that sometimes these random mutations end up being beneficial to the organism, and are therefore more likely to be copied than the unmutated genes in the surrounding population.
The fact that "evolution" (in the last few hundred years) hasn't demonstrated itself to the satisfaction of creationists--who frequently show themselves unwilling to accept clear evidence--only drives home the point that these forces work over geological timescales.
The demand by creationists for "intermediate species" is something of a joke. Say that creationists claim there is a "gap" between species X and species Z. Lo and behold, species Y is found.
Rather than admitting the possibility that they're wrong, they'll simply point to the freshly created "gaps" between X and Y, and between Y and Z.
They also ignore the fact that tiny changes in the genotype can sometimes have profound physical implications. For example, dwarfism in humans. So phenotypic "jumps" are possible.
You prefer "common sense" to facts? Thanks for reminding me why I put you on my foe list in the first place.
If this "warm == good, cold == bad" analysis of yours should be so terribly obvious to everyone, then there must be some sort of scientific literature to back you up. If, however, your opinion relies entirely on your own intuition, then that opinion has all the merit of, well, the average slashdot post.
Can you point to any sort of historical survey that shows this period was characterized by a remarkable absence of famines? If so, does the author show how the cause was climatic, rather than political? After all, having free trade and a good road system over large geographical areas, coupled with a notable absence of political disruptions (Pax Romana and whatnot) could have serious implications for the ability to produce food.
Regardless of how this story turns out, I doubt GIS will ever be the fastest, easiest way to locate somebody. While it's theoretically possible to associate locations with the names of their occupants, I haven't heard of anyone actually doing so.
Your other arguments against putting information out to the public are unimaginative. For every terrorist looking for a way to disrupt the natural gas supply, there are a million homeowners who just want to figure out where the freakin' gas line to their house is, without waiting two weeks for the gas company to come mark it. For every person who wants to blow up a power subsystem, there are a thousand real estate developers trying to figure out how to lay the wires for a new subdivision.
The point is, while there are costs and risks to putting the information out there for public consumption, there are also public benefits, which probably far outweigh those risks.
1) Are you trying to be a prick, or does it just come naturally?
2) The article can be applied to most of those "other professionals working long [shouldbeacommahere] thankless hours". Regardless of the nature of your job--so long as it requires even a modicum of creativity--overworking yourself may be less productive than working according to a sane schedule. In short, it's good advice for everybody, and doesn't amount to coders demanding special treatment.
3) Is it really "getting ahead" if it means we die of stress-induced coronaries before the age of 50? On the bright side, at that point we don't really lose much. A couple of decades of neglect should be enough to dump anyone's personal life down the toilet.
4) I think the major difference between you and me is that you appear to idolize the overachievers who put in 12-16 hour days to "get ahead", and seem to get really touchy when that lifestyle is called into question. Me, I consider them to be a bunch of morons who are driven by a mix of greed and ego.
I think it makes perfect sense. While the freedom to run software for any purpose does not literally exist, it isn't the responsibility of the software license itself to remind you that you cannot use the software to blow up federal buildings, run baby mulching machines, or mint counterfeit nickels. Furthermore, it is the acts themselves that are illegal, not the use of software in their commission.
The sort of restrictions on usage that could be put in a license agreement can be divided into two categories. The first are those which merely restate existing law (the sort of freedoms you rightfully claim don't exist), and those that go above and beyond existing law in limiting your ability to use the software. The first category aren't necessary to state. The second category are those which violate the so-called "freedom zero".
Any restrictions on usage found in a software license are either redundant or disempowering. Therefore, requesting that the creators of a piece of software place no restrictions on its use is sensible given the goals of the FSF.
Protectionism, as implemented by the United States over the last twenty years, apparently means, "You third world countries will open your markets, and your governments will cut back on taxes and social spending."
We demand that they open their financial markets, making it quick and easy for foreign investors to shovel money into a good economy (read: inflation) and then take it back out at the first sniff of trouble (read: bankruptcy).
We demand that their governments avoid deficit spending, making no distinction between money being spent on education and public infrastructure, or on facelifts for some dictator's harem. It's like John Maynard Keynes never existed.
We demand that they open up their markets. Now. There are no timetables for job retraining. There are no exemptions for classes of goods that, if left unprotected, would result in millions of people thrown immediately into poverty. There are no delays for giving local industries a chance to modernize and become competitive. They will "liberalize" their economies, and they will liberalize them now, or else we will have the IMF withhold loans and badmouth them to foreign investors.
Meanwhile, we protect and subsidize our agriculture, our steel and aluminum industries, and dozens of other industries. When Russia found that the one thing their economy could do really well was create aluminum, we responded by accusing them of "dumping," when in fact they simply had a competitive advantage because of their abundant cheap energy. Then when that failed, the Bush administration (those champions of the free market) latched onto the idea of creating a worldwide aluminum cartel. Hint: The entire point of a cartel is to keep the price of a given good higher than it would be if its sellers behaved competitively rather than cooperatively. They are also illegal as hell when done by anyone but the government.
The IMF has been pushing for decades for precisely these sorts of "reforms", though they haven't demonstrated any particular knack for improving economies or reducing poverty. But the policies are beloved by American banking institutions, as are the IMF "bailout" loans which are almost invariably targeted towards making sure that when a given economy collapses, foreign investors get their money back. So billionares can make bad investments and be protected from the risks of making those investments.
After the fall of communism, Russia's entire economy was ransacked. We encouraged its government to privatize in a Chinese fire drill* fashion, and threatened to withhold loans if there was any delay. Result? Many critical components of the Russian economy ended up in the hands of rich investors, usually friends of Yeltsin, who were more interested in stripping them of their assets for a quick buck than in taking the harder and riskier road of growing them into healthy and competitive companies.
Conclusion: simple-minded free trade policies, worshipped by so many on the right, hurt the poorest of the poor, while enriching a very few. Read more Stiglitz for a complete blow-by-blow of the incompetence and intellectual bankruptcy of the institutions we've charged with the task of reducing worldwide poverty.
* Besides being a bit racist, the phrase is a bit of a misnomer. China, which has basically told the IMF to go have relations with itself, seems to be doing quite well.
You're forgetting one thing. An MMORPG world is an optional diversion. People go there because it's more enjoyable than sitting around clipping their toenails. If it's a free-for-all where the people at the "top" (read: people who have nine hours a day to devote to leveling and camping for ph4t l00t) are allowed to abuse the people at the bottom, the people at the bottom go back to clipping their toenails.
So, if I'm an MMORPG publisher, what are my options? If I just let the attrition happen, if I just let the abuse continue until the abused leave, I'm going to lose my shirt. The vast majority of people just want to play and have a good time. From a purely financial perspective, I'd rather cater to the people you write off as "losers," because that's where the money is.
You sound like a born griefer, in which case, I'm sure that whatever MMORPGs you frequent would be better off for losing you as a customer.
I never said "if there is no such thing as a soul, then machines can think." Instead, I was arguing that "if a human mind can be correctly simulated, then that simulation is thinking." Your failure to grasp the difference, and obsess instead over this non-sequitor of your own devising, makes me somewhat apathetic towards your charge of moronhood.
Regarding the book I suggested: nowhere did I say that, by reading the book, you would have to fall and worship at the altar of the thinking machine. Rather, I said that it was a serious book with ideas that should be given serious consideration. I would also go so far as to say that no person not familiar with the ideas contained therein can be said to have an understanding of the philosophical underpinnings of the field of artificial intelligence. However, you ascribe to me a level of intellectual intolerance that I most certainly don't feel.
There are people far more talented than me, who understand the arguments far better than I do, and still reject the idea of conscious machines. I respect those people and their opinions. You just don't happen to be one of those people.
I've seen the error of my ways. When you're right, you're right. While I didn't find your arguments persuasive, there was something intellectually compelling about "No fucking 'soul' you fucking retarded fucking single argument mother fucking moron." The clarity of thought shown here convinces me that your wisdom and reason far surpasses my own.
Quite the contrary, I've now written you off as a buffoon. Before I figured that you were a reasonable person who simply didn't understand my argument. Having so written you off, I will proceed with my response, because annoying you gives me a warm fuzzy feeling inside.
You don't believe in a soul. Neither do I. Explain again how this is a failing in my position.
I've been using the term "soul" as a shorthand for "some ineffable 'essence' that gives humans their unique character, which cannot be captured by any simulation." Such an essence is the bedrock of just about every argument against the idea of a "thinking machine". That's why I've been hitting the idea so hard: Those who don't buy into Cartesian dualism mostly agree that a mind can be simulated in principle, and a great many believe that such a simulation is "thinking" in the sense that we understand it.
You've gone an unexpected route, hanging your objection to thinking machines on what I consider a flimsy reed: The Heisenberg Uncertainty Principle.
There is a great deal of your reasoning you haven't explicitly stated. I guess you were too busy deciding which precise string of expletives would carry the day. But if I understand your reasoning correctly, the information contained in the machine couldn't be a perfect replica, because there is no way for us to collect all the necessary data. Further, there is no way for the computer to correctly predict the future quantum states of its simulated particles so they correspond to the ones in the real world.
You're most likely right. An absolutely perfect simulation is impossible. To that extent, my argument has been wrongheaded.
But don't celebrate. I don't believe that this is enough to keep the thinking machines at bay. The simulation I propose, alas, will not be you. The differences between your behavior and the sim's behavior will grow over time. But that ignores the more relevant question: "Is the sim thinking?" When it comes to that question, I don't think quantum indeterminacy helps you at all. For one thing, when you move up to more complex structures like atoms and molecules, the quantum indeterminacy tends to "smear", and as a result the larger structures act with a great deal of regularity.
For another thing, quantum theory deals entirely in probabilities, and these probabilities are calculable. It's very likely that a simulation could behave realistically by simply making arbitrary choices from among those probabilities. Just as incorrect weather simulations still behave in very "weather-like" ways, a bad "you simulator" should still be behaving in ways that appear thoughtful.
Finally, your argument makes less and less sense the further removed the actual patterns of thought are from the subatomic realm. We don't know at precisely what level thought occurs, but the worst case scenario is that the neuron is the fundamental unit. In that case, you would be left arguing that thought cannot be simulated because sometimes QM causes one extra sodium ion among billions to jump the gap between neurons.
Now, if structures within the neuron can store and process information as well, then QM becomes more important. Since we don't know, it doesn't behoove you to hang on too certainly to this particular objection.
I would also like to point out, just as an aside, that if you'd ever read up on computational theory, you would understand the rich and fascinating theory behind the question, "What does it mean to say something is 'computable'?" When people start asking whether
You claim that there is no need to postulate a "soul". As far as I'm concerned, you've just closed the only loophole that would have made AI conceptually impossible.
Here is what I've been trying to get across: The alternative to having a "soul" is that our thinking processes arise entirely from the interactions between the physical matter in our bodies and the physical matter of our outside environment. This matter behaves according to the rules that govern subatomic interactions. A computer with sufficient memory could, given sufficient time, simulate having intelligence by simulating a known intelligent entity (a human being).
Therefore, it is possible for a computer to exhibit true intelligence.
The heuristic I've described for creating an intelligent computer is a stunningly bad one. It's slow, it requires a horrendous amount of computational power, and it can only simulate one type of intelligence (that can never be smarter than us). We cannot even change the simulation so that it "likes onions and dislikes Thai food" without a deeper understanding of how the structure of the brain creates such preferences. But it's enough to prove the point.
If there is something more that is required of the simulation, something about the person that the simulation fails to capture, then it is due to our misunderstanding of the laws of physics (in which case the simulation could be corrected) or there is something a-physical going on that the simulation cannot capture in principle. Like the action of a "soul".
I find the concept of a soul unlikely, but if there is nothing in principle barring us from simulating a human inside a computer, then there is nothing stopping a computer from exhibiting intelligence.
That's my central argument. "Godel, Escher, Bach" and Daniel Dennett's books make a comparable argument, but in a much more convincing and elegant manner. I say you're not informed on the subject of AI not simply because you disagree with me, but because you're blithely making the sorts of arguments that you could not make after having read those books.
A few random rejoinders:
"We can't even predict the weather for the next 24 hours with our simulations. Not even for one square mile. The best we can do is x% chance of rain or sun."
Show me that weather predictions fail even when simulated at the subatomic level, and then you might have a case. As it stands, weather predictions don't have nearly the predictive power they ought to because of a lack of information and a lack of computational power.
More to the point, even the coarse-grained simulations we do today have the power to behave in amazingly "weather-like" ways. If you were to make a comparable simulation of a human being, it wouldn't behave exactly like the person being simulated, but might exhibit intelligence anyways.
You're making a silly request. Figuring out how our minds turn the behavior of meaningless matter into meaningful thought is an ongoing process that engages many of mankind's best minds. The fact that nobody has written a piece of software to turn a supercomputer into an AI shows only that the problem isn't an easy one. I never claimed otherwise.
"Human brains are code. Imagine the following: You create a perfect of me and my environment, down to each and every subatomic particle. You put the whole shebang into a very fast computer, which predicts the future state of these particles in a trillionth of a second (or however small a delta you need to convince yourself that the predictions will be sufficiently accurate). It co
I'll admit that it's likely that you're much more intelligent and reasonable than your last twenty-odd posts would indicate. But if you don't have time to compose a good reply to every post that annoys you, then it's better to respond well to a few than badly to many.
As to your absolutely groundless and insulting insinuation that I have way too much time on my hands.... yeah, point taken.
I fully agree that, in the most fundamental sense, it is impossible for a computer to ignore its own programming. No computer, however intelligent, will ever be able to say, "You know, my thinking processes require me to perform a right bitshift right now. I'd rather do an integer multiply. Here it goes!"
In the same sense, we cannot sit down and decide which neurons we want to fire next. With some research, we might choose to take drugs that stimulate one part of the brain while supressing another, but that research and that decision would both be guided by the actions of the neurons themselves.
I don't believe that there is anything I can do to override the chemical and electrical processes going on inside my head. Since they are--in an essential way--me, then any decision I made to interfere with them would be due to their activity.
After reading some excellent books on the subject (Douglas Hofstadter, Daniel Dennett), I decided that if we exclude the possibility of supernatural intervention, then sentience must be a "computable" process. Given a large enough computer, it could be given a subatomic model of me and my environment, and simulate me just by calculating the future interactions of those particles. The computer itself wouldn't need to be intelligent; all it's doing is deterministically calculating the behavior of subatomic particles. But the data inside would have an awareness of itself, a capacity for irrationality and emotion, and--because it's me in there--an unhealthy Slashdot addiction.
If a machine behaves in such a way that it makes no sense not to call it sentient, then denying it a voice in the political process is simply wrong.
Furthermore, the problems with electronic voting have less to do with the medium itself (computers) than with a lack of transparency and accountability in the way voting machine manufacturers have decided to tabulate our votes. There are no implications to be drawn from electronic voting that can be applied to the competence of artificial intelligences.
[Banal, one-lined response in three... two... one...]
In one sense, at least, society has already admitted that such determinations are possible. The severely mentally retarded and clinically insane aren't generally given voting privileges. They have been judged unequipped to make such decisions.
Beneath that minimum, we've been rather loath to make judgments based on competence. The history of "poll tests" (tests to determine whether a person is competent to vote) are chock full of racism and unwarranted denial of voting rights. So even if they could be fair in theory, the idea has a lot of baggage to overcome. It also runs up against our egalitarian ideals; one man, one vote has been the tradition.
I think it would be possible to implement a basically fair poll test that judges a person's political competence based on generally agreed facts rather than subjective judgments. I also believe that one person's opinion can be more valuable than another's. Say you have three people: a person who doesn't know we have a bicameral legislature*, one who intends to vote for the candidate with the better hair, and one who is very familiar with how laws are made and how various government branches interact. Which would you rather trust to decide who should run the country?
My only question is whether the improvement in the political process would actually be worthwhile. Any system would have to guard against all sorts of potential abuses.
* The question would, of course, be asked without actually using the word "bicameral".
The smartest person in the world may have a much more valuable political opinion than average folks, but there are biologically imposed limits to the difference. I think machine intelligence, when it happens, could far outstrip our own the way the average adult's intelligence outstrips the average six year old's. And we don't even allow six year olds 1/100th of a vote.
I think I understand about as well as a layman can be expected to. You're wrong to think computers cannot be programmed to ignore logic, reason badly, or jump to unwarranted conclusions.
I would ask you to clarify what "more than that" the brain is supposed to be, but given your posting history consists entirely of short, dismissive, and gratuitously insulting posts, I don't figure I'll bother. I'm just not interested.
If, on November 5, 2004, every Republican in the U.S. (congresscritters and GWB excluded) had suddenly died of a disease known only as "right-wing cooties", would the remainder of the population accept the results?
Even if we accepted the idea of artificial intelligences, I don't think the voting system would be "hackable" in this way. Recognizing the ephemerality and reproducibility of intelligences, there would probably be various requirements for voting to address it. Lifespan, uniqueness, purpose, etc.
I'm sure that elaborate systems for determining voting eligibility would be deeply entrenched before the first machine was allowed to cast its first vote. So the paperwork alone might discourage this sort of attack.
Final thought: If it was clear that a machine was vastly smarter, had a vastly better understanding of the implications of an election, and gave every appearance of having interests in line with our own, would we still only give it one vote?
You say that an entity is self-aware only if it can act outside its programmed parameters. But since humans "don't run code" (an assertion I strongly deny) then we have no programmed parameters to compare our behaviors against. Therefore, there is no way to determine whether or not humans are self-aware.
You're ignoring the thrust of my "turkey sandwich" argument: There are two options: Either your mental processes--including desires, emotions, lusts, flights of fancy, etc.--are the product of your physical mind, or you believe that some heretofore undiscovered outside force (call it a "soul") is doing an end-run around the physical mind. This "soul" is causing you to behave in ways that your brain could not. So far, all the arguments I've seen for the latter are problematic and unnecessary, and most of them amount to "the physical mind alone cannot be responsible for thought because I cannot imagine that the physical mind alone can be responsible for thought."
If you believe that the physical mind alone is responsible for thought, then to deny the possibility of artificial intelligence is hopeless.
Human brains are code. Imagine the following: You create a perfect of me and my environment, down to each and every subatomic particle. You put the whole shebang into a very fast computer, which predicts the future state of these particles in a trillionth of a second (or however small a delta you need to convince yourself that the predictions will be sufficiently accurate). It continues to do this over and over, simulating what I will do for the next hour, the next day, the next week.
The model has an awareness of itself, because the molecular machinery it models (me) has an awareness of itself. The model has an awareness of its environment because I have an awareness of my environment, and the environment is part of the simulation as well.
But you're probably saying, "But the computer isn't aware of anything! All it's doing is simulating the motions of meaningless subatomic particles!"
Tell me, friend. If that's true, what is the universe doing if not moving about meaningless subatomic particles? We already have programs that simulate this with a great deal of accuracy, so imagining an entire universe simulated this way requires only changing the scale of the computer. To believe machines cannot think, you must believe there is something in the universe not captured by the sim-universe.
Please, read "Godel, Escher, Bach" and a couple of books by Daniel Dennett before weighing in with such certainty on the subject of AI. Even if you come away disagreeing with their conclusions, you'll be much better equipped to have this argument than you are now.
"Men only act on ANYTHING if they beleive it to be true."
Exactly. Even though just about everyone I walk past has the physical power to cause me real harm (I'm a good-sized and healthy male, but I'm not excluding the possibility of weapons), I feel pretty safe, because the vast majority of people abide by the social contracts that have evolved since we climbed down from the trees.
The government's enforcement of these social contracts aren't what keep them intact. What keeps them functioning is the fact that the people around me were raised to accept these contracts. If we didn't believe in them, we wouldn't have a government that enforced them.
My point was that, despite the more violent factions of the suffragist movement, it didn't achieve its victory by killing everyone who disagreed with its aims. It achieved it by convincing the majority that it was a good idea to allow women to vote.
Why do you imagine that it was impossible for the colonists to get the vote without resorting to violence? England had a tradition of representative democracy, which got exported to the colonies. Colonial legislatures were elected, not appointed. The idea of voting rights was already in place, and there was a healthy debate over whether and how the colonies should be represented in Parliament. In short, I think you're wrong to point to the American Revolution as a situation where rights could only be taken by force.
If you want rights, you have to kill whatever memes are prohibiting you from exercising those rights. Smashing the skull that holds those memes is just one option. There's also persuasion. To deny this is to deny our evolution as social creatures.
Yeah, I read in my history book about how women gained the right to vote. How they rose up, grabbed their double-barreled shotguns from above the fireplace, and shot everyone who thought women's sufferage was a bad idea.
Yep, blood ran in the streets.
Your assertions are only correct in the most literal sense. Rights don't exist if you are physically blocked from exercising those rights, but as a fundamental truth, it's about as helpful as pointing out that the only reason I'm alive is because the people around me haven't decided to kill me.
It would be more helpful to say that rights exist because we continue to believe that they exist. When we stop believing in them, they disappear. Like gods, in a way.
Slashdot Mirror
I think there are some important distinctions that can be made between "True Communism" as espoused by Karl Marx, Communism as practiced by China and the Countries Formerly Known as the Soviet Union, and simple property sharing.
I don't believe that Marx's initial conception of Communism was correct. While state ownership of resources can be a useful economic tool, he missed the boat when he decided that labor was the only thing that had economic value, and his belief that Communism was an inevitable force that drove history turns his conception into something more akin to religion than social science. Finally, if you read his work, you'll realize the whole thing is drawn from some really hare-brained speculations on physics.
Still, I think Marx would have been horrified to see the sort of brutality and supression of thought that marked our modern Communist regimes. Whatever his flaws, he spent his life formulating dialectic materialism because he believed it would bring freedom and happiness to an oppressed working class, and therefore he couldn't be satisfied with the way things turned out.
I'm still of the opinion that small, voluntary communes could work well for many people, as long as it was approached in a pragmatic rather than an ideological manner. That would be communism with a little 'c'.
Just as a tangent, I really doubt the U.S. Government would have valued the life of the dog over the winning of the space race. Therefore, I don't see how that example is useful as a comparator.
While you cannot disprove the possibility, it's impossible to do science unless you assume that conjecture is false.
No point in even doing science, if there's a meddlesome, omnipotent creator who has demonstrated a perfect willingness to mess with your test tubes. You can never be sure that the results of a given experiment are the result of natural laws, if you leave open the possibility that God is breaking in and introducing systematic errors to get you to draw wrong conclusions.
"Theory" is not speculation. In science, you start with a speculation, which is used to frame an initial hypothesis. Once this hypothesis has passed numerous attempts at disproof, then scientists start treating it as though it were likely to be correct. Only then does it become a theory in the sense that scientists use the word. No amount of further confirmation will elevate the idea any higher than "theory".
"Empirical observations" are where scientific laws come in. A law describes an apparent systematic, repeatable observation. For example, "The Law of Gravity" says that masses consistently attract each other. A Theory of Gravity might try to explain the law as the action of gravitons, or depressions in the fabric of space-time.
You say "empirical observation" means "speculative assignment of causational relationship in the absence of direct observation", which I would attempt to refute if only the statement didn't leave me confused and doubtful of your rationality.
Trying to claim equivalence between creationism and evolution is silly. Evolution has millions of observations across billions of years, while Creationism has a dusty old book and a few tired arguments based on misconceptions of evolution. Men wrote the Bible, God wrote the rocks. You choose.
Genes are "mutable" beyond the sort of mixing that comes from sexual reproduction. Viruses can implant their own genetic material. Radiation and other mutagens can change our genetic code.
Once you accept this, all that is required is to accept that sometimes these random mutations end up being beneficial to the organism, and are therefore more likely to be copied than the unmutated genes in the surrounding population.
The fact that "evolution" (in the last few hundred years) hasn't demonstrated itself to the satisfaction of creationists--who frequently show themselves unwilling to accept clear evidence--only drives home the point that these forces work over geological timescales.
The demand by creationists for "intermediate species" is something of a joke. Say that creationists claim there is a "gap" between species X and species Z. Lo and behold, species Y is found.
Rather than admitting the possibility that they're wrong, they'll simply point to the freshly created "gaps" between X and Y, and between Y and Z.
They also ignore the fact that tiny changes in the genotype can sometimes have profound physical implications. For example, dwarfism in humans. So phenotypic "jumps" are possible.
You prefer "common sense" to facts? Thanks for reminding me why I put you on my foe list in the first place.
If this "warm == good, cold == bad" analysis of yours should be so terribly obvious to everyone, then there must be some sort of scientific literature to back you up. If, however, your opinion relies entirely on your own intuition, then that opinion has all the merit of, well, the average slashdot post.
Can you point to any sort of historical survey that shows this period was characterized by a remarkable absence of famines? If so, does the author show how the cause was climatic, rather than political? After all, having free trade and a good road system over large geographical areas, coupled with a notable absence of political disruptions (Pax Romana and whatnot) could have serious implications for the ability to produce food.
Regardless of how this story turns out, I doubt GIS will ever be the fastest, easiest way to locate somebody. While it's theoretically possible to associate locations with the names of their occupants, I haven't heard of anyone actually doing so.
Your other arguments against putting information out to the public are unimaginative. For every terrorist looking for a way to disrupt the natural gas supply, there are a million homeowners who just want to figure out where the freakin' gas line to their house is, without waiting two weeks for the gas company to come mark it. For every person who wants to blow up a power subsystem, there are a thousand real estate developers trying to figure out how to lay the wires for a new subdivision.
The point is, while there are costs and risks to putting the information out there for public consumption, there are also public benefits, which probably far outweigh those risks.
A few random points:
1) Are you trying to be a prick, or does it just come naturally?
2) The article can be applied to most of those "other professionals working long [shouldbeacommahere] thankless hours". Regardless of the nature of your job--so long as it requires even a modicum of creativity--overworking yourself may be less productive than working according to a sane schedule. In short, it's good advice for everybody, and doesn't amount to coders demanding special treatment.
3) Is it really "getting ahead" if it means we die of stress-induced coronaries before the age of 50? On the bright side, at that point we don't really lose much. A couple of decades of neglect should be enough to dump anyone's personal life down the toilet.
4) I think the major difference between you and me is that you appear to idolize the overachievers who put in 12-16 hour days to "get ahead", and seem to get really touchy when that lifestyle is called into question. Me, I consider them to be a bunch of morons who are driven by a mix of greed and ego.
I think it makes perfect sense. While the freedom to run software for any purpose does not literally exist, it isn't the responsibility of the software license itself to remind you that you cannot use the software to blow up federal buildings, run baby mulching machines, or mint counterfeit nickels. Furthermore, it is the acts themselves that are illegal, not the use of software in their commission.
The sort of restrictions on usage that could be put in a license agreement can be divided into two categories. The first are those which merely restate existing law (the sort of freedoms you rightfully claim don't exist), and those that go above and beyond existing law in limiting your ability to use the software. The first category aren't necessary to state. The second category are those which violate the so-called "freedom zero".
Any restrictions on usage found in a software license are either redundant or disempowering. Therefore, requesting that the creators of a piece of software place no restrictions on its use is sensible given the goals of the FSF.
Protectionism, as implemented by the United States over the last twenty years, apparently means, "You third world countries will open your markets, and your governments will cut back on taxes and social spending."
We demand that they open their financial markets, making it quick and easy for foreign investors to shovel money into a good economy (read: inflation) and then take it back out at the first sniff of trouble (read: bankruptcy).
We demand that their governments avoid deficit spending, making no distinction between money being spent on education and public infrastructure, or on facelifts for some dictator's harem. It's like John Maynard Keynes never existed.
We demand that they open up their markets. Now. There are no timetables for job retraining. There are no exemptions for classes of goods that, if left unprotected, would result in millions of people thrown immediately into poverty. There are no delays for giving local industries a chance to modernize and become competitive. They will "liberalize" their economies, and they will liberalize them now, or else we will have the IMF withhold loans and badmouth them to foreign investors.
Meanwhile, we protect and subsidize our agriculture, our steel and aluminum industries, and dozens of other industries. When Russia found that the one thing their economy could do really well was create aluminum, we responded by accusing them of "dumping," when in fact they simply had a competitive advantage because of their abundant cheap energy. Then when that failed, the Bush administration (those champions of the free market) latched onto the idea of creating a worldwide aluminum cartel. Hint: The entire point of a cartel is to keep the price of a given good higher than it would be if its sellers behaved competitively rather than cooperatively. They are also illegal as hell when done by anyone but the government.
The IMF has been pushing for decades for precisely these sorts of "reforms", though they haven't demonstrated any particular knack for improving economies or reducing poverty. But the policies are beloved by American banking institutions, as are the IMF "bailout" loans which are almost invariably targeted towards making sure that when a given economy collapses, foreign investors get their money back. So billionares can make bad investments and be protected from the risks of making those investments.
After the fall of communism, Russia's entire economy was ransacked. We encouraged its government to privatize in a Chinese fire drill* fashion, and threatened to withhold loans if there was any delay. Result? Many critical components of the Russian economy ended up in the hands of rich investors, usually friends of Yeltsin, who were more interested in stripping them of their assets for a quick buck than in taking the harder and riskier road of growing them into healthy and competitive companies.
Conclusion: simple-minded free trade policies, worshipped by so many on the right, hurt the poorest of the poor, while enriching a very few. Read more Stiglitz for a complete blow-by-blow of the incompetence and intellectual bankruptcy of the institutions we've charged with the task of reducing worldwide poverty.
* Besides being a bit racist, the phrase is a bit of a misnomer. China, which has basically told the IMF to go have relations with itself, seems to be doing quite well.
You're forgetting one thing. An MMORPG world is an optional diversion. People go there because it's more enjoyable than sitting around clipping their toenails. If it's a free-for-all where the people at the "top" (read: people who have nine hours a day to devote to leveling and camping for ph4t l00t) are allowed to abuse the people at the bottom, the people at the bottom go back to clipping their toenails.
So, if I'm an MMORPG publisher, what are my options? If I just let the attrition happen, if I just let the abuse continue until the abused leave, I'm going to lose my shirt. The vast majority of people just want to play and have a good time. From a purely financial perspective, I'd rather cater to the people you write off as "losers," because that's where the money is.
You sound like a born griefer, in which case, I'm sure that whatever MMORPGs you frequent would be better off for losing you as a customer.
I never said "if there is no such thing as a soul, then machines can think." Instead, I was arguing that "if a human mind can be correctly simulated, then that simulation is thinking." Your failure to grasp the difference, and obsess instead over this non-sequitor of your own devising, makes me somewhat apathetic towards your charge of moronhood.
Regarding the book I suggested: nowhere did I say that, by reading the book, you would have to fall and worship at the altar of the thinking machine. Rather, I said that it was a serious book with ideas that should be given serious consideration. I would also go so far as to say that no person not familiar with the ideas contained therein can be said to have an understanding of the philosophical underpinnings of the field of artificial intelligence. However, you ascribe to me a level of intellectual intolerance that I most certainly don't feel.
There are people far more talented than me, who understand the arguments far better than I do, and still reject the idea of conscious machines. I respect those people and their opinions. You just don't happen to be one of those people.
I've seen the error of my ways. When you're right, you're right. While I didn't find your arguments persuasive, there was something intellectually compelling about "No fucking 'soul' you fucking retarded fucking single argument mother fucking moron." The clarity of thought shown here convinces me that your wisdom and reason far surpasses my own.
Quite the contrary, I've now written you off as a buffoon. Before I figured that you were a reasonable person who simply didn't understand my argument. Having so written you off, I will proceed with my response, because annoying you gives me a warm fuzzy feeling inside.
You don't believe in a soul. Neither do I. Explain again how this is a failing in my position.
I've been using the term "soul" as a shorthand for "some ineffable 'essence' that gives humans their unique character, which cannot be captured by any simulation." Such an essence is the bedrock of just about every argument against the idea of a "thinking machine". That's why I've been hitting the idea so hard: Those who don't buy into Cartesian dualism mostly agree that a mind can be simulated in principle, and a great many believe that such a simulation is "thinking" in the sense that we understand it.
You've gone an unexpected route, hanging your objection to thinking machines on what I consider a flimsy reed: The Heisenberg Uncertainty Principle.
There is a great deal of your reasoning you haven't explicitly stated. I guess you were too busy deciding which precise string of expletives would carry the day. But if I understand your reasoning correctly, the information contained in the machine couldn't be a perfect replica, because there is no way for us to collect all the necessary data. Further, there is no way for the computer to correctly predict the future quantum states of its simulated particles so they correspond to the ones in the real world.
You're most likely right. An absolutely perfect simulation is impossible. To that extent, my argument has been wrongheaded.
But don't celebrate. I don't believe that this is enough to keep the thinking machines at bay. The simulation I propose, alas, will not be you. The differences between your behavior and the sim's behavior will grow over time. But that ignores the more relevant question: "Is the sim thinking?" When it comes to that question, I don't think quantum indeterminacy helps you at all. For one thing, when you move up to more complex structures like atoms and molecules, the quantum indeterminacy tends to "smear", and as a result the larger structures act with a great deal of regularity.
For another thing, quantum theory deals entirely in probabilities, and these probabilities are calculable. It's very likely that a simulation could behave realistically by simply making arbitrary choices from among those probabilities. Just as incorrect weather simulations still behave in very "weather-like" ways, a bad "you simulator" should still be behaving in ways that appear thoughtful.
Finally, your argument makes less and less sense the further removed the actual patterns of thought are from the subatomic realm. We don't know at precisely what level thought occurs, but the worst case scenario is that the neuron is the fundamental unit. In that case, you would be left arguing that thought cannot be simulated because sometimes QM causes one extra sodium ion among billions to jump the gap between neurons.
Now, if structures within the neuron can store and process information as well, then QM becomes more important. Since we don't know, it doesn't behoove you to hang on too certainly to this particular objection.
I would also like to point out, just as an aside, that if you'd ever read up on computational theory, you would understand the rich and fascinating theory behind the question, "What does it mean to say something is 'computable'?" When people start asking whether
Here is what I've been trying to get across: The alternative to having a "soul" is that our thinking processes arise entirely from the interactions between the physical matter in our bodies and the physical matter of our outside environment. This matter behaves according to the rules that govern subatomic interactions. A computer with sufficient memory could, given sufficient time, simulate having intelligence by simulating a known intelligent entity (a human being).
Therefore, it is possible for a computer to exhibit true intelligence.
The heuristic I've described for creating an intelligent computer is a stunningly bad one. It's slow, it requires a horrendous amount of computational power, and it can only simulate one type of intelligence (that can never be smarter than us). We cannot even change the simulation so that it "likes onions and dislikes Thai food" without a deeper understanding of how the structure of the brain creates such preferences. But it's enough to prove the point.
If there is something more that is required of the simulation, something about the person that the simulation fails to capture, then it is due to our misunderstanding of the laws of physics (in which case the simulation could be corrected) or there is something a-physical going on that the simulation cannot capture in principle. Like the action of a "soul".
I find the concept of a soul unlikely, but if there is nothing in principle barring us from simulating a human inside a computer, then there is nothing stopping a computer from exhibiting intelligence.
That's my central argument. "Godel, Escher, Bach" and Daniel Dennett's books make a comparable argument, but in a much more convincing and elegant manner. I say you're not informed on the subject of AI not simply because you disagree with me, but because you're blithely making the sorts of arguments that you could not make after having read those books.
A few random rejoinders:
Show me that weather predictions fail even when simulated at the subatomic level, and then you might have a case. As it stands, weather predictions don't have nearly the predictive power they ought to because of a lack of information and a lack of computational power.
More to the point, even the coarse-grained simulations we do today have the power to behave in amazingly "weather-like" ways. If you were to make a comparable simulation of a human being, it wouldn't behave exactly like the person being simulated, but might exhibit intelligence anyways.
I looked. It's not up yet.
You're making a silly request. Figuring out how our minds turn the behavior of meaningless matter into meaningful thought is an ongoing process that engages many of mankind's best minds. The fact that nobody has written a piece of software to turn a supercomputer into an AI shows only that the problem isn't an easy one. I never claimed otherwise.
I'll admit that it's likely that you're much more intelligent and reasonable than your last twenty-odd posts would indicate. But if you don't have time to compose a good reply to every post that annoys you, then it's better to respond well to a few than badly to many.
As to your absolutely groundless and insulting insinuation that I have way too much time on my hands.... yeah, point taken.
I fully agree that, in the most fundamental sense, it is impossible for a computer to ignore its own programming. No computer, however intelligent, will ever be able to say, "You know, my thinking processes require me to perform a right bitshift right now. I'd rather do an integer multiply. Here it goes!"
In the same sense, we cannot sit down and decide which neurons we want to fire next. With some research, we might choose to take drugs that stimulate one part of the brain while supressing another, but that research and that decision would both be guided by the actions of the neurons themselves.
I don't believe that there is anything I can do to override the chemical and electrical processes going on inside my head. Since they are--in an essential way--me, then any decision I made to interfere with them would be due to their activity.
After reading some excellent books on the subject (Douglas Hofstadter, Daniel Dennett), I decided that if we exclude the possibility of supernatural intervention, then sentience must be a "computable" process. Given a large enough computer, it could be given a subatomic model of me and my environment, and simulate me just by calculating the future interactions of those particles. The computer itself wouldn't need to be intelligent; all it's doing is deterministically calculating the behavior of subatomic particles. But the data inside would have an awareness of itself, a capacity for irrationality and emotion, and--because it's me in there--an unhealthy Slashdot addiction.
Now you're just being bigoted.
If a machine behaves in such a way that it makes no sense not to call it sentient, then denying it a voice in the political process is simply wrong.
Furthermore, the problems with electronic voting have less to do with the medium itself (computers) than with a lack of transparency and accountability in the way voting machine manufacturers have decided to tabulate our votes. There are no implications to be drawn from electronic voting that can be applied to the competence of artificial intelligences.
[Banal, one-lined response in three... two... one...]
Me.
Oh, wait.
In one sense, at least, society has already admitted that such determinations are possible. The severely mentally retarded and clinically insane aren't generally given voting privileges. They have been judged unequipped to make such decisions.
Beneath that minimum, we've been rather loath to make judgments based on competence. The history of "poll tests" (tests to determine whether a person is competent to vote) are chock full of racism and unwarranted denial of voting rights. So even if they could be fair in theory, the idea has a lot of baggage to overcome. It also runs up against our egalitarian ideals; one man, one vote has been the tradition.
I think it would be possible to implement a basically fair poll test that judges a person's political competence based on generally agreed facts rather than subjective judgments. I also believe that one person's opinion can be more valuable than another's. Say you have three people: a person who doesn't know we have a bicameral legislature*, one who intends to vote for the candidate with the better hair, and one who is very familiar with how laws are made and how various government branches interact. Which would you rather trust to decide who should run the country?
My only question is whether the improvement in the political process would actually be worthwhile. Any system would have to guard against all sorts of potential abuses.
* The question would, of course, be asked without actually using the word "bicameral".
Probably. Not sure on that one.
The smartest person in the world may have a much more valuable political opinion than average folks, but there are biologically imposed limits to the difference. I think machine intelligence, when it happens, could far outstrip our own the way the average adult's intelligence outstrips the average six year old's. And we don't even allow six year olds 1/100th of a vote.
Them's my thoughts, anyhow.
I think I understand about as well as a layman can be expected to. You're wrong to think computers cannot be programmed to ignore logic, reason badly, or jump to unwarranted conclusions.
I would ask you to clarify what "more than that" the brain is supposed to be, but given your posting history consists entirely of short, dismissive, and gratuitously insulting posts, I don't figure I'll bother. I'm just not interested.
Disagree.
If, on November 5, 2004, every Republican in the U.S. (congresscritters and GWB excluded) had suddenly died of a disease known only as "right-wing cooties", would the remainder of the population accept the results?
Even if we accepted the idea of artificial intelligences, I don't think the voting system would be "hackable" in this way. Recognizing the ephemerality and reproducibility of intelligences, there would probably be various requirements for voting to address it. Lifespan, uniqueness, purpose, etc.
I'm sure that elaborate systems for determining voting eligibility would be deeply entrenched before the first machine was allowed to cast its first vote. So the paperwork alone might discourage this sort of attack.
Final thought: If it was clear that a machine was vastly smarter, had a vastly better understanding of the implications of an election, and gave every appearance of having interests in line with our own, would we still only give it one vote?
You say that an entity is self-aware only if it can act outside its programmed parameters. But since humans "don't run code" (an assertion I strongly deny) then we have no programmed parameters to compare our behaviors against. Therefore, there is no way to determine whether or not humans are self-aware.
You're ignoring the thrust of my "turkey sandwich" argument: There are two options: Either your mental processes--including desires, emotions, lusts, flights of fancy, etc.--are the product of your physical mind, or you believe that some heretofore undiscovered outside force (call it a "soul") is doing an end-run around the physical mind. This "soul" is causing you to behave in ways that your brain could not. So far, all the arguments I've seen for the latter are problematic and unnecessary, and most of them amount to "the physical mind alone cannot be responsible for thought because I cannot imagine that the physical mind alone can be responsible for thought."
If you believe that the physical mind alone is responsible for thought, then to deny the possibility of artificial intelligence is hopeless.
Human brains are code. Imagine the following: You create a perfect of me and my environment, down to each and every subatomic particle. You put the whole shebang into a very fast computer, which predicts the future state of these particles in a trillionth of a second (or however small a delta you need to convince yourself that the predictions will be sufficiently accurate). It continues to do this over and over, simulating what I will do for the next hour, the next day, the next week.
The model has an awareness of itself, because the molecular machinery it models (me) has an awareness of itself. The model has an awareness of its environment because I have an awareness of my environment, and the environment is part of the simulation as well.
But you're probably saying, "But the computer isn't aware of anything! All it's doing is simulating the motions of meaningless subatomic particles!"
Tell me, friend. If that's true, what is the universe doing if not moving about meaningless subatomic particles? We already have programs that simulate this with a great deal of accuracy, so imagining an entire universe simulated this way requires only changing the scale of the computer. To believe machines cannot think, you must believe there is something in the universe not captured by the sim-universe.
Please, read "Godel, Escher, Bach" and a couple of books by Daniel Dennett before weighing in with such certainty on the subject of AI. Even if you come away disagreeing with their conclusions, you'll be much better equipped to have this argument than you are now.
The government's enforcement of these social contracts aren't what keep them intact. What keeps them functioning is the fact that the people around me were raised to accept these contracts. If we didn't believe in them, we wouldn't have a government that enforced them.
My point was that, despite the more violent factions of the suffragist movement, it didn't achieve its victory by killing everyone who disagreed with its aims. It achieved it by convincing the majority that it was a good idea to allow women to vote.
Why do you imagine that it was impossible for the colonists to get the vote without resorting to violence? England had a tradition of representative democracy, which got exported to the colonies. Colonial legislatures were elected, not appointed. The idea of voting rights was already in place, and there was a healthy debate over whether and how the colonies should be represented in Parliament. In short, I think you're wrong to point to the American Revolution as a situation where rights could only be taken by force.
If you want rights, you have to kill whatever memes are prohibiting you from exercising those rights. Smashing the skull that holds those memes is just one option. There's also persuasion. To deny this is to deny our evolution as social creatures.
Thank you for that groundless assertion, followed by prophecies of doom. I'm positive that your post will bring a close to this decades-long debate.
Seriously, why do you say this?
Yeah, I read in my history book about how women gained the right to vote. How they rose up, grabbed their double-barreled shotguns from above the fireplace, and shot everyone who thought women's sufferage was a bad idea.
Yep, blood ran in the streets.
Your assertions are only correct in the most literal sense. Rights don't exist if you are physically blocked from exercising those rights, but as a fundamental truth, it's about as helpful as pointing out that the only reason I'm alive is because the people around me haven't decided to kill me.
It would be more helpful to say that rights exist because we continue to believe that they exist. When we stop believing in them, they disappear. Like gods, in a way.