Except that this wasn't in the public. This was on Apple's private property. Also, it appears to have been done at least partially without people's consent under conditions that they would not have expected to be photographed. You might not mind having hidden cameras taking pictures of you and posting them on the internet without your consent but a lot of (most?) other people do.
And why would the number of people unemployed and not looking for a job increase, but not those looking for a job?
Because a larger fraction of the potential labor market has given up. For example, if the only jobs that could not be outsourced or automated required a PhD, then a lot of people will simply leave the labor market because they may feel that they have no chance to get a job. For example, the unemployment rate can remain constant even though there are more people panhandling and begging on the streets. I'm not saying that this is what happened between the 70s and today but it makes sense given other trends, such as increased outsourcing, education costs, and automation.
The problem here is that Microsoft effectively made Android anything but free, which is exactly the opposite of what Google wanted to achieve with the OEM brand perception of Android as a platform, and that in and of itself is a fantastic business strategy. I can't even remotely justify it as either humiliating or desperate; it's well-played despite being immensely back-handed.
But that's the whole misunderstanding about Android. Who ever said that manufacturing and selling an Android device was supposed to be free anyway? As a hardware manufacturer, I don't see how paying this "Microsoft" tax is much different from the other telephony-related licensing fees that cellphone OEMs already have to pay. If there's a GSM radio, then there are also patent licensing fees with Nokia that have to be paid. It's just another per unit cost to building a physical device, just like it costs $X to use Gorilla Glass or a Sony lithium ion battery. This is just par for the course. I also don't see anything particularly special about the fact that these fees have to be paid to Microsoft. The only significance that I can see is that it is a unit cost to manufacturing an Android phone that (presumably) wouldn't have to be paid by a Windows Phone 7 manufacturer. Yet, it still doesn't mean that manufacturing and selling a Windows Phone 7 device is more cost effective than a comparable Android device simply because a Windows Phone 7 device will have its own unique unit costs. Plus, I imagine that this $5 fee essentially pays for itself simply from the relative strength of the Android platform vis-a-vis Windows Phone 7.
Exactly. Is it really that difficult to express legal contracts in plain language? Can't contracts (and the laws themselves, for that matter!) be expressed at the level of, say, a newspaper? Maybe the New York Times? One of the first lessons of writing is KNOW YOUR AUDIENCE. If the audience is other lawyers than go right ahead, use all the obfuscatory language you want. I'm sure it reads like poetry to some people. But you aren't trying to communicate properly if you're using lawyer-speak and you know the audience isn't made up of lawyers and isn't likely to fully understand the contract without one. And the problem starts at the top, with the laws themselves. It's like the laws are written by lawyers for lawyers; they're not intended to be understood by the general public (bingo?). It's funny how lawyers get such a bad rap here on Slashdot and yet, at least in the US, and at the Federal level, we are governed by lawyers. Why would they ever institute change that would hurt their own?
Worthless options are one thing but from the Businessweek article, it appears that the guy ended up OWING taxes for worthless options. Try figuring that one out.
I'm not a lawyer but it is called an "unconscionable" contract. Depending on jurisdiction, contracts that are deemed "unconscionable" can be declared void by a judge or in some cases, just the part that is "unconscionable".
Agreed. I love it how you have all these people saying, "JUST READ THE CONTRACT!" Well, hell, we wouldn't lawyers if it were that easy, now would we? The problem isn't just reading contract, it's reading the contract and understanding the contract. From TFA, it looks to me like this guy did read the contract, at least as well as most people probably do and it seemed to okay to him because he thought he understood it but didn't understand it completely. If the first page of the contract says you're fully vested after 1 year but 50 pages later in some footnote it says you're not fully vested in 1 year than I would certainly cry foul. I suspect that most people sign contracts not based on fully reading and understanding them but because they don't see other people complaining or raising a stir. If you don't have experience with hidden clauses and gotchas in contracts then you're probably not going to be looking for them -- and are likely to miss them even if you DO read the whole contract verbatim. Rightly or wrongly, much of our society is based on trust and an expectation that most people aren't being grossly dishonest.
If it prevents some real criminals from breaking into Sony in the future and actually using that info, they've done good.
Except that the "real criminals" don't have to break into Sony because these hacker groups (don't know if it was LulzSec or not) have already done it for them. That's the flaw in this type of argument. Hacking to teach a lesson when real people's data is exposed lowers the barrier to entry for identity thieves and similar criminals. Before, they may have needed some nominal level of competency with computers to break into Sony and get people's credit card info. Now, all they need to do is sit tight and wait for LulzSec and related groups to expose people's info and reap the ill-begotten rewards before the victims have a chance to respond. Sure, future victims might be avoided but they may have been avoided anyway since the "real" criminals might have been focusing on targets other than Sony. Hell, for all we know, Sony might have been planning to patch the vulnerabilities that were exploited. Now, I'm not excusing Sony's incompetence but it seems to me that there were a lot better ways to make a point about security holes than hurting innocent people.
If anything, the "man on the street" test is often the best one and usability testing is always best done with a total beginner.
I agree to some extent but I think the opinions of the average Joe only matter in aggregate. I wouldn't trust the untrained review by itself. You can't expect some random person off the street to do a proper review of a complex gadget like a tablet. A good thorough review is more powerful than just one guy's opinion; it will reveal strengths and weaknesses that aren't obvious to the average person. For example, I doubt that most people know how to properly compare the battery life of different gadgets. Did they try to equalize the screen brightness? Did they have the wifi radio turned on for one but not the other? How well does the UI scale to lots of apps versus a few? Due to impatience, they might favor a familiar user-interface over an unfamiliar but perhaps superior (if given the chance) interface. They may not correctly compare the web experience of one platform versus another because they are only comparing stock browsers versus the best of the third party browsers that are available. Due to their personal tastes, their review can be colored by an unconscious bias toward the few strengths of one platform even if the other platform has more strengths overall or can do things that the other can't. A good reviewer will be cognizant of which differences are probably important to most people and which are simply a matter of taste.
I sure did. But I was thinking more in terms of transhumanism and brain uploading, which isn't mentioned in the wikipedia article (perhaps, I should edit it). One of the key requirements for brain uploading is functionalism, and I think the original New York Times article presents evidence that this is possible. It shows that when one brain region communicates with another, all it cares about are gross inputs and outputs. Thus, if you replace one brain region with a computer that can produce the same outputs, the other one doesn't care. If you could gradually replace all important modules of the brain with software/hardware that is functionally equivalent, then you would be essentially uploading someone's brain. You don't need the biology at all, just the functions. Demonstrating that this works for memory is important since memory is one of the key components of consciousness. This is in stark contrast to the Penrose-types that believe that there are some spooky quantum mechanical effects that are essential to the human mind.
You're assuming that someone of any capability would want to work for them. Typically, the only people working for them are failures . ..
You may have a point when it comes to the FBI, but it's a mistake to think that only incompetent geeks work in law enforcement or, even more importantly, in national security. It really doesn't have anything to do with competency and everything to do with basic outlook on life. "Law and order" types exist at all intelligence levels. At a minimum, you get access (especially if you work for the NSA) to hardware, like supercomputers, that wouldn't normally be available to a civilian. Imagine what a determined person like Aaron Barr could do if he had a Watson-like supercomputer monitoring Facebook, chat-rooms and every major social networking site. Moreover, with the power of subpoena and warrantless wiretapping, a security expert working for the Federal government could potentially do a lot more damage to these hacker groups than someone in the private sector. Even if there aren't any hacker-superstars working for the government now, I could see this changing quickly with a sufficiently outrageous "stunt". Imagine what would happen if LulzSec took credit (no pun intended) for hacking a major credit card processing company and indiscriminately releasing people's financial info onto bittorrent. This is likely to hurt other hackers. I think many people can be called to arms if they perceive a personal threat or lots of innocents being harmed, especially people that they know. I know a guy from MIT who put school aside and joined the Marines after the September 11th attacks.
Re:Sony company culture of indifference won't chan
on
Sony Compromised, Again
·
· Score: 3, Insightful
I don't know. . . repeatedly losing this much customer data or really any customer data is a serious public relations blunder. Sony Computer Entertainment already lost this console generation. I don't know if it can handle too much more egg on its face. At some point this is going to start making a serious dent in the bottom line.
And I remember distinctly as a kid, when Darth Vader told Luke he was his father, thinking, "That's bullshit, Vader's lying." When RotJ came out and they acted like it was the god's-honest truth, I was like, "Whaaaaaat? That's so lame."
In ROTJ they acted like Vader had told the truth because he really had. Luke didn't believe it either at first but he felt that it was the truth, so it couldn't be denied.
Okay, I think I am all nerded out for today; I've already posted three times to this thread.
Why wouldn't they? You can't tell me that there's not a single network that wouldn't at least give it a try. I would think that there would be plenty of story material from the Old Republic alone. Just look at the success of the Knights of the Old Republic games. You could even go further than that. With time spans in the tens of thousands of years (the Republic is over twenty thousand years old!), imagine learning how hyperdrive was invented or how humans first reacted to extraterrestrial life. Star Wars is a gold mind of potential stories.
It's amazing how consistently Apple acts like a control freak. It can't just be Steve Jobs because this is a trait that seems to permeate the company at all levels. Any Apple employees want to confirm? I can just imagine what the hiring process for new managers would be like. No technical expertise required, just this: "Do you support a firm and controlling management style for customers, partners, friends, colleagues and those that report to you? Oh, you do? Well, then you're hired!"
I didn't mean to imply that Turing completeness is unimportant. Not by a long shot! I just hear people bringing up Turing completeness as if it is the end all and be all of computer science. In A New Kind of Science, Stephen Wolfram makes a lot out of the fact that Rule 30 is a universal Turing machine and the Church-Turing thesis. I can see how this is important from a theoretical perspective but I dont' see how it makes the types of problems that most developers face any easier. People frequently comment that the programming language you use to write a program doesn't matter because they're all Turing equivalent but I would disagree with this. This is because programs have to be written (for now) by humans and there are functional constraints on languages beyond just Turing completeness. Programs written in these languages have to be fast and the memory and execution cost of the support infrastructure (e.g,, interpreters, OS) can not be too high. Good program metaphors and syntax matter because they decrease the probability of error when transcribing an algorithm from your mind (where it originates) into code where it can be executed by a computer. Turing completeness is just the minimum requirement for any system that you want to program.
I think that language is what allows the human brain to be a Turing machine -- and a universal one at that. This is pretty self-evident to me since it is clear that you can teach a human being, with pencil and paper to execute any effective algorithm that can be programmed into a computer. It might be slow but the program behaviors would be functionally equivalent so long as run time doesn't matter. You might argue that this isn't entirely the brain, that it is really the entire system composed of a human being in conjunction with pencil and paper that is the universal Turing machine. I would concede this point but only because humans have limited memory, which requires the use of external tools like pencil and paper. If you could internalize those tools then you could say that it was entirely the human brain that was executing the algorithm in which case that brain would be a universal Turing machine. If we're going to say that human brains aren't universal Turing machines because they don't have enough memory then you'd have to say that Intel processors aren't either since they also have limited memory -- it's only a matter of what algorithms are practical to run entirely in the human brain versus those that -- due to time or memory constraints -- must be run on a traditional digital computer.
This, I think, leads to the crux of the problem which is that I think that Turing completeness is over-rated. Knowing that you can, with the right instructions, execute any effective algorithm on any universal Turing machine ignores the bigger practical considerations. Not all universal Turing machines are practical computational devices for the types of problems that we care about. An 8086 is a universal Turing machine as is a Core i7 processor. Does that mean that we should treat them as the same? In one sense, yes -- they can both be programmed to execute the same algorithms. But if there is a time or memory constraint then they are not equivalent, even in principle, since the Core i7 processor is absolutely faster then the 8086 and can execute algorithms that the 8086 could never execute within a short enough time to be usable.
Moreover, cellular automatons like John Conway's Game of Life are universal Turing machines as well, but that doesn't mean that we know how to program them to execute useful algorithms or that those algorithms will be executed within a reasonable amount of time. If all you know about a system is that it is a universal Turing machine then I think you know less about it than you might think. The hard questions are (1) How is data represented to this machine? (2) How are programs represented? (3) How is the output obtained? (4) How susceptible are the processing elements of this system to noise? These are some of the harder problems that have to be tackled before you can put Turing completeness to practical use.
If you flip a coin and get heads three times in a row, your chance of getting a head on the next flip is 50%.
Don't mean to nitpick, especially since I agree with the gist of everything else you wrote but this is not necessarily true. The probability of getting heads on the next flip is only 50% if the coin really is fair. Now, three coin tosses are really not enough to know whether the coin is fair or not but if you flipped the coin a hundred times and they all came out heads then that would be pretty solid evidence that there was some asymmetry in the characteristics of the coin or the way it was being tossed. Of course, even in this case, it is still possible that it was fair but less likely.
I would say that the biggest problem with common sense is drawing conclusions from too small of a sample. However, there is a logic to common sense. If you're in a situation where a decision has to be made then there might not be enough time to determine rigorously the probabilities of costs or benefits. A small number of samples may simply be all that you have to go on. So long as we accept common sense as ONLY a short-term heuristic, to be refined by more careful study, then we should be okay. The problem comes from when people refuse to accept scientific results simply because it contradicts their common sense notions.
The only thing the computer has done, the way we use it, is to make it quicker to come to the wrong conclusions.
Some people use the computer to make it easier to back up and try a new path having once come to the wrong end-point. That's a real improvement.
But it also makes it easier to just blindly try more wrong paths. Computers induce a lot of churn into our daily lives. I guess that's different. I'm still not convinced it's substantive. Too many of the important problems have too many paths to try, so many that you're probably going to die before you hit a right solution. And if you get used to the churn, I think you lose the ability to recognize a right path, so you often find yourself having backed off a real solution and started on a new wrong one, and by the time you can get back to what was a right path, well, you've changed, and the network has changed.
Interesting points. I think the bigger problem, though, isn't specific to computers at all. It's true, that computers make it easier to find wrong conclusions but that is simply because there are more wrong conclusions than right conclusions. I don't think that slowing things down will get us to correct solutions any faster. For any nontrivial system there will always be more ways to be wrong than right. Correct solutions will always be a small gem hidden in a forest of nonsense and bad ideas. On some fundamental level, I suspect this is the core principle behind the second law of thermodynamics. Any system, when left on its own, will act randomly within its natural degrees of freedom. In order to get the system to do something useful, i.e., to evolve toward a particular goal, you have to restrict some of its options so that it has no choice but to go in the direction that you want, toward order.
I think your problem with the way most people use computers is that, although computers allow to you quickly find both bad and good paths, a side effect of this is that it also allows you to find REALLY bad paths before you're prepared to deal with them. This is a valid concern.
I want a portable Unix workstation the size of a pocket calculator. I know it could be done.
Don't know about Unix, but you can certainly have Linux on a laptop. That should be as close to a portable Unix workstation as you would need. If you truly need something pocketsized (for cargo pants), try an Android phone with a folding bluetooth keyboard. This arrangement works quite well with the EVO 4G.
Slashdot Mirror
Except that this wasn't in the public. This was on Apple's private property. Also, it appears to have been done at least partially without people's consent under conditions that they would not have expected to be photographed. You might not mind having hidden cameras taking pictures of you and posting them on the internet without your consent but a lot of (most?) other people do.
And why would the number of people unemployed and not looking for a job increase, but not those looking for a job?
Because a larger fraction of the potential labor market has given up. For example, if the only jobs that could not be outsourced or automated required a PhD, then a lot of people will simply leave the labor market because they may feel that they have no chance to get a job. For example, the unemployment rate can remain constant even though there are more people panhandling and begging on the streets. I'm not saying that this is what happened between the 70s and today but it makes sense given other trends, such as increased outsourcing, education costs, and automation.
+1 Awesome -- wish I had mod points.
The problem here is that Microsoft effectively made Android anything but free, which is exactly the opposite of what Google wanted to achieve with the OEM brand perception of Android as a platform, and that in and of itself is a fantastic business strategy. I can't even remotely justify it as either humiliating or desperate; it's well-played despite being immensely back-handed.
But that's the whole misunderstanding about Android. Who ever said that manufacturing and selling an Android device was supposed to be free anyway? As a hardware manufacturer, I don't see how paying this "Microsoft" tax is much different from the other telephony-related licensing fees that cellphone OEMs already have to pay. If there's a GSM radio, then there are also patent licensing fees with Nokia that have to be paid. It's just another per unit cost to building a physical device, just like it costs $X to use Gorilla Glass or a Sony lithium ion battery. This is just par for the course. I also don't see anything particularly special about the fact that these fees have to be paid to Microsoft. The only significance that I can see is that it is a unit cost to manufacturing an Android phone that (presumably) wouldn't have to be paid by a Windows Phone 7 manufacturer. Yet, it still doesn't mean that manufacturing and selling a Windows Phone 7 device is more cost effective than a comparable Android device simply because a Windows Phone 7 device will have its own unique unit costs. Plus, I imagine that this $5 fee essentially pays for itself simply from the relative strength of the Android platform vis-a-vis Windows Phone 7.
Exactly. Is it really that difficult to express legal contracts in plain language? Can't contracts (and the laws themselves, for that matter!) be expressed at the level of, say, a newspaper? Maybe the New York Times? One of the first lessons of writing is KNOW YOUR AUDIENCE. If the audience is other lawyers than go right ahead, use all the obfuscatory language you want. I'm sure it reads like poetry to some people. But you aren't trying to communicate properly if you're using lawyer-speak and you know the audience isn't made up of lawyers and isn't likely to fully understand the contract without one. And the problem starts at the top, with the laws themselves. It's like the laws are written by lawyers for lawyers; they're not intended to be understood by the general public (bingo?). It's funny how lawyers get such a bad rap here on Slashdot and yet, at least in the US, and at the Federal level, we are governed by lawyers. Why would they ever institute change that would hurt their own?
Worthless options are one thing but from the Businessweek article, it appears that the guy ended up OWING taxes for worthless options. Try figuring that one out.
I'm not a lawyer but it is called an "unconscionable" contract. Depending on jurisdiction, contracts that are deemed "unconscionable" can be declared void by a judge or in some cases, just the part that is "unconscionable".
Agreed. I love it how you have all these people saying, "JUST READ THE CONTRACT!" Well, hell, we wouldn't lawyers if it were that easy, now would we? The problem isn't just reading contract, it's reading the contract and understanding the contract. From TFA, it looks to me like this guy did read the contract, at least as well as most people probably do and it seemed to okay to him because he thought he understood it but didn't understand it completely. If the first page of the contract says you're fully vested after 1 year but 50 pages later in some footnote it says you're not fully vested in 1 year than I would certainly cry foul. I suspect that most people sign contracts not based on fully reading and understanding them but because they don't see other people complaining or raising a stir. If you don't have experience with hidden clauses and gotchas in contracts then you're probably not going to be looking for them -- and are likely to miss them even if you DO read the whole contract verbatim. Rightly or wrongly, much of our society is based on trust and an expectation that most people aren't being grossly dishonest.
If it prevents some real criminals from breaking into Sony in the future and actually using that info, they've done good.
Except that the "real criminals" don't have to break into Sony because these hacker groups (don't know if it was LulzSec or not) have already done it for them. That's the flaw in this type of argument. Hacking to teach a lesson when real people's data is exposed lowers the barrier to entry for identity thieves and similar criminals. Before, they may have needed some nominal level of competency with computers to break into Sony and get people's credit card info. Now, all they need to do is sit tight and wait for LulzSec and related groups to expose people's info and reap the ill-begotten rewards before the victims have a chance to respond. Sure, future victims might be avoided but they may have been avoided anyway since the "real" criminals might have been focusing on targets other than Sony. Hell, for all we know, Sony might have been planning to patch the vulnerabilities that were exploited. Now, I'm not excusing Sony's incompetence but it seems to me that there were a lot better ways to make a point about security holes than hurting innocent people.
Why do you have to be a geek to judge technology?
If anything, the "man on the street" test is often the best one and usability testing is always best done with a total beginner.
I agree to some extent but I think the opinions of the average Joe only matter in aggregate. I wouldn't trust the untrained review by itself. You can't expect some random person off the street to do a proper review of a complex gadget like a tablet. A good thorough review is more powerful than just one guy's opinion; it will reveal strengths and weaknesses that aren't obvious to the average person. For example, I doubt that most people know how to properly compare the battery life of different gadgets. Did they try to equalize the screen brightness? Did they have the wifi radio turned on for one but not the other? How well does the UI scale to lots of apps versus a few? Due to impatience, they might favor a familiar user-interface over an unfamiliar but perhaps superior (if given the chance) interface. They may not correctly compare the web experience of one platform versus another because they are only comparing stock browsers versus the best of the third party browsers that are available. Due to their personal tastes, their review can be colored by an unconscious bias toward the few strengths of one platform even if the other platform has more strengths overall or can do things that the other can't. A good reviewer will be cognizant of which differences are probably important to most people and which are simply a matter of taste.
I sure did. But I was thinking more in terms of transhumanism and brain uploading, which isn't mentioned in the wikipedia article (perhaps, I should edit it). One of the key requirements for brain uploading is functionalism, and I think the original New York Times article presents evidence that this is possible. It shows that when one brain region communicates with another, all it cares about are gross inputs and outputs. Thus, if you replace one brain region with a computer that can produce the same outputs, the other one doesn't care. If you could gradually replace all important modules of the brain with software/hardware that is functionally equivalent, then you would be essentially uploading someone's brain. You don't need the biology at all, just the functions. Demonstrating that this works for memory is important since memory is one of the key components of consciousness. This is in stark contrast to the Penrose-types that believe that there are some spooky quantum mechanical effects that are essential to the human mind.
You're assuming that someone of any capability would want to work for them. Typically, the only people working for them are failures . . .
You may have a point when it comes to the FBI, but it's a mistake to think that only incompetent geeks work in law enforcement or, even more importantly, in national security. It really doesn't have anything to do with competency and everything to do with basic outlook on life. "Law and order" types exist at all intelligence levels. At a minimum, you get access (especially if you work for the NSA) to hardware, like supercomputers, that wouldn't normally be available to a civilian. Imagine what a determined person like Aaron Barr could do if he had a Watson-like supercomputer monitoring Facebook, chat-rooms and every major social networking site. Moreover, with the power of subpoena and warrantless wiretapping, a security expert working for the Federal government could potentially do a lot more damage to these hacker groups than someone in the private sector. Even if there aren't any hacker-superstars working for the government now, I could see this changing quickly with a sufficiently outrageous "stunt". Imagine what would happen if LulzSec took credit (no pun intended) for hacking a major credit card processing company and indiscriminately releasing people's financial info onto bittorrent. This is likely to hurt other hackers. I think many people can be called to arms if they perceive a personal threat or lots of innocents being harmed, especially people that they know. I know a guy from MIT who put school aside and joined the Marines after the September 11th attacks.
Epic comment fail by GP.
I don't know. . . repeatedly losing this much customer data or really any customer data is a serious public relations blunder. Sony Computer Entertainment already lost this console generation. I don't know if it can handle too much more egg on its face. At some point this is going to start making a serious dent in the bottom line.
And I remember distinctly as a kid, when Darth Vader told Luke he was his father, thinking, "That's bullshit, Vader's lying." When RotJ came out and they acted like it was the god's-honest truth, I was like, "Whaaaaaat? That's so lame."
In ROTJ they acted like Vader had told the truth because he really had. Luke didn't believe it either at first but he felt that it was the truth, so it couldn't be denied.
Okay, I think I am all nerded out for today; I've already posted three times to this thread.
Why wouldn't they? You can't tell me that there's not a single network that wouldn't at least give it a try. I would think that there would be plenty of story material from the Old Republic alone. Just look at the success of the Knights of the Old Republic games. You could even go further than that. With time spans in the tens of thousands of years (the Republic is over twenty thousand years old!), imagine learning how hyperdrive was invented or how humans first reacted to extraterrestrial life. Star Wars is a gold mind of potential stories.
There's a reason Hayden Christensen hasn't done anything of note before or since Star Wars.
Neither has Jake Lloyd.
It's amazing how consistently Apple acts like a control freak. It can't just be Steve Jobs because this is a trait that seems to permeate the company at all levels. Any Apple employees want to confirm? I can just imagine what the hiring process for new managers would be like. No technical expertise required, just this: "Do you support a firm and controlling management style for customers, partners, friends, colleagues and those that report to you? Oh, you do? Well, then you're hired!"
I didn't mean to imply that Turing completeness is unimportant. Not by a long shot! I just hear people bringing up Turing completeness as if it is the end all and be all of computer science. In A New Kind of Science, Stephen Wolfram makes a lot out of the fact that Rule 30 is a universal Turing machine and the Church-Turing thesis. I can see how this is important from a theoretical perspective but I dont' see how it makes the types of problems that most developers face any easier. People frequently comment that the programming language you use to write a program doesn't matter because they're all Turing equivalent but I would disagree with this. This is because programs have to be written (for now) by humans and there are functional constraints on languages beyond just Turing completeness. Programs written in these languages have to be fast and the memory and execution cost of the support infrastructure (e.g,, interpreters, OS) can not be too high. Good program metaphors and syntax matter because they decrease the probability of error when transcribing an algorithm from your mind (where it originates) into code where it can be executed by a computer. Turing completeness is just the minimum requirement for any system that you want to program.
I think that language is what allows the human brain to be a Turing machine -- and a universal one at that. This is pretty self-evident to me since it is clear that you can teach a human being, with pencil and paper to execute any effective algorithm that can be programmed into a computer. It might be slow but the program behaviors would be functionally equivalent so long as run time doesn't matter. You might argue that this isn't entirely the brain, that it is really the entire system composed of a human being in conjunction with pencil and paper that is the universal Turing machine. I would concede this point but only because humans have limited memory, which requires the use of external tools like pencil and paper. If you could internalize those tools then you could say that it was entirely the human brain that was executing the algorithm in which case that brain would be a universal Turing machine. If we're going to say that human brains aren't universal Turing machines because they don't have enough memory then you'd have to say that Intel processors aren't either since they also have limited memory -- it's only a matter of what algorithms are practical to run entirely in the human brain versus those that -- due to time or memory constraints -- must be run on a traditional digital computer.
This, I think, leads to the crux of the problem which is that I think that Turing completeness is over-rated. Knowing that you can, with the right instructions, execute any effective algorithm on any universal Turing machine ignores the bigger practical considerations. Not all universal Turing machines are practical computational devices for the types of problems that we care about. An 8086 is a universal Turing machine as is a Core i7 processor. Does that mean that we should treat them as the same? In one sense, yes -- they can both be programmed to execute the same algorithms. But if there is a time or memory constraint then they are not equivalent, even in principle, since the Core i7 processor is absolutely faster then the 8086 and can execute algorithms that the 8086 could never execute within a short enough time to be usable.
Moreover, cellular automatons like John Conway's Game of Life are universal Turing machines as well, but that doesn't mean that we know how to program them to execute useful algorithms or that those algorithms will be executed within a reasonable amount of time. If all you know about a system is that it is a universal Turing machine then I think you know less about it than you might think. The hard questions are (1) How is data represented to this machine? (2) How are programs represented? (3) How is the output obtained? (4) How susceptible are the processing elements of this system to noise? These are some of the harder problems that have to be tackled before you can put Turing completeness to practical use.
If you flip a coin and get heads three times in a row, your chance of getting a head on the next flip is 50%.
Don't mean to nitpick, especially since I agree with the gist of everything else you wrote but this is not necessarily true. The probability of getting heads on the next flip is only 50% if the coin really is fair. Now, three coin tosses are really not enough to know whether the coin is fair or not but if you flipped the coin a hundred times and they all came out heads then that would be pretty solid evidence that there was some asymmetry in the characteristics of the coin or the way it was being tossed. Of course, even in this case, it is still possible that it was fair but less likely.
I would say that the biggest problem with common sense is drawing conclusions from too small of a sample. However, there is a logic to common sense. If you're in a situation where a decision has to be made then there might not be enough time to determine rigorously the probabilities of costs or benefits. A small number of samples may simply be all that you have to go on. So long as we accept common sense as ONLY a short-term heuristic, to be refined by more careful study, then we should be okay. The problem comes from when people refuse to accept scientific results simply because it contradicts their common sense notions.
The only thing the computer has done, the way we use it, is to make it quicker to come to the wrong conclusions.
Some people use the computer to make it easier to back up and try a new path having once come to the wrong end-point. That's a real improvement.
But it also makes it easier to just blindly try more wrong paths. Computers induce a lot of churn into our daily lives. I guess that's different. I'm still not convinced it's substantive. Too many of the important problems have too many paths to try, so many that you're probably going to die before you hit a right solution. And if you get used to the churn, I think you lose the ability to recognize a right path, so you often find yourself having backed off a real solution and started on a new wrong one, and by the time you can get back to what was a right path, well, you've changed, and the network has changed.
Interesting points. I think the bigger problem, though, isn't specific to computers at all. It's true, that computers make it easier to find wrong conclusions but that is simply because there are more wrong conclusions than right conclusions. I don't think that slowing things down will get us to correct solutions any faster. For any nontrivial system there will always be more ways to be wrong than right. Correct solutions will always be a small gem hidden in a forest of nonsense and bad ideas. On some fundamental level, I suspect this is the core principle behind the second law of thermodynamics. Any system, when left on its own, will act randomly within its natural degrees of freedom. In order to get the system to do something useful, i.e., to evolve toward a particular goal, you have to restrict some of its options so that it has no choice but to go in the direction that you want, toward order.
I think your problem with the way most people use computers is that, although computers allow to you quickly find both bad and good paths, a side effect of this is that it also allows you to find REALLY bad paths before you're prepared to deal with them. This is a valid concern.
I want a portable Unix workstation the size of a pocket calculator. I know it could be done.
Don't know about Unix, but you can certainly have Linux on a laptop. That should be as close to a portable Unix workstation as you would need. If you truly need something pocketsized (for cargo pants), try an Android phone with a folding bluetooth keyboard. This arrangement works quite well with the EVO 4G.