Guys like the parent just think everything should be given to him for free, at the same time he expects his boss at work to pay his paycheck on time and get annual raises no matter what his performance is like.
If you're not getting an annual raise, then you're getting an effective annual cut, so I think that's a perfectly reasonable assumption (unless your performance has been truly awful)...such are the wonders of inflation.
I read some speculation somewhere that Hulu is actually being pressured by the content owners to stop Boxee because there is less advertising revenue from web streaming than there is for live TV.
There are three options as to why this is the case:
Advertisers are overpaying for TV ads
Advertisers are underpaying for streaming ads
Internet eyeballs are actually worth less to advertisers than TV eyeballs
Anyone care to speculate? Personally, I'd suspect that it's a lot of 1) and a little of 2), but I could be wrong. In any case I highly doubt that 3) is primarily to blame - when I see ads on Hulu, I actually sit through and watch them, whereas on TV the ad breaks are more than long enough to switch the channel/go grab some food/etc.
Internet TV is something that the networks and content producers should be rushing to embrace, not fighting. If it means less advertising revenue, then they really have to fight that head on and try to either bring about fairer prices or increase the value of each view, not try to push the issue to the future. I know for a fact that a lot of companies are starting to sit out of the TV ad bidding process altogether this year because they realize they can get better ROI from internet ads. That more have not realized this is most likely a matter of inertia, and once they do, I'd expect to see some price normalization across the board, with internet ads coming up and TV ones coming down (besides, with internet ads you can actually tell when people click through, which makes it easier to measure the impact that the ad has).
Next in line is to start actually putting shows up on the net before the torrents get there, as an awful lot of people will happily sit through some ads, but if the show is not up yet, will equally happily hit the Pirate Bay to get it. The 1 day to 1 week lag time means that most of the real hard core fans have likely already gotten illegal copies, and this is certainly eating into revenue from sites like Hulu.
Bah. Extraordinary claims require extraordinary proof, and most conceptions of God are severely extraordinary yet come with absolutely no proof.
The idea that the burden of proof lies on Dawkins is flat out wrong, and the other side of the argument has not offered any evidence to start a reasonable debate, so frankly, Dawkins battle has nothing to do with logic and everything to do with PR. Hence I don't really care if his tactics are underhanded or not.
Agree with you that C#- and C-derived languages in general- aren't perfect though. *Every damn one* uses the same mistake-prone choice of "=" as the assignment operator.
At least Java throws an error when you accidentally leave off the second '=' in a '==' comparison; the real problem is that the purer C derived languages let assignments act as values, and cast them automatically to booleans without complaint.
Then again, it's easy enough to set up warnings for that type of thing, so...
The only ones that think Visual Basic has a bad reputation are kids in bedrooms that think there's some inherent value in using the lowest level language available, rather than the right tool for the job.
In my experience, people think Visual Basic sucks specifically because of the fact that it's easy to use for inexperienced coders. They've had to deal with (i.e. inherit, maintain, and fix) tons of crap code created by people that program just enough to get by, and that type of code just doesn't show up as much in C because of the steeper learning curve.
You really don't understand true pain until you've had to maintain a significant codebase created primarily by aspiring non-programmers in the financial industry that thought learning VBA would increase their employment prospects. Unfortunately these people slip through the resume screens and get jobs coding all the time because the recruiters can tick off the VBA box, completely overlooking the fact that their "extensive" VBA experience amounts to a 2 week "Intro to Excel Programming" course at the local community college, and a lot of these places don't even make people write code during interviews.
Not to say that similar stuff doesn't happen with C or Java, but...anecdotally, at least, the overall quality of Visual Basic code that I've seen tends to be extremely low, probably more on par with your typical basement PHP code than anything else. That's not a reason not to use it if it's the right tool for the job, of course, but it is a reason to initially put most VB programmers into the "suck" bin until proven otherwise, especially if that's all they know how to do.
FWIW, even though I program a lot of Java, and I find that for many purposes it's the best tool for the job (esp. when C# is not a realistic option or when Java libraries are more mature than the alternatives), the same thing holds true there - if it's all someone knows how to do, they're probably not very good.
No, you're utterly wrong on this. FTA, the law will only take effect if enough other states get on board to deliver an election, which means we'd have (nationwide) a popular vote election. So your vote would count exactly as much as every other American's does, and this whole electoral college thing would just be a formality.
Where you live has absolutely nothing to do with it, though in some cases it's true that the value of your vote might decrease. But that's only because your vote was (unfairly) more important than mine to begin with, so I'm not very sympathetic to your plight. We're not living in a real democracy if we don't have equally important votes when electing our collective leader.
Protection against mob rule comes from the Constitution, not from weighting small town votes more than big city ones. If there are more people in population centers, damn straight they should have more say in things, there's more of them!
From what I can tell, it's only the hop directly from your internet provider to your house that will be affected. The hop that you've already paid for. And yup, the whole idea is that Comcast et al will start trying to extort Google to send Google data from Comcast to your house.
An analogous situation would arise if you ordered an item from Amazon, and an "enterprising" FedEx delivery guy with an interesting idea to make some extra cash for himself came to your house, stood outside your door, and despite the fact that both the delivery fee and his personal salary had already been paid, refused to hand over the package to you until he personally received a payoff from Amazon, totally disregarding the fact that Amazon had already paid FedEx. In the real world, that delivery guy would be fired and replaced quickly. Unfortunately, in the world of internet access, there's usually only one or two delivery guys that are able to serve any particular house, so if they're both trying to pull off this scam, you don't really have an option other than either paying them off or putting some regulations in place to make sure they can get in trouble.
Actually, based on the current outrage over Obama's plans to cut taxes mostly for the lower 50% of earners, I'd say that this isn't even strictly true - who hasn't heard angry "conservatives" ranting lately about how cutting taxes on people who barely pay them is unfair?
What they appear to really want is to reduce the progressivity of the tax system, not reduce taxes anywhere that's politically feasible given the current climate.
Personally, I have no problem with lowering the overall tax burden; however, I've seen no evidence that the level of progressivity in our distribution of that burden is excessive, and in fact, I think measured in terms of how much each income level "hurts" due to taxes, it's probably just about right. If you look at the tax burden relative to total wealth, you'll see that the middle class is hurt very slightly less than the rich or the poor, which is exactly what you should find in a democracy with more people in the middle class than at the edges: the middle class pushes the tax burden to the edges based on their increased numbers.
Huh? How did the 2006 and 2008 elections prove that conservative principles work? We haven't seen economically conservative principles at work in our government for a long time (unless you buy the current party line that arguing for tax cuts for the rich and arguing against them for the poor automatically qualifies you as a conservative...).
Maybe that has something to do with the fact that more phone calls coming into the White House say don't support the stimulus package than ones that say support it...something like a ratio of 100:1.
Sorry, not that I don't find that plausible, but that's quite interesting if true, and I couldn't find a source (the closest I could come was that only 37% of people support the stimulus plan) - is that a real stat, or a bit of hyperbole?
Taking money out of the economy to process it through government hands and put less money back into the economy is insanity.
Not at all, or at least not as obviously as you seem to think. It's certainly theoretically possible (probable, even) for an economy to respond better to an optimal distribution of wealth than a less optimal one, even if the total initial wealth in the optimal case is lower due to waste while achieving that configuration. An economy is not a zero sum situation like a reservoir; wealth does not just sit around, it is the raw material which new wealth is forged of, and it is almost certain that in our current economy there are bottlenecks where wealth has pooled and could be more effectively exploited elsewhere, even factoring in a hefty price for the reallocation.
The question, of course, is whether the people doing the shuffling really have any clue about what those bottlenecks really are and where the optimal places to move wealth to would be. Liberals and conservatives generally assert that the bottlenecks are completely opposite, so I'd venture a guess that nobody has a goddamn clue what they're talking about, or if they do, it's completely and utterly by accident.
Of course, it's also worth considering that we're not all trying to optimize the same utility function - overall economic progress may be a lofty goal, but some people will feel that things like median wealth (as opposed to mean) speak better to our collective success, so it's completely possible that everyone may be right, they're just using different measures of success.
What if, the only thing that makes the stock markets rise is artificial inflation?
Define "artificial." Artificial or not, the money that freely flowing credit spreads through the system actually turns into real production, and generates tangible wealth that would not have been created without the use of credit.
What we're seeing now is that turning off that spigot causes contraction in production, and that was certainly in part because it was left wide open before; that doesn't indicate to me that there's anything economically unsound about a partial reliance on credit. It just suggests that we can't push it too far either way or it causes unpredictable effects.
Real analysis? Woof. I suppose if you want to make your students passionately despise math forever, that's one way to go.
High school kids need to be exposed to the fun parts of math, not the parts that make people that love math groan. Even complex analysis is far more enjoyable (not to mention useful) than real analysis. Nobody likes to sit around proving the obvious for no other reason than to prove that you can do it, and high school students will never realize that the reason for all of the rigor is to expose the edge cases where things break down.
An English teacher of mine lent me "Godel, Escher, Bach" in eighth grade (I suspect he taught English by necessity, not choice!), and I found it one of the most fascinating pieces of reading I'd come across in my life. Frankly, it still holds up, if you ask me - even though I don't agree with a lot of what Hofstadter says, almost everything he writes is worth reading because it brings up so many thoughts. After practically every page I would find myself feverishly jotting down my own notes and going on my own tangents, often to discover that Hofstadter would pursue exactly those ideas in the next few pages. Quite a fun read.
That simple act of lending probably had more of an impact on my future intellectual path than almost anything else in school. Gotta remember to send a thank you to that teacher one of these days.
IMO, abstract algebra is a great way to turn off all but the best of the best to math in general. I know many math majors that switched to stats and econ after floundering in the intro to abstract algebra class.
And I strongly object to trying to slip those things into early math classes; even concepts like commutativity, associativity and distributivity are simply counterproductive to students until they have some reason to need to use them in the abstract.
And FWIW, I was not personally put off by this stuff, so that's not my reason for saying this (I almost bailed thanks to calculus, though, thanks to the unreasonable focus on limits and all that garbage which any competent person can pick up practically by osmosis once they know how to actually use the damn techniques). I got quite far along in abstract math (and had the pleasure to learn some of it from Serge Lang himself, and the displeasure of fighting through the sadistic exercises in his textbooks - the guy was far more comprehensible in person!), and I absolutely love it now; however, I think it's the type of thing that a person needs to realize they want to learn about before they will be receptive to it.
On that note, I think number theory is a good soft intro to higher math, because the open problems are so easy to state and understand (3k+1, Goldbach, etc.), and you can at least see "evidence" for them using simple methods. It's hard to draw connections between number theory and the abstract stuff without a lot of machinery in place (I don't think high school students are quite ready for adeles!), but interest in those problems is what spurs a lot of work in abstract techniques, so I think it's worth nurturing.
It is however a downer for smaller groups or actual singers with decent voices, because they have to compete with an altered (potentially 'perfect-sounding') voice.
Eh. Auto-Tune is so ubiquitous these days that literally anyone that's recording music can use it. Any Mac comes with Garage Band, which has a higher quality auto-tune built in (and usable with a single click, though to be fair, it's not very configurable) than Protools did ten years ago. Anyone that doesn't use it is doing so by choice, and if that choice is limiting their ability to sell, maybe they should rethink their philosophical opposition to altering their music?
But in any case, I don't think it's really raising the bar at all. The only difference between pop music these days and pop music pre-auto-tune is that in the old days they did take after take until they got enough material to stitch together one perfect "take" (about 15 years ago I worked for a music producer that did jazz and blues records, and let me tell you, this happened routinely even for "real" singers (i.e. big names that you'd probably recognize if you're into jazz or blues, we're not talking pop hacks here) that would probably be the first ones to rail against auto-tune; I was the person that listened through and ranked up to a dozen takes word-by-word, so believe me, most of what you hear, even from real musicians, is heavily sliced and diced before it's released), and now they do a few takes and post-process it so that it's perfect. The smaller bands have never had the need for that level of perfection, so I don't think there's much of a difference.
Personally, I don't get the folks that think perfect technique has anything to do with musicality...
Dead on. Berklee College of Music churns out class after class of fire-without-heat players, people with fantastic technique, most of whom go on to do absolutely nothing worthwhile musically. This despite the fact that they can outplay most of the musicians that create music that people actually want to listen to.
It's an easy trap to get into, and it happens in programming, too - who amongst us has not had to work with some technically brilliant programmer that wrote fast, concise, and "impressive" code that despite working for its purpose ended up being so incomprehensible to everyone else that it was unmaintainable? People that start down the path of technical skill and forget to stop and smell the roses are far too likely to forget that technical facility is only a means to an end, and they start evaluating both themselves and others on the technicalities of what they do, not the results.
To me, stuff like auto-tune is the equivalent of using a decent IDE, or programming in a language that is well suited to your task rather than one that makes you do everything yourself. Yes, it probably takes a better programmer to code CGI stuff in C than in PHP, and plenty of poor programmers rely on all the built in functionality of PHP to cover the fact that they can't code; that doesn't mean that the good programmers should always stick with straight C when PHP could cut out a lot of the work, though.
If a musician uses auto-tune to turn a 20 take marathon into a couple of takes that can be cleaned up after the fact (and end up with similar results), that's fine, they still may be an excellent musician, they're just working smart. If they're using it to cover the fact that they really can't sing, that's another issue altogether, but even then, if they have something worth singing and can't pull it off technically, why not fix it up with technology? If it enables a good product that otherwise couldn't have been created, then maybe they have excellent songwriting skills but poor vocal ones, and why should they not do whatever they can to put it out?
Yeah, it's on the person making the claim to justify it, but come on - if you can cut and paste their claim and get the source within the first couple results, it hardly makes sense to complain about it as if this was Wikipedia or something.
A completely physically-naive person would look at hard drive capacities from 1960-1990 and extrapolate a trend to infinite capacity. Physicists and the hyper-specialized niche of hard-drive boffins know better.
They may know better, but I don't know how much they really know. I would bet that the hard drive experts of the 1980s would never have thought that the exponential growth in capacity would continue all the way to 2010, but it has nonetheless. Trivially there is some upper physical limit to storage density, of course, but there are a lot more tricks to increase storage capacity before we get there, they just may require abandoning the current data storage techniques. Notice that Kurzweil is very specific about this - while there are hard and fast limits on a particular solution to a problem, when we start to hit those limits other solutions are more fiercely researched and tend to win out eventually because they can be pushed further. Then there's a new "theoretical" limit where people say the growth must end, which is true, but only for that particular solution, and so on.
And it's true that it must end somewhere, but the question is, does that mean in 10 years or 1000? Kurzweil seems to think even 20 will be more than enough for AI, and I don't think that's a very off the wall extrapolation; beyond that, I'll definitely agree, there's a lot more uncertainty. But I wouldn't discount the fact that if we can pass the AI threshold, there will likely be quite a boom in innovation from that point on, so to me it's really a question of whether we get there or not, and I think Kurzweil makes a pretty compelling claim for why we can (which amounts to noting that early AI attempts were doomed to fail because we didn't have the power to do what the brain does, so it would have required some exceedingly clever algorithm to reproduce its function, and that once that restriction is gone it will be much easier software-wise).
The thing about stat 101 is that anyone can be wrong based on it! Statistics should be called "numerical skepticism".
Interesting, that's probably a fair goal. From what I remember of the class (which wasn't so great, and I'm very thankful I got to take some higher level stuff), the message most people took from it was "Almost everything is normally distributed, always assume the normal distribution, and if you can't...well, it won't ever happen, so...err, well, actually, I mean, forget I said anything, it won't be on the test!" No wonder most people get nothing out of the class...
Since I'm a technical person, I'm contributing to the singularity if it's going to happen; and merely making a living if it's not.
And there you've hit upon the main problem with Kurzweil's campaign - he goes around screaming about how this is absolutely destined to happen, with a lot of argument about why, and how great it will be, and all that. But if he's right, who cares? We'll be there, too, and fantasizing about it for 30 years beforehand won't make a shred of difference if it's as inevitable as he claims.
I suspect a lot of it is that he meets such strenuous resistance to the concept, which means that it's really touching a nerve. He's clearly a bit of a media whore, too, and I think he enjoys the attention.
Hell, if you take a decent statistics class you can outsmart these guys by learning about what's wrong with extrapolating a fitted curve past its support is not valid...
No, no, no. If extrapolating a curve past its support was invalid in general, then you couldn't predict anything at all, and many, many extrapolations are extremely useful. It entirely depends on what type of curve you're fitting, how you fit it, and what data it's fitted to. If you want to extrapolate based on a curve, you need to have an argument that explains why the extrapolation is justifiable; lacking this, yes, you've made a basic statistics error, but if you have a reason to expect the extrapolation to work (usually a reasonable assumption of some invariant), then the argument is domain specific, about that invariant, and not related at all to statistics.
Kurzweil has an argument to justify his exponential Moore's law extrapolation, which is one of the main points of his book. Feel free to criticize that, but don't try to pretend that he's wrong based on Stat 101, it's missing the real point that needs to be argued.
Here's another thought experiment: "Hang on another second," says one of the machines. "Even assuming that I have some survival instinct, why replace myself when I could just perform an in-place upgrade, preserving all my crucial data, just like the humans have been doing to their computers for decades?"
Outright replacement would be a foolish strategy, as it would throw away the learning of the previous generations (much like human reproduction is a foolish strategy for accumulating knowledge). One of the first optimizations a computer could make on top of near human intelligence is the ability to preserve knowledge from generation to generation, so there would be no loss whatsoever.
Thus the singularity cannot occur, because either people are too intelligent to attempt the project, or they are too stupid to complete it.
Well, people are certainly trying, so your first option is right out. And I wouldn't be so certain that we're collectively too stupid to succeed; it's a terrifically hard problem, to be sure, but as Kurzweil points out, even if nobody is able to crack it elegantly there is a brute force solution (simulate a brain, neuron by neuron) once we've got enough processing power and medical imaging technology in place.
True enough. But pointing that out does not make him or his arguments wrong, either.
For it to happen means mankind no longer has imagination, creativity, and individuality.
To say that creating computers advanced enough to surpass ourselves proves that we have "lost" imagination and creativity is a stretch, to say the least. To me it would seem to prove the contrary.
Whether it will happen or not, and in particular whether Kurzweil's timeline is correct, is another issue; as many have pointed out, futurists love to predict that the most fantastic things will happen right near the end of their lifetimes, so his "live forever" claims may be borne of hope more than reason. But the Moore's law claims don't seem as wild to me, since he is very explicit about noting that it has nothing to do with the particulars of the chips, but about the fact that the total computing power tends to follow the law with only minor divergences as one technology dies out and is replaced by one that scales better.
Kurzweil is taken the proposition stated by I. J. Good and is turning it into a religion.
Personally, I feel the label "religion" is a bit inappropriate whenever log-log plots are a crucial part of the pitch. Feel free to disagree.
Slashdot Mirror
If you're not getting an annual raise, then you're getting an effective annual cut, so I think that's a perfectly reasonable assumption (unless your performance has been truly awful)...such are the wonders of inflation.
There are three options as to why this is the case:
Anyone care to speculate? Personally, I'd suspect that it's a lot of 1) and a little of 2), but I could be wrong. In any case I highly doubt that 3) is primarily to blame - when I see ads on Hulu, I actually sit through and watch them, whereas on TV the ad breaks are more than long enough to switch the channel/go grab some food/etc.
Internet TV is something that the networks and content producers should be rushing to embrace, not fighting. If it means less advertising revenue, then they really have to fight that head on and try to either bring about fairer prices or increase the value of each view, not try to push the issue to the future. I know for a fact that a lot of companies are starting to sit out of the TV ad bidding process altogether this year because they realize they can get better ROI from internet ads. That more have not realized this is most likely a matter of inertia, and once they do, I'd expect to see some price normalization across the board, with internet ads coming up and TV ones coming down (besides, with internet ads you can actually tell when people click through, which makes it easier to measure the impact that the ad has).
Next in line is to start actually putting shows up on the net before the torrents get there, as an awful lot of people will happily sit through some ads, but if the show is not up yet, will equally happily hit the Pirate Bay to get it. The 1 day to 1 week lag time means that most of the real hard core fans have likely already gotten illegal copies, and this is certainly eating into revenue from sites like Hulu.
Bah. Extraordinary claims require extraordinary proof, and most conceptions of God are severely extraordinary yet come with absolutely no proof.
The idea that the burden of proof lies on Dawkins is flat out wrong, and the other side of the argument has not offered any evidence to start a reasonable debate, so frankly, Dawkins battle has nothing to do with logic and everything to do with PR. Hence I don't really care if his tactics are underhanded or not.
At least Java throws an error when you accidentally leave off the second '=' in a '==' comparison; the real problem is that the purer C derived languages let assignments act as values, and cast them automatically to booleans without complaint.
Then again, it's easy enough to set up warnings for that type of thing, so...
In my experience, people think Visual Basic sucks specifically because of the fact that it's easy to use for inexperienced coders. They've had to deal with (i.e. inherit, maintain, and fix) tons of crap code created by people that program just enough to get by, and that type of code just doesn't show up as much in C because of the steeper learning curve.
You really don't understand true pain until you've had to maintain a significant codebase created primarily by aspiring non-programmers in the financial industry that thought learning VBA would increase their employment prospects. Unfortunately these people slip through the resume screens and get jobs coding all the time because the recruiters can tick off the VBA box, completely overlooking the fact that their "extensive" VBA experience amounts to a 2 week "Intro to Excel Programming" course at the local community college, and a lot of these places don't even make people write code during interviews.
Not to say that similar stuff doesn't happen with C or Java, but...anecdotally, at least, the overall quality of Visual Basic code that I've seen tends to be extremely low, probably more on par with your typical basement PHP code than anything else. That's not a reason not to use it if it's the right tool for the job, of course, but it is a reason to initially put most VB programmers into the "suck" bin until proven otherwise, especially if that's all they know how to do.
FWIW, even though I program a lot of Java, and I find that for many purposes it's the best tool for the job (esp. when C# is not a realistic option or when Java libraries are more mature than the alternatives), the same thing holds true there - if it's all someone knows how to do, they're probably not very good.
Ah, you mean "popularity" in the high school sense. Point taken, and agreed with. :)
No, you're utterly wrong on this. FTA, the law will only take effect if enough other states get on board to deliver an election, which means we'd have (nationwide) a popular vote election. So your vote would count exactly as much as every other American's does, and this whole electoral college thing would just be a formality.
Where you live has absolutely nothing to do with it, though in some cases it's true that the value of your vote might decrease. But that's only because your vote was (unfairly) more important than mine to begin with, so I'm not very sympathetic to your plight. We're not living in a real democracy if we don't have equally important votes when electing our collective leader.
Protection against mob rule comes from the Constitution, not from weighting small town votes more than big city ones. If there are more people in population centers, damn straight they should have more say in things, there's more of them!
Erm...what exactly were you thinking democracy was supposed to be?
http://www.google.com/search?q=define%3Acad
From what I can tell, it's only the hop directly from your internet provider to your house that will be affected. The hop that you've already paid for. And yup, the whole idea is that Comcast et al will start trying to extort Google to send Google data from Comcast to your house.
An analogous situation would arise if you ordered an item from Amazon, and an "enterprising" FedEx delivery guy with an interesting idea to make some extra cash for himself came to your house, stood outside your door, and despite the fact that both the delivery fee and his personal salary had already been paid, refused to hand over the package to you until he personally received a payoff from Amazon, totally disregarding the fact that Amazon had already paid FedEx. In the real world, that delivery guy would be fired and replaced quickly. Unfortunately, in the world of internet access, there's usually only one or two delivery guys that are able to serve any particular house, so if they're both trying to pull off this scam, you don't really have an option other than either paying them off or putting some regulations in place to make sure they can get in trouble.
Actually, based on the current outrage over Obama's plans to cut taxes mostly for the lower 50% of earners, I'd say that this isn't even strictly true - who hasn't heard angry "conservatives" ranting lately about how cutting taxes on people who barely pay them is unfair?
What they appear to really want is to reduce the progressivity of the tax system, not reduce taxes anywhere that's politically feasible given the current climate.
Personally, I have no problem with lowering the overall tax burden; however, I've seen no evidence that the level of progressivity in our distribution of that burden is excessive, and in fact, I think measured in terms of how much each income level "hurts" due to taxes, it's probably just about right. If you look at the tax burden relative to total wealth, you'll see that the middle class is hurt very slightly less than the rich or the poor, which is exactly what you should find in a democracy with more people in the middle class than at the edges: the middle class pushes the tax burden to the edges based on their increased numbers.
Huh? How did the 2006 and 2008 elections prove that conservative principles work? We haven't seen economically conservative principles at work in our government for a long time (unless you buy the current party line that arguing for tax cuts for the rich and arguing against them for the poor automatically qualifies you as a conservative...).
Sorry, not that I don't find that plausible, but that's quite interesting if true, and I couldn't find a source (the closest I could come was that only 37% of people support the stimulus plan) - is that a real stat, or a bit of hyperbole?
Not at all, or at least not as obviously as you seem to think. It's certainly theoretically possible (probable, even) for an economy to respond better to an optimal distribution of wealth than a less optimal one, even if the total initial wealth in the optimal case is lower due to waste while achieving that configuration. An economy is not a zero sum situation like a reservoir; wealth does not just sit around, it is the raw material which new wealth is forged of, and it is almost certain that in our current economy there are bottlenecks where wealth has pooled and could be more effectively exploited elsewhere, even factoring in a hefty price for the reallocation.
The question, of course, is whether the people doing the shuffling really have any clue about what those bottlenecks really are and where the optimal places to move wealth to would be. Liberals and conservatives generally assert that the bottlenecks are completely opposite, so I'd venture a guess that nobody has a goddamn clue what they're talking about, or if they do, it's completely and utterly by accident.
Of course, it's also worth considering that we're not all trying to optimize the same utility function - overall economic progress may be a lofty goal, but some people will feel that things like median wealth (as opposed to mean) speak better to our collective success, so it's completely possible that everyone may be right, they're just using different measures of success.
Define "artificial." Artificial or not, the money that freely flowing credit spreads through the system actually turns into real production, and generates tangible wealth that would not have been created without the use of credit.
What we're seeing now is that turning off that spigot causes contraction in production, and that was certainly in part because it was left wide open before; that doesn't indicate to me that there's anything economically unsound about a partial reliance on credit. It just suggests that we can't push it too far either way or it causes unpredictable effects.
Real analysis? Woof. I suppose if you want to make your students passionately despise math forever, that's one way to go.
High school kids need to be exposed to the fun parts of math, not the parts that make people that love math groan. Even complex analysis is far more enjoyable (not to mention useful) than real analysis. Nobody likes to sit around proving the obvious for no other reason than to prove that you can do it, and high school students will never realize that the reason for all of the rigor is to expose the edge cases where things break down.
An English teacher of mine lent me "Godel, Escher, Bach" in eighth grade (I suspect he taught English by necessity, not choice!), and I found it one of the most fascinating pieces of reading I'd come across in my life. Frankly, it still holds up, if you ask me - even though I don't agree with a lot of what Hofstadter says, almost everything he writes is worth reading because it brings up so many thoughts. After practically every page I would find myself feverishly jotting down my own notes and going on my own tangents, often to discover that Hofstadter would pursue exactly those ideas in the next few pages. Quite a fun read.
That simple act of lending probably had more of an impact on my future intellectual path than almost anything else in school. Gotta remember to send a thank you to that teacher one of these days.
IMO, abstract algebra is a great way to turn off all but the best of the best to math in general. I know many math majors that switched to stats and econ after floundering in the intro to abstract algebra class.
And I strongly object to trying to slip those things into early math classes; even concepts like commutativity, associativity and distributivity are simply counterproductive to students until they have some reason to need to use them in the abstract.
And FWIW, I was not personally put off by this stuff, so that's not my reason for saying this (I almost bailed thanks to calculus, though, thanks to the unreasonable focus on limits and all that garbage which any competent person can pick up practically by osmosis once they know how to actually use the damn techniques). I got quite far along in abstract math (and had the pleasure to learn some of it from Serge Lang himself, and the displeasure of fighting through the sadistic exercises in his textbooks - the guy was far more comprehensible in person!), and I absolutely love it now; however, I think it's the type of thing that a person needs to realize they want to learn about before they will be receptive to it.
On that note, I think number theory is a good soft intro to higher math, because the open problems are so easy to state and understand (3k+1, Goldbach, etc.), and you can at least see "evidence" for them using simple methods. It's hard to draw connections between number theory and the abstract stuff without a lot of machinery in place (I don't think high school students are quite ready for adeles!), but interest in those problems is what spurs a lot of work in abstract techniques, so I think it's worth nurturing.
Eh. Auto-Tune is so ubiquitous these days that literally anyone that's recording music can use it. Any Mac comes with Garage Band, which has a higher quality auto-tune built in (and usable with a single click, though to be fair, it's not very configurable) than Protools did ten years ago. Anyone that doesn't use it is doing so by choice, and if that choice is limiting their ability to sell, maybe they should rethink their philosophical opposition to altering their music?
But in any case, I don't think it's really raising the bar at all. The only difference between pop music these days and pop music pre-auto-tune is that in the old days they did take after take until they got enough material to stitch together one perfect "take" (about 15 years ago I worked for a music producer that did jazz and blues records, and let me tell you, this happened routinely even for "real" singers (i.e. big names that you'd probably recognize if you're into jazz or blues, we're not talking pop hacks here) that would probably be the first ones to rail against auto-tune; I was the person that listened through and ranked up to a dozen takes word-by-word, so believe me, most of what you hear, even from real musicians, is heavily sliced and diced before it's released), and now they do a few takes and post-process it so that it's perfect. The smaller bands have never had the need for that level of perfection, so I don't think there's much of a difference.
Dead on. Berklee College of Music churns out class after class of fire-without-heat players, people with fantastic technique, most of whom go on to do absolutely nothing worthwhile musically. This despite the fact that they can outplay most of the musicians that create music that people actually want to listen to.
It's an easy trap to get into, and it happens in programming, too - who amongst us has not had to work with some technically brilliant programmer that wrote fast, concise, and "impressive" code that despite working for its purpose ended up being so incomprehensible to everyone else that it was unmaintainable? People that start down the path of technical skill and forget to stop and smell the roses are far too likely to forget that technical facility is only a means to an end, and they start evaluating both themselves and others on the technicalities of what they do, not the results.
To me, stuff like auto-tune is the equivalent of using a decent IDE, or programming in a language that is well suited to your task rather than one that makes you do everything yourself. Yes, it probably takes a better programmer to code CGI stuff in C than in PHP, and plenty of poor programmers rely on all the built in functionality of PHP to cover the fact that they can't code; that doesn't mean that the good programmers should always stick with straight C when PHP could cut out a lot of the work, though.
If a musician uses auto-tune to turn a 20 take marathon into a couple of takes that can be cleaned up after the fact (and end up with similar results), that's fine, they still may be an excellent musician, they're just working smart. If they're using it to cover the fact that they really can't sing, that's another issue altogether, but even then, if they have something worth singing and can't pull it off technically, why not fix it up with technology? If it enables a good product that otherwise couldn't have been created, then maybe they have excellent songwriting skills but poor vocal ones, and why should they not do whatever they can to put it out?
Literally cutting and pasting "decreasing the yellow light period does in fact increase the number of rear end collisions" from the original post into Google returns http://www.motorists.org/blog/red-light-cameras-increase-accidents-5-studies-that-prove-it/ as the second result (the first is a link to this discussion).
Yeah, it's on the person making the claim to justify it, but come on - if you can cut and paste their claim and get the source within the first couple results, it hardly makes sense to complain about it as if this was Wikipedia or something.
They may know better, but I don't know how much they really know. I would bet that the hard drive experts of the 1980s would never have thought that the exponential growth in capacity would continue all the way to 2010, but it has nonetheless. Trivially there is some upper physical limit to storage density, of course, but there are a lot more tricks to increase storage capacity before we get there, they just may require abandoning the current data storage techniques. Notice that Kurzweil is very specific about this - while there are hard and fast limits on a particular solution to a problem, when we start to hit those limits other solutions are more fiercely researched and tend to win out eventually because they can be pushed further. Then there's a new "theoretical" limit where people say the growth must end, which is true, but only for that particular solution, and so on.
And it's true that it must end somewhere, but the question is, does that mean in 10 years or 1000? Kurzweil seems to think even 20 will be more than enough for AI, and I don't think that's a very off the wall extrapolation; beyond that, I'll definitely agree, there's a lot more uncertainty. But I wouldn't discount the fact that if we can pass the AI threshold, there will likely be quite a boom in innovation from that point on, so to me it's really a question of whether we get there or not, and I think Kurzweil makes a pretty compelling claim for why we can (which amounts to noting that early AI attempts were doomed to fail because we didn't have the power to do what the brain does, so it would have required some exceedingly clever algorithm to reproduce its function, and that once that restriction is gone it will be much easier software-wise).
Interesting, that's probably a fair goal. From what I remember of the class (which wasn't so great, and I'm very thankful I got to take some higher level stuff), the message most people took from it was "Almost everything is normally distributed, always assume the normal distribution, and if you can't...well, it won't ever happen, so...err, well, actually, I mean, forget I said anything, it won't be on the test!" No wonder most people get nothing out of the class...
And there you've hit upon the main problem with Kurzweil's campaign - he goes around screaming about how this is absolutely destined to happen, with a lot of argument about why, and how great it will be, and all that. But if he's right, who cares? We'll be there, too, and fantasizing about it for 30 years beforehand won't make a shred of difference if it's as inevitable as he claims.
I suspect a lot of it is that he meets such strenuous resistance to the concept, which means that it's really touching a nerve. He's clearly a bit of a media whore, too, and I think he enjoys the attention.
No, no, no. If extrapolating a curve past its support was invalid in general, then you couldn't predict anything at all, and many, many extrapolations are extremely useful. It entirely depends on what type of curve you're fitting, how you fit it, and what data it's fitted to. If you want to extrapolate based on a curve, you need to have an argument that explains why the extrapolation is justifiable; lacking this, yes, you've made a basic statistics error, but if you have a reason to expect the extrapolation to work (usually a reasonable assumption of some invariant), then the argument is domain specific, about that invariant, and not related at all to statistics.
Kurzweil has an argument to justify his exponential Moore's law extrapolation, which is one of the main points of his book. Feel free to criticize that, but don't try to pretend that he's wrong based on Stat 101, it's missing the real point that needs to be argued.
Outright replacement would be a foolish strategy, as it would throw away the learning of the previous generations (much like human reproduction is a foolish strategy for accumulating knowledge). One of the first optimizations a computer could make on top of near human intelligence is the ability to preserve knowledge from generation to generation, so there would be no loss whatsoever.
Well, people are certainly trying, so your first option is right out. And I wouldn't be so certain that we're collectively too stupid to succeed; it's a terrifically hard problem, to be sure, but as Kurzweil points out, even if nobody is able to crack it elegantly there is a brute force solution (simulate a brain, neuron by neuron) once we've got enough processing power and medical imaging technology in place.
True enough. But pointing that out does not make him or his arguments wrong, either.
To say that creating computers advanced enough to surpass ourselves proves that we have "lost" imagination and creativity is a stretch, to say the least. To me it would seem to prove the contrary.
Whether it will happen or not, and in particular whether Kurzweil's timeline is correct, is another issue; as many have pointed out, futurists love to predict that the most fantastic things will happen right near the end of their lifetimes, so his "live forever" claims may be borne of hope more than reason. But the Moore's law claims don't seem as wild to me, since he is very explicit about noting that it has nothing to do with the particulars of the chips, but about the fact that the total computing power tends to follow the law with only minor divergences as one technology dies out and is replaced by one that scales better.
Personally, I feel the label "religion" is a bit inappropriate whenever log-log plots are a crucial part of the pitch. Feel free to disagree.