In my first post, I said that 50MP is about enough. That we should probably invest in improving more than just more pixels. I did say that we should probably invest in better image delivery first, but that doesn't exclude investing in better color fidelity, which is probably also a better investment than in more pixels. And indeed I expect we'll get more ROI from investments in image delivery than in color fidelity, since we've already got very good color fidelity (if not nearly perfect, as I described), but our image delivery is really awful. For example, screens with limited viewing angles, low contrasts, glare... There's a lot more cheap improvements waiting in delivering images to the eye, probably at first with goggles or other delivery more directly to the retina.
That's what I've been saying since that first post. But there is indeed more to it than meets the eye.
My conclusion follows perfectly from my arguments. You're just not following the arguments. Because my conclusion states that more megapixels is probably not as good an investment now, after 50MP, as other investments, like better color precision and accuracy. As for the temporal limit (also not megapixels), the optic nerve signals at about 40Hz, so our fastest digital cameras can outperform that human bottleneck. However, the optic nerve is not only massively parallel, but is also "clockless": visual fields aren't carried to the visual cortex in discrete frames, but rather in parallel streams per "pixel" that arrive whenever they arrive, with a small latency (that cameras outperform). So not only is the retina's hexagonal grid more complex (and irregular) than our cameras' firmly regular rectangular grid (so we can see the static grid in our fluid visuals), but so too is the temporal grid in which the moving images arrive in our brains.
But none of that is megapixels. We've probably got about enough megapixels. As I concluded, we should invest in making the 3D (x,y,t) distribution of captured visual data match more closely the distribution of sensed visual signals, to get more fidelity return on our tech investment.
There's probably a complex argument to make about whether color fidelity is a better investment target than is improving the various "grid fidelities". But though that analysis requires a lot more complex economic and technical analysis than is appropriate to a Slashdot post, the industry has already backed away from 40bit color (that we delivered in the early 1990s) down to 32 or 24bit. Because humans distinguish (somewhat) fewer than the 16.8M total colors that 24 bits can specify. Again, the problem is the grid: the human colorspace is very complex, and there's no commercial colorspace that matches it exactly (even as an average of many tested humans). There's far too many reds and blues than greens in the artificial spaces, so there's many different greens our cameras can't capture, and many different reds and blues they do but we can't tell apart. And that is true across the gamut of colors in either the digital or natural spaces. Which again is a "grid problem". Not the resolution of the grid (this time the resolution of the gamut), but the distribution of the grid (to match the idiosyncratic distribution of the human colorspace in its natural grid).
So we have a long way to go. I didn't say we don't. The question I considered is how much further to go in the megapixels direction, rather than the others needed. My answer is accurate.
It's entirely possible that you have bad eyesight. Measuring lots of other people over long time periods has shown that the average human fovea has about 4000x3000 resolution. We all also use eye movement to gather more visual info.
No, the problem, as I described it, is that the real world is very complex, with many interacting factors. It's true that the delivery of the captured image to the eye is a whole other system, with its own resolution, fidelity, and complex factors. But the point, which I made, is that investing in more MPs probably isn't as important anymore after 50MP as is investing in the rest of that complex system (probably mainly the image grids).
You're just uninformed about the visual system (though I'm doing my best:). The visual sense is extremely well understood, including the brain, though of course there's plenty to know. Visual neuroscience is probably the best understood sense. And we've delivered quite a lot of results using it. Imaging and color use quite a lot of our large and detailed neuroscience to deliver the high fidelity we've already got, and engineering is doing more with the neuroscience all the time.
It makes perfect sense to make the comparison. I've been doing it for almost 20 years, since I took my undergrad neuroscience and light physics training to a digital camera company (32MP@40bits in the early 1990s). People who know what we're talking about make quite a lot of sense out of the biology and physics to make good engineering. If you can't keep up, that's not our problem - we're sharing as well as we can.
Humans have rights, including the rights to free expression. We make governments as our means of protecting those rights. When our governments don't protect our rights, we've failed.
Humans also make corporations to protect our rights. Sometimes these different rights conflict. That's the real work of getting along: negotiating those rights.
When human rights are abused for long enough, the people rebel. Along the way, those people might not rebel, at least not enough to succeed. But they underperform, resist in lame ways. They're ungovernable. The degree to which we create goverments that protect our rights, and form corporations that don't abuse them is the degree to which we can get along.
None of this is even the tiniest bit different on the Web. The Web is just the latest in spaces somewhere between public and private. Private property is used to admit the general public. And since the space is somewhere between public and private, the interplay of rights to public access with rights to control of private property are the subject of the negotiation. If that negotiation protects only the rights of the private corporations owning the property into which the public is admitted, and abuses the rights of that public, then the people will become ungovernable. Likewise if the public's rights are protected to the exclusion of the private rights, the people whose property is being used will become ungovernable.
Corporations will naturally tend to protect their own rights to the exclusion of others', of the public's. That's exactly why we have governments. The governments might also tend to prioritize the people's rights, except that our current governments are so much more interested in corporate people that such a conflict doesn't happen: the corporations have all the advantages. But if the people's rights aren't protected in the appropriate balance, the people will become ungovernable, and rebel.
The fovea of the human eye, the part that sees details, is approximately (in a hexagonal layout) 4000x3000 photoreceptive cells. To saturate the foveal field with data, the Nyquist rate says that an image must deliver 8000 x 6000 dots. Which is 48MP. 50MP is enough to cover that field. It's still not quite enough to completely fool the eye, until the 50MP is in a grid that exactly matches the eye - and no two eyes are the same, even in a single person, and not regularly hexagonal, but actually a stochastic distribution in a roughly six-axial surface across the inside of an uneven sphere.
And even then, the fovea is only about 1mm, capturing a 2-degree field in the middle of vision, about double the width of your thumbnail at arm's length. These 50MP cameras only capture the amount of info that's in the central 2 degrees, though the human eye captures data (though much less per degree outside the fovea) from a visual field with a 160 degree horizontal width and 135 degrees vertical height. Unless the image delivery can track the eye's movement to stay projected on the fovea, the image has to have foveal (over) density imagery across the entire scene for the fovea to track across.
But for images to stare at, 50MP is about the foveal (over) resolution. Further improvement is probably better off invested in image delivery technology, as we're sampling at about the limit of what we can actually see.
The language you know is better than the one you don't if they're otherwise equal. If the one you know is more than adequate, it's better than the one you don't know even if the other one could be better for someone else.
Now you're calling me silly. When you talk like that, you're snotty. But of course I've done more than just point out that you're snotty. I've pointed out that your argument is wrong. And you've conceded, so I've accepted.
But even that's not good enough for you. So you can now have your snotty argument with yourself, or someone else who cares.
There are no military applications for a tech that delivers huge amounts of energy suddenly, produced from a critial ingredient as plentiful as hydrogen?
Obviously there are many. Which fusion could put within reach of actors currently excluded from that scale of military action, but who could get access.
And suddenly, you're attacking me like an asshole, when it's you who obviously doens't know what the fuck you're talking about.
Wind and solar: less infrastructure than lots more nukes. BIG PROBLEMS. DEAD PEOPLE.
Yes, like Python. There is indeed a case to be made for a Python browser plugin, much like the case for Perl. If you actually made one - I never said there isn't.
You've not made any logical point in this whole discussion. All you're doing is denying that I've made the points about Perl that I've actually made. You're not arguing for anything, and your arguments against Perl are specious, while ignoring the valid points I'm making.
And now you're talking like an asshole. WHen in fact the "moron" in this discussion is surely the one who can't argue worth a damn. That would be you, in case you can't do the math on that one, either.
Anonymous Coward, just because you can't understand the posts enough to mouth a disagreement doesn't mean I should stop posting. It means you should stop "reading", and certainly don't post twaddle like that.
The point about wind capacity is that the limit is our engineering, not the resource or the science. A few years ago, most people would have said that there isn't even enough wind to power everything, no matter how good we improve the engineering - in fact, lots of people will still say that. Also, to generate 3.5TW we wouldn't need a billion windmills, we'd need about 550,000, and that's just the current generation of engineering. If we doubled their capacity to over 10MW each, and tapped only 20 states, that's 12,500 windmills per state. At an average windfarm installation of 50 turbines, that's only 250 farms per state. A lot, but doable, and producing our total output, more or less.
Solar doesn't need to take a long time. In the 10 years to permit, build and test a nuke plant, a large desert solar plant can be up and serving even more people, with lower financial, industrial/energy and security investments.
Unless you or I can present the comprehensive and complex economics of each of the nukes industry and the necessary alternative energy industries, we're just asserting that nukes are higher or lower on the list. But since nukes have had 50 years of supposed superiority, highly subsidized in every way, but still are competitively replaced by these new alternatives that haven't gotten anywhere near the subsidies, and have instead faced all kinds of supression, it's pretty clear that a level playing field would probably at least reverse their relative shares of the market (about 20% vs about 2%).
And yes, fusion would be great, but its military applications are probably even more destabilizing than fission is becoming, especially if it uses nonexotic feedstock and the kind of engineering that is available on the global market today. We'd be a lot better off if we could give energy independence to most countries, especially the equatorial ones that have been so volatile, for several years of constructive development before we unleash on everyone the power to incinerate their enemies from a far distance.
The extra interlinks will also relieve bottlenecks elsewhere. Since most Internet bandwidth now goes through the US, other links offloading from the segments tying the US together will also increase the spare capacity of those relieved internal links.
The telcos are going to have to lie a lot harder to pretend that there's not enough US bandwidth to retain Network Neutrality, and instead start the Net Doublecharge on bandwidth already paid for at the other ends.
The "Obfuscated Code" contest originated with C. Any language capable of succinct syntax therefore also has the power to be easily obfuscated. Perl is no different from C. But that's not necessary: the language's flexibility allows highly self documenting code. Show me how Javascript regexes can include literal comments, and then tell me that Perl can't be clearer than Javascript.
I prefer to enforce code's effective communication with humans through coding guidelines. And take the language's power to communicate with the machines unconstrained.
No, I asked you to send me your list. History doesn't have to come from just a few extremely partisan outfits - it's available to everyone. If you can't deliver a source with a reputation for truth, instead of obvious crazy liars, then those messengers demonstrate just what is the value of their message. Especially when that message is so obviously completely counter to the vast documentation to the contrary, it's obviously just a pack of lies.
Just because the message is more important than the messenger doesn't mean that you can reverse the logic to deliver a messenger who isn't credible, and expect their message to be believable. Of course, you rightwingers understand logic about as well as dogs understand "shake hands": it's a trick you've seen people do, so you go through the motions, but you're using your paws, not really shaking hands.
You rightwingers are so crazy that you don't even care that the sources you believe are obviously insane liars with vested interests in crazy lying. It's brains like that which lost the Vietnam War, are losing the Iraq War - and decided those wars would be good ideas to start in the first place. But since the true history of crazy, lying warmongers is so long, it's pretty clear that we're stuck with at least a few, very loud ones of you. So we've learned how to identify you and your same old lies, and just chain you up somewhere that you won't chew too much that's valuable.
The geothermal power in America's most accessible locations can be used to drive CO2 from the air into synthetic petrofuels, in large scales, with technologies like Green Freedom (which can also recover old nuke plant infrastructure without wasting it). There is so much geothermal so cheap around the Pacific Rim that we could power everything from it. But even elsewhere in the US the geothermal is plentiful enough that building it there is cost effective. Especially when compared to the pollution from digging up more hydrocarbon, which in the US is coal, which is filthy (both CO2 and radioactive particles). And compared to nukes, the risks are nil.
Before you get down the list to needing to indulge risks and costs like nukes and petrofuels, there's lots of other cheaper, cleaner and less risky alternatives. Wind alone has more power than we consume, if we tap it right. Solar in the Southwest could also make the US a net energy exporter again. And generators like this Anaconda device, or other hydroelectric, could also make more than we consume if deployed correctly with a properly efficient revision.
By the time we get to nukes to produce that much power, we're mining, importing, handling, storing and securing fuel and waste in such large amounts that those costs are much worse than those alternatives. Plus the nuke plants cost a lot to build and demolish. Plus, with such a large number of nukes, built so rapidly by a construction and management industry driven by so much greed and "growing pains", the chances of bad workmanship (especially by the nukes industry, which has such a mafioso record of cutting costs and screwing up) grow very large. And of course the price of any significant failure of a nuke plant is very high, compared to what happens when geothermal, wind, solar or hydroelectric go wrong. And then there's just the overall centralization of nukes, vs the distributed architecture of the alternatives - which also affects efficiency, as electric transmission losses are significant.
Nukes have a compelling power output, and they can indeed be made safer than the ones we've got now (which shows, conversely, that the ones we've got now are riskier than they have to be, yet we build and run them anyway). But they overall fit into an economic and industrial model that started going out a half-century ago: giant, centralized production, a series of single failurepoints stacked in a monopoly of the essential resource. The alternatives have the benefits of distributed systems, of diversity, of localization, as well as their own energetic advantages. Perhaps if the alternatives put us into an "Energy Age" where energy is so cheap that we can afford to invest the huge amounts in nuke tech and fuel management, then nukes could be a dividend. Especially if more plentiful energy makes the world more politically stable, rather than how petrofuels and nukes destabilized the world through the 20th Century. But in the meantime, we must prioritize the lower impact, more reliable alternatives that have better cost:benefit*risk values.
Except that the reasons I like Perl are the same as the reasons it's a good browser scripting language. How would those two arguments be different, anyway?
The main point is that Perl is better than Javascript. Which would also sound a lot like "why I like Perl".
One advantage is that lots of people, not just me, have liked Perl enough to have written a large, easily accessible library of code for doing all kinds of things.
Another advantage is that Perl is a much higher performance language than Javascript.
And then there's the flexibility of Perl's syntax and native complexity of data structures.
Perl is a much better language than Javascript in practically every way. The only limit is that Perl doesn't run in browsers against the browser and DOM APIs. That's why I like Perl, and why it would be a good browser plugin like I described.
See how someone actually forms an argument when they believe something? We back it up with facts and logic. Passive aggression like "whatever you say" doesn't cut it, nor mere assertions denying my argument, nor do empty promises to say the same thing for any language you like, when you can't even say it substantially about Javascript, which you're defending. So despite my giving you the respect of a reasonable argument, you've given me only a snotty attitude. So I accept your giving up, but it wasn't any fun to force it out of you.
Ha - who needs Limbo when we can learn "history" from you and your sources. Like AmericanThinker.com, your first offering, that features the lead story today as an expose of how Obama is really a Communist. Or VietQuoc.com, which is the Vietnamese National Party, which couldn't keep its own country with the US propping it up for generations, but has a pretty decent gig pretending that the US media lost the war.
Thanks for demonstrating just what a pack of lies you peddle.
There's no reason Perl can't do all that. In fact, dealing with files, processing text and gluing apps together are Perl's strengths. Perhaps you meant to type Perl instead of Javascript in that sentence, but then I don't follow you.
For traversing the DOM, Perl would be just as syntactically appropriate, with a Perl DOM API. There's nothing else in the functions and events of "web related" objects and actions that Perl isn't also suited to. It's a lot better suited to the tasks than just "adequate". Yes, ECMAScript had a lot of work to reconcile competing, incompatible scripts, but that wasted a lot of effort that didn't then go into making a language that's just the best.
So a Perl plugin with a DOM API would be even better. And use the large library of Perl, as well as the large community of Perl programmers. Perl still wins.
If someone wrote the Perl browser plugin I described, that could revolutionize Perl's perception into "savior language for everything". Which is exactly what happened with CGI programming, even though restarting (and therefore recompiling) a Perl script with each HTTP hit was exactly the worst use case for the language.
Are you a foreigner with a native language other than English who's learned a programming language? Before you learned English (or how are you reading this post)? What language is native for you?
How hard was it to learn the programming language without English as a starting point? Harder than it seems for English speakers with other skills similar to yours? Was learning English after learning the programming language easier for you than for people you know who share your language who don't program?
Perl is closer to C than Java is to C. They're each pretty close, but Java is always object-oriented, closer to C++ (but without a lot of C++ complexity and flexibility). Also, Perl includes references to objects and values, while Java does not. Java is a lot more strongly typed than is either C or Perl.
For the "standard pseudo-assembly bytecode" that is compiled from various languages, you're looking for Parrot, built to support Perl6, but also supporting other languages, that runs them as virtual machine bytecode.
Slashdot Mirror
In my first post, I said that 50MP is about enough. That we should probably invest in improving more than just more pixels. I did say that we should probably invest in better image delivery first, but that doesn't exclude investing in better color fidelity, which is probably also a better investment than in more pixels. And indeed I expect we'll get more ROI from investments in image delivery than in color fidelity, since we've already got very good color fidelity (if not nearly perfect, as I described), but our image delivery is really awful. For example, screens with limited viewing angles, low contrasts, glare... There's a lot more cheap improvements waiting in delivering images to the eye, probably at first with goggles or other delivery more directly to the retina.
That's what I've been saying since that first post. But there is indeed more to it than meets the eye.
My conclusion follows perfectly from my arguments. You're just not following the arguments. Because my conclusion states that more megapixels is probably not as good an investment now, after 50MP, as other investments, like better color precision and accuracy. As for the temporal limit (also not megapixels), the optic nerve signals at about 40Hz, so our fastest digital cameras can outperform that human bottleneck. However, the optic nerve is not only massively parallel, but is also "clockless": visual fields aren't carried to the visual cortex in discrete frames, but rather in parallel streams per "pixel" that arrive whenever they arrive, with a small latency (that cameras outperform). So not only is the retina's hexagonal grid more complex (and irregular) than our cameras' firmly regular rectangular grid (so we can see the static grid in our fluid visuals), but so too is the temporal grid in which the moving images arrive in our brains.
But none of that is megapixels. We've probably got about enough megapixels. As I concluded, we should invest in making the 3D (x,y,t) distribution of captured visual data match more closely the distribution of sensed visual signals, to get more fidelity return on our tech investment.
There's probably a complex argument to make about whether color fidelity is a better investment target than is improving the various "grid fidelities". But though that analysis requires a lot more complex economic and technical analysis than is appropriate to a Slashdot post, the industry has already backed away from 40bit color (that we delivered in the early 1990s) down to 32 or 24bit. Because humans distinguish (somewhat) fewer than the 16.8M total colors that 24 bits can specify. Again, the problem is the grid: the human colorspace is very complex, and there's no commercial colorspace that matches it exactly (even as an average of many tested humans). There's far too many reds and blues than greens in the artificial spaces, so there's many different greens our cameras can't capture, and many different reds and blues they do but we can't tell apart. And that is true across the gamut of colors in either the digital or natural spaces. Which again is a "grid problem". Not the resolution of the grid (this time the resolution of the gamut), but the distribution of the grid (to match the idiosyncratic distribution of the human colorspace in its natural grid).
So we have a long way to go. I didn't say we don't. The question I considered is how much further to go in the megapixels direction, rather than the others needed. My answer is accurate.
It's entirely possible that you have bad eyesight. Measuring lots of other people over long time periods has shown that the average human fovea has about 4000x3000 resolution. We all also use eye movement to gather more visual info.
No, the problem, as I described it, is that the real world is very complex, with many interacting factors. It's true that the delivery of the captured image to the eye is a whole other system, with its own resolution, fidelity, and complex factors. But the point, which I made, is that investing in more MPs probably isn't as important anymore after 50MP as is investing in the rest of that complex system (probably mainly the image grids).
You're just uninformed about the visual system (though I'm doing my best :). The visual sense is extremely well understood, including the brain, though of course there's plenty to know. Visual neuroscience is probably the best understood sense. And we've delivered quite a lot of results using it. Imaging and color use quite a lot of our large and detailed neuroscience to deliver the high fidelity we've already got, and engineering is doing more with the neuroscience all the time.
It makes perfect sense to make the comparison. I've been doing it for almost 20 years, since I took my undergrad neuroscience and light physics training to a digital camera company (32MP@40bits in the early 1990s). People who know what we're talking about make quite a lot of sense out of the biology and physics to make good engineering. If you can't keep up, that's not our problem - we're sharing as well as we can.
Humans have rights, including the rights to free expression. We make governments as our means of protecting those rights. When our governments don't protect our rights, we've failed.
Humans also make corporations to protect our rights. Sometimes these different rights conflict. That's the real work of getting along: negotiating those rights.
When human rights are abused for long enough, the people rebel. Along the way, those people might not rebel, at least not enough to succeed. But they underperform, resist in lame ways. They're ungovernable. The degree to which we create goverments that protect our rights, and form corporations that don't abuse them is the degree to which we can get along.
None of this is even the tiniest bit different on the Web. The Web is just the latest in spaces somewhere between public and private. Private property is used to admit the general public. And since the space is somewhere between public and private, the interplay of rights to public access with rights to control of private property are the subject of the negotiation. If that negotiation protects only the rights of the private corporations owning the property into which the public is admitted, and abuses the rights of that public, then the people will become ungovernable. Likewise if the public's rights are protected to the exclusion of the private rights, the people whose property is being used will become ungovernable.
Corporations will naturally tend to protect their own rights to the exclusion of others', of the public's. That's exactly why we have governments. The governments might also tend to prioritize the people's rights, except that our current governments are so much more interested in corporate people that such a conflict doesn't happen: the corporations have all the advantages. But if the people's rights aren't protected in the appropriate balance, the people will become ungovernable, and rebel.
The fovea of the human eye, the part that sees details, is approximately (in a hexagonal layout) 4000x3000 photoreceptive cells. To saturate the foveal field with data, the Nyquist rate says that an image must deliver 8000 x 6000 dots. Which is 48MP. 50MP is enough to cover that field. It's still not quite enough to completely fool the eye, until the 50MP is in a grid that exactly matches the eye - and no two eyes are the same, even in a single person, and not regularly hexagonal, but actually a stochastic distribution in a roughly six-axial surface across the inside of an uneven sphere.
And even then, the fovea is only about 1mm, capturing a 2-degree field in the middle of vision, about double the width of your thumbnail at arm's length. These 50MP cameras only capture the amount of info that's in the central 2 degrees, though the human eye captures data (though much less per degree outside the fovea) from a visual field with a 160 degree horizontal width and 135 degrees vertical height. Unless the image delivery can track the eye's movement to stay projected on the fovea, the image has to have foveal (over) density imagery across the entire scene for the fovea to track across.
But for images to stare at, 50MP is about the foveal (over) resolution. Further improvement is probably better off invested in image delivery technology, as we're sampling at about the limit of what we can actually see.
The language you know is better than the one you don't if they're otherwise equal. If the one you know is more than adequate, it's better than the one you don't know even if the other one could be better for someone else.
Now you're calling me silly. When you talk like that, you're snotty. But of course I've done more than just point out that you're snotty. I've pointed out that your argument is wrong. And you've conceded, so I've accepted.
But even that's not good enough for you. So you can now have your snotty argument with yourself, or someone else who cares.
Go and do whatever morons do. Goodbye.
There are no military applications for a tech that delivers huge amounts of energy suddenly, produced from a critial ingredient as plentiful as hydrogen?
Obviously there are many. Which fusion could put within reach of actors currently excluded from that scale of military action, but who could get access.
And suddenly, you're attacking me like an asshole, when it's you who obviously doens't know what the fuck you're talking about.
Wind and solar: less infrastructure than lots more nukes. BIG PROBLEMS. DEAD PEOPLE.
GOODBYE. Asshole.
Yes, like Python. There is indeed a case to be made for a Python browser plugin, much like the case for Perl. If you actually made one - I never said there isn't.
You've not made any logical point in this whole discussion. All you're doing is denying that I've made the points about Perl that I've actually made. You're not arguing for anything, and your arguments against Perl are specious, while ignoring the valid points I'm making.
And now you're talking like an asshole. WHen in fact the "moron" in this discussion is surely the one who can't argue worth a damn. That would be you, in case you can't do the math on that one, either.
Go to hell, jerk.
Anonymous Coward, just because you can't understand the posts enough to mouth a disagreement doesn't mean I should stop posting. It means you should stop "reading", and certainly don't post twaddle like that.
The point about wind capacity is that the limit is our engineering, not the resource or the science. A few years ago, most people would have said that there isn't even enough wind to power everything, no matter how good we improve the engineering - in fact, lots of people will still say that. Also, to generate 3.5TW we wouldn't need a billion windmills, we'd need about 550,000, and that's just the current generation of engineering. If we doubled their capacity to over 10MW each, and tapped only 20 states, that's 12,500 windmills per state. At an average windfarm installation of 50 turbines, that's only 250 farms per state. A lot, but doable, and producing our total output, more or less.
Solar doesn't need to take a long time. In the 10 years to permit, build and test a nuke plant, a large desert solar plant can be up and serving even more people, with lower financial, industrial/energy and security investments.
Unless you or I can present the comprehensive and complex economics of each of the nukes industry and the necessary alternative energy industries, we're just asserting that nukes are higher or lower on the list. But since nukes have had 50 years of supposed superiority, highly subsidized in every way, but still are competitively replaced by these new alternatives that haven't gotten anywhere near the subsidies, and have instead faced all kinds of supression, it's pretty clear that a level playing field would probably at least reverse their relative shares of the market (about 20% vs about 2%).
And yes, fusion would be great, but its military applications are probably even more destabilizing than fission is becoming, especially if it uses nonexotic feedstock and the kind of engineering that is available on the global market today. We'd be a lot better off if we could give energy independence to most countries, especially the equatorial ones that have been so volatile, for several years of constructive development before we unleash on everyone the power to incinerate their enemies from a far distance.
The extra interlinks will also relieve bottlenecks elsewhere. Since most Internet bandwidth now goes through the US, other links offloading from the segments tying the US together will also increase the spare capacity of those relieved internal links.
The telcos are going to have to lie a lot harder to pretend that there's not enough US bandwidth to retain Network Neutrality, and instead start the Net Doublecharge on bandwidth already paid for at the other ends.
The "Obfuscated Code" contest originated with C. Any language capable of succinct syntax therefore also has the power to be easily obfuscated. Perl is no different from C. But that's not necessary: the language's flexibility allows highly self documenting code. Show me how Javascript regexes can include literal comments, and then tell me that Perl can't be clearer than Javascript.
I prefer to enforce code's effective communication with humans through coding guidelines. And take the language's power to communicate with the machines unconstrained.
No, I asked you to send me your list. History doesn't have to come from just a few extremely partisan outfits - it's available to everyone. If you can't deliver a source with a reputation for truth, instead of obvious crazy liars, then those messengers demonstrate just what is the value of their message. Especially when that message is so obviously completely counter to the vast documentation to the contrary, it's obviously just a pack of lies.
Just because the message is more important than the messenger doesn't mean that you can reverse the logic to deliver a messenger who isn't credible, and expect their message to be believable. Of course, you rightwingers understand logic about as well as dogs understand "shake hands": it's a trick you've seen people do, so you go through the motions, but you're using your paws, not really shaking hands.
You rightwingers are so crazy that you don't even care that the sources you believe are obviously insane liars with vested interests in crazy lying. It's brains like that which lost the Vietnam War, are losing the Iraq War - and decided those wars would be good ideas to start in the first place. But since the true history of crazy, lying warmongers is so long, it's pretty clear that we're stuck with at least a few, very loud ones of you. So we've learned how to identify you and your same old lies, and just chain you up somewhere that you won't chew too much that's valuable.
Have a day.
The geothermal power in America's most accessible locations can be used to drive CO2 from the air into synthetic petrofuels, in large scales, with technologies like Green Freedom (which can also recover old nuke plant infrastructure without wasting it). There is so much geothermal so cheap around the Pacific Rim that we could power everything from it. But even elsewhere in the US the geothermal is plentiful enough that building it there is cost effective. Especially when compared to the pollution from digging up more hydrocarbon, which in the US is coal, which is filthy (both CO2 and radioactive particles). And compared to nukes, the risks are nil.
Before you get down the list to needing to indulge risks and costs like nukes and petrofuels, there's lots of other cheaper, cleaner and less risky alternatives. Wind alone has more power than we consume, if we tap it right. Solar in the Southwest could also make the US a net energy exporter again. And generators like this Anaconda device, or other hydroelectric, could also make more than we consume if deployed correctly with a properly efficient revision.
By the time we get to nukes to produce that much power, we're mining, importing, handling, storing and securing fuel and waste in such large amounts that those costs are much worse than those alternatives. Plus the nuke plants cost a lot to build and demolish. Plus, with such a large number of nukes, built so rapidly by a construction and management industry driven by so much greed and "growing pains", the chances of bad workmanship (especially by the nukes industry, which has such a mafioso record of cutting costs and screwing up) grow very large. And of course the price of any significant failure of a nuke plant is very high, compared to what happens when geothermal, wind, solar or hydroelectric go wrong. And then there's just the overall centralization of nukes, vs the distributed architecture of the alternatives - which also affects efficiency, as electric transmission losses are significant.
Nukes have a compelling power output, and they can indeed be made safer than the ones we've got now (which shows, conversely, that the ones we've got now are riskier than they have to be, yet we build and run them anyway). But they overall fit into an economic and industrial model that started going out a half-century ago: giant, centralized production, a series of single failurepoints stacked in a monopoly of the essential resource. The alternatives have the benefits of distributed systems, of diversity, of localization, as well as their own energetic advantages. Perhaps if the alternatives put us into an "Energy Age" where energy is so cheap that we can afford to invest the huge amounts in nuke tech and fuel management, then nukes could be a dividend. Especially if more plentiful energy makes the world more politically stable, rather than how petrofuels and nukes destabilized the world through the 20th Century. But in the meantime, we must prioritize the lower impact, more reliable alternatives that have better cost:benefit*risk values.
Except that the reasons I like Perl are the same as the reasons it's a good browser scripting language. How would those two arguments be different, anyway?
The main point is that Perl is better than Javascript. Which would also sound a lot like "why I like Perl".
One advantage is that lots of people, not just me, have liked Perl enough to have written a large, easily accessible library of code for doing all kinds of things.
Another advantage is that Perl is a much higher performance language than Javascript.
And then there's the flexibility of Perl's syntax and native complexity of data structures.
Perl is a much better language than Javascript in practically every way. The only limit is that Perl doesn't run in browsers against the browser and DOM APIs. That's why I like Perl, and why it would be a good browser plugin like I described.
See how someone actually forms an argument when they believe something? We back it up with facts and logic. Passive aggression like "whatever you say" doesn't cut it, nor mere assertions denying my argument, nor do empty promises to say the same thing for any language you like, when you can't even say it substantially about Javascript, which you're defending. So despite my giving you the respect of a reasonable argument, you've given me only a snotty attitude. So I accept your giving up, but it wasn't any fun to force it out of you.
Ha - who needs Limbo when we can learn "history" from you and your sources. Like AmericanThinker.com, your first offering, that features the lead story today as an expose of how Obama is really a Communist. Or VietQuoc.com, which is the Vietnamese National Party, which couldn't keep its own country with the US propping it up for generations, but has a pretty decent gig pretending that the US media lost the war.
Thanks for demonstrating just what a pack of lies you peddle.
There's no reason Perl can't do all that. In fact, dealing with files, processing text and gluing apps together are Perl's strengths. Perhaps you meant to type Perl instead of Javascript in that sentence, but then I don't follow you.
For traversing the DOM, Perl would be just as syntactically appropriate, with a Perl DOM API. There's nothing else in the functions and events of "web related" objects and actions that Perl isn't also suited to. It's a lot better suited to the tasks than just "adequate". Yes, ECMAScript had a lot of work to reconcile competing, incompatible scripts, but that wasted a lot of effort that didn't then go into making a language that's just the best.
So a Perl plugin with a DOM API would be even better. And use the large library of Perl, as well as the large community of Perl programmers. Perl still wins.
Who says?
I want a list.
When I see Rush Limbo at the top, I'll remember just what we're dealing with here.
Thank god Bush got us out of Iraq, though.
If someone wrote the Perl browser plugin I described, that could revolutionize Perl's perception into "savior language for everything". Which is exactly what happened with CGI programming, even though restarting (and therefore recompiling) a Perl script with each HTTP hit was exactly the worst use case for the language.
Are you a foreigner with a native language other than English who's learned a programming language? Before you learned English (or how are you reading this post)? What language is native for you?
How hard was it to learn the programming language without English as a starting point? Harder than it seems for English speakers with other skills similar to yours? Was learning English after learning the programming language easier for you than for people you know who share your language who don't program?
Yes, but I haven't noticed an emacs with a Firefox display mode.
Perl is closer to C than Java is to C. They're each pretty close, but Java is always object-oriented, closer to C++ (but without a lot of C++ complexity and flexibility). Also, Perl includes references to objects and values, while Java does not. Java is a lot more strongly typed than is either C or Perl.
For the "standard pseudo-assembly bytecode" that is compiled from various languages, you're looking for Parrot, built to support Perl6, but also supporting other languages, that runs them as virtual machine bytecode.
Yes, Perl's Parrot eats pythons.