I would agree that the precursors were all there, but the components themselves weren't. J2ME ran on postage-stamp displays, backed by processors so slow that yes, I could check my e-mail but it was painfully slow, and only for extenuating circumstances. J2ME made some money for the companies that sold bejeweled for kids to play on their Moto RAZR, but no one used it for productivity. Now, J2ME is synonymous with crappy feature phones for most consumers, so native applications are the way to go.
Apple won the "right place, right time" game, because they hit the market with the right product just as such products became possible. Having supported the precursors, though, I'd have to say that the components were coming, but weren't there yet.
True, you are just stealing power and cycles. I think the end result is that a) all BitCoins are assumed to be mined with stolen computing resources and b) BitCoin as a currency is devalued.
Your second paragraph hits the nail on the head. What basis is there for the presupposition that botnets do not account for at least 50% of the network? BitCoin, as I understand it, is predicated on the idea that the value of mined bitcoins is roughly equivalent to the cost of the processing resources needed for the computation. Thus, the best (meaning the method that makes the most money for you) way to mine bitcoins is to use someone else's computer!
People don't lose confidence in the dollar because the proportion of stolen dollars is fairly low, and there are agencies whose job it is to track down and prosecute those who stole the dollars in the first place (keeping the proportion of stolen dollars in circulation low). If, on the other hand, counterfeiting dollars (which, while these are "legit" bitcoins, the effect on the BitCoin market is essentially the same as the effect of counterfeit money) was more common, people would lose confidence, and someone would have to create an agency to do something about it. Based on what I've read, I have to assume that most bitcoins will be mined with stolen or coerced comutational resources as soon as they become valuable for any purposes besides counter-culture cred.
A lot of noise was made by the BitCoin Apologists that the value of any given BitCoin would match the cost of the resources it took to produce, making it the ultimate in fair, distributed currency, and no one could flood the market because it wouldn't be worth overproducing, right?
The catch is, of course, that it's possible to steal those resources pretty easily. I have to wonder why we wouldn't just assume that BitCoins are all mined by botnets, since the only way to make money mining bitcoins is to steal the resources from someone else.
You are correct. As an American with a manual transmission, I prefer to creep along in a low gear rather than stop and start. You raise an interesting point, though; perhaps if we simply tariff the heck out of automatic gearboxes, we can solve this problem.
So, you're in favor of removing all subsidies for oil companies? In a real market-driven situation, the government would not subsidies the oil companies, the price of fuel would be much higher, and the market would demand more efficient vehicles.
I think part of the problem is that it's too late to stop regulating. We have too many government-granted monopolies now, and while things may have been different had the market been unregulated from the beginning, suddenly de-regulating now results in huge market failures (such as the rolling brown-outs in CA a few years ago, when power was deregulated, or the recent banking melt-down, which happened as a result of banking de-regulation).
You're exactly right, this standard will only apply to new vehicles, so that guy with the late-80's buick that leaves a cloud of smoke every time he pulls away from a traffic light will still burn more fuel and dump more unburned hydrocarbons into the air than any five new cars.
The GP claims his car will last another 20 years, which would be impressive, and carries an implication that he will keep the car maintained (in order for it to last that long). However, unless he's a very fastidious vehicle owner, the fuel economy of that awesome car will start to decline, and emissions will degrade. And then, even though it was a great car when it was newer, it will be an environmental disaster. Bit-rot may be a myth, but exhaust-system rot is very, very real.
I can understand the extra weight argument, but I think it's a bit extreme. I had a '98 Civic sedan (also stick shift) that averaged 33 mpg (I don't tend to drive slowly, or at least didn't) and two cars before that, an '88 Corolla that easily made 40 mpg on the highway at 85+ mph (in a car with a carbureted engine). My father made 37 mpg in his 97 Accord (more efficient driver, also a stick shift). Yes, those cars were a bit lighter (and sure, the '88 Corolla had a pretty anemic engine), but you'd think that maybe someone, somewhere along the way would've done a little work on efficiency. I don't expect new cars to have better fuel economy than their earlier editions, but they could have at least tried not to lose so much ground.
It's not necessarily surprising that he couldn't explain what he heard. He was tapping some serious sub-conscious processing to make the distinctions between those two tracks, and trying to examine that with the conscious mind isn't always fruitful.
I would agree. Streaming is great for all those old movies or the ones you don't care too much about (just wanted to watch to see what the fuss was about, for example). However, if I want to actually sit down on a Friday night and watch a film with some friends, I'll take the disc. As I mentioned in another post, its even worse for foreign films; streaming is frequently dubbed-only, with no subtitles or original-language audio available!
I have long been frustrated by Netflix and their Streaming-in-English-only model. I prefer to watch foreign films and anime in their original languages with subtitles, since dubbed dialogue is typically sub-par (due to limitations of the format). With the DVD, I can choose whichever audio track I want, and enable or disable subtitles as needed. With streaming, that's not an option. I wouldn't think it's too high a technical hurdle to clear (a few radio buttons selected, you receive a different audio stream, and subtitles take very little bandwidth at all), but it hasn't been offered yet, meaning I'm stuck with DVDs for most of the content I want from Netflix.
The new pricing is causing me to seriously consider whether it's worthwhile, especially given how infrequently I have time to watch.
Refusing to learn is not really the same thing has having other priorities. I share your distaste for those who refuse to learn; I hate trying to assist people who proclaim "oh, I'm computer illiterate" with pride, as though it excuses anything that might be wrong with their computers, or other devices. However, there are plenty of people who literally have better things to do with their time than figure out how to do certain things with computers or handheld devices (e.g., heart surgery. I'm not exaggerating). At least part of Apple's target market are people in that category, who could learn, but would rather spend their time on other things.
Apparently I'm defending swaths of humanity against a cynic on Slashdot. I don't feel dizzy, but maybe I should go lie down...
Legally, you are correct. However, Weird Al has always had permissions to release parodies. It's not a question of legality, it's a question of being a nice guy.
Crime really started dropping about 17 years after Roe v. Wade. "Tough on crime" police tactics really don't factor into it [1]; reducing the causes of criminal behaviour (e.g., poverty) does.
[1] Dubner, Stephen J. and Steven D. Levitt, Freakonomics, HarperCollins, New York, pp. 137-144, 2005.
I don't know about "no way." It was a little unnerving the day I realized I could find (some) things in the Windows Registry without consulting documentation. Like an old manager used to say, "It's well known to those who know it well."
I definitely agree on the difficulty finding the keys which are omitted, because they don't have a default value in the registry. For whatever reason, those are also the keys whose names are a sentence with no spaces, like "AssumeUDPEncapsulationContextOnSendRule." (Anyone care to guess what that key is for?)
I call bollocks on your bollocks. I'm all for using gpedit or secpol when appropriate, but there are a lot of functions in the Windows registry which are not accessible except by manually editing the registry. Microsoft's online KB is full of articles about what keys you have to edit in the registry, from allowing Windows XP to use L2TP VPN through a NAT device, to specifying the compressed attachment size limits for ForeFront/Exchange.
The Windows registry is frustrating, for (at least) the following two reasons:
1) Long strings of hexadecimal numbers, and duplicate entries: There are a lot of programs that are identified by these. I understand the need for unique identifiers, but when you need to go dig something out because an uninstall failed, you're comparing 32-character hexadecimal strings. Additionally, any given application may (or may not!) have several of those keys associated with it. The multiple-key issues are probably a result of legacy support (have to keep the same keys for old versions of the application, but need to add new keys for new features in the new one), but it's still maddening when you don't know which of the half-dozen similarly named keys you need to delete or edit.
2) Lack of compartmentalization: Someone mentioned that the central database is fine, as long as everyone follows certain conventions. That's sort-of true, except that not everyone follows those conventions, and they're not great conventions to start. For example, one of the best troubleshooting features in MacOS (I mean 7-9, less-so in X) was that if an application was misbehaving, you could just open your preferences folder (hmm, centralized management?) and delete the preference file for that application. Doing so was nigh guaranteed to a) reset that application to its default state, and b) not effect any other application on the system. If I have to dig into the registry both of the opposite things are true: a) if I remove registry keys, I have to run the installer to recreate the defaults, since applications typically lose their minds if you rip out their keys, and b) I can never be quite sure if I have all the keys for the application I want to fix, or if I may have deleted a key that something else uses.
As an addendum, one of the nicest features of (most) configuration files on well-constructed open source software is that the defaults are listed, commented, for reference. This means that all I need to do to change a directive is uncomment the line and change the setting; in the registry, I find that the settings I want to change have omitted keys, which I need to type precisely for it to work. It's not a huge problem, but it is a nuisance.
Two questions/comments: Firstly, exponentially increasing computational requirements don't seem relevant to preventing nefarious miners. Someone has to mine bitcoins, or there won't be any. Either a company does it at scale and can recoup the investment by selling the bitcoins, or a botnet does it and doesn't need to recoup anything because there is no investment. I think in that case it amounts to serial theft (more like skimming the till than holding up the liquor store). What makes difficult computation a deterrent to someone using stolen computing power? Except for opportunity cost (e.g., BitCoins would have to be worth more than the computational equivalent in spam e-mail sent or extorted dollars in a DDoS blackmail scheme) I can't see any.
Secondly, and I guess this is related to the first, if someone holds up a liquor store, they're liable to be pursued and prosecuted. Conversely, there is almost no enforcement against botnet operators (for various reasons - separate discussion I suppose). I'd be inclined to compare botnet bitcoin mining to counterfeiting; someone is creating money without putting in the appropriate investment (e.g., paying for the utilities or computational cost, because it's distributed across an unwilling and unwitting botnet). Who plays the role of Abe Lincoln's Secret Service? Based on the admittedly limited reading I've done so far, I get the impression that the creators and proponents of bitcoin are so hung up on the comparative cost of the computational resources required to do the mining that they've ignored the relative ease with which those resources can effectively be hijacked or stolen.
#6 is the piece that breaks the system, from an economic standpoint, because it distributes the cost of computation to un-compensated and typically unwilling participants.
In the BitCoin model there is essentially perfect competition for production, so the value of a BitCoin should be roughly equal to the cost of the processing power needed to produce one (computers, electricity, cooling, etc.). Thus, it is only ever marginally profitable to mine BitCoins, and no one can really game the system by devoting tons of processor time to it, because it comes out even in the end (MR=MC).
However, the people who control the botnets didn't buy them, they stole them; malware of some kind (virus, worm, trojan, social engineering, etc.) was used to hijack the cycles from thousands of computers that the botnet controllers never paid for, and don't pay to maintain (electricity, cooling, etc.). Therefore, it is hugely profitable for a botnet controller to mine bitcoins, because they don't have to pay for any of the infrastructure to do so.
In the end, spammers win, good-guys are left holding the bag. I don't think I want to participate in that economy.
The difference is, botnets allow a few to harness the "labor" of many for their own gain, without compensating (and sometimes without even notifying) the many. Thus, some scammers who run a botnet could be using the computing power of thousands of zombie PCs to mine bitcoins, but only paying for their own workstations; the rest of the cost is foisted off on the owners of the zombie machines.
DeBeers aside, when mining precious metals you can't usually force someone else to do the mining for you without compensating them, and you certainly can't steal their effort without their knowledge.
Slashdot Mirror
It's like watching a Tea Party activist debate a Leninist.
Where may I purchase tickets to this event? Or the URL of the stream? Please?
I would agree that the precursors were all there, but the components themselves weren't. J2ME ran on postage-stamp displays, backed by processors so slow that yes, I could check my e-mail but it was painfully slow, and only for extenuating circumstances. J2ME made some money for the companies that sold bejeweled for kids to play on their Moto RAZR, but no one used it for productivity. Now, J2ME is synonymous with crappy feature phones for most consumers, so native applications are the way to go.
Apple won the "right place, right time" game, because they hit the market with the right product just as such products became possible. Having supported the precursors, though, I'd have to say that the components were coming, but weren't there yet.
True, you are just stealing power and cycles. I think the end result is that a) all BitCoins are assumed to be mined with stolen computing resources and b) BitCoin as a currency is devalued.
Your second paragraph hits the nail on the head. What basis is there for the presupposition that botnets do not account for at least 50% of the network? BitCoin, as I understand it, is predicated on the idea that the value of mined bitcoins is roughly equivalent to the cost of the processing resources needed for the computation. Thus, the best (meaning the method that makes the most money for you) way to mine bitcoins is to use someone else's computer!
People don't lose confidence in the dollar because the proportion of stolen dollars is fairly low, and there are agencies whose job it is to track down and prosecute those who stole the dollars in the first place (keeping the proportion of stolen dollars in circulation low). If, on the other hand, counterfeiting dollars (which, while these are "legit" bitcoins, the effect on the BitCoin market is essentially the same as the effect of counterfeit money) was more common, people would lose confidence, and someone would have to create an agency to do something about it. Based on what I've read, I have to assume that most bitcoins will be mined with stolen or coerced comutational resources as soon as they become valuable for any purposes besides counter-culture cred.
Here: http://slashdot.org/comments.pl?sid=2159376&cid=36158738
A lot of noise was made by the BitCoin Apologists that the value of any given BitCoin would match the cost of the resources it took to produce, making it the ultimate in fair, distributed currency, and no one could flood the market because it wouldn't be worth overproducing, right?
The catch is, of course, that it's possible to steal those resources pretty easily. I have to wonder why we wouldn't just assume that BitCoins are all mined by botnets, since the only way to make money mining bitcoins is to steal the resources from someone else.
You are correct. As an American with a manual transmission, I prefer to creep along in a low gear rather than stop and start. You raise an interesting point, though; perhaps if we simply tariff the heck out of automatic gearboxes, we can solve this problem.
So, you're in favor of removing all subsidies for oil companies? In a real market-driven situation, the government would not subsidies the oil companies, the price of fuel would be much higher, and the market would demand more efficient vehicles.
I think part of the problem is that it's too late to stop regulating. We have too many government-granted monopolies now, and while things may have been different had the market been unregulated from the beginning, suddenly de-regulating now results in huge market failures (such as the rolling brown-outs in CA a few years ago, when power was deregulated, or the recent banking melt-down, which happened as a result of banking de-regulation).
You're exactly right, this standard will only apply to new vehicles, so that guy with the late-80's buick that leaves a cloud of smoke every time he pulls away from a traffic light will still burn more fuel and dump more unburned hydrocarbons into the air than any five new cars.
The GP claims his car will last another 20 years, which would be impressive, and carries an implication that he will keep the car maintained (in order for it to last that long). However, unless he's a very fastidious vehicle owner, the fuel economy of that awesome car will start to decline, and emissions will degrade. And then, even though it was a great car when it was newer, it will be an environmental disaster. Bit-rot may be a myth, but exhaust-system rot is very, very real.
I can understand the extra weight argument, but I think it's a bit extreme. I had a '98 Civic sedan (also stick shift) that averaged 33 mpg (I don't tend to drive slowly, or at least didn't) and two cars before that, an '88 Corolla that easily made 40 mpg on the highway at 85+ mph (in a car with a carbureted engine). My father made 37 mpg in his 97 Accord (more efficient driver, also a stick shift). Yes, those cars were a bit lighter (and sure, the '88 Corolla had a pretty anemic engine), but you'd think that maybe someone, somewhere along the way would've done a little work on efficiency. I don't expect new cars to have better fuel economy than their earlier editions, but they could have at least tried not to lose so much ground.
It's not necessarily surprising that he couldn't explain what he heard. He was tapping some serious sub-conscious processing to make the distinctions between those two tracks, and trying to examine that with the conscious mind isn't always fruitful.
I would agree. Streaming is great for all those old movies or the ones you don't care too much about (just wanted to watch to see what the fuss was about, for example). However, if I want to actually sit down on a Friday night and watch a film with some friends, I'll take the disc. As I mentioned in another post, its even worse for foreign films; streaming is frequently dubbed-only, with no subtitles or original-language audio available!
I have long been frustrated by Netflix and their Streaming-in-English-only model. I prefer to watch foreign films and anime in their original languages with subtitles, since dubbed dialogue is typically sub-par (due to limitations of the format). With the DVD, I can choose whichever audio track I want, and enable or disable subtitles as needed. With streaming, that's not an option. I wouldn't think it's too high a technical hurdle to clear (a few radio buttons selected, you receive a different audio stream, and subtitles take very little bandwidth at all), but it hasn't been offered yet, meaning I'm stuck with DVDs for most of the content I want from Netflix.
The new pricing is causing me to seriously consider whether it's worthwhile, especially given how infrequently I have time to watch.
I remember hearing some stories about the unfortunate experiences of shoe salesmen from that era.
Refusing to learn is not really the same thing has having other priorities. I share your distaste for those who refuse to learn; I hate trying to assist people who proclaim "oh, I'm computer illiterate" with pride, as though it excuses anything that might be wrong with their computers, or other devices. However, there are plenty of people who literally have better things to do with their time than figure out how to do certain things with computers or handheld devices (e.g., heart surgery. I'm not exaggerating). At least part of Apple's target market are people in that category, who could learn, but would rather spend their time on other things.
Apparently I'm defending swaths of humanity against a cynic on Slashdot. I don't feel dizzy, but maybe I should go lie down...
So how can we harvest that CH4 to use as additional fuel?
Legally, you are correct. However, Weird Al has always had permissions to release parodies. It's not a question of legality, it's a question of being a nice guy.
Crime really started dropping about 17 years after Roe v. Wade. "Tough on crime" police tactics really don't factor into it [1]; reducing the causes of criminal behaviour (e.g., poverty) does.
[1] Dubner, Stephen J. and Steven D. Levitt, Freakonomics, HarperCollins, New York, pp. 137-144, 2005.
Excellent bumper sticker (actually saw this one):
"Militant Agnostic: I don't know, and neither do you!"
Unless he meant the talk command.
I don't know about "no way." It was a little unnerving the day I realized I could find (some) things in the Windows Registry without consulting documentation. Like an old manager used to say, "It's well known to those who know it well."
I definitely agree on the difficulty finding the keys which are omitted, because they don't have a default value in the registry. For whatever reason, those are also the keys whose names are a sentence with no spaces, like "AssumeUDPEncapsulationContextOnSendRule." (Anyone care to guess what that key is for?)
I call bollocks on your bollocks. I'm all for using gpedit or secpol when appropriate, but there are a lot of functions in the Windows registry which are not accessible except by manually editing the registry. Microsoft's online KB is full of articles about what keys you have to edit in the registry, from allowing Windows XP to use L2TP VPN through a NAT device, to specifying the compressed attachment size limits for ForeFront/Exchange.
The Windows registry is frustrating, for (at least) the following two reasons:
1) Long strings of hexadecimal numbers, and duplicate entries: There are a lot of programs that are identified by these. I understand the need for unique identifiers, but when you need to go dig something out because an uninstall failed, you're comparing 32-character hexadecimal strings. Additionally, any given application may (or may not!) have several of those keys associated with it. The multiple-key issues are probably a result of legacy support (have to keep the same keys for old versions of the application, but need to add new keys for new features in the new one), but it's still maddening when you don't know which of the half-dozen similarly named keys you need to delete or edit.
2) Lack of compartmentalization: Someone mentioned that the central database is fine, as long as everyone follows certain conventions. That's sort-of true, except that not everyone follows those conventions, and they're not great conventions to start. For example, one of the best troubleshooting features in MacOS (I mean 7-9, less-so in X) was that if an application was misbehaving, you could just open your preferences folder (hmm, centralized management?) and delete the preference file for that application. Doing so was nigh guaranteed to a) reset that application to its default state, and b) not effect any other application on the system. If I have to dig into the registry both of the opposite things are true: a) if I remove registry keys, I have to run the installer to recreate the defaults, since applications typically lose their minds if you rip out their keys, and b) I can never be quite sure if I have all the keys for the application I want to fix, or if I may have deleted a key that something else uses.
As an addendum, one of the nicest features of (most) configuration files on well-constructed open source software is that the defaults are listed, commented, for reference. This means that all I need to do to change a directive is uncomment the line and change the setting; in the registry, I find that the settings I want to change have omitted keys, which I need to type precisely for it to work. It's not a huge problem, but it is a nuisance.
Two questions/comments:
Firstly, exponentially increasing computational requirements don't seem relevant to preventing nefarious miners. Someone has to mine bitcoins, or there won't be any. Either a company does it at scale and can recoup the investment by selling the bitcoins, or a botnet does it and doesn't need to recoup anything because there is no investment. I think in that case it amounts to serial theft (more like skimming the till than holding up the liquor store). What makes difficult computation a deterrent to someone using stolen computing power? Except for opportunity cost (e.g., BitCoins would have to be worth more than the computational equivalent in spam e-mail sent or extorted dollars in a DDoS blackmail scheme) I can't see any.
Secondly, and I guess this is related to the first, if someone holds up a liquor store, they're liable to be pursued and prosecuted. Conversely, there is almost no enforcement against botnet operators (for various reasons - separate discussion I suppose). I'd be inclined to compare botnet bitcoin mining to counterfeiting; someone is creating money without putting in the appropriate investment (e.g., paying for the utilities or computational cost, because it's distributed across an unwilling and unwitting botnet). Who plays the role of Abe Lincoln's Secret Service? Based on the admittedly limited reading I've done so far, I get the impression that the creators and proponents of bitcoin are so hung up on the comparative cost of the computational resources required to do the mining that they've ignored the relative ease with which those resources can effectively be hijacked or stolen.
#6 is the piece that breaks the system, from an economic standpoint, because it distributes the cost of computation to un-compensated and typically unwilling participants.
In the BitCoin model there is essentially perfect competition for production, so the value of a BitCoin should be roughly equal to the cost of the processing power needed to produce one (computers, electricity, cooling, etc.). Thus, it is only ever marginally profitable to mine BitCoins, and no one can really game the system by devoting tons of processor time to it, because it comes out even in the end (MR=MC).
However, the people who control the botnets didn't buy them, they stole them; malware of some kind (virus, worm, trojan, social engineering, etc.) was used to hijack the cycles from thousands of computers that the botnet controllers never paid for, and don't pay to maintain (electricity, cooling, etc.). Therefore, it is hugely profitable for a botnet controller to mine bitcoins, because they don't have to pay for any of the infrastructure to do so.
In the end, spammers win, good-guys are left holding the bag. I don't think I want to participate in that economy.
The difference is, botnets allow a few to harness the "labor" of many for their own gain, without compensating (and sometimes without even notifying) the many. Thus, some scammers who run a botnet could be using the computing power of thousands of zombie PCs to mine bitcoins, but only paying for their own workstations; the rest of the cost is foisted off on the owners of the zombie machines.
DeBeers aside, when mining precious metals you can't usually force someone else to do the mining for you without compensating them, and you certainly can't steal their effort without their knowledge.
Be reasonable: there's no reason someone couldn't be all three.