Adaptation to longer yellow lights is severely misrepresented in the studies cited above. I have no idea how you could measure that adaptation without lengthening ALL yellows in a certain area, and limit the study ONLY to those driving in the respective area. Otherwise the driver is adjusted to the average duration of a yellow light in his area - of course he will break within a 6 seconds yellow, he expects red after 4 seconds ! So changing the yellow duration in only a couple of intersections in the city is unlikely to trigger adaptation, thus it grossly understates the effectiveness of changing all traffic lights to longer yellows.
The whole idea that longer yellows increase safety is specious. It's effectively claiming that a 4 second period is insufficient for stopping safely when driving at city speeds - and that's clearly bollocks.
The inter-light timing is horrible. In areas where there are several traffic lights in a row, a driver who has to stop at a red light at an intersection can be ceratin of one thing: they will have to stop for the next 4 intersections because those will also turn red, just as they approach. But if they either get a green light or manage to run the first red light, they won't have to stop at any of the next several intersections.
This idea is stupid, and you are a stupid person to bring it up. It's physically impossible for someone to stop at red light, get the green, start the car early in the green cycle, and reach the other intersection just in time to catch the next red, while at the same time someone who arrives late at the same 1st semaphore and barely makes a yellow to somehow catch the green in the next intersection before guy 1. In other words, assuming both drivers have similar speed, if you can get a string of yellows you can also get a string of greens, and the first red you catch will sync you up to the green. The scenario where one driver makes yellows in a row, while another driver stops at red in a row can't happen on a given road segment at a given speed.
The nice thing about this Patent Stimulus Plan is that it will cost only a small fraction of the amount of money we have already wasted on failed economic stimulus. What we need to do is have President Obama issue an Executive Order directing the Patent Office to start allowing patents. A 42% allowance rate during the first quarter of 2009 is wholly unacceptable. So while you are at it President Obama, order the Patent Office to issue a patent UNLESS there is a reason to deny it.
Obesity also increased from virtually unknown to over half of the western population. I think we are evolving to a super race of fat and smelly geniuses.
They are just as genuine as the ones generated by people infected with the get-rich-quick-bitcoin-bug, al miners compete for a fixed number of bitcoins.
As it turns out, the "immense power of an infected machine's graphical processing unit" is not that large when up against a bunch of people determined to make it big. The current hash rate of the network is 12Thash/second while a typical infected machine will have a low end graphics card, say 20 MHash/second (a vast majority will not have OpenCL-capable cards, and a tiny minority will have cards capable of 100 - 500 Mhash/sec). So this virus needs to infect on the order of 500.000 typical machines before it has enough power to compromise the security of the monetary system via double spend. At that time the author would earn about 1 million dollars/month (at current bitcoin prices), so he will have little incentive to destroy the network.
But why take money only when they screw up, when you can abuse your market position to ask for the maximum amount of money in the beginning, if they want to be in your root ? You are not making any extra money if they are competent, so why bother? Ah, the joys of free market.
This is a common theme among failing business. They become large bureaucracies hungry for cash and they will cannibalize their customer base in a vain attempt to cling to profitability. Can't wait for a failing Facebook, I need my own Database of Everyone® available now for the low low price of $100.
All those dollar bills in your wallet are only as valuable as we collectively agree they are.
Well then, let's collectively agree they are worth a mansion and a yacht for each citizen, shall we ? As it turns out, the long term value of money is not the result of psychology, rather a simple division of the economic output of the country by the outstanding monetary base. If some guy in Washington keeps printing those bills, they will eventually drop to zero, no collective agreement needed.
But who knows what that balanced economy would even look like, or whether it would even work.
Yup, we should stick to what we know works: spending money that will be paid by the future generation. I can't see any reason why this can't go on for ever.
Who are any of the rating agencies to say how much debt a given country should carry?
They do no such thing. What they speculate about is your ability to pay your existing debt. If you have trouble meeting the monthly mortgage payments, do you think you should get another credit card ? Well, if you can find somebody willing to give you one, sure.
The "2 billion dollar mistake" is not a mistake per se, but an acknowledged choice: the agency based it's calculations on a less optimistic growth scenario then the official one. This is not an actuarial calculation, it's a value judgement that involves macroeconomic speculation for the next decade. The agency does not believe the growth factors are realistic.
I find the "why should we trust an industry that failed to predict the recession and junk MBS" argument specious. It's not we who have to trust them, it's the investors, and they have the undeniable right to act on any information they believe relevant - it's their money after all. Yes, the rating agencies have lost credibility in the MBS debacle. This should suggest that things are really bad if even S&P sees it.
Your rant is ruined by some factual inaccuracies. You seem to lack of understanding of how "domain validation only" certificates work, and how EV certificates came to be. You are presuming that all certificates do the offline identity check, when in fact only EV issuers do that nowadays. Leveraging the DNSSEC tree to do the domain validation is actualy more secure than involving a 3rd party via the classic DNS system.
I've never proposed to make SSL work like Putty, and I clearly stated in my message that accepting self-signed certificates today way will open PayPal to attack. Self signed certificates would become usable, I was saying, only when they could be authenticated via the secure DNS tree, at which point the system would become equivalent to the existing non-EV certs that validate only the domain.
I concede that 15$ is indeed a low price, last time I checked it was something like 70 less inflated dollars, and when I had multiple vhosts on multiple domains it added up quite quickly. But there's also a huge benefit with free: you can automate it as a step in a webserver setup script, making ALL sites secure by default. Since the domain is verified by Google's or Mozilla's automated interface, you would get full end-to-end encryption and authentication without DNSSEC extensions, not just protection against passive eavesdropping .
On a side note, you should understand that accepting a self-signed certificate still protects you from passive eavesdropping, a major attack scenario. A man-the-middle is forced to fake all your SSL traffic with certificates generated by him. If applied on a large scale this can easily be detected by informed users, so a democratically elected government will think twice before doing it.
I would also actually run a HTTPS server everywhere if I didn't have to deal with the certificate mafia, and if major browsers would silently accept self-signed without drowning the user in a storm of "RUN FOREST, RUN !!!" messages. This is currently pretty tricky to do on the browser side without opening PayPal to attack (cache the sites that use real certs ? have a hardcoded master list for first connect ?). But it would be very nice if I could publish a flag in DNSSEC that could say "This is my certificate thumbprint, use it", and leverage the secure DNS tree instead of the insecure and bogus certificate industry.
Why again should I have to fork a pile of cash to obtain a bit string that says that I actually own the domain I'm using ? Generating this bit string seems like a task that could easily be automated to the point of being free. I can understand why Microsoft would be against this (and claim tens of thousand to add you to their root zone), but for example Mozilla or Google could create such an automated certification authority, and add it to their trusted root zone since they know they can trust themselves. Such certificates would work just as the "real thing" on Mozilla or Chrome, but would of course get the usual prompts in Internet Explorer.
That sounds like a grid that is not smart enough; if everybody charges his Chevy at 2AM, then 2AM will be the new peak hour and it will cost an arm and a leg to charge at that time. If the price information is delayed versus the instantaneous power consumption, then yes, a spike should be expected when the the price drops, but this could be countered by distributing the price information with random delays and only in some areas.
I mean, we're posting on a site that seems to handle anonymity just fine.
But I'm not anonymous, I'm Stellian, well known slashdot lurker and kook. I have a nice karma and care about my reputation, so I try my best to behave. See, no childporn or viruses in this post.
If I disclose my real identity complete with full name and postal address:
- it will not improve the quality of my posts; facebook is a perfect example on non-anonymous people incapable, on average, to produce any useful content
- it will not stop other anonymous people to do illegal things, in fact criminals will always try to remain anonymous when operating, just like in the real world
- it will allow an anonymous stock owner of facebook/slashdot/etc. to make a few bucks more by farming my data
No, anonymity is not going anywhere and I will refuse to use any service does not respect my privacy
How exactly one goes about building a nuclear reactor from mail order uranium (presumably depleted) and smoke detectors (about 1 microgram of Americium 241 each) ? The critical mass of Am 241 is over 50 Kg, so he would need 50 million smoke detectors to build a bomb. For a controlled, moderated reaction, much more, maybe hundreds of Kg. The technology to enrich natural uranium up to reactor-grade level is barely in the hands of states.
The fact that someone took him seriously and actually sent a guy with a detector AND a police squad to his house shows just how ridiculously incompetent the regulators are, and how paranoid people get when the word "radiation" is uttered.
Yeah, well good luck with that. If your are approached by a small company in a booming market and you make a great impression at the interview it might fly. It won't work with a multinational corporation - even if you can convince the person making the hire decision, they have no authority to change the contract, and any such change must go through legal and 3rd party auditors that certify the change does not endanger the company; this process takes months in a large corporation - there's simply to much at stake. A single wrong word and you might be able to sue the corporation for 2.6 billion dollars for your Java. You might say you don't want to work for such a corporation any way, but then again you might not have another option not evolving a steep pay-cut.
It also drives the point home that you are a peculiar person and a non-conformist. A small shop working on the next big-thing ? Sure. A corporation staffed full with bosses who did sign the same contract ? No way. You actually weed yourself out before getting a chance to put your non-conformism to the work and shine in a sea of mediocrity (or make a complete fool out of yourself and have people avoid eye contact in the hallways in fear of being gunned down in your imminent killing spree).
if you are honestly interested in claiming your copyrights this is the best (and arguable only) way to enforce the license.
Assuming you actually own that copyright. 99% of the work contracts out there have a clause where you are ceding all intellectual property to the employer. It's so standard that you should always ask for permission from your employer before writing and releasing open source software - you might not have the right to do so, even if the software is not related to the business of your employer and even if developed in your spare time; the language in my contract is unambiguous about that.
Assuming that in this case the permission to write and release open source software was implicit, it still does not mean the company has lost it's control of it's intellectual property - they can always dual-license it under a proprietary license. They can't "take back" the already released GPL software, and they can't grab any contribution of 3rd parties to that lineage, but they can chose to develop the original codebase in an entirely closed source fashion - it's theirs.
So spending 10 minutes to read your contract might save a butt-load of lawyer fees.
The vast majority of servers support CRAM-MD5, a simple protocol that is immune to replay attacks and allows the server and client to store only an MD5 internal context of the password (128 bit, before finalizing the hash), and not the plain-text password. But it's still an unsalted hash that is fairly easy to break in the age of GPU brute-force crackers.
Not an effective sample of the general population but still likely to be indicative of attitudes held by a particular subset.
"I think people should be free to chose a religious curricula for their children should they so desire, but they shouldn't be allowed to force religious bullcrap onto other people as an equal option to scientific fact"
Way to piss off your voters, miss-nowhere-to-be-seen-in-this-video. The girls are shallow and stupid because they are the product of a selection process orchestrated by the shallow and stupid. The smarter girls don't engage the issue directly to not offend the stupid.
What if the name Dennis was fashionable among the upper-middle class during the years the current generation of dentist was born, leading to a significantly higher propensity for a high-investment, high-income career for the children ? The guys of Freakonomics explained how children names become fashionable among the upper classes, and are then emulated by the lower classes; the upper classes then move to new names as the old names become mundane. Slutty names like Bambi and Brandy were at one point all the rage. Incidentally, they point to a study where identical CVs get 50% less callbacks when they belong to people named Tyrone and Lakisha.
I only had an account at all so people would stop bothering me to make one, and thanks to G+ I don't use it anymore.
This works best if Google+ flops. "Why, my retarded friend, I'd love to add you to my Facebook profile and see your inane drivel on my wall, but unfortunately I use Google+". I love how you don't even need a Google+ account.
Another green fad. PV are only marginally cost effective when they tracking the sun. A PV cell that is engineered to look nice and has 3 hours/day angled sunlight will probably break even when hell freezes over.
In other words by the time you have developed a thorium reactor renewables will have taken away much of the demand
We know how to build advanced nuclear reactors today. If fully committed, they could come online in less than a decade, and be one order of magnitude cheaper than any renewables. What's preventing them is: a. NIMBY-type ecologists and fear-mongers b. Proliferation concerns c. Increasingly, green industry lobby, makeshift "job creation" and other assorted economic fallacies
I don't dispute your conclusion that the free market will chose renewables over nuclear. But that's not because of engineering concerns or risks of an unproven technology. It all boils down to political pressure on the market against nuclear, no one will sink billions into nuclear when there is massive risk that they will not be able to deploy it.
Rolling the transactions back is a huger blow to that interesting experiment, and basically undermines the attempt to get bitcoins accepted as a form of currency.
Trades on the exchange do not impact the Bitcoin blockchain (transaction history) directly, in the exact same way as money is not directly transferred to/from your bank when you trade. Any market event is buffered into the virtual accounts that traders hold with Mt.gox, while the actual bitcoins are in Mt.gox's wallet and the actual dollars are in Mt.gox's bank account. You need to specifically request a transfer to get either money or bitcoins out of the system. So the event is in no way relevant for Bitcoin. It's just a bad case of unsanitized inputs.
The passwords before ID 3000 that were not changed are plain md5 hashes. Almost all are easily cracked. Example: id: 642 name: shlax hash: de434a6e3a01de06657454e07349535c password: pretorian
The ones starting with $ are MD5 crypt passwords. The 1000 MD5 iterations add about 10 bits of apparent entropy, and the salts prevent parallelisation.
The point was that even if you know someone is involved in crime, it's very hard to prove the crime as long as you don't know what it is. It's orders of magnitude harder than, say, raiding a postal address where drug money are funneled. So if criminal operations are marginally effective using cash and regular bank accounts, some become orders of magnitude more effective with cryptocurrency.
The notion of the government watching Bitcoin transactions for which tax is not paid makes little sense, cryptocurrencies are transnational and anonymous. Someone can trade bitcoins for dollars in a jurisdictions that allows it, bring the revenue in US and be 100% legal as long as he pays taxes. There's no way of tracing money to point of exit from the Bitcoin system short of strong-arming every country in the world to report all transactions happening at Bitcoin exchanges.
Slashdot Mirror
Adaptation to longer yellow lights is severely misrepresented in the studies cited above. I have no idea how you could measure that adaptation without lengthening ALL yellows in a certain area, and limit the study ONLY to those driving in the respective area. Otherwise the driver is adjusted to the average duration of a yellow light in his area - of course he will break within a 6 seconds yellow, he expects red after 4 seconds ! So changing the yellow duration in only a couple of intersections in the city is unlikely to trigger adaptation, thus it grossly understates the effectiveness of changing all traffic lights to longer yellows.
The whole idea that longer yellows increase safety is specious. It's effectively claiming that a 4 second period is insufficient for stopping safely when driving at city speeds - and that's clearly bollocks.
The inter-light timing is horrible. In areas where there are several traffic lights in a row, a driver who has to stop at a red light at an intersection can be ceratin of one thing: they will have to stop for the next 4 intersections because those will also turn red, just as they approach. But if they either get a green light or manage to run the first red light, they won't have to stop at any of the next several intersections.
This idea is stupid, and you are a stupid person to bring it up. It's physically impossible for someone to stop at red light, get the green, start the car early in the green cycle, and reach the other intersection just in time to catch the next red, while at the same time someone who arrives late at the same 1st semaphore and barely makes a yellow to somehow catch the green in the next intersection before guy 1.
In other words, assuming both drivers have similar speed, if you can get a string of yellows you can also get a string of greens, and the first red you catch will sync you up to the green. The scenario where one driver makes yellows in a row, while another driver stops at red in a row can't happen on a given road segment at a given speed.
All those patents are useful and unique - right?
Right. According to IP industry insiders, what we need is a Patent Stimulus to end the recession
The nice thing about this Patent Stimulus Plan is that it will cost only a small fraction of the amount of money we have already wasted on failed economic stimulus. What we need to do is have President Obama issue an Executive Order directing the Patent Office to start allowing patents. A 42% allowance rate during the first quarter of 2009 is wholly unacceptable. So while you are at it President Obama, order the Patent Office to issue a patent UNLESS there is a reason to deny it.
Obesity also increased from virtually unknown to over half of the western population. I think we are evolving to a super race of fat and smelly geniuses.
They are just as genuine as the ones generated by people infected with the get-rich-quick-bitcoin-bug, al miners compete for a fixed number of bitcoins.
As it turns out, the "immense power of an infected machine's graphical processing unit" is not that large when up against a bunch of people determined to make it big. The current hash rate of the network is 12Thash/second while a typical infected machine will have a low end graphics card, say 20 MHash/second (a vast majority will not have OpenCL-capable cards, and a tiny minority will have cards capable of 100 - 500 Mhash/sec).
So this virus needs to infect on the order of 500.000 typical machines before it has enough power to compromise the security of the monetary system via double spend. At that time the author would earn about 1 million dollars/month (at current bitcoin prices), so he will have little incentive to destroy the network.
financially liable
But why take money only when they screw up, when you can abuse your market position to ask for the maximum amount of money in the beginning, if they want to be in your root ? You are not making any extra money if they are competent, so why bother?
Ah, the joys of free market.
They are somehow worse than Facebook.
This is a common theme among failing business. They become large bureaucracies hungry for cash and they will cannibalize their customer base in a vain attempt to cling to profitability. Can't wait for a failing Facebook, I need my own Database of Everyone® available now for the low low price of $100.
All those dollar bills in your wallet are only as valuable as we collectively agree they are.
Well then, let's collectively agree they are worth a mansion and a yacht for each citizen, shall we ? As it turns out, the long term value of money is not the result of psychology, rather a simple division of the economic output of the country by the outstanding monetary base. If some guy in Washington keeps printing those bills, they will eventually drop to zero, no collective agreement needed.
But who knows what that balanced economy would even look like, or whether it would even work.
Yup, we should stick to what we know works: spending money that will be paid by the future generation. I can't see any reason why this can't go on for ever.
Who are any of the rating agencies to say how much debt a given country should carry?
They do no such thing. What they speculate about is your ability to pay your existing debt. If you have trouble meeting the monthly mortgage payments, do you think you should get another credit card ? Well, if you can find somebody willing to give you one, sure.
The "2 billion dollar mistake" is not a mistake per se, but an acknowledged choice: the agency based it's calculations on a less optimistic growth scenario then the official one. This is not an actuarial calculation, it's a value judgement that involves macroeconomic speculation for the next decade. The agency does not believe the growth factors are realistic.
I find the "why should we trust an industry that failed to predict the recession and junk MBS" argument specious. It's not we who have to trust them, it's the investors, and they have the undeniable right to act on any information they believe relevant - it's their money after all. Yes, the rating agencies have lost credibility in the MBS debacle. This should suggest that things are really bad if even S&P sees it.
Your rant is ruined by some factual inaccuracies. You seem to lack of understanding of how "domain validation only" certificates work, and how EV certificates came to be. You are presuming that all certificates do the offline identity check, when in fact only EV issuers do that nowadays. Leveraging the DNSSEC tree to do the domain validation is actualy more secure than involving a 3rd party via the classic DNS system.
I've never proposed to make SSL work like Putty, and I clearly stated in my message that accepting self-signed certificates today way will open PayPal to attack. Self signed certificates would become usable, I was saying, only when they could be authenticated via the secure DNS tree, at which point the system would become equivalent to the existing non-EV certs that validate only the domain.
I concede that 15$ is indeed a low price, last time I checked it was something like 70 less inflated dollars, and when I had multiple vhosts on multiple domains it added up quite quickly. But there's also a huge benefit with free: you can automate it as a step in a webserver setup script, making ALL sites secure by default. Since the domain is verified by Google's or Mozilla's automated interface, you would get full end-to-end encryption and authentication without DNSSEC extensions, not just protection against passive eavesdropping .
On a side note, you should understand that accepting a self-signed certificate still protects you from passive eavesdropping, a major attack scenario. A man-the-middle is forced to fake all your SSL traffic with certificates generated by him. If applied on a large scale this can easily be detected by informed users, so a democratically elected government will think twice before doing it.
Another good reason to install HTTPS Everywhere
I would also actually run a HTTPS server everywhere if I didn't have to deal with the certificate mafia, and if major browsers would silently accept self-signed without drowning the user in a storm of "RUN FOREST, RUN !!!" messages. This is currently pretty tricky to do on the browser side without opening PayPal to attack (cache the sites that use real certs ? have a hardcoded master list for first connect ?). But it would be very nice if I could publish a flag in DNSSEC that could say "This is my certificate thumbprint, use it", and leverage the secure DNS tree instead of the insecure and bogus certificate industry.
Why again should I have to fork a pile of cash to obtain a bit string that says that I actually own the domain I'm using ? Generating this bit string seems like a task that could easily be automated to the point of being free. I can understand why Microsoft would be against this (and claim tens of thousand to add you to their root zone), but for example Mozilla or Google could create such an automated certification authority, and add it to their trusted root zone since they know they can trust themselves. Such certificates would work just as the "real thing" on Mozilla or Chrome, but would of course get the usual prompts in Internet Explorer.
That sounds like a grid that is not smart enough; if everybody charges his Chevy at 2AM, then 2AM will be the new peak hour and it will cost an arm and a leg to charge at that time. If the price information is delayed versus the instantaneous power consumption, then yes, a spike should be expected when the the price drops, but this could be countered by distributing the price information with random delays and only in some areas.
I mean, we're posting on a site that seems to handle anonymity just fine.
But I'm not anonymous, I'm Stellian, well known slashdot lurker and kook. I have a nice karma and care about my reputation, so I try my best to behave. See, no childporn or viruses in this post.
If I disclose my real identity complete with full name and postal address:
- it will not improve the quality of my posts; facebook is a perfect example on non-anonymous people incapable, on average, to produce any useful content
- it will not stop other anonymous people to do illegal things, in fact criminals will always try to remain anonymous when operating, just like in the real world
- it will allow an anonymous stock owner of facebook/slashdot/etc. to make a few bucks more by farming my data
No, anonymity is not going anywhere and I will refuse to use any service does not respect my privacy
How exactly one goes about building a nuclear reactor from mail order uranium (presumably depleted) and smoke detectors (about 1 microgram of Americium 241 each) ? The critical mass of Am 241 is over 50 Kg, so he would need 50 million smoke detectors to build a bomb. For a controlled, moderated reaction, much more, maybe hundreds of Kg. The technology to enrich natural uranium up to reactor-grade level is barely in the hands of states.
The fact that someone took him seriously and actually sent a guy with a detector AND a police squad to his house shows just how ridiculously incompetent the regulators are, and how paranoid people get when the word "radiation" is uttered.
Yeah, well good luck with that. If your are approached by a small company in a booming market and you make a great impression at the interview it might fly.
It won't work with a multinational corporation - even if you can convince the person making the hire decision, they have no authority to change the contract, and any such change must go through legal and 3rd party auditors that certify the change does not endanger the company; this process takes months in a large corporation - there's simply to much at stake. A single wrong word and you might be able to sue the corporation for 2.6 billion dollars for your Java. You might say you don't want to work for such a corporation any way, but then again you might not have another option not evolving a steep pay-cut.
It also drives the point home that you are a peculiar person and a non-conformist. A small shop working on the next big-thing ? Sure. A corporation staffed full with bosses who did sign the same contract ? No way. You actually weed yourself out before getting a chance to put your non-conformism to the work and shine in a sea of mediocrity (or make a complete fool out of yourself and have people avoid eye contact in the hallways in fear of being gunned down in your imminent killing spree).
if you are honestly interested in claiming your copyrights this is the best (and arguable only) way to enforce the license.
Assuming you actually own that copyright. 99% of the work contracts out there have a clause where you are ceding all intellectual property to the employer. It's so standard that you should always ask for permission from your employer before writing and releasing open source software - you might not have the right to do so, even if the software is not related to the business of your employer and even if developed in your spare time; the language in my contract is unambiguous about that.
Assuming that in this case the permission to write and release open source software was implicit, it still does not mean the company has lost it's control of it's intellectual property - they can always dual-license it under a proprietary license. They can't "take back" the already released GPL software, and they can't grab any contribution of 3rd parties to that lineage, but they can chose to develop the original codebase in an entirely closed source fashion - it's theirs.
So spending 10 minutes to read your contract might save a butt-load of lawyer fees.
The vast majority of servers support CRAM-MD5, a simple protocol that is immune to replay attacks and allows the server and client to store only an MD5 internal context of the password (128 bit, before finalizing the hash), and not the plain-text password. But it's still an unsalted hash that is fairly easy to break in the age of GPU brute-force crackers.
Not an effective sample of the general population but still likely to be indicative of attitudes held by a particular subset.
"I think people should be free to chose a religious curricula for their children should they so desire, but they shouldn't be allowed to force religious bullcrap onto other people as an equal option to scientific fact"
Way to piss off your voters, miss-nowhere-to-be-seen-in-this-video. The girls are shallow and stupid because they are the product of a selection process orchestrated by the shallow and stupid. The smarter girls don't engage the issue directly to not offend the stupid.
What if the name Dennis was fashionable among the upper-middle class during the years the current generation of dentist was born, leading to a significantly higher propensity for a high-investment, high-income career for the children ?
The guys of Freakonomics explained how children names become fashionable among the upper classes, and are then emulated by the lower classes; the upper classes then move to new names as the old names become mundane. Slutty names like Bambi and Brandy were at one point all the rage. Incidentally, they point to a study where identical CVs get 50% less callbacks when they belong to people named Tyrone and Lakisha.
I only had an account at all so people would stop bothering me to make one, and thanks to G+ I don't use it anymore.
This works best if Google+ flops. "Why, my retarded friend, I'd love to add you to my Facebook profile and see your inane drivel on my wall, but unfortunately I use Google+". I love how you don't even need a Google+ account.
Another green fad. PV are only marginally cost effective when they tracking the sun. A PV cell that is engineered to look nice and has 3 hours/day angled sunlight will probably break even when hell freezes over.
In other words by the time you have developed a thorium reactor renewables will have taken away much of the demand
We know how to build advanced nuclear reactors today. If fully committed, they could come online in less than a decade, and be one order of magnitude cheaper than any renewables. What's preventing them is:
a. NIMBY-type ecologists and fear-mongers
b. Proliferation concerns
c. Increasingly, green industry lobby, makeshift "job creation" and other assorted economic fallacies
I don't dispute your conclusion that the free market will chose renewables over nuclear. But that's not because of engineering concerns or risks of an unproven technology. It all boils down to political pressure on the market against nuclear, no one will sink billions into nuclear when there is massive risk that they will not be able to deploy it.
Rolling the transactions back is a huger blow to that interesting experiment, and basically undermines the attempt to get bitcoins accepted as a form of currency.
Trades on the exchange do not impact the Bitcoin blockchain (transaction history) directly, in the exact same way as money is not directly transferred to/from your bank when you trade. Any market event is buffered into the virtual accounts that traders hold with Mt.gox, while the actual bitcoins are in Mt.gox's wallet and the actual dollars are in Mt.gox's bank account. You need to specifically request a transfer to get either money or bitcoins out of the system.
So the event is in no way relevant for Bitcoin. It's just a bad case of unsanitized inputs.
Here's the leaked account list: http://bit.ly/kE3Q4D
The passwords before ID 3000 that were not changed are plain md5 hashes. Almost all are easily cracked. Example:
id: 642
name: shlax
hash: de434a6e3a01de06657454e07349535c
password: pretorian
The ones starting with $ are MD5 crypt passwords. The 1000 MD5 iterations add about 10 bits of apparent entropy, and the salts prevent parallelisation.
The point was that even if you know someone is involved in crime, it's very hard to prove the crime as long as you don't know what it is. It's orders of magnitude harder than, say, raiding a postal address where drug money are funneled. So if criminal operations are marginally effective using cash and regular bank accounts, some become orders of magnitude more effective with cryptocurrency.
The notion of the government watching Bitcoin transactions for which tax is not paid makes little sense, cryptocurrencies are transnational and anonymous. Someone can trade bitcoins for dollars in a jurisdictions that allows it, bring the revenue in US and be 100% legal as long as he pays taxes. There's no way of tracing money to point of exit from the Bitcoin system short of strong-arming every country in the world to report all transactions happening at Bitcoin exchanges.