Except that in normal systems the size of the money supply adjusts to the size of the economy. You do some work, produce something of value, and then the reserve banking system uses this to construct a corresponding amount of money (and then, 60 or so years later, people say 'OMG! Money is created from debt!' as if this is some awesome revelation and not simply the way that you move assets into the system where value is tracked by money). With Bitcoin, there is no correlation between the amount of Bitcoin and the size of the economy and so the system is inherently unstable, just as gold-based currencies were.
Nonsense. There are lots of competent people all across the world. Unless you need a hundred or so people in the same office, there's no reason to be offering California salaries. The FreeBSD Foundation has contracted a lot of work to people in the Ukraine this year for that reason: they're competent, but not too expensive.
And you're probably not alone. I tend to read papers on whatever the default PDF reader for the platform I'm on is, but a lot of my colleagues have preferred tools that track the PDF and citation metadata for papers and index their annotations in searchable form. They sync the data between computer and tablet. If you have a custom proprietary format, then it won't integrate with any of this, so your article won't be found (if it's read at all) by people who search their local repositories of papers that they've read to try to find the thing that they remember reading six months ago.
Are they? I was under the impression that most stuff that they publish is the pop-science puff piece that's supposed to encourage interested readers to go to the real articles. I've never cited an article from either of them and wouldn't think to look at them unless someone sends me a link, whereas I'll skim the top tier journals and conference proceedings in my field regularly to keep up to date.
Linux? What about Android, Kindle, and iOS? A lot of academics read papers on tablets now. If your platform doesn't support all of those (including integrating with the apps that people use for bibliography management and annotation on these devices) then it's dead in the water.
No, they mean being able to effectively collaborate. There's a limit to how much someone can achieve on their own and it's far more valuable to have someone who improves everyone's productivity than someone that drains a little bit from everyone who has to work with them by not communicating what they're doing, or even what they've done, causing duplicated work and wasted effort.
Social skills are learned just like your coding skills are. You may never be a world-class entertainer and storyteller, but being able to meet someone, shake their hand, and engage in a few minutes of small talk and crack a joke or two is a learnable skill for anybody who decides to try.
Neil deGrasse Tyson made a great point in a (relatively old) video someone posted here last week: You can always become better at something. You might not ever become great at it, but if you practice then you can always become better.
a self signed cert does *not* protect your connection from anything, unless the client already knows what to look for to ensure the cert they have is the cert you intended them to use
Yes it does, it protects you against passive adversaries. Compromising an SSL connection with a self-signed certificate requires an active adversary (i.e. one who will modify traffic, not just sniff it). This is still possible with a signed cert if you're sufficiently large as the trust model for SSL (in the absence of certificate transparency, which isn't yet widely deployed) means that if any registrar is compromised (e.g. the one owned by the Turkish intelligence agency that all major browsers trust) then they can sign certificates for any domain.
A self-signed cert that is silently accept it is much much worse than no SSL at all, because it allows the user to make assumptions about their use of the website which are absolutely not true
No, displaying a user interface element indicating the site is secure when it only has a self-signed certificate is worse than no SSL. Rendering self-signed SSL certs in exactly the same way as unencrypted connections (as I suggested) is better, because it allows people to roll out SSL cheaply and makes the world no worse, just raises the costs for interception.
Even initial versions were actually more heavyweight and leaked more memory than mozilla suite
That's not quite what I remember. Phoenix (then Firebird, then Firefox) and Thunderbird (or whatever it was called back then) between them used more memory than the Mozilla suite, but Firefox was lighter than using the Mozilla suite and just the browser. The big appeal of separating the two was that the browser was a buggy piece of crap and every time it crashed it took out everything else sharing the XUL runtime, including the mail client (which then had to spend time recovering corrupted databases on next launch). With Firefox, only the browser crashed and restarted quite quickly. Given that the browser crashed at least once an hour back then, it was a bit advantage. No one cared about memory leaks, because the browser didn't stay up long enough for them to become apparent. It was only after they fixed the stability issues that people started noticing.
No. A self-signed SSL certificate is no worse than no SSL. The correct thing for browsers to do when encountering a self-signed SSL certificate is accept it silently, but not display any of the UI elements indicating that the site is secure (and to have a prominent red {insecure} label for every site that doesn't have the padlock). A self-signed cert protects your connection from passive adversaries and, as we have learned recently, that's a very important threat model to care about. Making it appear to users that a self-signed SSL cert is less secure than an unencrypted connection is not helpful.
It's amazing how many Americans seem not to understand the difference between direct democracy and representative democracy, nor the fact that democracy and republic are orthogonal concepts. You post makes as much sense as saying emacs isn't a text editor, it's a religion - just because it is one, doesn't mean that it isn't the other.
They can be a lot faster if you live somewhere with dedicated bus lanes. The bus is starting and stopping a lot, but in between stops it can actually move, whereas the cars are stuck in traffic jams...
Which means that minibuses make sense. 20-30 seats will seat those 9 comfortably, but still handle the other side of the bell curve nicely (don't forget, if 9 is the average, that doesn't mean that they only have 9 at peak times). The driver is a big part of the cost and the reason that they're less common. Add some more intelligent routing and the ability to just specify where you want to get on and off, and have the system readjust the routs to accommodate you.
Being able to write with a pen and paper is a sign of intelligence and education
The same can be said for being able to sight-read music, compose poetry, or read Latin, and I suspect that all three are about as good indicators of job performance as carrying a pen and being good at handwriting.
If writing down the equations is anywhere near a bottleneck, then you're probably wasting your time writing them down at all. If they're in any way difficult, then you probably want to have them in a form that's machine readable and the time spent with computer-assisted evaluation / theorem proving will vastly outweigh any inefficiency in the input method.
The reason to learn and use cursive is because it's much more efficient in terms of writing large amounts.
And that's the reason why there's little point in teaching it anymore. I spent years at school being forced to do writing exercises with a pen. At age 14, when we were allowed to submit type-written essays, my average grade in English jumped from a C to an A. After a year of typing regularly, I was faster typing than I was at writing after 12 years of practicing. I've now had four books published and type more in a typical day than I write with a pen in a year.
The algorithm is dumb. Explaining it is smart. We had to do thousands of long division problems when I was a kid, and it was utterly useless after, say, the first hundred. Boring as hell. Math should be fun.
Exactly. At school, I had to solve huge numbers of differential equations. The useful learning outcomes were recognising the kind of problem that a differential equation can solve, being able to express problems in that form, and understanding (at least vaguely) the steps that are involved in solving them. Then there was the practice, which (after months of this repetition) meant that I went from taking about half an hour to solve a differential equation to taking about 10 minutes. Meanwhile, a computer can solve them in about the time that it takes to type the problem in.
For simple arithmetic, it is useful to be able to solve things in less time than it takes to pull out a calculator, but other than that the benefits of practicing an obsolete skill are negligible.
Bankruptcy technically means that your debts are worth more than your assets. Being declared bankrupt means that someone has called you on it, forcing you to repay loans that you don't have the capital for (you can get away with being technically bankrupt for a while if your creditors have enough faith that your income will increase to cover the difference). If he bought the medallions with 20% his own money and 80% borrowed money, then he'll be close to bankruptcy now. If his lender decides that they want their money back, he can't sell the medallions for enough to cover their 80%, so he'd have a problem. If he's running 100 taxis and making a reasonable profit from them, then it may be that his company will keep existing, but with some difficulties with future investment (it's difficult to get more loans when the value of your assets is negative). I'm not sure what the law is in the US in general and NY in particular, but the company may also need to file for one of the many forms of bankruptcy when its finances are audited at the end of the year (some forms just say 'technically we have negative assets, but we're doing okay as an operating concern and the value of our assets will increase, so there are some restrictions on what we can do, but the company can keep operating').
That's exactly what Uber is doing - building a worldwide taxi dispatcher service that can be this 'undifferentiated platform' that you want for routing driverless taxis.
Slashdot Mirror
Individuals aren't taxes based on their profit but income
Yes they are. You can (as an individual in the UK) offset most business expenses against tax. It's a tax on net income, not gross.
Can't you mount them south facing with fish-eye lenses on them so that they get a good amount of the sunlight when it's coming from any direction?
As you yourself observed this "normal" system is only a very recent "innovation".
Yes, an innovation by learning from past mistakes.
Which makes bitcoin more or a throw-back to the "old normal" rather than anything new.
Indeed. Innovation involves making new mistakes, not repeating old ones.
How are they producing iOS apps? Can Shockwave output iOS apps now?
Except that in normal systems the size of the money supply adjusts to the size of the economy. You do some work, produce something of value, and then the reserve banking system uses this to construct a corresponding amount of money (and then, 60 or so years later, people say 'OMG! Money is created from debt!' as if this is some awesome revelation and not simply the way that you move assets into the system where value is tracked by money). With Bitcoin, there is no correlation between the amount of Bitcoin and the size of the economy and so the system is inherently unstable, just as gold-based currencies were.
Nonsense. There are lots of competent people all across the world. Unless you need a hundred or so people in the same office, there's no reason to be offering California salaries. The FreeBSD Foundation has contracted a lot of work to people in the Ukraine this year for that reason: they're competent, but not too expensive.
And you're probably not alone. I tend to read papers on whatever the default PDF reader for the platform I'm on is, but a lot of my colleagues have preferred tools that track the PDF and citation metadata for papers and index their annotations in searchable form. They sync the data between computer and tablet. If you have a custom proprietary format, then it won't integrate with any of this, so your article won't be found (if it's read at all) by people who search their local repositories of papers that they've read to try to find the thing that they remember reading six months ago.
Are they? I was under the impression that most stuff that they publish is the pop-science puff piece that's supposed to encourage interested readers to go to the real articles. I've never cited an article from either of them and wouldn't think to look at them unless someone sends me a link, whereas I'll skim the top tier journals and conference proceedings in my field regularly to keep up to date.
Linux? What about Android, Kindle, and iOS? A lot of academics read papers on tablets now. If your platform doesn't support all of those (including integrating with the apps that people use for bibliography management and annotation on these devices) then it's dead in the water.
Sounds like a good reason for money-conscious non-profits to hire people outside of California...
No, they mean being able to effectively collaborate. There's a limit to how much someone can achieve on their own and it's far more valuable to have someone who improves everyone's productivity than someone that drains a little bit from everyone who has to work with them by not communicating what they're doing, or even what they've done, causing duplicated work and wasted effort.
Social skills are learned just like your coding skills are. You may never be a world-class entertainer and storyteller, but being able to meet someone, shake their hand, and engage in a few minutes of small talk and crack a joke or two is a learnable skill for anybody who decides to try.
Neil deGrasse Tyson made a great point in a (relatively old) video someone posted here last week: You can always become better at something. You might not ever become great at it, but if you practice then you can always become better.
And Apple licenses that trademark from Cisco. They also license the trademark on iPhone from Cisco, who previously had a SIP phone with the same name.
a self signed cert does *not* protect your connection from anything, unless the client already knows what to look for to ensure the cert they have is the cert you intended them to use
Yes it does, it protects you against passive adversaries. Compromising an SSL connection with a self-signed certificate requires an active adversary (i.e. one who will modify traffic, not just sniff it). This is still possible with a signed cert if you're sufficiently large as the trust model for SSL (in the absence of certificate transparency, which isn't yet widely deployed) means that if any registrar is compromised (e.g. the one owned by the Turkish intelligence agency that all major browsers trust) then they can sign certificates for any domain.
A self-signed cert that is silently accept it is much much worse than no SSL at all, because it allows the user to make assumptions about their use of the website which are absolutely not true
No, displaying a user interface element indicating the site is secure when it only has a self-signed certificate is worse than no SSL. Rendering self-signed SSL certs in exactly the same way as unencrypted connections (as I suggested) is better, because it allows people to roll out SSL cheaply and makes the world no worse, just raises the costs for interception.
Even initial versions were actually more heavyweight and leaked more memory than mozilla suite
That's not quite what I remember. Phoenix (then Firebird, then Firefox) and Thunderbird (or whatever it was called back then) between them used more memory than the Mozilla suite, but Firefox was lighter than using the Mozilla suite and just the browser. The big appeal of separating the two was that the browser was a buggy piece of crap and every time it crashed it took out everything else sharing the XUL runtime, including the mail client (which then had to spend time recovering corrupted databases on next launch). With Firefox, only the browser crashed and restarted quite quickly. Given that the browser crashed at least once an hour back then, it was a bit advantage. No one cared about memory leaks, because the browser didn't stay up long enough for them to become apparent. It was only after they fixed the stability issues that people started noticing.
No. A self-signed SSL certificate is no worse than no SSL. The correct thing for browsers to do when encountering a self-signed SSL certificate is accept it silently, but not display any of the UI elements indicating that the site is secure (and to have a prominent red {insecure} label for every site that doesn't have the padlock). A self-signed cert protects your connection from passive adversaries and, as we have learned recently, that's a very important threat model to care about. Making it appear to users that a self-signed SSL cert is less secure than an unencrypted connection is not helpful.
It's amazing how many Americans seem not to understand the difference between direct democracy and representative democracy, nor the fact that democracy and republic are orthogonal concepts. You post makes as much sense as saying emacs isn't a text editor, it's a religion - just because it is one, doesn't mean that it isn't the other.
They can be a lot faster if you live somewhere with dedicated bus lanes. The bus is starting and stopping a lot, but in between stops it can actually move, whereas the cars are stuck in traffic jams...
Which means that minibuses make sense. 20-30 seats will seat those 9 comfortably, but still handle the other side of the bell curve nicely (don't forget, if 9 is the average, that doesn't mean that they only have 9 at peak times). The driver is a big part of the cost and the reason that they're less common. Add some more intelligent routing and the ability to just specify where you want to get on and off, and have the system readjust the routs to accommodate you.
Being able to write with a pen and paper is a sign of intelligence and education
The same can be said for being able to sight-read music, compose poetry, or read Latin, and I suspect that all three are about as good indicators of job performance as carrying a pen and being good at handwriting.
If writing down the equations is anywhere near a bottleneck, then you're probably wasting your time writing them down at all. If they're in any way difficult, then you probably want to have them in a form that's machine readable and the time spent with computer-assisted evaluation / theorem proving will vastly outweigh any inefficiency in the input method.
The reason to learn and use cursive is because it's much more efficient in terms of writing large amounts.
And that's the reason why there's little point in teaching it anymore. I spent years at school being forced to do writing exercises with a pen. At age 14, when we were allowed to submit type-written essays, my average grade in English jumped from a C to an A. After a year of typing regularly, I was faster typing than I was at writing after 12 years of practicing. I've now had four books published and type more in a typical day than I write with a pen in a year.
The algorithm is dumb. Explaining it is smart. We had to do thousands of long division problems when I was a kid, and it was utterly useless after, say, the first hundred. Boring as hell. Math should be fun.
Exactly. At school, I had to solve huge numbers of differential equations. The useful learning outcomes were recognising the kind of problem that a differential equation can solve, being able to express problems in that form, and understanding (at least vaguely) the steps that are involved in solving them. Then there was the practice, which (after months of this repetition) meant that I went from taking about half an hour to solve a differential equation to taking about 10 minutes. Meanwhile, a computer can solve them in about the time that it takes to type the problem in.
For simple arithmetic, it is useful to be able to solve things in less time than it takes to pull out a calculator, but other than that the benefits of practicing an obsolete skill are negligible.
Bankruptcy technically means that your debts are worth more than your assets. Being declared bankrupt means that someone has called you on it, forcing you to repay loans that you don't have the capital for (you can get away with being technically bankrupt for a while if your creditors have enough faith that your income will increase to cover the difference). If he bought the medallions with 20% his own money and 80% borrowed money, then he'll be close to bankruptcy now. If his lender decides that they want their money back, he can't sell the medallions for enough to cover their 80%, so he'd have a problem. If he's running 100 taxis and making a reasonable profit from them, then it may be that his company will keep existing, but with some difficulties with future investment (it's difficult to get more loans when the value of your assets is negative). I'm not sure what the law is in the US in general and NY in particular, but the company may also need to file for one of the many forms of bankruptcy when its finances are audited at the end of the year (some forms just say 'technically we have negative assets, but we're doing okay as an operating concern and the value of our assets will increase, so there are some restrictions on what we can do, but the company can keep operating').
That's exactly what Uber is doing - building a worldwide taxi dispatcher service that can be this 'undifferentiated platform' that you want for routing driverless taxis.