That would just force it underground by disincentivizing victims from reporting the crime, and make it even harder for law enforcement to catch the crooks. Not every problem is a nail that needs to be hammered.
extortion, only encourages these criminals to go after more lucrative targets
That is not necessarily a bad thing. Many security breaches result in spam botnets or customer data leaks, that harm people not that were not responsible. The nice thing about ransomware is that the cost of bad security lands directly in the lap of the people that can actually do something about it.
I don't see how Yahoo has $39 billion in market cap.
Yahoo was an early investor in Alibaba, and owns about 15% of Alibaba's stock. If you subtract out the value of that stock, the rest of Yahoo actually had negative value prior to Verizon's offer.
I do. They are a great site for throwaway email accounts, so I can make one-time-use of sites that insist that I "create an account". Of the billion compromised accounts, I suspect that only a small percentage are currently used by real people.
Elimination of cash means that VISA and MASTERCARD know everything you're doing, and take 2% off the top of every transaction.
Cashless transactions in India became widespread when a rice distribution scheme was replaced by direct payments to debit cards issued to the poor. Under the old system, about 80% of the rice was stolen before it reached the final recipients. Compared to that, 2% is nothing.
AM has publicly admitted that most of the women on the site are fake. The fine print in their TOS says they can engage their users with artificial profiles for "entertainment purposes". Anyone wasting their time on AM is a fool, which is further demonstrated by the fact that many of them were stupid enough to use their real names and contact info.
Sadly, in the UK the law in this area is anything but straightforward.
In America, the IRS publishes a list of criteria, but none of them are specifically necessary or sufficient to determine if someone is an employee or a contractor. The determination is usually ad hoc, with a lot of discretion left to employers, IRS agents, and judges. Anyone claiming that contractor law is "straightforward" has no idea what they are talking about.
Most drivers do NOT rely on Uber to make a living. The majority do it part time, and it is not the main source of income for their household.
... minimum wage, sick pay and paid holidays.
Uber drivers, on average, make about $19/hour in America. That is more than double the minimum wage. They can take a day off any time they like. Those days are unpaid, of course, but "sick days" and vacation are not free in a regular job either: the cost of benefits is just incorporated into lower pay spread over the year.
traditional taxi drivers have these rights.
TFA is about the UK, but at least in America, most taxi drivers do NOT have these rights. Most taxi drivers are either contractors or fully self-employed, and either own their taxi/medallion or lease it on a per-shift basis.
I can probably see a hundred or so from my house in the UK. Is Amercia really so far behind with renewables?
Nope. America is ahead of Britain in both total capacity and per capita wind power generation. Texas alone has more installed wind capacity than all of the UK. However, China has us both beat in total capacity, and Denmark has us both beat in per capita generation.
Israel and Korea spend the most on R&D, at about 4% of GDP each. But America is still in the top ten. China's R&D spending is rising the fastest. Countries in Latin America and Eastern Europe tend to invest the least.
That may have been illegal, but I don't see how it demonstrates technical incompetence. Since there is no evidence it was hacked, I would say it demonstrates the opposite.
the huge thing that everyone should know is never to use an email to log into an account.
Technology is failing if it falls on individuals to remember rules like this. They won't. Instead, services like Gmail, Yahoo, etc. should detect when emails contain fake links to login pages for email accounts or financial institutions, and warn the user that they are about to do something dumb.
This is just CYA bullshit designed to make them look less incompetent.
I am confused. Up till now, I thought they were the victims of sophisticated Russian ex-KGB agents using quantum cryptanalysis. But it turns out they fell for a common phishing scam written by some script kiddie. How does this make them look less incompetent?
I work in government IT. No one is worried about Trump. No one.
The last president that promised to cut government and eliminate entire departments was Ronald Regan. He eliminated zero departments and greatly expanded the size of government. The only post-war president to cut the size of government, as a percentage of GDP, was Bill Clinton. But he had it easy after the bloat of the Reagan years. Conservatives didn't love Reagan because of what he did, but because of what he said. Same for liberals and Bill Clinton.
The lesson Donald should learn from this is that what he says means a lot more than what he does. Talk big, do little, blame others, be popular. If he sticks to this formula, he should coast to reelection in 2020.
If the institution has turned against what you believe is right, then the odds of making any positive change "from the inside" are extremely low.
I don't believe this is true. Bureaucracies have tremendous resilience, and are notorious for their institutional resistance to change. For once, this force may be used for good.
Bureaucracies also have a long history of turning "reformers" into allies of the status quo as they "go native". It is easy to advocate destruction of, say, the EPA, but once you are the chief of the EPA, then is is your organization, and there is a natural desire to defend your turf. The people you wanted to fire are now real people sitting in your office and doing your bidding.
Nope. Back in the 19th century WAY more people worked in retail, because customers were not allowed near the merchandise. A customer would enter the store, walk up to the counter, and give a clerk their list of items. The clerk would then go into the storage room and retrieve the items, bring them to the counter, and tally the bill.
This changed when "self-shopping" was introduced in the 1880s by F. W. Woolworth. This revolutionized retail, and created a whole new pastime of recreational shopping without an apriori shopping list.
So, no, the farmers did not become store clerks, because the clerks were getting laid off at the same time. It is also silly to refuse to use self-checkout because "that's the employee's job" when fetching the merchandise used to be their job too. Times change.
...What? It's analog. It's got precision going down into the quantum scale...
That is not true in any meaningful sense. If you give the same inputs to the same biological neuron, there is no way that you are going to get the same output down to the planck scale. In fact, it is unlikely that you are even going to get 8 bit precision (an output difference of 1/256th).
Sure, but it doesn't matter. Backprop, learning rate, denoising, etc. all just heuristics anyway. So what if your mantissa is off by one bit? You get better accuracy by going wider, adding layers, and (most importantly) using more data. But you can't afford to do that if half your bandwidth is sucked up transmitting meaningless precision.
Also, do you have a good citation that FP16 neural networks are, overall, more effective than FP32 networks, as you've described?
They are not necessarily more effective, just more efficient. If you have infinite resources, you might even get better results using FP32. But resources are never infinite. Here is a guy who claims that even 8 bits is enough for deep NNs.
Once you start talking about a deep learning network the updates to individual perceptrons can be very small and 32 bits are needed.
You can get the same flexibility and more just by going wider and deeper. The bottleneck for NNs is not the math ops, but getting data in and out of the GPU. By using FP16, you cut the per-neuron data in half.
So they're all excited about the lowest-precision, smallest-size floating point math in IEEE 754?
FP16 is good enough for neural nets. Do you really think the output voltage of a biological neurons has 32 bits of precision and range? For any given speed, FP16 allows you to run NNs that are wider and deeper, and/or to use bigger datasets That is way more important than the precision of individual operations.
I was thinking Slashdot would be the crowd that I wouldn't have to add the sarcasm tag (/s) but it appears a few people took it literally.
Many Aspies have difficulty understanding understanding sarcasm. We take everything literally. Slashdot tends to have more "whooshes" than other online forums.
Sure, he's flipped on a lot of issues, but opposition to free trade, along with restricting immigration, were his two big signature issues. If he flips on those, then he stands for absolutely nothing.
Disclaimer: As someone who didn't vote for him, I have been happy to see all his flipping to positions that actually make sense, and I would be even happier if he flips on these as well. But his supporters would rightfully feel betrayed.
But the ROI is more than zero, and increases over time.
That is not necessarily true, unless you assume your cost of capital is zero, your investment will last forever, and require no maintenance. But if you assume that, then there is never a bad investment.
In the real world, any investment has to have returns sufficient to cover interest, depreciation, and on-going expenses such as maintenance and repairs. And even then, it needs to be better than alternative investments.
Slashdot Mirror
Paying ransom should be a felony.
That would just force it underground by disincentivizing victims from reporting the crime, and make it even harder for law enforcement to catch the crooks. Not every problem is a nail that needs to be hammered.
extortion, only encourages these criminals to go after more lucrative targets
That is not necessarily a bad thing. Many security breaches result in spam botnets or customer data leaks, that harm people not that were not responsible. The nice thing about ransomware is that the cost of bad security lands directly in the lap of the people that can actually do something about it.
I don't see how Yahoo has $39 billion in market cap.
Yahoo was an early investor in Alibaba, and owns about 15% of Alibaba's stock. If you subtract out the value of that stock, the rest of Yahoo actually had negative value prior to Verizon's offer.
I see no reason for them to exist.
I do. They are a great site for throwaway email accounts, so I can make one-time-use of sites that insist that I "create an account". Of the billion compromised accounts, I suspect that only a small percentage are currently used by real people.
Elimination of cash means that VISA and MASTERCARD know everything you're doing, and take 2% off the top of every transaction.
Cashless transactions in India became widespread when a rice distribution scheme was replaced by direct payments to debit cards issued to the poor. Under the old system, about 80% of the rice was stolen before it reached the final recipients. Compared to that, 2% is nothing.
Or a pro. But most probably a chatbot.
AM has publicly admitted that most of the women on the site are fake. The fine print in their TOS says they can engage their users with artificial profiles for "entertainment purposes". Anyone wasting their time on AM is a fool, which is further demonstrated by the fact that many of them were stupid enough to use their real names and contact info.
Sadly, in the UK the law in this area is anything but straightforward.
In America, the IRS publishes a list of criteria, but none of them are specifically necessary or sufficient to determine if someone is an employee or a contractor. The determination is usually ad hoc, with a lot of discretion left to employers, IRS agents, and judges. Anyone claiming that contractor law is "straightforward" has no idea what they are talking about.
But if drivers rely on Uber to make a living ...
Most drivers do NOT rely on Uber to make a living. The majority do it part time, and it is not the main source of income for their household.
... minimum wage, sick pay and paid holidays.
Uber drivers, on average, make about $19/hour in America. That is more than double the minimum wage. They can take a day off any time they like. Those days are unpaid, of course, but "sick days" and vacation are not free in a regular job either: the cost of benefits is just incorporated into lower pay spread over the year.
traditional taxi drivers have these rights.
TFA is about the UK, but at least in America, most taxi drivers do NOT have these rights. Most taxi drivers are either contractors or fully self-employed, and either own their taxi/medallion or lease it on a per-shift basis.
I can probably see a hundred or so from my house in the UK. Is Amercia really so far behind with renewables?
Nope. America is ahead of Britain in both total capacity and per capita wind power generation. Texas alone has more installed wind capacity than all of the UK. However, China has us both beat in total capacity, and Denmark has us both beat in per capita generation.
The R&D starvation is pandemic in the US.
Israel and Korea spend the most on R&D, at about 4% of GDP each. But America is still in the top ten. China's R&D spending is rising the fastest. Countries in Latin America and Eastern Europe tend to invest the least.
Podesta used G-fucking-mail...
What's wrong with Gmail?
HRC used a homebrew server for convienience...
That may have been illegal, but I don't see how it demonstrates technical incompetence. Since there is no evidence it was hacked, I would say it demonstrates the opposite.
the huge thing that everyone should know is never to use an email to log into an account.
Technology is failing if it falls on individuals to remember rules like this. They won't. Instead, services like Gmail, Yahoo, etc. should detect when emails contain fake links to login pages for email accounts or financial institutions, and warn the user that they are about to do something dumb.
This is just CYA bullshit designed to make them look less incompetent.
I am confused. Up till now, I thought they were the victims of sophisticated Russian ex-KGB agents using quantum cryptanalysis. But it turns out they fell for a common phishing scam written by some script kiddie. How does this make them look less incompetent?
Who talks like that anyway?
Lawyers. If you are paid to obfuscate, it eventually becomes second nature, and you fail to communicate clearly even to your friends and family.
I work in government IT. No one is worried about Trump. No one.
The last president that promised to cut government and eliminate entire departments was Ronald Regan. He eliminated zero departments and greatly expanded the size of government. The only post-war president to cut the size of government, as a percentage of GDP, was Bill Clinton. But he had it easy after the bloat of the Reagan years. Conservatives didn't love Reagan because of what he did, but because of what he said. Same for liberals and Bill Clinton.
The lesson Donald should learn from this is that what he says means a lot more than what he does. Talk big, do little, blame others, be popular. If he sticks to this formula, he should coast to reelection in 2020.
If the institution has turned against what you believe is right, then the odds of making any positive change "from the inside" are extremely low.
I don't believe this is true. Bureaucracies have tremendous resilience, and are notorious for their institutional resistance to change. For once, this force may be used for good.
Bureaucracies also have a long history of turning "reformers" into allies of the status quo as they "go native". It is easy to advocate destruction of, say, the EPA, but once you are the chief of the EPA, then is is your organization, and there is a natural desire to defend your turf. The people you wanted to fire are now real people sitting in your office and doing your bidding.
They check out groceries in convenience stores.
Nope. Back in the 19th century WAY more people worked in retail, because customers were not allowed near the merchandise. A customer would enter the store, walk up to the counter, and give a clerk their list of items. The clerk would then go into the storage room and retrieve the items, bring them to the counter, and tally the bill.
This changed when "self-shopping" was introduced in the 1880s by F. W. Woolworth. This revolutionized retail, and created a whole new pastime of recreational shopping without an apriori shopping list.
So, no, the farmers did not become store clerks, because the clerks were getting laid off at the same time. It is also silly to refuse to use self-checkout because "that's the employee's job" when fetching the merchandise used to be their job too. Times change.
...What? It's analog. It's got precision going down into the quantum scale...
That is not true in any meaningful sense. If you give the same inputs to the same biological neuron, there is no way that you are going to get the same output down to the planck scale. In fact, it is unlikely that you are even going to get 8 bit precision (an output difference of 1/256th).
There's a lot of rounding error with FP16.
Sure, but it doesn't matter. Backprop, learning rate, denoising, etc. all just heuristics anyway. So what if your mantissa is off by one bit? You get better accuracy by going wider, adding layers, and (most importantly) using more data. But you can't afford to do that if half your bandwidth is sucked up transmitting meaningless precision.
Also, do you have a good citation that FP16 neural networks are, overall, more effective than FP32 networks, as you've described?
They are not necessarily more effective, just more efficient. If you have infinite resources, you might even get better results using FP32. But resources are never infinite. Here is a guy who claims that even 8 bits is enough for deep NNs.
More pragmatically, they are a network of perceptrons with sigmoidal output functions.
Today, most bleeding edge NNs use rectified linear activation functions. Sigmoids are soooo 2014.
Once you start talking about a deep learning network the updates to individual perceptrons can be very small and 32 bits are needed.
You can get the same flexibility and more just by going wider and deeper. The bottleneck for NNs is not the math ops, but getting data in and out of the GPU. By using FP16, you cut the per-neuron data in half.
So they're all excited about the lowest-precision, smallest-size floating point math in IEEE 754?
FP16 is good enough for neural nets. Do you really think the output voltage of a biological neurons has 32 bits of precision and range? For any given speed, FP16 allows you to run NNs that are wider and deeper, and/or to use bigger datasets That is way more important than the precision of individual operations.
I was thinking Slashdot would be the crowd that I wouldn't have to add the sarcasm tag (/s) but it appears a few people took it literally.
Many Aspies have difficulty understanding understanding sarcasm. We take everything literally. Slashdot tends to have more "whooshes" than other online forums.
He's flip-flopped on all kinds of stuff already
Sure, he's flipped on a lot of issues, but opposition to free trade, along with restricting immigration, were his two big signature issues. If he flips on those, then he stands for absolutely nothing.
Disclaimer: As someone who didn't vote for him, I have been happy to see all his flipping to positions that actually make sense, and I would be even happier if he flips on these as well. But his supporters would rightfully feel betrayed.
But the ROI is more than zero, and increases over time.
That is not necessarily true, unless you assume your cost of capital is zero, your investment will last forever, and require no maintenance. But if you assume that, then there is never a bad investment.
In the real world, any investment has to have returns sufficient to cover interest, depreciation, and on-going expenses such as maintenance and repairs. And even then, it needs to be better than alternative investments.
If you're going to automate a $15/hr job, then why wouldn't you automate a $6/hr job?
Because the return on investment is less than half as much.
If you ever decide to start a business, you should partner with someone that can do math.