Why is it society's responsibility to teach you job skills?
Because long-term unemployment is a societal burden, not just an individual one? And it's a missed economic opportunity for society as well as the individual?
It is a shared responsibility because mismatches between worker skills and opportunities is a shared economic burden.
When you subdivide a problem, each core works on a smaller subset. If those subsets fit into a cache that the bigger problem didn't, you can easily get superlinear increase as a result. In many cases you could actually rewrite the bigger problem to be more cache-friendly and get a similar speedup, so you generally don't make much of such "extra" performance increases.
never having to worry about being in a car accident and thusly never having to pay for collision insurance will save you a lot of money.
If you don't have or can get the money to actually buy the thing in the first place, it doesn't matter if it saves you money in the long run. This is part reason why it tends to cost more to be poor than to be rich.
One one hand, revitalizing city centers is not necessarily a bad thing. On the other, this starts to smell a little of Shadowrun-style megacorporations (or of industrial-era company towns).
Live and work your entire life within the protective confines of your employer. Go to the company school, work at the company office, live in company housing paid for with a company-bank supplied mortgage, dine at your choice of company restaurants, vacation at the company resort, get a company funeral...
And the end result is another Android phone, except with small compatibility issues and without the actual app store. You'd be left with the worst of two words: Users would rather get a proper Android device with all the apps; and developers would rather develop for the billions of people using the Android ecosystem and would not bother rewriting and submitting their stuff to MS own app store variant.
Well, duh. But F77 is obsolete and MPI is now only used for distributed systems (which are harder to write for than shared-memory systems). You want to see equally painful, write the same MPI code using a pre-ANSI C-compiler because that's the equivalent of what you're complaining about.
The modern way to do this is OpenMP and Fortran 95 (which is supported by all major compiler vendors) or Fortran 2003 (which is 90% supported by all major compiler vendors). Just tell the compiler what you want, and you have parallel code. No fuss, no muss.
I certainly love OpenMP, and I much prefer it over MPI. But big machines are all distributed, not shared memory. If your code is going to run on anything on the top100 list, for instance, or any university cluster, then OpenMP is not enough. All the big systems I've worked with need a hybrid design where you use OpenMP for within-node parallelism, and MPI across the nodes.
With that said, there are some encouraging signs. Things like XScalableMP attempt to implement OpenMP-like semantics on top of MPI, and mostly succeeds. Although you still need to do explicit synchronization on occasion. The offloading infrastructure in the latest GCC versions could possibly also abstract away this in the future, but again, we're not there yet.
The parallel array stuff in later Fortrans are really neat. But last time I looked, doing it explicitly yourself with MPI in FORTRAN77 was still quite a bit faster. Have compilers improved to the point where that speed difference is mostly gone?
For large-scale simulations you need them to be pseudo-random, as in repeatable. If you are running a parallel simulation across hundreds or thousands of nodes, you can't send random data to all the nodes. You'd spend all your time sending that, not getting anywhere with your simulation.
Instead, what you do is that each node runs its own random generator, but seeded with the same state. That way they all have the same random source, with no need to flood the interconnect with (literally) random data.
Another reason is repeatability of course. You may often want to run the exact same simulation many times when you're developing and debugging your code.
Helicopter money is a temporary intervention, though, while UBI is meant to be permanent. At best, you could use the helicopter money angle to cover the initial startup and adjustment costs of an UBI implementation. But the long-term financing will need to be from elsewhere.
It's called "Google Maps"; though, not "Google Driving Instructions". You note yourself that finding the route from A to B is only one possible use of a map. Google Maps is increasingly failing at most of those other uses. perhaps they should rename it to "Google GPS" and leave the actual map field to other companies.
But what does the IoT bit have to do with it, though? Could just have a local thermostat with sensors in every room. Optionally passively download prices from your electricity provider if you want to get really fancy with scheduling your water heater.
You don't need the thing to be accessible over the net, and it most certainly doesn't need to synchronize all your usage data with some IT company somewhere. Our fridge manages to save us quite a lot of money simply by learning what time of day we're likely to open the door and when we are not, and adjust itself accordingly. No need for a connection of any kind.
"Things" are fine. "Internet of" is what people are questioning.
If you look at the labs homepage you see they're doing all kinds of things related to human augmentation, telepresence and human-machine interaction. The stuff ranges from the silly ("hoverball": use a small drone as an active ball in sports) to the useful ("Expressive typing": read the accelerometer in hard drives to estimate keyboard typing force). This fork is probably more about exploring what is possible to achieve rather than trying to solve a particular problem.
With that said, there are people that really can't eat salt; some people with kidney diseases for instance. For such people, this may help make their food a bit more palatable.
We already have a Linux for limited devices which works gracefully under those circumstances, it's called Android.
Much as I like Android - and already have several Android devices - it's not a good platform for content creation. Try finding a graceful way to run ipython with scipy and matplotlib (for simple analysis and plotting); something like Inkscape to edit SVG diagrams and illustrations; or Impress to make and edit presentations using that data while travelling.
The price is about normal for a device this class. And you're not going to use this for all the same things as a big laptop. I certainly want at least 16GB on a "real" machine, but I'm not going to do any of the things that require it on this.
My next-to-last laptop would have been 8 years this year. And performance-wise it would still be a perfectly viable machine for the kind of stuff I do on the road. Surf the web, lightweight programming, a few SSH sessions, writing and creating graphs, creating and editing presentations and so on.
This one is thin and lightweight enough - similar to my current Xperia tablet - that it really is unnoticeable in the bag. And, of course, the price makes it a "safe" buy, and something you don't need to be excessively worried about when bringing it anywhere.
I want one of these. It's plenty capable enough as a lightweight laptop replacement and companion device to a real desktop. And the price is low enough that I don't feel worried about wasting my money.
"Section 1050.7 Disorderly conduct. No person on or in any facility or conveyance shall:... (g) drink any alcoholic beverage or possess any opened or unsealed container of alcoholic beverage, except on premises duly licensed for the sale of alcoholic beverages, such as bars and restaurants;"
Didn't realize drinking beer in US subways is according to the rules. Wouldn't want a self-styled vigilante grab your beer and pour it all over you or something, now would we?
Yep. I always use LibreOffice to edit and send back documents for work. It usually works OK, but with frequent glitches. I worried about that, so I once asked our admin if she had a problem with the docs I sent back. She said mine were no worse than those she got from everybody else, and she had never realized I wasn't actually using MS word to edit them. Glitches and formatting errors is apparently completely normal even with the same version of MS word on different computers.
Another way is to find a book that represents your company culture, and give it to each new arrival to read.
Give a different book to each new arrival. Put them all into the same, new greenfield project and let them fight it out. Adopt the winners' methodology as the new company culture.
John Cook (put his blog in your RSS feed if you don't already have it) made a very good point recently: The speed gains from Moore's Law are dwarfed by the speed gains from algorithmic improvements. And unlike Moore's Law, we're not yet seeing a limit approaching for better ways to solve stuff. The post in question: http://www.johndcook.com/blog/...
A lot of tasks intrinsically don't scale, or scale only up to some limit. Some people are running into this already in the HPC world, were we have big parallel machines that they can't take full advantage of. Their simulations simply don't scale above a certain number of cores.
This problem is becoming steadily worse, since people want to make models with more detail (that tends to not parallelize well), and simulate much longer timeframes than before. If you're simulating protein interactions over one millisecond, then it might not matter if it takes an hour or two. But if you want to use that to understand LTP in neurons and simulate a second or two, then it becomes a very major problem if your model can't parallelize further and the per-core speed stays put.
Just remember that a real neuron is nothing like the "neurons" in neural networks. Each one is really computing a fairly complex set of functions. A single real neuron would be best represented by a decent-sized recurrent neural network all by itself.
Slashdot Mirror
Because long-term unemployment is a societal burden, not just an individual one? And it's a missed economic opportunity for society as well as the individual?
It is a shared responsibility because mismatches between worker skills and opportunities is a shared economic burden.
When you subdivide a problem, each core works on a smaller subset. If those subsets fit into a cache that the bigger problem didn't, you can easily get superlinear increase as a result. In many cases you could actually rewrite the bigger problem to be more cache-friendly and get a similar speedup, so you generally don't make much of such "extra" performance increases.
If you don't have or can get the money to actually buy the thing in the first place, it doesn't matter if it saves you money in the long run. This is part reason why it tends to cost more to be poor than to be rich.
One one hand, revitalizing city centers is not necessarily a bad thing. On the other, this starts to smell a little of Shadowrun-style megacorporations (or of industrial-era company towns).
Live and work your entire life within the protective confines of your employer. Go to the company school, work at the company office, live in company housing paid for with a company-bank supplied mortgage, dine at your choice of company restaurants, vacation at the company resort, get a company funeral...
And the end result is another Android phone, except with small compatibility issues and without the actual app store. You'd be left with the worst of two words: Users would rather get a proper Android device with all the apps; and developers would rather develop for the billions of people using the Android ecosystem and would not bother rewriting and submitting their stuff to MS own app store variant.
I certainly love OpenMP, and I much prefer it over MPI. But big machines are all distributed, not shared memory. If your code is going to run on anything on the top100 list, for instance, or any university cluster, then OpenMP is not enough. All the big systems I've worked with need a hybrid design where you use OpenMP for within-node parallelism, and MPI across the nodes.
With that said, there are some encouraging signs. Things like XScalableMP attempt to implement OpenMP-like semantics on top of MPI, and mostly succeeds. Although you still need to do explicit synchronization on occasion. The offloading infrastructure in the latest GCC versions could possibly also abstract away this in the future, but again, we're not there yet.
The parallel array stuff in later Fortrans are really neat. But last time I looked, doing it explicitly yourself with MPI in FORTRAN77 was still quite a bit faster. Have compilers improved to the point where that speed difference is mostly gone?
For large-scale simulations you need them to be pseudo-random, as in repeatable. If you are running a parallel simulation across hundreds or thousands of nodes, you can't send random data to all the nodes. You'd spend all your time sending that, not getting anywhere with your simulation.
Instead, what you do is that each node runs its own random generator, but seeded with the same state. That way they all have the same random source, with no need to flood the interconnect with (literally) random data.
Another reason is repeatability of course. You may often want to run the exact same simulation many times when you're developing and debugging your code.
Helicopter money is a temporary intervention, though, while UBI is meant to be permanent. At best, you could use the helicopter money angle to cover the initial startup and adjustment costs of an UBI implementation. But the long-term financing will need to be from elsewhere.
It's called "Google Maps"; though, not "Google Driving Instructions". You note yourself that finding the route from A to B is only one possible use of a map. Google Maps is increasingly failing at most of those other uses. perhaps they should rename it to "Google GPS" and leave the actual map field to other companies.
Install an adblocker. Free, unlike getting a new phone, and it'll help your monthly data usage as well.
But what does the IoT bit have to do with it, though? Could just have a local thermostat with sensors in every room. Optionally passively download prices from your electricity provider if you want to get really fancy with scheduling your water heater.
You don't need the thing to be accessible over the net, and it most certainly doesn't need to synchronize all your usage data with some IT company somewhere. Our fridge manages to save us quite a lot of money simply by learning what time of day we're likely to open the door and when we are not, and adjust itself accordingly. No need for a connection of any kind.
"Things" are fine. "Internet of" is what people are questioning.
If you look at the labs homepage you see they're doing all kinds of things related to human augmentation, telepresence and human-machine interaction. The stuff ranges from the silly ("hoverball": use a small drone as an active ball in sports) to the useful ("Expressive typing": read the accelerometer in hard drives to estimate keyboard typing force). This fork is probably more about exploring what is possible to achieve rather than trying to solve a particular problem.
With that said, there are people that really can't eat salt; some people with kidney diseases for instance. For such people, this may help make their food a bit more palatable.
Much as I like Android - and already have several Android devices - it's not a good platform for content creation. Try finding a graceful way to run ipython with scipy and matplotlib (for simple analysis and plotting); something like Inkscape to edit SVG diagrams and illustrations; or Impress to make and edit presentations using that data while travelling.
The price is about normal for a device this class. And you're not going to use this for all the same things as a big laptop. I certainly want at least 16GB on a "real" machine, but I'm not going to do any of the things that require it on this.
My next-to-last laptop would have been 8 years this year. And performance-wise it would still be a perfectly viable machine for the kind of stuff I do on the road. Surf the web, lightweight programming, a few SSH sessions, writing and creating graphs, creating and editing presentations and so on.
This one is thin and lightweight enough - similar to my current Xperia tablet - that it really is unnoticeable in the bag. And, of course, the price makes it a "safe" buy, and something you don't need to be excessively worried about when bringing it anywhere.
I want one of these. It's plenty capable enough as a lightweight laptop replacement and companion device to a real desktop. And the price is low enough that I don't feel worried about wasting my money.
"Section 1050.7 ...
Disorderly conduct.
No person on or in any facility or conveyance shall:
(g) drink any alcoholic beverage or possess any opened or unsealed container of alcoholic beverage, except on premises duly licensed for the sale of alcoholic beverages, such as bars and restaurants;"
Didn't realize drinking beer in US subways is according to the rules. Wouldn't want a self-styled vigilante grab your beer and pour it all over you or something, now would we?
I would have used proper dental tools myself. To each their own of course.
Yep. I always use LibreOffice to edit and send back documents for work. It usually works OK, but with frequent glitches. I worried about that, so I once asked our admin if she had a problem with the docs I sent back. She said mine were no worse than those she got from everybody else, and she had never realized I wasn't actually using MS word to edit them. Glitches and formatting errors is apparently completely normal even with the same version of MS word on different computers.
Give a different book to each new arrival. Put them all into the same, new greenfield project and let them fight it out. Adopt the winners' methodology as the new company culture.
John Cook (put his blog in your RSS feed if you don't already have it) made a very good point recently: The speed gains from Moore's Law are dwarfed by the speed gains from algorithmic improvements. And unlike Moore's Law, we're not yet seeing a limit approaching for better ways to solve stuff. The post in question: http://www.johndcook.com/blog/...
A lot of tasks intrinsically don't scale, or scale only up to some limit. Some people are running into this already in the HPC world, were we have big parallel machines that they can't take full advantage of. Their simulations simply don't scale above a certain number of cores.
This problem is becoming steadily worse, since people want to make models with more detail (that tends to not parallelize well), and simulate much longer timeframes than before. If you're simulating protein interactions over one millisecond, then it might not matter if it takes an hour or two. But if you want to use that to understand LTP in neurons and simulate a second or two, then it becomes a very major problem if your model can't parallelize further and the per-core speed stays put.
Just remember that a real neuron is nothing like the "neurons" in neural networks. Each one is really computing a fairly complex set of functions. A single real neuron would be best represented by a decent-sized recurrent neural network all by itself.