Does this also apply to throwing bricks through windows to demonstrate that they should all use tougher windows? We have the tech to make brick resistant windows but it would cost a lot for every business to upgrade their windows. Once they do that do you then show that their windows are still able to be broken with a rifle?
Going around breaking things just to show they can be broken is not civil disobedience. The systems are already known to be vulnerable and we count on people not to be jackasses and breaking them and then punish people when they can't behave.
I understand that we make NO effort to stop people from putting nails in the street, throwing bricks through windows etc. The VAST majority of our society has no or very minimal security. Expecting that every system that can be hardened should be hardened is not reasonable. The costs to society to harden everything is extremely high and it makes more sense to go after people that abuse these systems.
I don't think it is actually reasonable or possible to secure all of these systems and at some point we need to go after the people that abuse them. We don't require that all windows are brick proof. No matter what security is added to these systems in time it will be outdated and keep needing more upgrades to just keep up with newer security standards and the money needed to do all of that has to come from somewhere. Something will have to be cut or taxes will have to go up to cover it.
We don't live in hardened societies and most people would hate to live in a hardened society where everything had to be designed to protect itself against all kinds of abuse. Cars would be more like tanks since you never know when someone is going to drop something on your car from an overpass or things are spread on the road to destroy tires.
At some point we need to just go after the people that do this kind of thing and make them stop abusing the system the same way we would if they went through an area and broke all the windows.
As opposed to the rest of the 1st world countries which manage to have affordable and working health care systems. Canada, France, UK etc all have single payer systems with working hospitals and better health outcomes at a fraction of the price.
Curing cancer is a little different and would be highly profitable. You would wipe out all your competitors that also compete against you in other drug areas. Cancer is also something where curing it once will not prevent you from getting it again and again and again and keep needing a different customized cure. Also a cure from cancer could be sold at a truly staggering price and it would still be worth it so you manufacture far less of a complex drug substance at a MUCH higher price per unit.
I love all the updates to Visual Studio. With VS2017 I reported problems with CMake and all have been fixed in updates. The software I work on is now much simpler to build on Windows than it was before. Almost all the issues I reported where new feature requests.
Of course if that was true of electricity we would have made no real progress yet since we still don't know what an electron is and our understanding of electricity is still very primitive. Flight is also quite primitive still.
It turns out that humans do many things without really understanding it. There are some things that neural nets do a VERY good job of (high dimensional interpolation far better than any spline).
The idea that we have to understand 100% of how the brain works to build a real AI is nonsense. Basically no technology has ever developed that way. I think we are a long ways from real AI right now but what we have right now is still useful.
The last few laptops I have gotten game with electronic manuals. They were nicely formatted, complete, on the usb stick that came with the laptop so I could view it offline along with a complete set of drivers and other recovery information.
The whole IDE thing was crazy. Had to jumper the drives to set their master or slave status. Some would work as master or slave, some only as master, some only in certain combinations. CDrom master and HD slave often did not work but the other way around worked much better.
Part of the reason I think we ended up with boards that had 6+ slots on them is that you could just make every device a master and the system worked much more reliably.
Many motherboards had primary and secondary slots and each slot had a master and a slave connector.
There are still a number of things like to DNS and other network services where primary and secondary are not the same as master and slave and a primary or secondary dns server can have slave servers.
I understand coming up with different terms but we also need to make sure the new terms are not more confusing than what we have already.
Organic should have EXACTLY the same burden. Using chemical mutagens and radiation to cross-breed plants and to mutate the genome and selecting for desirable traits is FAR more likely to have nasty side effects. You should look and see what organic actually allows. For a long time people have used certain crushed rocks or other chemicals to get plants to cross-breed. We now know that most of those are chemical mutagens or radiation sources. I trust gene editing far more than that and all of the methods should be tested.
It looks like the SSDs that stuck to the standard drivers and interfaces did not run into any problems at all. Basically this only impacted drives that used custom drivers.
Right now everything is parallelized using TBB. I did look into parallelizing with a GPU however the switching from solver to matrix construction happens quite quickly and matrix construction looks very inefficient on a GPU. On the whole it looks like the system runs slower if we try to run it with a GPU compared to highly optimized CPU code. I do have a usage for GPUs later though for a different part of the process.
If you are working with things like molecular dynamics there is a limit to how much speed you can trade off to make the code easier to write and maintain. For many of these problems we can't just buy more hardware to solve the problem since we are already at the limits of hardware. I work on a chromatography simulator which takes several minutes to do a single simulation and millions of simulations could be needed. In the end many of our runs takes a few days to work. Most of the simpler simulators run 1/100th the speed or less. Taking something from a few days to most of a year is just not going to work.
Languages like C++ are just inherently harder than languages like Python. If you want high performance you have to pay for it with something like C++ and your code will be harder to write and maintain. We work to make our code as clear and maintainable as possible in C++ but still not as simple as Python.
Biotech has had amazing benefits. Protein based drugs have massively increased cancer survival rates along with many other diseases like hemophilia.
CPUs have gotten much faster than they where. For HPC applications cpus have increased about 10x in the last 5 years or so and it looks like AVX-512 is continuing that trend. Neural nets are also proving to be extremely useful in science as surrogate models. Neural nets as surrogate models often execute tens of thousands of times faster than the base models.
Most of the reason you don't see regular software benefiting much from faster CPUs is most software is not written to take advantage of what CPUs can do. Most software is designed to be easier to write and maintain and not to run fast. It also turns out that if you want software to run REALLY fast you do need to understand linear algebra.
I really don't think the media is liberal or conservative. They just want conflict and Trump generates conflict so they cover it non-stop. Even with what they know now about how much of a problem they created they still can't stop themselves from doing it.
If Fox loses too many of its viewers because they have died of old age they will switch to something else in order to keep making money. They don't really care about being conservative.
At the core I see the media companies as thriving on conflict. If aliens invaded Earth and we were kicking their asses then the media companies would give the aliens weapons and teach them how to use the weapons in order to make covering the conflict more interesting.
That is also why one presidential candidate never really get very far ahead of the other one. The media companies will bring up all the negative things they can find on whoever is ahead so they can make the race close and then disclaim any responsibility for it.
I have done some work with neural networks to operate as a surrogate model for more complex simulations. There is no analytical solution but some function exists such that for the given inputs the simulated outputs are the result to some determined precision. Neural networks are quite good at these kinds of problems. The system is in no way intelligent and it is not capable of doing anything beyond giving the same output as the simulation based on a given input.
I don't see any reason that just having a neural network involved should involve regulation.
For HPC (high performance computing) I don't see C++ going anywhere. For HPC the only viable languages I see are C, C++ and Fortran since they have the best optimizing compilers.
Mostly I use python for command and control with a simulator written in C++ and this seems to be a pretty common setup for HPC applications.Command and control often has a lot of code but is 1% of the compute time so write it in a high level language and then do the simulator is something that is FAST.
Studies can and have trivially checked for herbicide and pesticides in the food and from what I remember it was not significantly different than organic foods. Remember there are many old herbicides and pesticides that are classified as organic that farmers use and some of them are really not that safe.
The difference with GMO is we don't know of any mechanism that would allow a GMO plant to harm you in general. When a plant is modified it generates slightly different proteins, dna and rna than normal. However, we would not survive as a species if we just absorbed those things directly into us. It would be a HUGE security compromise for the immune system. What our digestion process does is break them all down into simple molecules and that process destroys anything we do with GMO. Outside of creating a direct toxin with GMO there is almost no way to harm someone with them and even then that is quite difficult to do on accident and it is trivial to screen for known toxic proteins.
It is just inherent to how your body is built that GMO represents essentially zero risk to it. Certainly the same or less than organic food.
If you seriously think that GMO is bad for you then you would need to propose a mechanism that would allow GMO to be bad for you that Organic food does not share and then test that mechanism. So far studies have found no different in health outcomes or any such mechanism.
The sad thing is that this should not qualify as a shining example of morality. There are so many fun games available that you can never play more than a tiny percentage of them. If you try to get a game illegally and get infected from it and have to rebuild the machine and restore from backups you easily lose far more in time than the game was worth.
My general view is that you are free to set any price you want for your product. If I don't like the price I will go somewhere else. I don't have the right to illegally copy your product just because I want it for a price you don't want to sell it at. There is also ample competition at this point. There is more entertainment than any human could ever consume. If someone charges too much go find something else or wait for a sale.
There will be a huge steam sale soon for Black Friday in the USA. That is a good time to get games.
I pretty much only buy games on steam that are on-sale. I think the last full price game I bought was Fallout 4 and the next one I buy at full price will probably be Elder Scrolls 6 or Fallout 5.
There are two games I will probably get in the next steam sale but I have no real interest in going to the pirate sites to try and find them, hope nothing gets infected etc.
Slashdot Mirror
Does this also apply to throwing bricks through windows to demonstrate that they should all use tougher windows? We have the tech to make brick resistant windows but it would cost a lot for every business to upgrade their windows. Once they do that do you then show that their windows are still able to be broken with a rifle?
Going around breaking things just to show they can be broken is not civil disobedience. The systems are already known to be vulnerable and we count on people not to be jackasses and breaking them and then punish people when they can't behave.
I understand that we make NO effort to stop people from putting nails in the street, throwing bricks through windows etc. The VAST majority of our society has no or very minimal security. Expecting that every system that can be hardened should be hardened is not reasonable. The costs to society to harden everything is extremely high and it makes more sense to go after people that abuse these systems.
I don't think it is actually reasonable or possible to secure all of these systems and at some point we need to go after the people that abuse them. We don't require that all windows are brick proof. No matter what security is added to these systems in time it will be outdated and keep needing more upgrades to just keep up with newer security standards and the money needed to do all of that has to come from somewhere. Something will have to be cut or taxes will have to go up to cover it.
We don't live in hardened societies and most people would hate to live in a hardened society where everything had to be designed to protect itself against all kinds of abuse. Cars would be more like tanks since you never know when someone is going to drop something on your car from an overpass or things are spread on the road to destroy tires.
At some point we need to just go after the people that do this kind of thing and make them stop abusing the system the same way we would if they went through an area and broke all the windows.
As opposed to the rest of the 1st world countries which manage to have affordable and working health care systems. Canada, France, UK etc all have single payer systems with working hospitals and better health outcomes at a fraction of the price.
Curing cancer is a little different and would be highly profitable. You would wipe out all your competitors that also compete against you in other drug areas. Cancer is also something where curing it once will not prevent you from getting it again and again and again and keep needing a different customized cure. Also a cure from cancer could be sold at a truly staggering price and it would still be worth it so you manufacture far less of a complex drug substance at a MUCH higher price per unit.
I love all the updates to Visual Studio. With VS2017 I reported problems with CMake and all have been fixed in updates. The software I work on is now much simpler to build on Windows than it was before. Almost all the issues I reported where new feature requests.
Of course if that was true of electricity we would have made no real progress yet since we still don't know what an electron is and our understanding of electricity is still very primitive. Flight is also quite primitive still.
It turns out that humans do many things without really understanding it. There are some things that neural nets do a VERY good job of (high dimensional interpolation far better than any spline).
The idea that we have to understand 100% of how the brain works to build a real AI is nonsense. Basically no technology has ever developed that way. I think we are a long ways from real AI right now but what we have right now is still useful.
That one I can ABSOLUTELY agree with.
The last few laptops I have gotten game with electronic manuals. They were nicely formatted, complete, on the usb stick that came with the laptop so I could view it offline along with a complete set of drivers and other recovery information.
I much prefer electronic manuals to paper ones.
The whole IDE thing was crazy. Had to jumper the drives to set their master or slave status. Some would work as master or slave, some only as master, some only in certain combinations. CDrom master and HD slave often did not work but the other way around worked much better.
Part of the reason I think we ended up with boards that had 6+ slots on them is that you could just make every device a master and the system worked much more reliably.
IDE cables are a great way to show the problem.
Many motherboards had primary and secondary slots and each slot had a master and a slave connector.
There are still a number of things like to DNS and other network services where primary and secondary are not the same as master and slave and a primary or secondary dns server can have slave servers.
I understand coming up with different terms but we also need to make sure the new terms are not more confusing than what we have already.
Organic should have EXACTLY the same burden. Using chemical mutagens and radiation to cross-breed plants and to mutate the genome and selecting for desirable traits is FAR more likely to have nasty side effects. You should look and see what organic actually allows. For a long time people have used certain crushed rocks or other chemicals to get plants to cross-breed. We now know that most of those are chemical mutagens or radiation sources. I trust gene editing far more than that and all of the methods should be tested.
Nuclear physics
It looks like the SSDs that stuck to the standard drivers and interfaces did not run into any problems at all. Basically this only impacted drives that used custom drivers.
Right now everything is parallelized using TBB. I did look into parallelizing with a GPU however the switching from solver to matrix construction happens quite quickly and matrix construction looks very inefficient on a GPU. On the whole it looks like the system runs slower if we try to run it with a GPU compared to highly optimized CPU code. I do have a usage for GPUs later though for a different part of the process.
If you are working with things like molecular dynamics there is a limit to how much speed you can trade off to make the code easier to write and maintain. For many of these problems we can't just buy more hardware to solve the problem since we are already at the limits of hardware. I work on a chromatography simulator which takes several minutes to do a single simulation and millions of simulations could be needed. In the end many of our runs takes a few days to work. Most of the simpler simulators run 1/100th the speed or less. Taking something from a few days to most of a year is just not going to work.
Languages like C++ are just inherently harder than languages like Python. If you want high performance you have to pay for it with something like C++ and your code will be harder to write and maintain. We work to make our code as clear and maintainable as possible in C++ but still not as simple as Python.
Biotech has had amazing benefits. Protein based drugs have massively increased cancer survival rates along with many other diseases like hemophilia.
CPUs have gotten much faster than they where. For HPC applications cpus have increased about 10x in the last 5 years or so and it looks like AVX-512 is continuing that trend. Neural nets are also proving to be extremely useful in science as surrogate models. Neural nets as surrogate models often execute tens of thousands of times faster than the base models.
Most of the reason you don't see regular software benefiting much from faster CPUs is most software is not written to take advantage of what CPUs can do. Most software is designed to be easier to write and maintain and not to run fast. It also turns out that if you want software to run REALLY fast you do need to understand linear algebra.
I really don't think the media is liberal or conservative. They just want conflict and Trump generates conflict so they cover it non-stop. Even with what they know now about how much of a problem they created they still can't stop themselves from doing it.
If Fox loses too many of its viewers because they have died of old age they will switch to something else in order to keep making money. They don't really care about being conservative.
At the core I see the media companies as thriving on conflict. If aliens invaded Earth and we were kicking their asses then the media companies would give the aliens weapons and teach them how to use the weapons in order to make covering the conflict more interesting.
That is also why one presidential candidate never really get very far ahead of the other one. The media companies will bring up all the negative things they can find on whoever is ahead so they can make the race close and then disclaim any responsibility for it.
I have done some work with neural networks to operate as a surrogate model for more complex simulations. There is no analytical solution but some function exists such that for the given inputs the simulated outputs are the result to some determined precision. Neural networks are quite good at these kinds of problems. The system is in no way intelligent and it is not capable of doing anything beyond giving the same output as the simulation based on a given input.
I don't see any reason that just having a neural network involved should involve regulation.
For HPC (high performance computing) I don't see C++ going anywhere. For HPC the only viable languages I see are C, C++ and Fortran since they have the best optimizing compilers.
Mostly I use python for command and control with a simulator written in C++ and this seems to be a pretty common setup for HPC applications.Command and control often has a lot of code but is 1% of the compute time so write it in a high level language and then do the simulator is something that is FAST.
The python version of this works really well.
foo = 1 if condition else 0
Studies can and have trivially checked for herbicide and pesticides in the food and from what I remember it was not significantly different than organic foods. Remember there are many old herbicides and pesticides that are classified as organic that farmers use and some of them are really not that safe.
The difference with GMO is we don't know of any mechanism that would allow a GMO plant to harm you in general. When a plant is modified it generates slightly different proteins, dna and rna than normal. However, we would not survive as a species if we just absorbed those things directly into us. It would be a HUGE security compromise for the immune system. What our digestion process does is break them all down into simple molecules and that process destroys anything we do with GMO. Outside of creating a direct toxin with GMO there is almost no way to harm someone with them and even then that is quite difficult to do on accident and it is trivial to screen for known toxic proteins.
It is just inherent to how your body is built that GMO represents essentially zero risk to it. Certainly the same or less than organic food.
If you seriously think that GMO is bad for you then you would need to propose a mechanism that would allow GMO to be bad for you that Organic food does not share and then test that mechanism. So far studies have found no different in health outcomes or any such mechanism.
The sad thing is that this should not qualify as a shining example of morality. There are so many fun games available that you can never play more than a tiny percentage of them. If you try to get a game illegally and get infected from it and have to rebuild the machine and restore from backups you easily lose far more in time than the game was worth.
My general view is that you are free to set any price you want for your product. If I don't like the price I will go somewhere else. I don't have the right to illegally copy your product just because I want it for a price you don't want to sell it at. There is also ample competition at this point. There is more entertainment than any human could ever consume. If someone charges too much go find something else or wait for a sale.
There will be a huge steam sale soon for Black Friday in the USA. That is a good time to get games.
I pretty much only buy games on steam that are on-sale. I think the last full price game I bought was Fallout 4 and the next one I buy at full price will probably be Elder Scrolls 6 or Fallout 5.
There are two games I will probably get in the next steam sale but I have no real interest in going to the pirate sites to try and find them, hope nothing gets infected etc.