Basically what I'm saying that eventually just putting solar panels everywhere, plus storage for the night will do the trick.
The Greens will fight you tooth and nail to prevent so much of the countryside being covered with ugly solar panels, and they'll have lots of company. In Japan, putting them up also won't be cheap, given the limited unused land is largely hilly or mountainous.
As others have noted, your IT orientation is leading you astray, we're not talking about doing stuff in a few isolated machine rooms. Try this analogy which works for a place like the US with so many single detached family dwellings: demanding everyone allow you to put up a shack in their back yard to house some of your computers, supplying it with power and telecom, and regularly letting people on you land to do maintenance. Which won't be a small thing, putting this stuff out in the environment means things will regularly fail, the panels will have to routinely be cleaned of dirt and dust blocking sunlight, etc. etc. The Japanese in particular are very sensitive about how little unspoiled land they have, BTW.
Most importantly, Japan doesn't have western defeatism. When a problem comes up, it gets worked on and solved.
I don't think you can solve something as fundamental as their cultural approach to safety. It's sufficiently bad in general that they should never have anything to do with nuclear power, full stop, something that was very clear before the tsunami from just their nuclear operation mishaps and management of those, let alone things like JAL Flight 123 were people noticed for years the plane was making weird noises from Boeing's faulty repair, but did nothing to address them.
And that's why eventually it will go away. When faced with the choice of dealing all the stuff above, or just putting a bunch of solar panels/mirrors/wind turbines pretty much anywhere you please with a cheap metal fence around it, it's clear what is the most convenient option.
Given that "solar panels/mirrors/wind turbines" provide neither baseline nor peaking power, the only required forms for a reliable grid, where nuclear excels at the former, and can do the latter for the easily predicted general peaks like morning when people wake up and after people get home from work, no, they're not "convenient" at all. They have to be backstopped by other forms of generation, or maybe some decades in the future we'll have batteries cheap and long lived enough to store their output for when the sun goes down, is occluded, the wind doesn't blow, etc.
Which is not to say that some countries like Japan have safety cultures so poor they should never touch nuclear power, something that was very clear long before the tsunami.
All the steel, copper, aluminum, and so on needed for the wires and structures for that wind and solar has to come from somewhere.
Well, after a wind tower reaches its end of life in 20, maybe 25 years, you can recycle the metals pretty easily. The plastics used to insulate wires, etc. not so much. Doubt much of a solar panel can be recycled, but I don't have any figures off the top of my head on their lifecycle, might not be as ultra short as wind towers, which are subject to lots of stress, with many parts needing to be as light as possible, forcing engineers into tough yield (of power) and longevity tradeoffs, and they're badly exposed to the elements. Solar cells, you ought to be able to seal them up pretty well, but I have no idea how much they're subject to degradation over time.
Airborne invasions indeed have some poor records, but Crete still succeeded. A politically desperate Xi might still try it, hope it works for him. And it would still trash TSMC production.
China could go nuclear, but that would likely bring American retaliation.
I don't see any scenario where we retaliate with nukes. To quote that PLA general from memory, we care more about Los Angeles that we care about Taipei.
If Taiwan ever feels like they can't count on America, they could build their own nukes in, maybe, a month. Remember, every country that has ever made a serious attempt to build a nuke has succeeded on the first try. As one of the most technologically advanced countr^H^H^H^H^Hregions in the world, Taiwan would have no problem.
Unless they get there hands on some plans, like the Khan network's, it would take them much longer, there are a fair number of tricks and experimentation needed to get implosion devices to work, and the much bigger problem would be procuring the fissionables, can't use civilian reactor "waste" for that. But in geopolitical time frames, maybe, but the PRC wouldn't react well to such an effort, not well at all.
The PRC might try to finish off the ROC before we can get a carrier there, but worse than the carriers for their schemes are our hunter-killer subs. Unless they figure out something the Soviet never could, it won't take many of them to turn it into an all air affair. Which might be part of their plan, why they've emplaced so many short range missiles, they might hope to beat down the ROC's defenses and get enough airheads in place that they sort of win by default, maybe?
Getting back to the topic at hand, any war which freely uses those missiles would almost certainly shut down TSMC's production on the island for a long time. It could be far easier to exfiltrate their people and set up shop somewhere else (think Operation Paperclip), although of course the PRC would insist they are PRC citizens, and they would be subject to the current sorts of kidnapping and disappearing that's already happening.
Xi doesn't need a war
Right now he doesn't, but if he starts losing his grip.... That's why you always evaluate countries by their capabilities first, you never know when something externally "stupid" like a faction fight will cause a country to lash out at another.
Speculative execution is not intrinsically vulnerable. Intel's implementation could be better....
Maybe not "intrinsically vulnerable", but certainly very difficult to get right. I mean, ever other designer of high end speculative execution CPUs but AMD also had Meltdown problems, and all, Intel, AMD, ARM, IBM mainframes and POWER, are plagued with Spectre bugs, and almost certainly will be for a long time.
The stock price did get a nice momentary bump due to Intels utterly pathetic failing on the security front, but as anyone who owned AMD stock in the 2000s when it hit 38 a share can remember, it can and will come crashing down in a moment.
I read some time ago, probably after they completely dropped the ball after totally slaughtering Intel with their K8 microarchitecture and interconnect technologies, that over time, AMD has never made money for their shareholders. They create a big hit, like their 486 based DX4-100, and then reliably drop the ball again.
But Intel's unique for them fab fumble, at least for a very long time, like going back to the early 1980s when they were forced out of the DRAM business because Japanese yields were so high, also has to be a significant factor. Between the lack of "10nm" production, and their shifting production around in anticipation of it, normally they transition older nodes to chipsets etc., which they're undoing, they can't make enough larger node CPUs to meet current demands. Kinda glad I have a surplus of still functioning fine Sandy and Ivy Bridge systems for my personal use right now....
Indeed, but are any of those chips within 10' of you competing with Intel for the desktop and server markets, or could they in theory do so (especially for the desktop)?
BTW, I remember reading that there's pretty much one ARM chip in every SD/SDHC/SDetc. card. I assume this is one way we get "billions and billions" of ARM CPUs shipped every year.
In fact, when it comes to rocket engines, the USA also relies on the Russians.
Utterly false, except for that for first stages, one batch of mothballed NK-33s Orbital Sciences used/is still using for a while for their Antares craft, one of which caused a nasty total failure in 2014, and is to be replaced with the RD-181.
ULA uses the RD-180 for the Atlas V, a tremendously successful albeit now expensive rocket, with 78 successful launches and one partial but not complete low orbit failure for the NRO, an arrangement which figuratively blew up also in 2014 with the mess with Ukraine. In theory to be replaced with an Aerojet and/or Blue Origin rocket.
If there's any other rocket engine we use that's not Made in the USA I'm unaware of it. We're currently launching Delta IVs, 36 successes, one partial low orbit failure, Falcon 9/Heavy, Minotaurs (well, they're just converted Minuteman and Peacekeeper ICBMs), and the Pegasus apparently is still hanging in there.
As for your first point, that's a NASA problem, likely soon to be solved by SpaceX and/or Boeing/ULA, both under the heavy thumb of NASA. And right now officially no one, or are you forgetting last month's Soyuz failure? (Well, the escape system worked!) Is that fleet still grounded while they figure out what happened?
Plus you're studiously ignoring the wild success of SpaceX, which by focusing on economics, and e.g. not starting with a military missile design, is putting the rest of the international space industry at risk of losing most of their business.
Quick question if I may impose? In very general terms, you and your (would be) startup competitors, SaaS offerings, who uses 3rd party cloud vendors and who in-house, and "why?" for all the combinations except the obvious of startups using the cloud.
The concentration of power of TSMC is already a bit worrying.
It would be very, very bad if their high end fabs got trashed or destroyed in a PRC attempt to take over the ROC. Everything would be messed up for a while due to inevitable trade disruptions with the mainland, but other countries can spin up discrete parts and low end silicon fab lines (for discrete transistors and the like, and for those a lot of outside China fabs could probably just up their output) faster than high end fabs.
And doesn't Intel still make what are pretty much the best Ethernet chips. They work, and are well and openly documented? And if you don't need the highest performance, aren't their on chip GPUs also good on both metrics?
Anyone have some real facts, or pointers to the quality of AMD's current chipsets? I've heard ugly mutterings, but have no knowledge, still using stuff that's many years old now.
ARM has the same exact security problems, right down to one of their latest designs having a Meltdown variant. Of course, the overwhelmingly vast majority of ARM chips are in-order, some superscalar like the Pentium before you get to their fastest designs, which of course are the only ones that can compete with Intel for processing power. No real joy for Intel haters, everyone in the industry has failed, AMD except for Meltdown, IBM both mainframe and POWER including Meltdown.
Also, I've never heard that ARM beats Intel on electrical power consumed for computing power, for servers and the like where you want the absolute maximum computer power you can afford.
The big question is: can TSMC produce processor chips without massive gaping security holes?
As others have noted, it depends on who's designs they're fabbing, but the answer is that everybody who's designing high end, state of the art out-of-order with speculative execution chips has screwed up royalty. All of Intel, ARM, and IBM, both zSeries (lastest mainframes derived from the System/360) and POWER have Meltdown problems. All including AMD have much more sinister Spectre problems.
So everyone in the industry has a lot of work to do, and TSMC... well, this might actually hurt them if this delays some designs, or in general slows the design process down.
The policy, an executive branch memorandum, was announced by President Barack Obama on June 15, 2012.
Before that, many efforts in the Congress failed. And now the Federal courts, Ninth District and New York ones, have ordered it can't be undone.
Does any of this sound like the functioning of a "democracy", aside of course from it not passing in the Congress? Do you think there's any way this will end well?
Only a few OpenJDK distributions are Java Technical Compatibility Kit (TCK) certified, Sun didn't like handing that capability out as you might remember, refused to give it to Apache. Azul's Zulu is the only other one I know for sure.
That the total of "Defensive Gun Uses" per year; obviously not all would result in the death of the defender, but plenty would. Given our somewhat similar demographics, our murder rate would perhaps more resemble Brazil's.
Only the stupid and/or evil believe or claim guns are the underlying problem, there's always been plenty of people willing and able to use lethal force to achieve the goals, see for example the obscure guy name Abel in the Bible. The major question is a society's ability and willingness to suppress this, which is by no means a given, see anarcho-tyranny where the ruling class allies themselves with the criminal class to keep the rest in line. Something we're seeing more and more in the open in the US. Ditto outright Marxism and its universal killing fields if and when they seize power.
Nationwide, ~2.5 million per year, and that figure is from a while ago, quite a few more people are legally carrying concealed, while ~27% of the population isn't allowed to. Broken down by state, you'd need to see if that was one of the states the CDC included in their Behavioral Risk Factor Surveillance System (BRFSS) 1996, 1997, and 1998 surveys which recently came to light, or see if one of the others breaks it down by state.
Heard it was available for 20 people. If so, that's not a lot of incentive to apply.
That's an appropriate number for a pilot program.
Tulsa's too high crime for my tastes, but I'm sure its much lower that the rate in which lots of the targeted population live. And it's a shall issue concealed carry state, so you've at least got a fighting chance if you so choose.
His executive orders tend to be reversed by the courts in SF or Seattle right away.
Exactly. Even if he's undoing a pure (and illegal) Obama executive order like DACA, it gets put on hold until it makes it to the Supreme court, should they deign to hear the case.
But quoting the author:
Trump promised, but he still hasn't even reverse Obama's executive order letting H1-B spouses work. He could do that with a stroke of a pen
As we're noted, he cannot in practice end it with "a stroke of a pen", so his administration is doing it through the normal rules making process. It should come into effect in 1-2 months based on my searching just now and memory.
Of course, the 9th Circuit might nullify it anyway, but it'll be in theory harder to sustain.
Thanks! That does put a slightly different cast on what they did, and likely absolves ISRO/India of most of the blame (I'm assuming the company just plain lied after they got their denial, by omission if nothing else), but doesn't make them look any better. "Move fast and break things", well, only the first is a useful motto in rocketry.
Slashdot Mirror
The Greens will fight you tooth and nail to prevent so much of the countryside being covered with ugly solar panels, and they'll have lots of company. In Japan, putting them up also won't be cheap, given the limited unused land is largely hilly or mountainous.
As others have noted, your IT orientation is leading you astray, we're not talking about doing stuff in a few isolated machine rooms. Try this analogy which works for a place like the US with so many single detached family dwellings: demanding everyone allow you to put up a shack in their back yard to house some of your computers, supplying it with power and telecom, and regularly letting people on you land to do maintenance. Which won't be a small thing, putting this stuff out in the environment means things will regularly fail, the panels will have to routinely be cleaned of dirt and dust blocking sunlight, etc. etc. The Japanese in particular are very sensitive about how little unspoiled land they have, BTW.
I don't think you can solve something as fundamental as their cultural approach to safety. It's sufficiently bad in general that they should never have anything to do with nuclear power, full stop, something that was very clear before the tsunami from just their nuclear operation mishaps and management of those, let alone things like JAL Flight 123 were people noticed for years the plane was making weird noises from Boeing's faulty repair, but did nothing to address them.
Given that "solar panels/mirrors/wind turbines" provide neither baseline nor peaking power, the only required forms for a reliable grid, where nuclear excels at the former, and can do the latter for the easily predicted general peaks like morning when people wake up and after people get home from work, no, they're not "convenient" at all. They have to be backstopped by other forms of generation, or maybe some decades in the future we'll have batteries cheap and long lived enough to store their output for when the sun goes down, is occluded, the wind doesn't blow, etc.
Which is not to say that some countries like Japan have safety cultures so poor they should never touch nuclear power, something that was very clear long before the tsunami.
Well, after a wind tower reaches its end of life in 20, maybe 25 years, you can recycle the metals pretty easily. The plastics used to insulate wires, etc. not so much. Doubt much of a solar panel can be recycled, but I don't have any figures off the top of my head on their lifecycle, might not be as ultra short as wind towers, which are subject to lots of stress, with many parts needing to be as light as possible, forcing engineers into tough yield (of power) and longevity tradeoffs, and they're badly exposed to the elements. Solar cells, you ought to be able to seal them up pretty well, but I have no idea how much they're subject to degradation over time.
I don't see any scenario where we retaliate with nukes. To quote that PLA general from memory, we care more about Los Angeles that we care about Taipei.
Unless they get there hands on some plans, like the Khan network's, it would take them much longer, there are a fair number of tricks and experimentation needed to get implosion devices to work, and the much bigger problem would be procuring the fissionables, can't use civilian reactor "waste" for that. But in geopolitical time frames, maybe, but the PRC wouldn't react well to such an effort, not well at all.
Sorry if I wasn't clear about AMD, I was saying everyone else, Intel, ARM, and IBM's two macroarchitectures, have Meltdown.
The PRC might try to finish off the ROC before we can get a carrier there, but worse than the carriers for their schemes are our hunter-killer subs. Unless they figure out something the Soviet never could, it won't take many of them to turn it into an all air affair. Which might be part of their plan, why they've emplaced so many short range missiles, they might hope to beat down the ROC's defenses and get enough airheads in place that they sort of win by default, maybe?
Getting back to the topic at hand, any war which freely uses those missiles would almost certainly shut down TSMC's production on the island for a long time. It could be far easier to exfiltrate their people and set up shop somewhere else (think Operation Paperclip), although of course the PRC would insist they are PRC citizens, and they would be subject to the current sorts of kidnapping and disappearing that's already happening.
Right now he doesn't, but if he starts losing his grip.... That's why you always evaluate countries by their capabilities first, you never know when something externally "stupid" like a faction fight will cause a country to lash out at another.
Maybe not "intrinsically vulnerable", but certainly very difficult to get right. I mean, ever other designer of high end speculative execution CPUs but AMD also had Meltdown problems, and all, Intel, AMD, ARM, IBM mainframes and POWER, are plagued with Spectre bugs, and almost certainly will be for a long time.
I read some time ago, probably after they completely dropped the ball after totally slaughtering Intel with their K8 microarchitecture and interconnect technologies, that over time, AMD has never made money for their shareholders. They create a big hit, like their 486 based DX4-100, and then reliably drop the ball again.
But Intel's unique for them fab fumble, at least for a very long time, like going back to the early 1980s when they were forced out of the DRAM business because Japanese yields were so high, also has to be a significant factor. Between the lack of "10nm" production, and their shifting production around in anticipation of it, normally they transition older nodes to chipsets etc., which they're undoing, they can't make enough larger node CPUs to meet current demands. Kinda glad I have a surplus of still functioning fine Sandy and Ivy Bridge systems for my personal use right now....
Indeed, but are any of those chips within 10' of you competing with Intel for the desktop and server markets, or could they in theory do so (especially for the desktop)?
BTW, I remember reading that there's pretty much one ARM chip in every SD/SDHC/SDetc. card. I assume this is one way we get "billions and billions" of ARM CPUs shipped every year.
Utterly false, except for that for first stages, one batch of mothballed NK-33s Orbital Sciences used/is still using for a while for their Antares craft, one of which caused a nasty total failure in 2014, and is to be replaced with the RD-181.
ULA uses the RD-180 for the Atlas V, a tremendously successful albeit now expensive rocket, with 78 successful launches and one partial but not complete low orbit failure for the NRO, an arrangement which figuratively blew up also in 2014 with the mess with Ukraine. In theory to be replaced with an Aerojet and/or Blue Origin rocket.
If there's any other rocket engine we use that's not Made in the USA I'm unaware of it. We're currently launching Delta IVs, 36 successes, one partial low orbit failure, Falcon 9/Heavy, Minotaurs (well, they're just converted Minuteman and Peacekeeper ICBMs), and the Pegasus apparently is still hanging in there.
As for your first point, that's a NASA problem, likely soon to be solved by SpaceX and/or Boeing/ULA, both under the heavy thumb of NASA. And right now officially no one, or are you forgetting last month's Soyuz failure? (Well, the escape system worked!) Is that fleet still grounded while they figure out what happened?
Plus you're studiously ignoring the wild success of SpaceX, which by focusing on economics, and e.g. not starting with a military missile design, is putting the rest of the international space industry at risk of losing most of their business.
Quick question if I may impose? In very general terms, you and your (would be) startup competitors, SaaS offerings, who uses 3rd party cloud vendors and who in-house, and "why?" for all the combinations except the obvious of startups using the cloud.
It would be very, very bad if their high end fabs got trashed or destroyed in a PRC attempt to take over the ROC. Everything would be messed up for a while due to inevitable trade disruptions with the mainland, but other countries can spin up discrete parts and low end silicon fab lines (for discrete transistors and the like, and for those a lot of outside China fabs could probably just up their output) faster than high end fabs.
And doesn't Intel still make what are pretty much the best Ethernet chips. They work, and are well and openly documented? And if you don't need the highest performance, aren't their on chip GPUs also good on both metrics?
Anyone have some real facts, or pointers to the quality of AMD's current chipsets? I've heard ugly mutterings, but have no knowledge, still using stuff that's many years old now.
ARM has the same exact security problems, right down to one of their latest designs having a Meltdown variant. Of course, the overwhelmingly vast majority of ARM chips are in-order, some superscalar like the Pentium before you get to their fastest designs, which of course are the only ones that can compete with Intel for processing power. No real joy for Intel haters, everyone in the industry has failed, AMD except for Meltdown, IBM both mainframe and POWER including Meltdown.
Also, I've never heard that ARM beats Intel on electrical power consumed for computing power, for servers and the like where you want the absolute maximum computer power you can afford.
As others have noted, it depends on who's designs they're fabbing, but the answer is that everybody who's designing high end, state of the art out-of-order with speculative execution chips has screwed up royalty. All of Intel, ARM, and IBM, both zSeries (lastest mainframes derived from the System/360) and POWER have Meltdown problems. All including AMD have much more sinister Spectre problems.
So everyone in the industry has a lot of work to do, and TSMC ... well, this might actually hurt them if this delays some designs, or in general slows the design process down.
Even better example, the largest components of the KGB's emblem are literally a shield, with a sword on top, that's significant, see how little changed the graphical components of the successor FSB's emblem are, the sword is now behind the shield because they're no longer literally trying to take over the world, red star etc. replaced with appropriate Russian images.
Before that, many efforts in the Congress failed. And now the Federal courts, Ninth District and New York ones, have ordered it can't be undone.
Does any of this sound like the functioning of a "democracy", aside of course from it not passing in the Congress? Do you think there's any way this will end well?
Only a few OpenJDK distributions are Java Technical Compatibility Kit (TCK) certified, Sun didn't like handing that capability out as you might remember, refused to give it to Apache. Azul's Zulu is the only other one I know for sure.
That the total of "Defensive Gun Uses" per year; obviously not all would result in the death of the defender, but plenty would. Given our somewhat similar demographics, our murder rate would perhaps more resemble Brazil's.
Only the stupid and/or evil believe or claim guns are the underlying problem, there's always been plenty of people willing and able to use lethal force to achieve the goals, see for example the obscure guy name Abel in the Bible. The major question is a society's ability and willingness to suppress this, which is by no means a given, see anarcho-tyranny where the ruling class allies themselves with the criminal class to keep the rest in line. Something we're seeing more and more in the open in the US. Ditto outright Marxism and its universal killing fields if and when they seize power.
Nationwide, ~2.5 million per year, and that figure is from a while ago, quite a few more people are legally carrying concealed, while ~27% of the population isn't allowed to. Broken down by state, you'd need to see if that was one of the states the CDC included in their Behavioral Risk Factor Surveillance System (BRFSS) 1996, 1997, and 1998 surveys which recently came to light, or see if one of the others breaks it down by state.
The "the 'penumbras' and 'emanations' cast by the Constitution and the Bill of Rights", for one thing. Except for the 2nd Amendment, it's a black hole.
That's an appropriate number for a pilot program.
Tulsa's too high crime for my tastes, but I'm sure its much lower that the rate in which lots of the targeted population live. And it's a shall issue concealed carry state, so you've at least got a fighting chance if you so choose.
Exactly. Even if he's undoing a pure (and illegal) Obama executive order like DACA, it gets put on hold until it makes it to the Supreme court, should they deign to hear the case.
But quoting the author:
Trump promised, but he still hasn't even reverse Obama's executive order letting H1-B spouses work. He could do that with a stroke of a pen
As we're noted, he cannot in practice end it with "a stroke of a pen", so his administration is doing it through the normal rules making process. It should come into effect in 1-2 months based on my searching just now and memory.
Of course, the 9th Circuit might nullify it anyway, but it'll be in theory harder to sustain.
Thanks! That does put a slightly different cast on what they did, and likely absolves ISRO/India of most of the blame (I'm assuming the company just plain lied after they got their denial, by omission if nothing else), but doesn't make them look any better. "Move fast and break things", well, only the first is a useful motto in rocketry.