With all the voter suppression happening and voting machines flipping votes, you will not have to worry about it any more.
Showing you're on the left. If you were on the right you'd have said:
With all the illegal immigration happening and the fake voters created by the combination of unexamined mail-in registration and no-appearance-no-hardship absentee (mail in) ballots, you will not have to worry about it any more.
Both ideas are a hazard, whether or not they're true. The purpose of an election is to convince the losing side they'd have also lost the war, so they aren't tempted to start one.
They don't have to believe it's rock-solid-honest for that to work, because a close race means they'd probably lose the war because a lot of people who otherwise could care less WOULD care about trying to overturn the election and fight against them. But if they believe it's massively corrupted and the cheaters are now the tyrants, that stops working.
Look at The Battle of Athens, for instance. Or any of several other cases in US history where a corrupt and election-hacking political machine was thrown out by vigilanties. Or the left's tantrum after the 2016 election.
And to think all this fuss doesn't change the fact we still have 24 hrs. in a day.
Yes, it does.
Last Saturday had 25 hours. A Saturday in the spring had 23.
A standard time day is a legal construct, not a physical one.
And do you have ANY IDEA how much that complicates automated systems (like building energy management) that have to deal with getting things to happen at the right time for production schedules? *I* do. I had to write some of those programs. B-b doesn't even begin to express it.
It also substantially improved the reliability and maintainability. (For starters, you can shut down the magnet, warm it up, open it up, disconnecting the windings, replace the reaction chamber liner, put it back together, cool them down, and resrart it. You don't have to replace the winding after one use.)
I'd love to see a diesel engine running on fusion.
But Tokamaks aren't dead end any more. Recent improvements in superconducting ribbon have improved the strength of confinement magnets by a factor of several (about five, IARC)
It also substantially improved the reliability and maintainability. (For starters,he reaction chamber liner, put them back together, cool them down, and resrart them. You don't have to replace the winding after one use.)
This moves Tokamaks from "breakeven is a job for a multi-state-actor decades-long research project" to "past breakeven to substantial net output is a several year job for a university or sartup, with a few million in funding".
- interconnections to the rest of the world via paving the remaining four faces with fiber optic interconnections. (The bandgap puts the natural frequency of a diamond LED/laser in the near ultraviolet, and the carrier mobility allows extreme bit rates.)
- Enclosed in a container filled with a vacuum or oxygen-free gas (so it can be run at temperatures of, say, an orange glow without catching fire.) High temperature gradients can push a lot of heat across the coolling paths, while the strength of the diamond would allow it to survive the stresses.
The joy of the scenario, of course, is that this would look like the sort of ship's brain you might find in a golden-age-of-SF novel - say the Skylark series by E. E. Smith. (Six foot cube of diamond in a vacuum or gas tube enclosure, glowing orange, supported by water cooled silver bus bars, etc. B-) )
Why don't we have it already? Because of the difficulty of fabricating a collection of transistors that size with the necessary perfection (or even "good enough" to function with error correcting codes and redundant logic). I had, and still have, a possible solution to that, as well, applicable to technologies other than diamond semiconductors as well. But the technology is JUST NOW bringing it within reach - half a century later. So I'll withhold that secret-sauce recipe, just in case I get a chance to develop and patent it. B-)
Learn some more about thermal dissipation, and then ask yourself why your brilliant idea hasn't been implemented already...
I knew about that and took it into account even at the time (before making part of my career designing things like network processor chips - a field in which I have several patents.)
The original Preposterous Scale Integration pipe dream included:
- A six-foot cube of semiconductor. (That was before I knew about Amdahl's elevator hack. Cutting it down to 3'x7'x5' only cuts the volume about in half.)
- Diamond for the semiconductor. (It has EXTREME thermal conductivity, and charge carrier mobility, while the 5.47 eV bandgap allows it to run VERY hot before thermal noise is an issue- though it increases dissipation due to the high switching energy.
- Power and cooling provided across two of the six faces via water-cooled silver bus bars. (There are better alternative now, but wait for it...) In the Amdahl rectangular hack dimensionality these would be the two broadest and closest faces.
- Interconnections with the rest of the world
The end of moore's law is the problem.... We need 8k per eye, and graphics cards need to be at least 4-8x if not 16x as fast for that to be viable, and given the end of moore's law this seems unlikely.
Like the multi-decade running gag "Imminent death of the Internet" predicted, film at 11:00", the end of Moore's Law has been long predicted and slow in coming.
Lately we're starting to hit a wall on the "just make the features smaller and being closer together makes things run faster" approach: As features near the scale of the size of the electron wave function keeping the signals separate becomes an issue.
But Moore's law is really about the number of transistors you can build into a chip, and we've literally "just scratched the surface" there. Until recently the transistors have been essentially a two-dimensional layer on the surface of a chip. Now we're starting to use the third dimension - up to eight layers of it, last time I looked.
The limit to that is what I called "Preposterous Scale Integration" back in the mid-sixites (when the buzzwords were "small...", "medium..." and "large scale integration", rather than, say, the "deep sub-micron" of the millennium or the current "xx nanometer process"). When the chip has layers of logic elements spaced along the z axis as closely as the feature size in the x y plane, and acceptable yields are obtained for "chips" the size of the old IBM mainframe cabinets - just small enough to fit through a standard elevator door - THEN you're approaching the end of Moore's Law. That's still many decades out.
Now for a single thread you're currently starting to fall off the speed-doubles-every-1.5 or so-years interpretation of Moore's law. Having lots of transistors doesn't get you faster once you're simultaneously running into limits of feature size AND switching speed AND propagation time. But for computations that are massively parallelizable, you can continue to throw more and more elements at them. Graphics rendering, combined with speculatively pre-computing multiple viewpoints for head rotation and translation options (i.e. the core of VR and AR), is such a problem.
if it comes packaged as a service and you access a website, portal, or online content to use it, then you dont own it. Read the terms of service...
Even without terms of service that claim ownership of the data, when the data IS yours and nobody claims otherwise: If it's stored on an external service the supreme court has repeatedly ruled that that you have no "reasonable expectation of privacy".
And if the FCC thinks that the internet is not telecommunications it's packed full of incompetent idiots.
You're thinking the plain-English meaning of telecommunications. The FCC is working with the terms of legal art, as defined by the congress in the enabling acts and interpreted by the court.
In those terms, there is a distinction between "telecommunication services" (the default when one-way or two-way information transmission, like telegraph, telephone, radio, television, etc. are involved) and information services (a carve-out for computer networking, defined by later federal legislation which also tells both the FCC and the states to keep their hands off from it).
To say the FCC has authority over computer networking, adequate to impose network neutrality, because carrying phone calls and the like makes it a "telecommunications service", is also to say that computer networking is NOT an "information service" (a category created just for it) and all the protections against meddling by federal and state governments are moot.
Is that what you want?
If it's an "information service", both the FCC and the states must keep their hands off. (Though there's nothing to stop the FCC from choosing to ask the federal and state courts to block state meddling.)
But that doesn't stop the FTC from regulating the companies' business practices when they fall afoul of federal business law. (At the moment there IS a clause in federal law that the FTC interprets as depowering them, too. But congress could fix that with an almost trivial legislative tweak.)
I've been saying for years that:
- There are good TECHNICAL reasons for treating different packets differently. (For starters, TCP file transfers and streaming don't "play well together" without such QoS management, but can both work just great if it's present and done correctly.)
- The problems we see with non-net-neutrality are actually business pathologies that are already violations of (the spirit, if not the letter, of) laws banning anti-competitive practices, consumer fraud, "tying" (and other pathologies of vertical integration), and the like.
The FTC is very good at going after companies that break such laws. It has some big hammers and is not afraid to use them on big companies - even to the point of breaking them up. (See Standard Oil and mama-bell-era AT&T. IBM "foreign attachments". Microsoft v. Mozilla. Carterfone. And on and on...)
Meanwhile, the FCC is good at many tech issues, but horrible at business competition regulation. (Look at telephony, where they defined "competition" as "at least two players" or even "one player and the threat of another" - and even built that into channel assignment for analog cellphones. Two players aren't competition - they're a duopoly that doesn't even have to communicate to fix prices. You need at least three, and preferably four or more, before market forces drive prices down rather than up.) Trying to apply a tech fix to a business practice issue is a recipe for just breaking something technical (like streaming through network congestion) without solving the underlying problem.
For years I've been saying this is a job for the FTC and NOT the FCC. It's nice to see that the head of the FCC is now saying it, too.
The feed conversion ratio typically compares dry weight of feed to gross live weight of the animal. If you'd compare by nutrients, the numbers can be 3x worse because of water, bones, and intestines. Even for poultry it'd be better to eat the poultry feed directly instead of converting it into chicken.
Bird performance in the commercial poultry industry has shown a staggering improvement over recent decades. Modern broilers weigh about 2.5 kg at 39 days, with a live-weight feed conversion ratio of 1.6 kg of feed per kilogram of body weight gain. Laying hens in modern commercial flocks typically produce about 330 eggs per year with a FCR of 2 kg of feed per kilogram of eggs produced.
For white leghorns the FCR is closer to 1.5.
Maybe that is somewhat inflated by water that isn't counted in the feed weight. But 3x? Hardly. (Birds get a lot of their water from the waste of their energy metabolism.)
Meanwhile: Have you looked at what chickens eat? (I have - in detail - because my wife and I raise the birds.) If you want to make porridge of layer chow, grower chow, insects, grass, etc. you're welcome to try it. (Don't forget to include the fuel and other inputs of any cooking and/or processing you have to do to it to make it digestible by a human.) But I bet you'd have to eat a LOT more than 3x the weight of that gorp to actually absorb and utilize the nutrients you'd get from a roast chicken or a plate of eggs.
I'll let the birdies do the chemical magic of turning that low-grade veggie junk into the raw material for tasty and nutritious meals.
Twenty five plant calories for every one beef calorie. There's no way to rationalize your way around that, no matter how many whatabouts you throw at it.
Yeah? Let's see you digest the calories from the cellulose in sage and clump-grass, without a four-chambered stomach full of symbiotic bacteria. Then tell me it's more efficient to eat desert plants directly than to process them by passing them through a cow.
A shortage of calories is not the issue - as anyone trying to lose weight will tell you. The issue is nutrients: Amino acids, vitamins, trace elements,... Animals synthesize some vital compounds that we don't, and some of those aren't even found in plants. Others aren't found in plants in the levels we need, but are concentrated even by beasts that don't make more.
One example is vitamin B12, a shortage of which causes nerve damage. It's virtually impossible to get it from a pure plant diet. (Fortunately for vegans, some plants have enough insect contamination to keep them alive.)
B12, along with substantial numbers of various sorts of vegetarians in the population, is the reason that the US government limits the amount of folic acid in vitamin supplements - dooming women to morning sickness in early pregnancy: Folic Acid masks the early, reversible, symptoms of B12 deficiency, so vegetarians don't notice it until irreversible nerve damage has occurred.
Another is the essential amino acid Lysine, which is short in things like maize. (Shortage of any essential amino acid stalls protein synthesis, causing a ribosome to be "stuck" waiting for the missing piece when it's time to add it to the growing chain.) There is some question whether the cannibalism in the Inca empire (where corn was the main vegetable staple) was partly driven by the need to conserve and recycle the available lysine. Though the plants may have proteins that average low lysine and there may be less muscle tissue on a lysine-starved animal, the muscle tissue that IS there has the normal number of LYS residues.
For the same amount of calories there would be fewer overall plants consumed if you ate the plants directly rather than going through the meat middleman.
Not necessarily. Herbivores' digestive systems are more efficient, especially after thousands of years of selective breeding. (I hear some domestic animals (e.g. meat chickens) can turn feed into meat at nearly 2:1.) Pushing a pure veggie diet through our non-optimized omnivore digestive tracts might end up wasting more due to our inefficiency than is lost on the extra step between the veggies and us.
Also: Some regions (e.g. western range land) can't grow anything we can eat without massive reengineering and water importation. But it can grow lots of stuff that cattle can eat (much of which is invasive plants that the native animals DON'T eat and which thus take over and wipe out the native fodder if you don't add cattle, who consider it a treat). The most efficient - indeed, nearly the only, way to get human nutrition off this land is to let cattle and/or native quadrupeds eat the low-water weeds and then eat THEM.
People are dying right now in the worst storm since Andrew and Camille and you are posting about some STEM nonsense.
So?
Talking about nerd stuff on a nerd board doesn't in any way prevent talking, or doing something, about such a storm and the people in harm's way.
Even on the same site.
Meanwhile, just because, for the people in the storm, everything else stops while they work on staying alive, doesn't mean that everything stops for everybody else. Quite the contrary: If we let it make everything else stop we've just massively increased the harm done by the storm.
Sure we help out where we can. But we don't drop everything else we do.
This same guy and [others in the US "intelligence community"] have been able to do the same thing w... for over a decade and a half
Quite. They can, and do, do everything this alleged hardware hack is alleged to enable, and more. Since Snowden that's solidly on the public record, manuals and all. Since the Shadow Brokers, lots of others have been able to do some of it and/or see how it works.
Seems to me they are trying to tone down the outrage - because if it really gets going, it might (finally) be turned on them.
What's the big deal if the Chinese came up with the capability, but had to put a chip on the boards to make it happen, rather than get Intel and AMD build it into their own chip sets?
Natural gas is always cheaper than coal - assuming a good supply.
It's also the opposite end of the fossil fuel energy vs carbon scale from coal.
Fossil fuels are mostly either hydrocarbons or carbon with no hydrogen. Hydrogen provides most of the energy when burned, the carbon some, but mostly it's a backbone to stick the hydrogens together into a liquid or gas.
Hydrocarbons have two hydrogens per carbon, plus two extra (unless some carbon bonds are used up making rings or double-bonds). So the smaller the molecule the higher the hydrogen fraction,with methane (CH4), the bulk of natural gas, having the most energy for the least carbon. Coal, of course, is just carbon (plus traces of impurities).
Why put the chip on the Ethernet connector? You know this doesn't decrypt encrypted traffic.
To give it the ability to exchange command-and-control traffic with a remote controller while keeping it from the rest of the system (by "eating" the incoming packets for itself without handing them to the processor's stack, and sending outbound packets directly, again without processing them through the rest of the system.)
This is both convenient, and lets the C&C communicate with the victim box even when the bulk of the victim is shut down.
The Ethernet controller has lots of processing power to play with once it's subverted, control-channel access to the board management system, and already has power-when-the-system-is-down specifically so it can hear the wake-on-LAN packets and bring the machine up to full function - one less mechanism to build.
That's exactly what Intel did when they first started doing Management engines. It was only later versions where they moved it in deeper.
Funny thing, the fossil fuel energy companies have about $107 TRILLION in fuel reserves that have not yet been pumped/mined
Not to mention unproved reserves. When you have enough found for the next 25 years or so you lose drastically if you spend current money looking for more, which you won't use for decades, rather than waiting and using the money for something that will pay off sooner. So you only have more than that amount when a big discovery has pushed the total above that limit by accident.
(That's why it always looks like we'll run out of mined/drilled resources in about 30 years, and has for well over 50.)
For instance: With something close to 300 miles of range an electric in Silicon Valley could handle both commutes and weekend excursions...
Los Angels - Las Vegas similarly, with a minimum of 350 miles range.
SF - LA is nearly 400 miles (383.1 center-to-center), the cities are spread out and the traffic is snared at both ends. So you need about 500 mile range for an electric to be practical for trips between them.
Range of gas/diesel cars has historically grown to save you driving to a gas station every day or every few days.
Within the constraint of being able to make cross-country trips without long pauses for refueling.
Because battery charging, even when blazingly fast with recent cell types, is still longer than a human-maintenance pit stop, they don't do well on long trips. So they need to have larger range than a fuel car or they are limited to a subset of the service types and you need two vehicles.
For instance: With something close to 300 miles of range an electric in Silicon Valley could handle both commutes and weekend excursions to Tahoe-area ski resorts or Reno-area entertainment and gaming, with a safety margin for contingencies like getting stuck in traffic jams during foul mountain weather. Under 250 miles of range and forget it.
(Don't say "rent something else for those non-commute trips". The time and hassle comes out of the precious vacation time, the cost from the budget, a breakdown is more likely and kills the vacation, and it leaves you driving a strange vehicle, a safety issue. You might consider that for a two-week vacation. But it kills weekenders, which is a drastic drop in quality of life.)
I don't run chrome on my desktop and laptop for several reasons, but on my android smartphone, which includes Chrome as the default browser, I don't run it because of the licensing.
The Chrome license includes an Adobe license, due to their using an Adobe product as a component.
The Adobe license, while apparently intended as a no-reverse-engineering prohibition, amounts to a non-compete that would forever taint my ability to work on software similar to Adobe's.
(Clicking "accept" on a cellphone can be expected to leave a database record that is personally identifiable by its connection to my cellphone account with the carrier.)
So I haven't activated it. At the moment, having not gotten around to installing a replacement browser (and figuring out how to do so without use of the un-enabled Chrome browser) I don't have browser-based functions on the cellphone. This largely cripples it as a smartphone.
(Fortunately I can get by using it as a "dumb phone" and tethering it to my laptop (currently using Firefox) when I really need to get on the net. The latter works great. But it's a pain if I'm on the move.)
Energy storage solutions alone will not save solar power because such storage doesn't care if it is charged from solar power or anything else.
But the cost of getting the energy to the storage must be included.
Large fuel-powered steam generation plants and their distribution networks have had a century of intense engineering and are currently nibbling away at the last slivers of inefficiency between their current state and theoretical limits, such as the Carnot cycle and conservation of energy. Any nontrivial cost-of-power improvements there will come from fuel prices and construction/operation cost improvements. Solar photovoltaic is primarily a semiconductor technology and is still following its own version of Moore's Law, with intermittent breakthroughs (like some announced here recently) combining into a rapid improvement in manufacturing cost-per-watt and with some pad left between current efficiency and theoretical limits.
At a good solar site, as of a couple years ago, using current moderately-OK storage, photovoltaic power was cheaper than grid power by a factor of three or better before the recent anti-dumping tariffs With the tariffs in place it's still better and by a factor of about 1.5. Add in improved storage and/or continued improvement of panels, and grid, which has already fallen behind, gets left in the dust. This doesn't kill it, but it does leave it powering mostly sites that don't have good solar resources, require larger amounts of power than is available from the sun on the site's building roofing, where capital isn't available for installing the solar equipment, or serving as a backup/peaking supply.
Solar doesn't need "saving" from natural gas. To a large extent it's the other way around.
Slashdot Mirror
With all the voter suppression happening and voting machines flipping votes, you will not have to worry about it any more.
Showing you're on the left. If you were on the right you'd have said:
With all the illegal immigration happening and the fake voters created by the combination of unexamined mail-in registration and no-appearance-no-hardship absentee (mail in) ballots, you will not have to worry about it any more.
Both ideas are a hazard, whether or not they're true. The purpose of an election is to convince the losing side they'd have also lost the war, so they aren't tempted to start one.
They don't have to believe it's rock-solid-honest for that to work, because a close race means they'd probably lose the war because a lot of people who otherwise could care less WOULD care about trying to overturn the election and fight against them. But if they believe it's massively corrupted and the cheaters are now the tyrants, that stops working.
Look at The Battle of Athens, for instance. Or any of several other cases in US history where a corrupt and election-hacking political machine was thrown out by vigilanties. Or the left's tantrum after the 2016 election.
And to think all this fuss doesn't change the fact we still have 24 hrs. in a day.
Yes, it does.
Last Saturday had 25 hours. A Saturday in the spring had 23.
A standard time day is a legal construct, not a physical one.
And do you have ANY IDEA how much that complicates automated systems (like building energy management) that have to deal with getting things to happen at the right time for production schedules? *I* do. I had to write some of those programs. B-b doesn't even begin to express it.
Make that: ... (about five, IIRC).
It also substantially improved the reliability and maintainability. (For starters, you can shut down the magnet, warm it up, open it up, disconnecting the windings, replace the reaction chamber liner, put it back together, cool them down, and resrart it. You don't have to replace the winding after one use.)
I'd love to see a diesel engine running on fusion.
But Tokamaks aren't dead end any more. Recent improvements in superconducting ribbon have improved the strength of confinement magnets by a factor of several (about five, IARC)
It also substantially improved the reliability and maintainability. (For starters,he reaction chamber liner, put them back together, cool them down, and resrart them. You don't have to replace the winding after one use.)
This moves Tokamaks from "breakeven is a job for a multi-state-actor decades-long research project" to "past breakeven to substantial net output is a several year job for a university or sartup, with a few million in funding".
- interconnections to the rest of the world via paving the remaining four faces with fiber optic interconnections. (The bandgap puts the natural frequency of a diamond LED/laser in the near ultraviolet, and the carrier mobility allows extreme bit rates.)
- Enclosed in a container filled with a vacuum or oxygen-free gas (so it can be run at temperatures of, say, an orange glow without catching fire.) High temperature gradients can push a lot of heat across the coolling paths, while the strength of the diamond would allow it to survive the stresses.
The joy of the scenario, of course, is that this would look like the sort of ship's brain you might find in a golden-age-of-SF novel - say the Skylark series by E. E. Smith. (Six foot cube of diamond in a vacuum or gas tube enclosure, glowing orange, supported by water cooled silver bus bars, etc. B-) )
Why don't we have it already? Because of the difficulty of fabricating a collection of transistors that size with the necessary perfection (or even "good enough" to function with error correcting codes and redundant logic). I had, and still have, a possible solution to that, as well, applicable to technologies other than diamond semiconductors as well. But the technology is JUST NOW bringing it within reach - half a century later. So I'll withhold that secret-sauce recipe, just in case I get a chance to develop and patent it. B-)
Learn some more about thermal dissipation, and then ask yourself why your brilliant idea hasn't been implemented already...
I knew about that and took it into account even at the time (before making part of my career designing things like network processor chips - a field in which I have several patents.)
The original Preposterous Scale Integration pipe dream included: ...) In the Amdahl rectangular hack dimensionality these would be the two broadest and closest faces.
- A six-foot cube of semiconductor. (That was before I knew about Amdahl's elevator hack. Cutting it down to 3'x7'x5' only cuts the volume about in half.)
- Diamond for the semiconductor. (It has EXTREME thermal conductivity, and charge carrier mobility, while the 5.47 eV bandgap allows it to run VERY hot before thermal noise is an issue- though it increases dissipation due to the high switching energy.
- Power and cooling provided across two of the six faces via water-cooled silver bus bars. (There are better alternative now, but wait for it
- Interconnections with the rest of the world
The end of moore's law is the problem. ... We need 8k per eye, and graphics cards need to be at least 4-8x if not 16x as fast for that to be viable, and given the end of moore's law this seems unlikely.
Like the multi-decade running gag "Imminent death of the Internet" predicted, film at 11:00", the end of Moore's Law has been long predicted and slow in coming.
Lately we're starting to hit a wall on the "just make the features smaller and being closer together makes things run faster" approach: As features near the scale of the size of the electron wave function keeping the signals separate becomes an issue.
But Moore's law is really about the number of transistors you can build into a chip, and we've literally "just scratched the surface" there. Until recently the transistors have been essentially a two-dimensional layer on the surface of a chip. Now we're starting to use the third dimension - up to eight layers of it, last time I looked.
The limit to that is what I called "Preposterous Scale Integration" back in the mid-sixites (when the buzzwords were "small ...", "medium ..." and "large scale integration", rather than, say, the "deep sub-micron" of the millennium or the current "xx nanometer process"). When the chip has layers of logic elements spaced along the z axis as closely as the feature size in the x y plane, and acceptable yields are obtained for "chips" the size of the old IBM mainframe cabinets - just small enough to fit through a standard elevator door - THEN you're approaching the end of Moore's Law. That's still many decades out.
Now for a single thread you're currently starting to fall off the speed-doubles-every-1.5 or so-years interpretation of Moore's law. Having lots of transistors doesn't get you faster once you're simultaneously running into limits of feature size AND switching speed AND propagation time. But for computations that are massively parallelizable, you can continue to throw more and more elements at them. Graphics rendering, combined with speculatively pre-computing multiple viewpoints for head rotation and translation options (i.e. the core of VR and AR), is such a problem.
if it comes packaged as a service and you access a website, portal, or online content to use it, then you dont own it. Read the terms of service ...
Even without terms of service that claim ownership of the data, when the data IS yours and nobody claims otherwise: If it's stored on an external service the supreme court has repeatedly ruled that that you have no "reasonable expectation of privacy".
Actually, [...] with nukes, hydro, and Geothermal, would be the best places to locate battery plants. They are already clean.
Their current clenliness doesn't matter.
What matters is the clenliness of the power supply capacity they ADD to run the new loads.
And if the FCC thinks that the internet is not telecommunications it's packed full of incompetent idiots.
You're thinking the plain-English meaning of telecommunications. The FCC is working with the terms of legal art, as defined by the congress in the enabling acts and interpreted by the court.
In those terms, there is a distinction between "telecommunication services" (the default when one-way or two-way information transmission, like telegraph, telephone, radio, television, etc. are involved) and information services (a carve-out for computer networking, defined by later federal legislation which also tells both the FCC and the states to keep their hands off from it).
To say the FCC has authority over computer networking, adequate to impose network neutrality, because carrying phone calls and the like makes it a "telecommunications service", is also to say that computer networking is NOT an "information service" (a category created just for it) and all the protections against meddling by federal and state governments are moot.
Is that what you want?
If it's an "information service", both the FCC and the states must keep their hands off. (Though there's nothing to stop the FCC from choosing to ask the federal and state courts to block state meddling.)
But that doesn't stop the FTC from regulating the companies' business practices when they fall afoul of federal business law. (At the moment there IS a clause in federal law that the FTC interprets as depowering them, too. But congress could fix that with an almost trivial legislative tweak.)
I've been saying for years that:
- There are good TECHNICAL reasons for treating different packets differently. (For starters, TCP file transfers and streaming don't "play well together" without such QoS management, but can both work just great if it's present and done correctly.)
- The problems we see with non-net-neutrality are actually business pathologies that are already violations of (the spirit, if not the letter, of) laws banning anti-competitive practices, consumer fraud, "tying" (and other pathologies of vertical integration), and the like.
The FTC is very good at going after companies that break such laws. It has some big hammers and is not afraid to use them on big companies - even to the point of breaking them up. (See Standard Oil and mama-bell-era AT&T. IBM "foreign attachments". Microsoft v. Mozilla. Carterfone. And on and on...)
Meanwhile, the FCC is good at many tech issues, but horrible at business competition regulation. (Look at telephony, where they defined "competition" as "at least two players" or even "one player and the threat of another" - and even built that into channel assignment for analog cellphones. Two players aren't competition - they're a duopoly that doesn't even have to communicate to fix prices. You need at least three, and preferably four or more, before market forces drive prices down rather than up.) Trying to apply a tech fix to a business practice issue is a recipe for just breaking something technical (like streaming through network congestion) without solving the underlying problem.
For years I've been saying this is a job for the FTC and NOT the FCC. It's nice to see that the head of the FCC is now saying it, too.
The feed conversion ratio typically compares dry weight of feed to gross live weight of the animal. If you'd compare by nutrients, the numbers can be 3x worse because of water, bones, and intestines. Even for poultry it'd be better to eat the poultry feed directly instead of converting it into chicken.
From WattAgNet.com:
For white leghorns the FCR is closer to 1.5.
Maybe that is somewhat inflated by water that isn't counted in the feed weight. But 3x? Hardly. (Birds get a lot of their water from the waste of their energy metabolism.)
Meanwhile: Have you looked at what chickens eat? (I have - in detail - because my wife and I raise the birds.) If you want to make porridge of layer chow, grower chow, insects, grass, etc. you're welcome to try it. (Don't forget to include the fuel and other inputs of any cooking and/or processing you have to do to it to make it digestible by a human.) But I bet you'd have to eat a LOT more than 3x the weight of that gorp to actually absorb and utilize the nutrients you'd get from a roast chicken or a plate of eggs.
I'll let the birdies do the chemical magic of turning that low-grade veggie junk into the raw material for tasty and nutritious meals.
Twenty five plant calories for every one beef calorie. There's no way to rationalize your way around that, no matter how many whatabouts you throw at it.
Yeah? Let's see you digest the calories from the cellulose in sage and clump-grass, without a four-chambered stomach full of symbiotic bacteria. Then tell me it's more efficient to eat desert plants directly than to process them by passing them through a cow.
A shortage of calories is not the issue - as anyone trying to lose weight will tell you. The issue is nutrients: Amino acids, vitamins, trace elements, ... Animals synthesize some vital compounds that we don't, and some of those aren't even found in plants. Others aren't found in plants in the levels we need, but are concentrated even by beasts that don't make more.
One example is vitamin B12, a shortage of which causes nerve damage. It's virtually impossible to get it from a pure plant diet. (Fortunately for vegans, some plants have enough insect contamination to keep them alive.)
B12, along with substantial numbers of various sorts of vegetarians in the population, is the reason that the US government limits the amount of folic acid in vitamin supplements - dooming women to morning sickness in early pregnancy: Folic Acid masks the early, reversible, symptoms of B12 deficiency, so vegetarians don't notice it until irreversible nerve damage has occurred.
Another is the essential amino acid Lysine, which is short in things like maize. (Shortage of any essential amino acid stalls protein synthesis, causing a ribosome to be "stuck" waiting for the missing piece when it's time to add it to the growing chain.) There is some question whether the cannibalism in the Inca empire (where corn was the main vegetable staple) was partly driven by the need to conserve and recycle the available lysine. Though the plants may have proteins that average low lysine and there may be less muscle tissue on a lysine-starved animal, the muscle tissue that IS there has the normal number of LYS residues.
For the same amount of calories there would be fewer overall plants consumed if you ate the plants directly rather than going through the meat middleman.
Not necessarily. Herbivores' digestive systems are more efficient, especially after thousands of years of selective breeding. (I hear some domestic animals (e.g. meat chickens) can turn feed into meat at nearly 2:1.) Pushing a pure veggie diet through our non-optimized omnivore digestive tracts might end up wasting more due to our inefficiency than is lost on the extra step between the veggies and us.
Also: Some regions (e.g. western range land) can't grow anything we can eat without massive reengineering and water importation. But it can grow lots of stuff that cattle can eat (much of which is invasive plants that the native animals DON'T eat and which thus take over and wipe out the native fodder if you don't add cattle, who consider it a treat). The most efficient - indeed, nearly the only, way to get human nutrition off this land is to let cattle and/or native quadrupeds eat the low-water weeds and then eat THEM.
May the Lord have mercy on your soul
If one wants freedom one must ignore the petty totalitarians who would manipulate via social pressure.
My soul and The Lord's mercy on it are an issue between the two of us. Third parties need not bother meddling.
People are dying right now in the worst storm since Andrew and Camille and you are posting about some STEM nonsense.
So?
Talking about nerd stuff on a nerd board doesn't in any way prevent talking, or doing something, about such a storm and the people in harm's way.
Even on the same site.
Meanwhile, just because, for the people in the storm, everything else stops while they work on staying alive, doesn't mean that everything stops for everybody else. Quite the contrary: If we let it make everything else stop we've just massively increased the harm done by the storm.
Sure we help out where we can. But we don't drop everything else we do.
This same guy and [others in the US "intelligence community"] have been able to do the same thing w... for over a decade and a half
Quite. They can, and do, do everything this alleged hardware hack is alleged to enable, and more. Since Snowden that's solidly on the public record, manuals and all. Since the Shadow Brokers, lots of others have been able to do some of it and/or see how it works.
Seems to me they are trying to tone down the outrage - because if it really gets going, it might (finally) be turned on them.
What's the big deal if the Chinese came up with the capability, but had to put a chip on the boards to make it happen, rather than get Intel and AMD build it into their own chip sets?
Natural gas is always cheaper than coal - assuming a good supply.
It's also the opposite end of the fossil fuel energy vs carbon scale from coal.
Fossil fuels are mostly either hydrocarbons or carbon with no hydrogen. Hydrogen provides most of the energy when burned, the carbon some, but mostly it's a backbone to stick the hydrogens together into a liquid or gas.
Hydrocarbons have two hydrogens per carbon, plus two extra (unless some carbon bonds are used up making rings or double-bonds). So the smaller the molecule the higher the hydrogen fraction ,with methane (CH4), the bulk of natural gas, having the most energy for the least carbon. Coal, of course, is just carbon (plus traces of impurities).
TFA says: "Unusual communications from a Supermicro server ..." and on inspection the Ethernet hardware looked odd.
Maybe they just saw some Intel AMT traffic and components. B-)
Why put the chip on the Ethernet connector? You know this doesn't decrypt encrypted traffic.
To give it the ability to exchange command-and-control traffic with a remote controller while keeping it from the rest of the system (by "eating" the incoming packets for itself without handing them to the processor's stack, and sending outbound packets directly, again without processing them through the rest of the system.)
This is both convenient, and lets the C&C communicate with the victim box even when the bulk of the victim is shut down.
The Ethernet controller has lots of processing power to play with once it's subverted, control-channel access to the board management system, and already has power-when-the-system-is-down specifically so it can hear the wake-on-LAN packets and bring the machine up to full function - one less mechanism to build.
That's exactly what Intel did when they first started doing Management engines. It was only later versions where they moved it in deeper.
Funny thing, the fossil fuel energy companies have about $107 TRILLION in fuel reserves that have not yet been pumped/mined
Not to mention unproved reserves. When you have enough found for the next 25 years or so you lose drastically if you spend current money looking for more, which you won't use for decades, rather than waiting and using the money for something that will pay off sooner. So you only have more than that amount when a big discovery has pushed the total above that limit by accident.
(That's why it always looks like we'll run out of mined/drilled resources in about 30 years, and has for well over 50.)
For instance: With something close to 300 miles of range an electric in Silicon Valley could handle both commutes and weekend excursions...
Los Angels - Las Vegas similarly, with a minimum of 350 miles range.
SF - LA is nearly 400 miles (383.1 center-to-center), the cities are spread out and the traffic is snared at both ends. So you need about 500 mile range for an electric to be practical for trips between them.
Range of gas/diesel cars has historically grown to save you driving to a gas station every day or every few days.
Within the constraint of being able to make cross-country trips without long pauses for refueling.
Because battery charging, even when blazingly fast with recent cell types, is still longer than a human-maintenance pit stop, they don't do well on long trips. So they need to have larger range than a fuel car or they are limited to a subset of the service types and you need two vehicles.
For instance: With something close to 300 miles of range an electric in Silicon Valley could handle both commutes and weekend excursions to Tahoe-area ski resorts or Reno-area entertainment and gaming, with a safety margin for contingencies like getting stuck in traffic jams during foul mountain weather. Under 250 miles of range and forget it.
(Don't say "rent something else for those non-commute trips". The time and hassle comes out of the precious vacation time, the cost from the budget, a breakdown is more likely and kills the vacation, and it leaves you driving a strange vehicle, a safety issue. You might consider that for a two-week vacation. But it kills weekenders, which is a drastic drop in quality of life.)
I don't run chrome on my desktop and laptop for several reasons, but on my android smartphone, which includes Chrome as the default browser, I don't run it because of the licensing.
The Chrome license includes an Adobe license, due to their using an Adobe product as a component.
The Adobe license, while apparently intended as a no-reverse-engineering prohibition, amounts to a non-compete that would forever taint my ability to work on software similar to Adobe's.
(Clicking "accept" on a cellphone can be expected to leave a database record that is personally identifiable by its connection to my cellphone account with the carrier.)
So I haven't activated it. At the moment, having not gotten around to installing a replacement browser (and figuring out how to do so without use of the un-enabled Chrome browser) I don't have browser-based functions on the cellphone. This largely cripples it as a smartphone.
(Fortunately I can get by using it as a "dumb phone" and tethering it to my laptop (currently using Firefox) when I really need to get on the net. The latter works great. But it's a pain if I'm on the move.)
(Lenovo's touchpad hits "[Submit]" once again...)
Energy storage solutions alone will not save solar power because such storage doesn't care if it is charged from solar power or anything else.
But the cost of getting the energy to the storage must be included.
Large fuel-powered steam generation plants and their distribution networks have had a century of intense engineering and are currently nibbling away at the last slivers of inefficiency between their current state and theoretical limits, such as the Carnot cycle and conservation of energy. Any nontrivial cost-of-power improvements there will come from fuel prices and construction/operation cost improvements. Solar photovoltaic is primarily a semiconductor technology and is still following its own version of Moore's Law, with intermittent breakthroughs (like some announced here recently) combining into a rapid improvement in manufacturing cost-per-watt and with some pad left between current efficiency and theoretical limits.
At a good solar site, as of a couple years ago, using current moderately-OK storage, photovoltaic power was cheaper than grid power by a factor of three or better before the recent anti-dumping tariffs With the tariffs in place it's still better and by a factor of about 1.5. Add in improved storage and/or continued improvement of panels, and grid, which has already fallen behind, gets left in the dust. This doesn't kill it, but it does leave it powering mostly sites that don't have good solar resources, require larger amounts of power than is available from the sun on the site's building roofing, where capital isn't available for installing the solar equipment, or serving as a backup/peaking supply.
Solar doesn't need "saving" from natural gas. To a large extent it's the other way around.
are a cheap and efficient way to turn that into electricity.
At grid scale, yes. At residential scale, not so much.
Energy storage solutions alone will not save solar power because such storage doesn't care if it is charged from solar power or anything else.
But the cost of getting the energy to the storage must be included.
Solar pa