And how does that help you with data structures in which the access sequence is data-dependent even over smaller pieces of data?
Generally, if you're scattering over different row selects in RAM, you stall the CPU about 200 FSB cycles or 2,000 cycles for a 10x multiplier when you jump around in RAM. That means if the data is all in RAM to begin with and you spend 20 cycles processing, then jump to some data 40 megabytes away, you spend roughly 99.0099% of your time stalled waiting for CPU cache miss. To get around this, you'd have to use CPU prefetch instructions to load the upcoming data into L1.
Access structures as such tend to be packed to consider CPU cache for that reason--see Judy Trees--and, as such, you'd be fetching blocks of at least 4K if not stuff packed localized to 4M (huge pages) to take advantage of the required packing.
In other words: structures that aren't decently optimizable by reading the data in from disk ahead of time are hella slow even if in RAM.
Since I don't see how madvise could tweak CPU cache logic which is apparently hard-coded
madvise() lets you mmap() 128TiB of disk-backed data (files) and then tell the kernel that span A is no longer needed (so discard it from memory) and span B will be needed very soon (so read it into RAM now instead of waiting for the upcoming read() call). You'll need a separate CPU-level prefetch instruction after that read due to cache flushing on the context switch, except on modern architectures which can context switch to supervisor and back without a thread switch and not flush all user-context cache.
No, you're doing this wrong. That "really expensive computer-controlled device" has a driver, right? The supplier of that device is the vendor. The vendor needs to update the driver they wrote.
Someone didn't sell you a medical device with a PCI card someone sold them that didn't have drivers. If they did, well, they wrote the driver, and can update it; otherwise, they can go to the vendor of the PCI card and tell them they need to update the driver.
Read-ahead protocols allow you to identify further data sets and bring them in and out of memory faster than algorithmic performance. The fastest pattern is a giant linear read, and you can issue a DMA to read in the next several hundred megabytes and expire the prior without the CPU being further involved.
Algorithms that process more-complex data sets generally need instrumentation code to identify where the next addresses are, which can be ordered to occur before processing: instead of identify an array of 300, process it, then read off the next address and move your attention there, you would identify the array of 300, skip it, read the next address, issue the read-ahead, and process. This ordering only really adds the call for read-ahead (an OS madvise() call, really) on top of all other work.
For sufficiently-small data sets, issuing readaheads at every junction is brutal. For large datasets, scattering across RAM would incur an incredible amount of lag due to long RAS/CAS cycles, and CPU prefetch calls are required for performance. A "large data set" is one that your CPU is not sufficiently fast enough to process in a small time.
A cashier scans 980 items per hour; scan 2,000 items instead of 3,000 and you only need 2 cashiers.
A 40-foot truck carries 20,000 pairs of pants or 20,000 jackets. If you sell 20,000,000 jackets per week, you need 1,000 trucks driving between the distribution centers and the retail centers each week; if you sell 10,000 jackets per week, you only need 500 trucks driving those deliveries per week.
Fewer trucks driven means fewer inventory workers unloading the trucks onto store shelves. It means fewer warehouse workers loading and unloading pallets into trucks.
Moving less product means fewer cashiers and stockers. If in an area you reduce the load by 50%, you can just close 50% of your stores, meaning infrastructure support to maintain the building, deliver electricity, and provide Internet connectivity goes away. Since we don't need as much of such infrastructure support, somebody's sitting idle and getting paid; fire him, because he does nothing, and we're not receiving a revenue stream to pay him since we're not selling products which he would be making if we actually need him.
We're not allowed to say things like that because it exposes that Manning is potentially the same kind of person as Cosmo or whatever they're calling themselves now: a mentally-imbalanced individual with severe cognitive and emotional control problems due to unaddressed psychiatric issues.
GID is generally handled by gender reassignment. That's usually mostly-benign, although there's a lot of social stress that's secondary. GID is frequently comorbid with serious psychiatric disturbances that look an awful damned lot like schizophrenia or schyzotypical personality disorder.
It's theoretically possible for a schizotypal to manifest their paranoid delusions as GID, along with a belief that Obama is plotting to lead the Muslim invasion from the secret military installation under Denver Airport and implement Shariah Law in the US, and that FEMA has coffins everywhere to put the bodies when they release the plague and kill off 99% of the US population so they can herd the rest into concentration camps for forced labor. Nobody talks publicly about that because identifying a subset of GID cases as downstream from blunt insanity is impolitic.
So yeah, everyone is going to laud this person as some kind of immaculate hero because it's not stylish to do otherwise.
why can't businesses go back to the model of taking care of their employees and figuring out how to build better products to increase profits
"Taking care of their employees" means keeping people on payroll ("instead of always going for the short-term solution of axing payroll"), paying higher wages, paying increased benefits, and so forth.
You point out that Ford can't compete in the market. Remember 2004? Remember when Ford tried to close 10 factories due to a lack of demand for cars, and the UAT forced them to bench the workers? Ford was trying to run a round of layoffs, and was legally-obligated to instead have ten factories which opened up, marched workers in, sat their asses down to chatter and produce nothing, and pay them to stand around playing with their dicks. That is a historical fact.
All of that "taking care of their employees" business means money. Pay them more. Keep them on payroll even if you don't need them. Pay for expensive benefits. Pay them after they stop working (pensions, severance). Where do you think the money comes from?
Who do you think needs to pay for employers to keep people on payroll when people aren't buying their stuff, or when they've figured out how to make their stuff better and with fewer hands?
Your analysis ignores that those lower wages paid to workers also reduce the purchasing power of those workers
Not really. Everyone can't be a higher-paid worker; you get inflation that way. To be clear: if we double everyone's wage, then everyone gets $10,000 instead of $5,000, and they get it for the same hours worked; to pay that wage, the price of products must double; and thus everyone spends the same hours worked on a product and has the same purchasing power. No change.
Wage inequality is a matter of marginalization: some portion of the population has less buying power than some other portion. If you raise that portion of the population's buying power, then you reduce the buying power of the rest of the market.
In practice, this means that raising a subset of 10% of wages by 10% makes those wage earners 9% richer, while making the other 90% poorer by 1%. You lose.1% and gain.099%, a net loss of.001%. Shrug.
The main impact is the shrinking of a market: that big 90% span is mostly able to afford a service--say 70% of the market can afford it. Now you've adjusted it in such a way that the service is 30% more-expensive. People who didn't get the wage adjustment are less-capable of buying it; the ones barely on the demand curve will fall off it, and so the jobs supporting that product are reduced by that much. Maybe 55% of your market can afford it.
If you move a middle-class $60,000 job market to SF and get a middle-class $140,000 job market, then you impact the entire lower market segment flatly, to no real gain anywhere (SF is already a high-wage, expensive locale). If we're talking about a minimum wage increase, then the bottom part of your market gets 9% richer; in the above hypothetical, those people are facing a product that they couldn't afford before, they're 9% richer, and the product costs 30% more, so they still can't afford it. That means you get an actual reduction in ability of them to purchase that product in particular (they can purchase other products which weren't affected).
In other words: raising a subset of wages without raising all wages just concentrates wealth into fewer hands. You get a reduction of overall purchasing power, and an increase in purchasing power in fewer hands.
When you drive down wages across the board, there is no one else paying those higher wages.
When you drive down wages across the board, you get simple deflation.
Now, you ignored two major factors.
The first is that Silicone Valley, San Francisco, and other high-cost areas are faced with prolific economic rent. For example: landlords are taking immense profit margins charging $4,000 for an apartment that would cost $600 in another area. The difference isn't all profit margins: maintenance workers charge higher rates to those landlords, so some of that revenue stream goes to expenses. They might make 50% or 80% profit margins instead of 33%, rather than taking 200%-300% profits, because all the service providers have also jacked up their prices.
In other words: we're concentrating wealth into the hands of landlords, landlord service providers, and the like. The actual workers also have a decent take, since they're still taking home enough to buy a $50,000 car instead of a $25,000 car like we do here on the east coast; and they're still just somewhat-upper-middle-class and not a full 2x as rich as those of us making $60k over here. Richer, yes, just also faced with more parasites sucking them dry.
Second, these jobs produce products purchased by a majority. That, again, means that a tiny, tiny fraction of people are recipients of the personal wages involved here, and a massive proportion of people are impacted down the line. That's the problem with applying microeconomics to macroeconomics problems: you have a distorted myopia on lowerin
They're economizing their time. They could spend hours repairing their clothes, or buy new clothes when theirs start to fray. My clothes fray after a couple years and need replacement or reconstruction.
Here's the thing: I spent $18 on a shirt from China. I don't want to spend $157 on a shirt from America 2 years later when that shirt has a small hole worn in it from being worn two days a week and washed once per week. I also don't want to spend $15 on a spool of thread that's going to last forever, $900 on a sewing machine that can stitch way better than I can hand stitch, and 20-30 minutes doing patch work on my clothing every few weeks (I did that for a while--with $3 spools of thread in 5 colors matched to the clothes and a $120 sewing machine largely made in China, with patch work at least once a month).
Somewhere, a spending habit has to go away. Domestic shipping and its support goes away with it. A reduction in item scans (at 980 scans per hour) means retail cashier jobs go away.
"Dear Sir" isn't used anymore because it indicates you didn't do 5 minutes of googling to find the name of the person you're mailing.
We also still don't call our boss Mr. Jobberson, but instead just call him Dan. I'm sure your grandfather called his boss by a title; nobody does that anymore, and anyone who does tends to make their boss uncomfortable.
Because consumers have to pay the wages. Auto makers have to hold onto employees well-beyond their usefulness because of union contracts, and you have to pay the charity that keeps those people benched instead of out working other jobs making stuff you're buying with the money you're not spending on cars.
Consumers pay wages, as the revenue pays your wage.
Trade tends to improve wealth. I've done the analysis for eliminating Chinese imports of pants and it's pretty hard to not net-loose American jobs unless you pay about minimum wage; that's not the problem, though. The loss or gain of unemployment statistic will buff itself out with labor force growth in a few short years (like 1-3). In all cases, however, you end up with more than doubling the cost of pants--meaning consumers are able to buy less stuff.
Think about what happens when you have a place like Silicon Valley where people pull $150k salaries, and you have to buy a product that takes some hours of work to produce. It's fortunate that technology cuts out so much of the actual labor--that way there aren't as many wage-hours sunk in per delivered good, and we don't have to pay all those wages.
Ultimately, lowering the product of wage times labor means a reduction in prices. Making goods at a lower cost allows you to target broader markets, which lowers the barrier for entry into a market. This increases competition and causes smaller net profit margins. In the case of Apple specifically, they want to position themselves as a luxury good provider, narrowing their market to gain wide margins--the broader market for the iPhone, for example, is smart phones, and Android phones pull one hell of a narrow margin across the board compared to iPhones.
The net effect across all markets is a lowering of costs, lowering of prices, and increase in accessibility of goods. That means wealth among the full span of incomes: the lowest- and middle-class consumers benefit from this.
Understanding all of that requires a grasp of history and an ability to actually look at what happens in the real world, though. Most people prefer to live in a politically-driven fantasy.
Vendors don't need to charge a lot; the cost of doing such an upgrade is fractional compared to the entire development process. "Charging a lot" would be a giant amount of profit.
We can't afford the economics of people having a "moral obligation" to support software for all eternity. When you move from WinXP to Win10, that's one iteration; when you have to support 37 versions of Windows, that's an enormous amount of cross-testing with geometric growth. It's expensive as hell and requires commitment of labor that could do something more-useful.
It is our economic obligation to expire old shit that has been around for 20 years.
So this will probably appeal into a Federal court, where it will get tossed out, and Oculus probably won't think to push for sanctions against Zenimax for frivolous lawsuits and abuse of the court system?
It's been suggested that Zenimax has no interest in VR, and so Carmack quit and used a modified Doom 3 to demo some VR, and implemented some 3D stenciling that he also implemented separately for Zenimax, and so is being sued on tenuous basis.
Like, you got paid to draw a shadow once, so you can never draw another, different shadow again.
BSP 3D rendering, which allowed Quake 1 to run full 3D rendering on a 33MHz 486SX (no math coprocessor) with 8MB RAM and no 3D acceleration, and allows modern graphics cards to produce the 3D renderings they do from modern software.
Shadow volume stencil rendering, which reduces the complexity of rendering shadows in realtime to something reasonable.
Second-hand smoke is not scientifically-demonstrated to cause any health problems whatsoever in adults. Every once in a while, someone comes up with a study showing a weak potential link; these typically fail peer review, and generally only go around in circles while scientists conclude that they're still not able to show any real impact from second-hand smoke at all.
Children in smoking households who breathe the stuff constantly may have trouble from smoke particles. They don't get to inhale high-temperature hot ash, concentrated carbon monoxide, or high amounts of nicotine, most of which is absorbed into the smoker's lungs or dispersed into the air; they just inhale a lot of carbon particulate. Newborn babies don't handle that quite as well as teenagers.
The economy as a whole is a labor trade with a maximum carry capacity. When you increase production rate, you eventually hit a maximum scale, beyond which you require more labor per unit good produced (scarcity), and the cost goes up. Create a technology that scales better (e.g. GMOs produce more yield on the same land area with the same fertilizer, pesticides, irrigation, planting, and harvesting) and your carry capacity goes up.
So your entire economy can produce and buy a certain amount of stuff.
Then: resources are dedicated to producing unnecessary stuff like healthcare and children.
When a laborer dies, the carry capacity doesn't change. A laborer usually consumes for about 20 years, then enters the workforce and consumes and produces for 40 years, then retires and consumes for 15 years. Children and retirees consume less because active workers can buy more luxuries. If the laborer dies at 40, then (eventually) you have a situation of producing new laborers which consume (children) for ~20 years, losing out on 20 years of production and saving yourself 20 years of consumption and 15 years of retiree consumption.
Now you live in a more-poor economy. Whereas we can devote our resources to things like Spotify and Netflix, we instead devote those resources to excessive healthcare and child support. Maybe you still have Netflix, but the market is too small to attract early releases of new movies, and too small for competition, so the licensing fees are high, the prices are high, and there are even fewer Netflix subscribers and so even less market pressure to help push prices down and increase the scope of service. You can buy Netflix for $15.99 instead of $11.99, and you get fewer movies, and you don't get parallel airing of TV series on cable and Netflix.
Weird shit, huh?
On top of that, you pay more for health insurance; that's because your personal risk of illness is higher, along with everyone else's. The costs are leveled out: either you pay a lot and don't get sick, or you pay a lot and do get sick and other people are now paying for your healthcare. In that way, you actually do end up paying for the proportional increase in economic activity devoted to healthcare.
The generating plants are more-efficient, such that a certain lump amount of fuel burned provides more miles driven. It's not one oil plant per car--that would just be a car with a diesel engine.
The problem with all that is real-world concentrations of NOx in the atmosphere are below the toxicity level. There's an LC0 and an LD50; LC0 indicates the level where zero of a population experience any toxic effect. Humans have a pretty good antioxidant delivery system--if Sprague-Dawley rats had similar, they wouldn't live long enough to get cancer (they only live 2 years, although they might live 3-5 instead with less NOx-induced CNS damage)--hence why so much shit doesn't bother us.
By the way, ascorbic acid is an excellent electron donor, and can donate two electrons without gaining enough of a charge to become radicalized. Vitamin C can act as a powerful antioxidant in the body, regulate prolactin, contribute to catecholamine production, and then go on to bind to lead and mercury and eliminate them from the body as lead ascorbate and mercury ascorbate.
Ozone is similarly short-lived (smog is ozone, and we've all seen at least one or two days of smog in our lives, although it looks similar to fog) and generally low in concentration. Soot, however, is a persistent irritant: it leaves the air quickly enough, and instead carries the concern of sticking particles in your lungs. A little ozone won't do any damage (smog will) and will clear out of your body with antioxidants or with exposure to cleaner air; a little soot from a dirty tailpipe passing by will go into your lungs and stay there.
So it's quite possible for NOx and ozone emissions at current levels to contribute to zero deaths. There's some debate around soot: particles larger than 5 micron cause irritation and coughing, while smaller particles don't get coughed out; a reduction in soot is good in any case, and there's some discussion over whether small particles are even-more-harmful than large particles. The current consensus leads to regulations restricting heavy particulate--clean diesel engines need to reduce large soot particles to something smaller than 2.5 micron--and the ideal case is to oxidize more carbon so you get CO2 instead of particulate; reducing heavy particulate necessarily involves processes which oxidize more carbon, so the net result may be positive regardless.
Soot is an actual concern, and a complex concern. Lumping all three together is bad reporting; assuming any amount of NOx or ozone has a non-zero impact on health is bad science.
Slashdot Mirror
And how does that help you with data structures in which the access sequence is data-dependent even over smaller pieces of data?
Generally, if you're scattering over different row selects in RAM, you stall the CPU about 200 FSB cycles or 2,000 cycles for a 10x multiplier when you jump around in RAM. That means if the data is all in RAM to begin with and you spend 20 cycles processing, then jump to some data 40 megabytes away, you spend roughly 99.0099% of your time stalled waiting for CPU cache miss. To get around this, you'd have to use CPU prefetch instructions to load the upcoming data into L1.
Access structures as such tend to be packed to consider CPU cache for that reason--see Judy Trees--and, as such, you'd be fetching blocks of at least 4K if not stuff packed localized to 4M (huge pages) to take advantage of the required packing.
In other words: structures that aren't decently optimizable by reading the data in from disk ahead of time are hella slow even if in RAM.
Since I don't see how madvise could tweak CPU cache logic which is apparently hard-coded
madvise() lets you mmap() 128TiB of disk-backed data (files) and then tell the kernel that span A is no longer needed (so discard it from memory) and span B will be needed very soon (so read it into RAM now instead of waiting for the upcoming read() call). You'll need a separate CPU-level prefetch instruction after that read due to cache flushing on the context switch, except on modern architectures which can context switch to supervisor and back without a thread switch and not flush all user-context cache.
No, you're doing this wrong. That "really expensive computer-controlled device" has a driver, right? The supplier of that device is the vendor. The vendor needs to update the driver they wrote.
Someone didn't sell you a medical device with a PCI card someone sold them that didn't have drivers. If they did, well, they wrote the driver, and can update it; otherwise, they can go to the vendor of the PCI card and tell them they need to update the driver.
I thought 3-way cats got rid of NOx, too.
Doubtful.
Read-ahead protocols allow you to identify further data sets and bring them in and out of memory faster than algorithmic performance. The fastest pattern is a giant linear read, and you can issue a DMA to read in the next several hundred megabytes and expire the prior without the CPU being further involved.
Algorithms that process more-complex data sets generally need instrumentation code to identify where the next addresses are, which can be ordered to occur before processing: instead of identify an array of 300, process it, then read off the next address and move your attention there, you would identify the array of 300, skip it, read the next address, issue the read-ahead, and process. This ordering only really adds the call for read-ahead (an OS madvise() call, really) on top of all other work.
For sufficiently-small data sets, issuing readaheads at every junction is brutal. For large datasets, scattering across RAM would incur an incredible amount of lag due to long RAS/CAS cycles, and CPU prefetch calls are required for performance. A "large data set" is one that your CPU is not sufficiently fast enough to process in a small time.
False.
A cashier scans 980 items per hour; scan 2,000 items instead of 3,000 and you only need 2 cashiers.
A 40-foot truck carries 20,000 pairs of pants or 20,000 jackets. If you sell 20,000,000 jackets per week, you need 1,000 trucks driving between the distribution centers and the retail centers each week; if you sell 10,000 jackets per week, you only need 500 trucks driving those deliveries per week.
Fewer trucks driven means fewer inventory workers unloading the trucks onto store shelves. It means fewer warehouse workers loading and unloading pallets into trucks.
Moving less product means fewer cashiers and stockers. If in an area you reduce the load by 50%, you can just close 50% of your stores, meaning infrastructure support to maintain the building, deliver electricity, and provide Internet connectivity goes away. Since we don't need as much of such infrastructure support, somebody's sitting idle and getting paid; fire him, because he does nothing, and we're not receiving a revenue stream to pay him since we're not selling products which he would be making if we actually need him.
It's hard to claim lesser-evil when one is incompetent and the other is a retard, and yet both have managed to con their way into shitloads of money.
We're not allowed to say things like that because it exposes that Manning is potentially the same kind of person as Cosmo or whatever they're calling themselves now: a mentally-imbalanced individual with severe cognitive and emotional control problems due to unaddressed psychiatric issues.
GID is generally handled by gender reassignment. That's usually mostly-benign, although there's a lot of social stress that's secondary. GID is frequently comorbid with serious psychiatric disturbances that look an awful damned lot like schizophrenia or schyzotypical personality disorder.
It's theoretically possible for a schizotypal to manifest their paranoid delusions as GID, along with a belief that Obama is plotting to lead the Muslim invasion from the secret military installation under Denver Airport and implement Shariah Law in the US, and that FEMA has coffins everywhere to put the bodies when they release the plague and kill off 99% of the US population so they can herd the rest into concentration camps for forced labor. Nobody talks publicly about that because identifying a subset of GID cases as downstream from blunt insanity is impolitic.
So yeah, everyone is going to laud this person as some kind of immaculate hero because it's not stylish to do otherwise.
No, wrong problem. The question was:
why can't businesses go back to the model of taking care of their employees and figuring out how to build better products to increase profits
"Taking care of their employees" means keeping people on payroll ("instead of always going for the short-term solution of axing payroll"), paying higher wages, paying increased benefits, and so forth.
You point out that Ford can't compete in the market. Remember 2004? Remember when Ford tried to close 10 factories due to a lack of demand for cars, and the UAT forced them to bench the workers? Ford was trying to run a round of layoffs, and was legally-obligated to instead have ten factories which opened up, marched workers in, sat their asses down to chatter and produce nothing, and pay them to stand around playing with their dicks. That is a historical fact.
All of that "taking care of their employees" business means money. Pay them more. Keep them on payroll even if you don't need them. Pay for expensive benefits. Pay them after they stop working (pensions, severance). Where do you think the money comes from?
Who do you think needs to pay for employers to keep people on payroll when people aren't buying their stuff, or when they've figured out how to make their stuff better and with fewer hands?
Your analysis ignores that those lower wages paid to workers also reduce the purchasing power of those workers
Not really. Everyone can't be a higher-paid worker; you get inflation that way. To be clear: if we double everyone's wage, then everyone gets $10,000 instead of $5,000, and they get it for the same hours worked; to pay that wage, the price of products must double; and thus everyone spends the same hours worked on a product and has the same purchasing power. No change.
Wage inequality is a matter of marginalization: some portion of the population has less buying power than some other portion. If you raise that portion of the population's buying power, then you reduce the buying power of the rest of the market.
In practice, this means that raising a subset of 10% of wages by 10% makes those wage earners 9% richer, while making the other 90% poorer by 1%. You lose .1% and gain .099%, a net loss of .001%. Shrug.
The main impact is the shrinking of a market: that big 90% span is mostly able to afford a service--say 70% of the market can afford it. Now you've adjusted it in such a way that the service is 30% more-expensive. People who didn't get the wage adjustment are less-capable of buying it; the ones barely on the demand curve will fall off it, and so the jobs supporting that product are reduced by that much. Maybe 55% of your market can afford it.
If you move a middle-class $60,000 job market to SF and get a middle-class $140,000 job market, then you impact the entire lower market segment flatly, to no real gain anywhere (SF is already a high-wage, expensive locale). If we're talking about a minimum wage increase, then the bottom part of your market gets 9% richer; in the above hypothetical, those people are facing a product that they couldn't afford before, they're 9% richer, and the product costs 30% more, so they still can't afford it. That means you get an actual reduction in ability of them to purchase that product in particular (they can purchase other products which weren't affected).
In other words: raising a subset of wages without raising all wages just concentrates wealth into fewer hands. You get a reduction of overall purchasing power, and an increase in purchasing power in fewer hands.
When you drive down wages across the board, there is no one else paying those higher wages.
When you drive down wages across the board, you get simple deflation.
Now, you ignored two major factors.
The first is that Silicone Valley, San Francisco, and other high-cost areas are faced with prolific economic rent. For example: landlords are taking immense profit margins charging $4,000 for an apartment that would cost $600 in another area. The difference isn't all profit margins: maintenance workers charge higher rates to those landlords, so some of that revenue stream goes to expenses. They might make 50% or 80% profit margins instead of 33%, rather than taking 200%-300% profits, because all the service providers have also jacked up their prices.
In other words: we're concentrating wealth into the hands of landlords, landlord service providers, and the like. The actual workers also have a decent take, since they're still taking home enough to buy a $50,000 car instead of a $25,000 car like we do here on the east coast; and they're still just somewhat-upper-middle-class and not a full 2x as rich as those of us making $60k over here. Richer, yes, just also faced with more parasites sucking them dry.
Second, these jobs produce products purchased by a majority. That, again, means that a tiny, tiny fraction of people are recipients of the personal wages involved here, and a massive proportion of people are impacted down the line. That's the problem with applying microeconomics to macroeconomics problems: you have a distorted myopia on lowerin
Read my words: they'll purchase fewer things, which means fewer jobs.
Formation of atmospheric nitric acid is a new one on me; and atmospheric acid is no joke.
They're economizing their time. They could spend hours repairing their clothes, or buy new clothes when theirs start to fray. My clothes fray after a couple years and need replacement or reconstruction.
Here's the thing: I spent $18 on a shirt from China. I don't want to spend $157 on a shirt from America 2 years later when that shirt has a small hole worn in it from being worn two days a week and washed once per week. I also don't want to spend $15 on a spool of thread that's going to last forever, $900 on a sewing machine that can stitch way better than I can hand stitch, and 20-30 minutes doing patch work on my clothing every few weeks (I did that for a while--with $3 spools of thread in 5 colors matched to the clothes and a $120 sewing machine largely made in China, with patch work at least once a month).
Somewhere, a spending habit has to go away. Domestic shipping and its support goes away with it. A reduction in item scans (at 980 scans per hour) means retail cashier jobs go away.
"Dear Sir" isn't used anymore because it indicates you didn't do 5 minutes of googling to find the name of the person you're mailing.
We also still don't call our boss Mr. Jobberson, but instead just call him Dan. I'm sure your grandfather called his boss by a title; nobody does that anymore, and anyone who does tends to make their boss uncomfortable.
Because consumers have to pay the wages. Auto makers have to hold onto employees well-beyond their usefulness because of union contracts, and you have to pay the charity that keeps those people benched instead of out working other jobs making stuff you're buying with the money you're not spending on cars.
Consumers pay wages, as the revenue pays your wage.
Trade tends to improve wealth. I've done the analysis for eliminating Chinese imports of pants and it's pretty hard to not net-loose American jobs unless you pay about minimum wage; that's not the problem, though. The loss or gain of unemployment statistic will buff itself out with labor force growth in a few short years (like 1-3). In all cases, however, you end up with more than doubling the cost of pants--meaning consumers are able to buy less stuff.
Think about what happens when you have a place like Silicon Valley where people pull $150k salaries, and you have to buy a product that takes some hours of work to produce. It's fortunate that technology cuts out so much of the actual labor--that way there aren't as many wage-hours sunk in per delivered good, and we don't have to pay all those wages.
Ultimately, lowering the product of wage times labor means a reduction in prices. Making goods at a lower cost allows you to target broader markets, which lowers the barrier for entry into a market. This increases competition and causes smaller net profit margins. In the case of Apple specifically, they want to position themselves as a luxury good provider, narrowing their market to gain wide margins--the broader market for the iPhone, for example, is smart phones, and Android phones pull one hell of a narrow margin across the board compared to iPhones.
The net effect across all markets is a lowering of costs, lowering of prices, and increase in accessibility of goods. That means wealth among the full span of incomes: the lowest- and middle-class consumers benefit from this.
Understanding all of that requires a grasp of history and an ability to actually look at what happens in the real world, though. Most people prefer to live in a politically-driven fantasy.
Vendors don't need to charge a lot; the cost of doing such an upgrade is fractional compared to the entire development process. "Charging a lot" would be a giant amount of profit.
We can't afford the economics of people having a "moral obligation" to support software for all eternity. When you move from WinXP to Win10, that's one iteration; when you have to support 37 versions of Windows, that's an enormous amount of cross-testing with geometric growth. It's expensive as hell and requires commitment of labor that could do something more-useful.
It is our economic obligation to expire old shit that has been around for 20 years.
So this will probably appeal into a Federal court, where it will get tossed out, and Oculus probably won't think to push for sanctions against Zenimax for frivolous lawsuits and abuse of the court system?
SCO did.
It's been suggested that Zenimax has no interest in VR, and so Carmack quit and used a modified Doom 3 to demo some VR, and implemented some 3D stenciling that he also implemented separately for Zenimax, and so is being sued on tenuous basis.
Like, you got paid to draw a shadow once, so you can never draw another, different shadow again.
BSP 3D rendering, which allowed Quake 1 to run full 3D rendering on a 33MHz 486SX (no math coprocessor) with 8MB RAM and no 3D acceleration, and allows modern graphics cards to produce the 3D renderings they do from modern software.
Shadow volume stencil rendering, which reduces the complexity of rendering shadows in realtime to something reasonable.
New cars are for fools.
Second-hand smoke is not scientifically-demonstrated to cause any health problems whatsoever in adults. Every once in a while, someone comes up with a study showing a weak potential link; these typically fail peer review, and generally only go around in circles while scientists conclude that they're still not able to show any real impact from second-hand smoke at all.
Children in smoking households who breathe the stuff constantly may have trouble from smoke particles. They don't get to inhale high-temperature hot ash, concentrated carbon monoxide, or high amounts of nicotine, most of which is absorbed into the smoker's lungs or dispersed into the air; they just inhale a lot of carbon particulate. Newborn babies don't handle that quite as well as teenagers.
The economy as a whole is a labor trade with a maximum carry capacity. When you increase production rate, you eventually hit a maximum scale, beyond which you require more labor per unit good produced (scarcity), and the cost goes up. Create a technology that scales better (e.g. GMOs produce more yield on the same land area with the same fertilizer, pesticides, irrigation, planting, and harvesting) and your carry capacity goes up.
So your entire economy can produce and buy a certain amount of stuff.
Then: resources are dedicated to producing unnecessary stuff like healthcare and children.
When a laborer dies, the carry capacity doesn't change. A laborer usually consumes for about 20 years, then enters the workforce and consumes and produces for 40 years, then retires and consumes for 15 years. Children and retirees consume less because active workers can buy more luxuries. If the laborer dies at 40, then (eventually) you have a situation of producing new laborers which consume (children) for ~20 years, losing out on 20 years of production and saving yourself 20 years of consumption and 15 years of retiree consumption.
Now you live in a more-poor economy. Whereas we can devote our resources to things like Spotify and Netflix, we instead devote those resources to excessive healthcare and child support. Maybe you still have Netflix, but the market is too small to attract early releases of new movies, and too small for competition, so the licensing fees are high, the prices are high, and there are even fewer Netflix subscribers and so even less market pressure to help push prices down and increase the scope of service. You can buy Netflix for $15.99 instead of $11.99, and you get fewer movies, and you don't get parallel airing of TV series on cable and Netflix.
Weird shit, huh?
On top of that, you pay more for health insurance; that's because your personal risk of illness is higher, along with everyone else's. The costs are leveled out: either you pay a lot and don't get sick, or you pay a lot and do get sick and other people are now paying for your healthcare. In that way, you actually do end up paying for the proportional increase in economic activity devoted to healthcare.
The generating plants are more-efficient, such that a certain lump amount of fuel burned provides more miles driven. It's not one oil plant per car--that would just be a car with a diesel engine.
The problem with all that is real-world concentrations of NOx in the atmosphere are below the toxicity level. There's an LC0 and an LD50; LC0 indicates the level where zero of a population experience any toxic effect. Humans have a pretty good antioxidant delivery system--if Sprague-Dawley rats had similar, they wouldn't live long enough to get cancer (they only live 2 years, although they might live 3-5 instead with less NOx-induced CNS damage)--hence why so much shit doesn't bother us.
By the way, ascorbic acid is an excellent electron donor, and can donate two electrons without gaining enough of a charge to become radicalized. Vitamin C can act as a powerful antioxidant in the body, regulate prolactin, contribute to catecholamine production, and then go on to bind to lead and mercury and eliminate them from the body as lead ascorbate and mercury ascorbate.
Ozone is similarly short-lived (smog is ozone, and we've all seen at least one or two days of smog in our lives, although it looks similar to fog) and generally low in concentration. Soot, however, is a persistent irritant: it leaves the air quickly enough, and instead carries the concern of sticking particles in your lungs. A little ozone won't do any damage (smog will) and will clear out of your body with antioxidants or with exposure to cleaner air; a little soot from a dirty tailpipe passing by will go into your lungs and stay there.
So it's quite possible for NOx and ozone emissions at current levels to contribute to zero deaths. There's some debate around soot: particles larger than 5 micron cause irritation and coughing, while smaller particles don't get coughed out; a reduction in soot is good in any case, and there's some discussion over whether small particles are even-more-harmful than large particles. The current consensus leads to regulations restricting heavy particulate--clean diesel engines need to reduce large soot particles to something smaller than 2.5 micron--and the ideal case is to oxidize more carbon so you get CO2 instead of particulate; reducing heavy particulate necessarily involves processes which oxidize more carbon, so the net result may be positive regardless.
Soot is an actual concern, and a complex concern. Lumping all three together is bad reporting; assuming any amount of NOx or ozone has a non-zero impact on health is bad science.