New GPUs don't necessarily have to drive performance forward or even have a demand for yet more performance. You can use process improvements due to Moore's law and improved designs to produce a smaller GPU capable of the same level of performance as the current parts. This makes your product cheaper and increases profits assuming the price remains fixed. Alternatively it allows you to drop the price as well, which may increase consumption and overall net profit.
The real reason that NVidia feels no pressure to get new product to market is that AMD doesn't have a strong competitor at the high-end of the market and due to the mining craze, you can't even find mid-range AMD GPUs for anywhere near MSRP. This means NVidia is better off spending time and wafers on other product lines where there's more profit to be had.
and there's no shortage of unused land in the world.
Not all "unused" land is equally useless though. If it gets plenty of sunlight, but also plenty of rain, it's being used as farm land and the value of that land is what could have instead been earned if it were being used for farming. Alternatively if it gets plenty of sun and no rain, it's practically perfect for solar and the land will be cheap unless there are a lot of natural resources underneath of it.
The only other issue to consider is proximity to where the power will be used. Long distance transmission isn't is deal-breaker, but we'd like to avoid it if possible since you need to spend more money on lines. We get the same kind of problem where the best land (the least distance) is also more valuable since it's closer to a population center and the value of the land is the houses or businesses that could have been built there instead. If you're already next to existing power lines, that can cut cost considerably.
I suppose you have to factor in environmental costs. There's always some asshole complaining about wind turbines spoiling their view or about how many birds will end up being killed by them. I think the solution to this is easy though if you just move those people next to the dirtiest coal plant imaginable so they're too busy coughing to complain about a few birds.
This prevents unauthorized access. There's no guarantee that it's the police or some other lawful agency that's attempting to unlock your phone without your consent. If the police want access, they can get a warrant. Failure to comply at that point puts you in prison in most jurisdictions so from the perspective of the police, they don't really need to care if they can't actually get into the device.
The problem with your proposed solution is that you assume that these same tracking companies would be wholly incapable of cleaning the polluted data. All they would need is access to the browser that does the polluting and enough time to see how it works and they could probably get above 95% accuracy in terms of removing the fake, polluted data.
It's always a game of cat and mouse. The only way to really stop it is to make a user's data so worthless as to remove the economic incentive to attempt to track them. Unfortunately, that doesn't seem all that likely either.
I'd like to think there was more to it than this because this looks like research that got stitched together from some failed experiment. This AI isn't psychopathic in any way and doesn't even understand that its looking at dead bodies or why that might be considered why that should be viewed in a negative way. If we fed this algorithm internet porn, I'm guessing it would see tits and dicks in the ink blots.
Psychopathy seems to be a condition where a person lacks empathy towards others. It doesn't mean that they will be fascinated with death, merely that they won't feel particularly disgusted at hundreds of people dying, especially if it was in service of furthering their own goals. You can't create a psychopathic AI until you've managed to create an AI that has emotions of some kind.
I don't think this is particularly new or novel either. I remember the Microsoft chatbot from a few years ago that 4chan turned into a shit spouting racist, anti-semite within a day or two. That's pretty much the same thing.
The scary part is that even if you hadn't created an account, they probably have all of that information anyways if enough people you know use Facebook and have tagged you in photographs on their service.
The creepy part is that a lot of information can be derived about you based on the people you have friended. If it's only old acquaintances or classmates, they don't get much, but for regular users they can easily and accurately predict political affiliation and sexual orientation even if you don't fill in either of those categories yourself and don't consume news stories that would give that information away. That information could be sold to third parties or used for other nefarious purposes.
Most of this isn't world ending or life threatening, but it's invasive. And that's just what we know that they're capable of doing. The reality is probably far worse.
It's also the reference design that almost none of the big players actually use. All of Apple, Qualcomm, and NVidia have custom cores. Even Samsung has moved away from the reference design for their Exynos line. I wouldn't be surprised to see one or more of these companies catch or surpass Intel in the next few years. Apple might even do it with their next chip.
Also, for what it's worth Intel's 14 nm process (14++) is also performing better than their own first generation 10 nm process, which is about the equivalent (if you start measuring feature size instead of just looking at the marketing label) of the 7 nm processes from TSMC and Global Foundries. Intel has been stuck at 14 nm for so long that they've refined the hell out of it. Even if they didn't have terrible yield problems with their 10 nm node, the 14 nm parts would still perform better. See the recently announced i3 8181U which has lower clocks than Kaby Lake processors and the same TDP, despite having the GPU fused off.
China is nominally communist at best these days. The central government still likes to exert a lot of authority, but they're perfectly happy having a market economy as long as everyone recognizes their authority and doesn't speak out against the government.
They've allowed for several special economic zones which have some of the least government economic interference on the planet and these places are wildly popular and productive. Shenzhen went from a tiny town (by Chinese standards) to one of the largest financial and technology centers on the planet in last ~40 years. The per capita GDP there (when adjusted for purchasing power parity) is similar to major US cities with a lot of tech companies such as Seattle or San Francisco.
Or someone complained that applying more force in order to brake more quickly wore down the brake pads more quickly or that it increased the amount of harmful particulate matter released into the environment, so they reduced the maximum amount of pressure that could be applied.
The moral of the story is you don't let penny pinchers or greenies design safety critical systems. They're more concerned about their own financial safety or the safety of a damned tree as opposed to yours.
I'm not so sure about that. Certainly back when I went, if you didn't apply in your senior year of high school you missed out on certain opportunities. Even one year off unless you were in community college put you in the "non-traditional" crowd and effectively limited your choices of schools that would accept you.
It doesn't surprise me that schools would attempt to push people towards a path that results in the most money for the school even if it is a disservice to the individuals involved.
In a perfect world where everyone knew what they wanted to do with their life at 18 and was mature enough to take advantage of all of the opportunities they have with regards to education, I'd agree with you completely. However, we know that isn't the case from the data: Only 19% graduate on time.. Worse still 30% will not finish at all essentially dropping or failing out. That is not good and we as a society need to be aware that what we might want and what happens in the real world are two very different things. Kicking and screaming at reality are unlikely to yield different results.
Even minimum wage labor is often a major part of expenses. Consider restaurants where employees are minimum wage or potentially below that in jurisdictions where tips can compensate for wages paid out if they're above a threshold. The leading expense for a restaurant will still be staff, and even if you include the cost of sales with expenses, the labor will be as much or more than the food in most cases. That's why there's any number of restaurants that offer $10 all-you-can-eat buffets. It's no more expensive for a restaurant to give Americans (who can shovel down enough food to make one third of the country obese) as much as they want to eat if it means they don't need to involve wait staff or servers in the transaction. The automated system to replace that human labor isn't going to get high and not show up for work either.
Also, not all jobs can command a living wage for a given area. You make the mistake that assumes all labor is valuable. Should I be guaranteed a living wage if I want to fashion life-sized busts of president Trump from cow manure that I sell by the side of the road in western Oklahoma? You can only pay someone as much as consumers are willing to pay for their labor. If no one in western Oklahoma wants to buy a Trump dung-head, then my labor is not valuable at any price.
Money is merely a commodity and attempts to shuffle it around do not change the underlying reality that there is a certain amount of productivity and that the relative value of any labor cannot be established by fiat. Were that the case, the Soviets would have won the cold war and many Venezuelans wouldn't be facing starvation at the current moment. If you want the most impoverished individuals to be more well off, the only effective method is to increase the overall amount of productivity and wealth in the world. They will still be about as poor relatively speaking, but they can get a tiny slice of a bigger pie. That's why it's not uncommon to see homeless people with cell phones. Productivity improvements and technological advancements have made them so ubiquitous that their within reach of almost all of society.
University education has become quite dysfunctional. There're large chunks of it that server little in the way of educational value and too many students who are either incapable of even those low bars or have no real interest in attending. It's warped into a perverse kind of monster that is incentivized to attract as many students as possible to get their loan money with no real concern for the quality of education. This has led to an adult daycare type of situation where schools will spend large sums of money on student centers, athletics, and anything else that will attract more students. All manner of useless degrees are offered, because the university doesn't care if it does students no good as long as it attracts students and their tuition payments. Worse (or better if you're the college) yet, if you give a student a useless degree, they'll eventually figure out they need to come back for even more college at some point in the future.
We've done today's youth a massive disservice by telling them that they need a college education. If you're going to work retail for the rest of your life, a college education is useless. Similarly, too many kids overlook trade schools at the expense of chasing a four year degree. There are all together too many young people entering college at 18 that don't have any idea what they want to do with their life yet and invariable fail or drop out due to disinterest or confusion. We need to tell people that there's no shame in getting a minimum wage job and figuring out how to be an adult and what you want to do with your life before going to school. I think that message would prevent a large part of the problem.
Removing government backed loans would probably fix the other half. When banks are on the hook, the actuaries will crunch the numbers and quickly realize that loaning $100,000 to someone for a degree in underwater basket weaving is a good investment. If rich kids want to get useless degrees on mommy and daddy's dime, that's their own business, but letting an 18 year old run up six figure debts that they can't hope to repay is just irresponsible on society's part.
This didn't even appear to attempt to do that, but as you point out, even if they had the result would be just as useless for the reasons you've described.
All that I got from the shiny infographic was that they considered two pieces of data which were sex of streamer and whether the stream had received any amount of payment. Both of these are binary classifiers in this case. What I'm suggesting is that at a bare minimum there are other factors that likely influence whether you get paid beyond gender. A few obvious ones that jump out to me are audience size (popularity), time spent streaming, and game(s) streamed. There's no reason to believe that those are consistent across the sexes so you need to control for those factors.
Here's another example to illustrate at why such naive interpretations of data lead to shitty conclusions. Take the FBI statistics for murder based on perpetrator's race and the U.S. population demographics. If you simply use those two factors and don't control for anything else, you end up with a conclusion that looks like "black men are three times more likely to be murders than white men". We've arrived at using the same technique that we used to demonstrate this supposed wage gap, so if you're will to accept that as valid then you should probably accept what I've posited as valid as well. Typically the people who love their wage gap arguments are abhorred by this example of where their lack of detailed analysis leads and will start suggesting that there's a need to control for things like socioeconomic status, family environment, education, and plenty of other things along those lines. It turns out that once you do those things, you explain away the vast majority of that 200% increase, because the black people who are committing murder are disproportionately from single parent households and impoverished and those factors are far better predictors of likelihood to commit murder than skin color.
Let's substitute "older" with "black" or "Jewish" and see if it changes your feelings on the matter. I find that's a useful way to gauge something like this.
Personally, I think that companies that limit their potential pool of applicants and talent for any reason are just shooting themselves in the foot, unless you want to argue that older applicants are less qualified. In some cases they may not have exposure to the latest languages, frameworks, etc. but 20 years of experience is worth a lot in its own way.
The other part of me thinks that these companies want young employees fresh out of college that they can work like a dog for a few years before casting them off. Older employees aren't anywhere near as willing to put up with that shit and likely have life commitments outside of work that don't afford them the opportunity to slog through 60 work weeks.
I'm curious to what extent the data was analyzed. It seems like we've just been given a raw statistic as it just indicates the % of streamers who have never been paid, which outside of the U.S. isn't even that large. Like the pay gap itself, I wouldn't be surprised if after performing multivariate analysis or a linear regression that most of this gap disappears for the same reasons such as men streaming for larger amounts of time and game selection when streaming which likely contribute to audience size. I'm assuming that the viewer base for this is predominantly male and that influences what streams get watched and contributed to.
The overall market is probably very feast or famine judging by the stats that ~40% of streamers don't get paid. I went to Twitch and picked the most popular game (Fortnite) to look at the viewer counts. The game itself has a little over 100,000 at the current moment and of those, about 60% are watching the 4 most popular streamers. After the top 50 streams, viewer counts are only ~180 people. After the top 100, viewer counts are down to ~60. I don't know how total streams there are as they have a scroll to load more style and I quit counting, but it keeps going on and on. I wouldn't be surprised if there were more than 1000, but once you get down somewhere past the top 200, everyone's down to 4 viewers.
Apple will probably remove the app, but does it really matter? I thought that Apple's messaging service had encryption to prevent snooping. Maybe the government doesn't care if they can still find out who you're messaging, but I don't think this is going to stop the people who are really serious about their privacy.
On the other hand, maybe Apple doesn't care either. I don't know what their market share looks like in Russia, but I'd imagine it to be lower than most other countries. Apple might care when it's China come knocking since they need to manufacture their phones their, but if Russia just outright bans them they might not lose all that much.
However, no company refuses revenue no matter how small and Apple is glad to sell you $20 accessories that probably cost them a quarter to manufacture so I can't see them turning down even what little pittance the Russian market might represent. However, stonewalling for a while might make them look tough and concerned about privacy, which might make them more marketable.
I wonder how long until we start seeing more devices or service providers start to implement more defensive mechanisms against this. There's no reason my phone can't misreport its location to the services that ask for it or I can't get an ISP that will anonymize or obfuscate my traffic. I don't think it needs to be something that's always on, just on enough to discourage and disincentivize this kind of behavior.
Unfortunately I don't see it becoming widespread since most people don't value their privacy.
I don't think the people who are starving were going to be spending money on $27,000 digital purchases. You may as well complain about the fact that there are hundred million dollar yachts, hundred thousand dollar cars, and hundred dollar steaks while you're at it. You yourself probably have some luxury device that you overpaid for too. Possibly the very same one you're using to post your comments here.
Also, North America has an obesity problem several times larger than it does with undernourishment. I believe that even the rate for morbid obesity is twice as large. If you're starving in the U.S. or Canada it's either because your out in the sticks and completely cut off from most support systems (and probably your immediate neighbors) or because you're mentally ill and wandering the streets.
There's such an abundance of food here that no one need starve. The real issue is that people would rather lament about it online instead of actually doing anything in the real world themselves to solve even a tiny part of the problem.
Just get a cheap portable Bluetooth speaker that you can cart around for stuff like that instead of putting speakers in every room. Then you can take it with you out to the garage, park, lake, etc. as well and it's going to be more than good enough for those situations. I'm comparing a set of dedicated stereo speakers to a pair of Apple HomePods. If you're getting even just one of those for every room in your house (so you can have music throughout the entire house) you've probably got more money than sense.
I think it's cool technology, but it's overpriced for what it is even if you do consider the usual Apple tax.
From what I saw regarding the design of the speakers, it looks as though each pod has got an array of speakers that is putting sound out in every direction. You should expect that to be able to produce a better overall sound all else equal. If you're pitting it against some cheap speakers that came as part of a surround package, it's probably no contest. I suspect that if you've got a decent pair of quality speakers though that you can probably get better sound quality from them.
Getting two HomePod units is $700 though. You can get a great set of floor standing stereo speakers for that price though and even a pair of really good ones for the cost of just one HomePod. You're also going to get something that can fill a much larger space because each of your really good speakers is going to be able to handle far more power individual than a single HomePod so you'll be able to get loud without getting distorted.
So your argument is that young boys experience less anxiety about math, which means that they are less likely to become disinterested in it or averse to it which leads to higher likelihood of employment in fields that require larger amounts of math, but also that the levels of anxiety experienced by boys/girls in relation to math is not biological? The first part of that sounds perfectly reasonable, but the last half of it is a rejection of reality in favor of constructing a hypothesis to fit a conclusion that has already been drawn.
I'll never be able to wrap my head around why some people believe that men and women can have vastly different biology that leads to differences in sex organs, bone density and skeletal structure, muscle mass, and plenty of other aspects but that those differences will stop at the neck. You can take an MRI scan of a brain and reliably categorize a person as male or female based on what it looks like.
To believe that there's no biological basis for this difference is ignoring common sense on top of a mountain of evidence. The number of studies into infant toy preference show that there are some obvious differences from a very young age. Similar studies conducted with other primates that have found similar results seem to suggest that these differences date quite far back into our evolutionary past.
I'm not even sure why it matters either. If women are less interested in something than men (or vice versa) what does it matter as long as you ensure that the women and men who are interested in pursuing some field of study are able to do so? Women (and by extension men) shouldn't be forced into careers that they don't prefer just to appease people with idiotic notions related to sex and biology. If I go to a mechanic, a dentist, or any other professional, I want someone who's passionate about their job and don't care about whether they're male/female, black/white, gay/straight, etc. Give people the opportunity to do what makes them happiest and they're going to lead more satisfying lives.
Slashdot Mirror
New GPUs don't necessarily have to drive performance forward or even have a demand for yet more performance. You can use process improvements due to Moore's law and improved designs to produce a smaller GPU capable of the same level of performance as the current parts. This makes your product cheaper and increases profits assuming the price remains fixed. Alternatively it allows you to drop the price as well, which may increase consumption and overall net profit.
The real reason that NVidia feels no pressure to get new product to market is that AMD doesn't have a strong competitor at the high-end of the market and due to the mining craze, you can't even find mid-range AMD GPUs for anywhere near MSRP. This means NVidia is better off spending time and wafers on other product lines where there's more profit to be had.
and there's no shortage of unused land in the world.
Not all "unused" land is equally useless though. If it gets plenty of sunlight, but also plenty of rain, it's being used as farm land and the value of that land is what could have instead been earned if it were being used for farming. Alternatively if it gets plenty of sun and no rain, it's practically perfect for solar and the land will be cheap unless there are a lot of natural resources underneath of it.
The only other issue to consider is proximity to where the power will be used. Long distance transmission isn't is deal-breaker, but we'd like to avoid it if possible since you need to spend more money on lines. We get the same kind of problem where the best land (the least distance) is also more valuable since it's closer to a population center and the value of the land is the houses or businesses that could have been built there instead. If you're already next to existing power lines, that can cut cost considerably.
I suppose you have to factor in environmental costs. There's always some asshole complaining about wind turbines spoiling their view or about how many birds will end up being killed by them. I think the solution to this is easy though if you just move those people next to the dirtiest coal plant imaginable so they're too busy coughing to complain about a few birds.
This prevents unauthorized access. There's no guarantee that it's the police or some other lawful agency that's attempting to unlock your phone without your consent. If the police want access, they can get a warrant. Failure to comply at that point puts you in prison in most jurisdictions so from the perspective of the police, they don't really need to care if they can't actually get into the device.
The problem with your proposed solution is that you assume that these same tracking companies would be wholly incapable of cleaning the polluted data. All they would need is access to the browser that does the polluting and enough time to see how it works and they could probably get above 95% accuracy in terms of removing the fake, polluted data.
It's always a game of cat and mouse. The only way to really stop it is to make a user's data so worthless as to remove the economic incentive to attempt to track them. Unfortunately, that doesn't seem all that likely either.
I'd like to think there was more to it than this because this looks like research that got stitched together from some failed experiment. This AI isn't psychopathic in any way and doesn't even understand that its looking at dead bodies or why that might be considered why that should be viewed in a negative way. If we fed this algorithm internet porn, I'm guessing it would see tits and dicks in the ink blots.
Psychopathy seems to be a condition where a person lacks empathy towards others. It doesn't mean that they will be fascinated with death, merely that they won't feel particularly disgusted at hundreds of people dying, especially if it was in service of furthering their own goals. You can't create a psychopathic AI until you've managed to create an AI that has emotions of some kind.
I don't think this is particularly new or novel either. I remember the Microsoft chatbot from a few years ago that 4chan turned into a shit spouting racist, anti-semite within a day or two. That's pretty much the same thing.
That's why you have to use Uber. The only way to be safe from them is to be inside the vehicle.
Fuck everything, we're doing five cameras.
The scary part is that even if you hadn't created an account, they probably have all of that information anyways if enough people you know use Facebook and have tagged you in photographs on their service.
The creepy part is that a lot of information can be derived about you based on the people you have friended. If it's only old acquaintances or classmates, they don't get much, but for regular users they can easily and accurately predict political affiliation and sexual orientation even if you don't fill in either of those categories yourself and don't consume news stories that would give that information away. That information could be sold to third parties or used for other nefarious purposes.
Most of this isn't world ending or life threatening, but it's invasive. And that's just what we know that they're capable of doing. The reality is probably far worse.
It's also the reference design that almost none of the big players actually use. All of Apple, Qualcomm, and NVidia have custom cores. Even Samsung has moved away from the reference design for their Exynos line. I wouldn't be surprised to see one or more of these companies catch or surpass Intel in the next few years. Apple might even do it with their next chip.
Also, for what it's worth Intel's 14 nm process (14++) is also performing better than their own first generation 10 nm process, which is about the equivalent (if you start measuring feature size instead of just looking at the marketing label) of the 7 nm processes from TSMC and Global Foundries. Intel has been stuck at 14 nm for so long that they've refined the hell out of it. Even if they didn't have terrible yield problems with their 10 nm node, the 14 nm parts would still perform better. See the recently announced i3 8181U which has lower clocks than Kaby Lake processors and the same TDP, despite having the GPU fused off.
China is nominally communist at best these days. The central government still likes to exert a lot of authority, but they're perfectly happy having a market economy as long as everyone recognizes their authority and doesn't speak out against the government.
They've allowed for several special economic zones which have some of the least government economic interference on the planet and these places are wildly popular and productive. Shenzhen went from a tiny town (by Chinese standards) to one of the largest financial and technology centers on the planet in last ~40 years. The per capita GDP there (when adjusted for purchasing power parity) is similar to major US cities with a lot of tech companies such as Seattle or San Francisco.
Or someone complained that applying more force in order to brake more quickly wore down the brake pads more quickly or that it increased the amount of harmful particulate matter released into the environment, so they reduced the maximum amount of pressure that could be applied.
The moral of the story is you don't let penny pinchers or greenies design safety critical systems. They're more concerned about their own financial safety or the safety of a damned tree as opposed to yours.
I'm not so sure about that. Certainly back when I went, if you didn't apply in your senior year of high school you missed out on certain opportunities. Even one year off unless you were in community college put you in the "non-traditional" crowd and effectively limited your choices of schools that would accept you.
It doesn't surprise me that schools would attempt to push people towards a path that results in the most money for the school even if it is a disservice to the individuals involved.
In a perfect world where everyone knew what they wanted to do with their life at 18 and was mature enough to take advantage of all of the opportunities they have with regards to education, I'd agree with you completely. However, we know that isn't the case from the data: Only 19% graduate on time.. Worse still 30% will not finish at all essentially dropping or failing out. That is not good and we as a society need to be aware that what we might want and what happens in the real world are two very different things. Kicking and screaming at reality are unlikely to yield different results.
Even minimum wage labor is often a major part of expenses. Consider restaurants where employees are minimum wage or potentially below that in jurisdictions where tips can compensate for wages paid out if they're above a threshold. The leading expense for a restaurant will still be staff, and even if you include the cost of sales with expenses, the labor will be as much or more than the food in most cases. That's why there's any number of restaurants that offer $10 all-you-can-eat buffets. It's no more expensive for a restaurant to give Americans (who can shovel down enough food to make one third of the country obese) as much as they want to eat if it means they don't need to involve wait staff or servers in the transaction. The automated system to replace that human labor isn't going to get high and not show up for work either.
Also, not all jobs can command a living wage for a given area. You make the mistake that assumes all labor is valuable. Should I be guaranteed a living wage if I want to fashion life-sized busts of president Trump from cow manure that I sell by the side of the road in western Oklahoma? You can only pay someone as much as consumers are willing to pay for their labor. If no one in western Oklahoma wants to buy a Trump dung-head, then my labor is not valuable at any price.
Money is merely a commodity and attempts to shuffle it around do not change the underlying reality that there is a certain amount of productivity and that the relative value of any labor cannot be established by fiat. Were that the case, the Soviets would have won the cold war and many Venezuelans wouldn't be facing starvation at the current moment. If you want the most impoverished individuals to be more well off, the only effective method is to increase the overall amount of productivity and wealth in the world. They will still be about as poor relatively speaking, but they can get a tiny slice of a bigger pie. That's why it's not uncommon to see homeless people with cell phones. Productivity improvements and technological advancements have made them so ubiquitous that their within reach of almost all of society.
University education has become quite dysfunctional. There're large chunks of it that server little in the way of educational value and too many students who are either incapable of even those low bars or have no real interest in attending. It's warped into a perverse kind of monster that is incentivized to attract as many students as possible to get their loan money with no real concern for the quality of education. This has led to an adult daycare type of situation where schools will spend large sums of money on student centers, athletics, and anything else that will attract more students. All manner of useless degrees are offered, because the university doesn't care if it does students no good as long as it attracts students and their tuition payments. Worse (or better if you're the college) yet, if you give a student a useless degree, they'll eventually figure out they need to come back for even more college at some point in the future.
We've done today's youth a massive disservice by telling them that they need a college education. If you're going to work retail for the rest of your life, a college education is useless. Similarly, too many kids overlook trade schools at the expense of chasing a four year degree. There are all together too many young people entering college at 18 that don't have any idea what they want to do with their life yet and invariable fail or drop out due to disinterest or confusion. We need to tell people that there's no shame in getting a minimum wage job and figuring out how to be an adult and what you want to do with your life before going to school. I think that message would prevent a large part of the problem.
Removing government backed loans would probably fix the other half. When banks are on the hook, the actuaries will crunch the numbers and quickly realize that loaning $100,000 to someone for a degree in underwater basket weaving is a good investment. If rich kids want to get useless degrees on mommy and daddy's dime, that's their own business, but letting an 18 year old run up six figure debts that they can't hope to repay is just irresponsible on society's part.
This didn't even appear to attempt to do that, but as you point out, even if they had the result would be just as useless for the reasons you've described.
All that I got from the shiny infographic was that they considered two pieces of data which were sex of streamer and whether the stream had received any amount of payment. Both of these are binary classifiers in this case. What I'm suggesting is that at a bare minimum there are other factors that likely influence whether you get paid beyond gender. A few obvious ones that jump out to me are audience size (popularity), time spent streaming, and game(s) streamed. There's no reason to believe that those are consistent across the sexes so you need to control for those factors.
Here's another example to illustrate at why such naive interpretations of data lead to shitty conclusions. Take the FBI statistics for murder based on perpetrator's race and the U.S. population demographics. If you simply use those two factors and don't control for anything else, you end up with a conclusion that looks like "black men are three times more likely to be murders than white men". We've arrived at using the same technique that we used to demonstrate this supposed wage gap, so if you're will to accept that as valid then you should probably accept what I've posited as valid as well. Typically the people who love their wage gap arguments are abhorred by this example of where their lack of detailed analysis leads and will start suggesting that there's a need to control for things like socioeconomic status, family environment, education, and plenty of other things along those lines. It turns out that once you do those things, you explain away the vast majority of that 200% increase, because the black people who are committing murder are disproportionately from single parent households and impoverished and those factors are far better predictors of likelihood to commit murder than skin color.
Let's substitute "older" with "black" or "Jewish" and see if it changes your feelings on the matter. I find that's a useful way to gauge something like this.
Personally, I think that companies that limit their potential pool of applicants and talent for any reason are just shooting themselves in the foot, unless you want to argue that older applicants are less qualified. In some cases they may not have exposure to the latest languages, frameworks, etc. but 20 years of experience is worth a lot in its own way.
The other part of me thinks that these companies want young employees fresh out of college that they can work like a dog for a few years before casting them off. Older employees aren't anywhere near as willing to put up with that shit and likely have life commitments outside of work that don't afford them the opportunity to slog through 60 work weeks.
I'm curious to what extent the data was analyzed. It seems like we've just been given a raw statistic as it just indicates the % of streamers who have never been paid, which outside of the U.S. isn't even that large. Like the pay gap itself, I wouldn't be surprised if after performing multivariate analysis or a linear regression that most of this gap disappears for the same reasons such as men streaming for larger amounts of time and game selection when streaming which likely contribute to audience size. I'm assuming that the viewer base for this is predominantly male and that influences what streams get watched and contributed to.
The overall market is probably very feast or famine judging by the stats that ~40% of streamers don't get paid. I went to Twitch and picked the most popular game (Fortnite) to look at the viewer counts. The game itself has a little over 100,000 at the current moment and of those, about 60% are watching the 4 most popular streamers. After the top 50 streams, viewer counts are only ~180 people. After the top 100, viewer counts are down to ~60. I don't know how total streams there are as they have a scroll to load more style and I quit counting, but it keeps going on and on. I wouldn't be surprised if there were more than 1000, but once you get down somewhere past the top 200, everyone's down to 4 viewers.
I think it's because the world is full of narcissists hunting for new ways to say "look at me!"
As the old saying goes, you don't pay a hooker for sex, you pay her to leave.
Apple will probably remove the app, but does it really matter? I thought that Apple's messaging service had encryption to prevent snooping. Maybe the government doesn't care if they can still find out who you're messaging, but I don't think this is going to stop the people who are really serious about their privacy.
On the other hand, maybe Apple doesn't care either. I don't know what their market share looks like in Russia, but I'd imagine it to be lower than most other countries. Apple might care when it's China come knocking since they need to manufacture their phones their, but if Russia just outright bans them they might not lose all that much.
However, no company refuses revenue no matter how small and Apple is glad to sell you $20 accessories that probably cost them a quarter to manufacture so I can't see them turning down even what little pittance the Russian market might represent. However, stonewalling for a while might make them look tough and concerned about privacy, which might make them more marketable.
I wonder how long until we start seeing more devices or service providers start to implement more defensive mechanisms against this. There's no reason my phone can't misreport its location to the services that ask for it or I can't get an ISP that will anonymize or obfuscate my traffic. I don't think it needs to be something that's always on, just on enough to discourage and disincentivize this kind of behavior.
Unfortunately I don't see it becoming widespread since most people don't value their privacy.
I don't think the people who are starving were going to be spending money on $27,000 digital purchases. You may as well complain about the fact that there are hundred million dollar yachts, hundred thousand dollar cars, and hundred dollar steaks while you're at it. You yourself probably have some luxury device that you overpaid for too. Possibly the very same one you're using to post your comments here.
Also, North America has an obesity problem several times larger than it does with undernourishment. I believe that even the rate for morbid obesity is twice as large. If you're starving in the U.S. or Canada it's either because your out in the sticks and completely cut off from most support systems (and probably your immediate neighbors) or because you're mentally ill and wandering the streets.
There's such an abundance of food here that no one need starve. The real issue is that people would rather lament about it online instead of actually doing anything in the real world themselves to solve even a tiny part of the problem.
Just get a cheap portable Bluetooth speaker that you can cart around for stuff like that instead of putting speakers in every room. Then you can take it with you out to the garage, park, lake, etc. as well and it's going to be more than good enough for those situations. I'm comparing a set of dedicated stereo speakers to a pair of Apple HomePods. If you're getting even just one of those for every room in your house (so you can have music throughout the entire house) you've probably got more money than sense.
I think it's cool technology, but it's overpriced for what it is even if you do consider the usual Apple tax.
From what I saw regarding the design of the speakers, it looks as though each pod has got an array of speakers that is putting sound out in every direction. You should expect that to be able to produce a better overall sound all else equal. If you're pitting it against some cheap speakers that came as part of a surround package, it's probably no contest. I suspect that if you've got a decent pair of quality speakers though that you can probably get better sound quality from them.
Getting two HomePod units is $700 though. You can get a great set of floor standing stereo speakers for that price though and even a pair of really good ones for the cost of just one HomePod. You're also going to get something that can fill a much larger space because each of your really good speakers is going to be able to handle far more power individual than a single HomePod so you'll be able to get loud without getting distorted.
So your argument is that young boys experience less anxiety about math, which means that they are less likely to become disinterested in it or averse to it which leads to higher likelihood of employment in fields that require larger amounts of math, but also that the levels of anxiety experienced by boys/girls in relation to math is not biological? The first part of that sounds perfectly reasonable, but the last half of it is a rejection of reality in favor of constructing a hypothesis to fit a conclusion that has already been drawn.
I'll never be able to wrap my head around why some people believe that men and women can have vastly different biology that leads to differences in sex organs, bone density and skeletal structure, muscle mass, and plenty of other aspects but that those differences will stop at the neck. You can take an MRI scan of a brain and reliably categorize a person as male or female based on what it looks like.
To believe that there's no biological basis for this difference is ignoring common sense on top of a mountain of evidence. The number of studies into infant toy preference show that there are some obvious differences from a very young age. Similar studies conducted with other primates that have found similar results seem to suggest that these differences date quite far back into our evolutionary past.
I'm not even sure why it matters either. If women are less interested in something than men (or vice versa) what does it matter as long as you ensure that the women and men who are interested in pursuing some field of study are able to do so? Women (and by extension men) shouldn't be forced into careers that they don't prefer just to appease people with idiotic notions related to sex and biology. If I go to a mechanic, a dentist, or any other professional, I want someone who's passionate about their job and don't care about whether they're male/female, black/white, gay/straight, etc. Give people the opportunity to do what makes them happiest and they're going to lead more satisfying lives.