The U.S. causes of death statistics are readily available from the CDC website. For 2015, the leading causes of death for the 15-19 year old demographic were:
3,919 deaths - Accidents (mostly automobile accidents and drug overdoses)
2.061 deaths - Suicide
1,587 deaths - Homicide (mostly outside school, and gang related)
583 deaths - Malignant neoplasms (cancer)
306 deaths - Heart disease
195 deaths - Birth defects
72 deaths - Influenza (the flu)
63 deaths - Chronic lower respiratory diseases
61 deaths - Cerebrovascular diseases
52 deaths - Diabetes
41 deaths - Complications from pregnancy and childbirth
Where do school shootings rank? There have been about 250 deaths in school shootings over 18 years, or about 14 per year.(and K-12, not just ages 15-19). Since there are approximately 51 million K-12 students in the U.S., a student's chances of being killed in a non-gang, non-suicide school shooting in any given year are about 1 in 3.6 million. You are roughly 3x more likely to be struck by lightning (1 in 1.08 million).
Like airliner crashes, school shootings are one of these extremely rare, statistically insignificant events whose emotional impact creates a large amount of social interest. This causes a disproportionate amount of press coverage, leading people to wildly overestimate the actual danger. If you really want to save high schoolers' lives, teach them to: drive safely and buckle their seat belts, not to abuse drugs, seek counseling for depression, stay out of gangs, use sunscreen, eat healthy and exercise, get the flu shot, don't smoke, don't eat too many sweets, and avoid teen pregnancy. Each of these will save many more lives than all the hand-wringing over school shootings, some (like suicide-prevention) around a hundred times more.
Texas is #1 only because it has a large population. If you combine the data in your first link with a list of state populations and rates of gun ownership, and calculate the number of guns stolen per owner ( (# of guns stolen) / (state population * % who are owners) ), Texas is 17th, and is statistically almost exactly at the national average (35.9% of Texans own guns vs 32.1% for the U.S., and 0.18% have a gun stolen or 1.03x the national rate). The District of Columbia ends up topping the list with a theft rate a staggering 162 times the U.S. national average.
The huge deviation of DC from the national average (29.3% of DC gun owners have a gun stolen vs 0.18% for the nation) makes me think gun theft is primarily a problem in urban areas, not rural. Further supporting this hypothesis is that Alaska, Montana, Idaho, and North and South Dakota all have gun ownership rates over 50%, but their rate of guns stolen is below the national average.
Also, the low rate for gun theft nationwide (0.18% per owner per year, vs 0.47% burglaries and 1.75% larceny) makes me think outside of certain cities, gun theft is not really a serious issue, and is more incidental property theft rather than targeted, and for the most part gun owners do a pretty good job keeping their guns safe from theft.
Which sort order you want isn't what's important. What's important is that different people will want different sort orders. The correct solution to this problem is to let user pick which sort order they want applied to their feed. But the people making the UI for these things seem to be on an ego trip, and enjoy playing god by deciding how millions of users must use an app, so won't give their users the decency of choosing what they want.
You see the same problem in Apple's walled garden, cellular carriers making certain apps on your phone undeletable, and printer manufacturers trying to restrict which ink cartridges you're allowed to use. VLC had the same problem - I and lots of others wanted to use the mouse wheel to seek (FF/RW) through the video. The main programmer thought it should control the volume, and refused to allow users to use it to seek. It was years before he finally relented and allowed an option to change the wheel's function to seek.
The real problem here is lack of respect for the user. Treating them like cattle instead of customers, just because you can.
LIDAR (and radar, and sonar) is one of those things which sounds great when you consider the car in isolation by itself. But isn't so great once you have multiple cars on the road all using it. I'm already noticing the problem with the sonar-based parking sensors. Despite using CHIRP (sonar frequency which varies over time), there's still enough random overlap from nearby cars that the parking sensor will occasionally trigger due to other cars which also have parking sensors. Usually it's at a red light when a lot of cars are bunched up. But I've had it trigger during normal driving and even a couple times on the freeway.
A passive system, like generating a 3D model of the surroundings based on the slight parallax between two eyeballs, does not interfere with other similar or identical systems sharing the road. So unless someone can come up with a bulletproof way of preventing two laser / radar / sonar systems from interfering with each other, the end goal should be an entirely passive environmental mapping system.
You're just deliberately conflating self-defence with guns.
Actually, I think conflating the two is very relevant to the debate. Gun control is strongly favored in cities. Gun ownership is strongly favored in rural areas. I don't think this is a coincidence or due to any of the silly reasons proffered by both sides (that rural people are ignorant hicks, or urban people are sheep). Police response times are shorter in densely packed cities, so urban residents are more comfortable with the idea of calling 911 and waiting for police to show up. Police response times are a lot longer in rural areas, so rural residents will feel more comfortable having some means to defend themselves rather than wait 20-45 minutes for police to arrive.
You don't need guns for self-defence.
Other means of defending yourself rely your physical strength exceeding that of an attacker. You have to be able to physically fight off an assailant(s). Guns are an equalizer - they remove physical strength from the equation. They're also fairly effective at equalizing an imbalance in numbers.
That's from an editorial in Mother Jones about 5 years back which made a big splash on the Internet, but didn't provide much evidence (mainly it just asserted the correlation was strong enough to be irrefutable).
A perusal of recent research on the topic turns up this recent JAMA paper which concluded that after controlling for childhood socioeconomic status, "Findings failed to support a dose-response association between BLL and consequential criminal offending." That would suggest that it's growing up in poverty which leads to future criminal behavior. And indeed if you look at the historical poverty rate, it dropped substantially right around 1970, around the time leaded gasoline began being phased out. And if you compare poverty rate by race, you find that the two races with the highest crime rates (black and hispanic) also have the highest childhood poverty rates.
This study which states "The consequences of lead exposure for later crime are theoretically compelling, but direct evidence from representative, longitudinal samples is sparse," reaches pretty much the same conclusion, but may be more useful as it provides direct links to other Google Scholar papers on the topic.
This isn't to say lead is safe. It's known to depress IQ, and though the link with future criminal behavior is weak, it is more strongly linked to antisocial behavior and delinquency. Just that the "irrefutable" link between leaded gasoline and crime presented in the Mother Jones article may in fact just have been a random correlation, not causation.
The bigger problem is that electricity pricing is done locally, while production is done nationally (or even internationally in the case of the U.S. and Canada). That is, your electricity rates depend on the cost of the plants which are nearby you. But the power grid is national. Shortages in one locale are made up by diverting power from a different location. The net result is that if there's a marginal increase in the consumption of electricity, the price the country pays for it is the marginal price for whatever power plant ends up generating that extra electricity. Hydro (and solar and wind) are supply-limited, so can never provide this marginal increase in electricity production. Nuclear plants are slow to ramp up or down, so are usually run at full capacity 24/7.
So the extra electricity used by crytocurrency mining comes entirely from fossil fuel plants. Even if they're located in an area which gets its electricity from hydro, their extra power consumption means there's less hydro power available to send to neighboring locales. That neighboring locale has to make up that electricity shortfall somehow, so a coal or gas plant near them ends up burning more fuel to generate it.
This is why it's pointless building your company next to a renewable power source just so you can advertise that your company is being green. Unless that renewable plant was built specifically to generate power for you (i.e. it wouldn't have been built otherwise), all you're doing is depriving someone else of renewable energy that they would've gotten if you hadn't built your company there. You haven't reduced the country's fossil fuel consumption, you've just pushed your fossil fuel consumption onto someone else just so you can claim the bragging rights of being green when in fact you're having zero net effect on the nation's pollution generation.
It's also why you should try to conserve electricity even if you live in an area with cheap electricity rates (e.g. Pacific Northwest, home of U.S. hydro power). Every kWh of hydro power you don't use is a kWh which gets transmitted to another part of the country, meaning some coal or gas plant somewhere has to generate a kWh less energy, and the air is that much cleaner for it.
Real reduction in fossil fuel emissions comes only two ways - reducing the country's overall power consumption, and increasing the percentage of power generated by nuclear and renewables. Cryptocurrency mining violates the first, so is just bad for the country regardless of where you do it. (Though there is an exception if you can do it during winter in an area which would've used electricity for heating anyway. It doesn't matter if the heat comes from an electric radiator or from a massive bank of GPUs. Both are electricity-in, heat-out at 100% conversion efficiency.)
Amazon could be dominating (more than they already are), except their search sucks. It's a large part of the reason I still buy most of my computer components from Newegg. Newegg has really useful and easy to use topical include/exclude options for pretty much every search I do, and it's easy to narrow down the results list to a handful of products which are exactly what I want. Amazon's searches seem to be fuzzy - even the include/exclude options seem to be polluted by vendors misrepresenting their products. Their "best match" algorithm seems to work best, except you can't sort it by ratings. If you try, you end up with a bunch of products which are seemingly only vaguely related to your search at the top of the list, or only have 1-2 ratings which are probably paid for. You often have to drill down 3-5 pages before you find a highly-rated product that you're actually searching for.
A lot of times I actually find it quicker to search for a product on Google, then follow the Amazon link in the search results. I mean I do that too for other sites (e.g. Best Buy, Staples) because they're intolerably slow. But I do it for Amazon simply because their search engine plain sucks. I'm pretty sure it all stems from Amazon trying to satisfy both sides - buyers and sellers. Buyers want highly-rated products that lots of other users have bought and reviewed. Sellers want to be able to break into a market with a new product. So Amazon feels compelled to return search results with few reviews even if that's not what customers want - to encourage more customers to try out new products instead of sticking with the safe choice. The problem is, many product markets are flooded with hundreds of cheap Chinese knockoffs of dubious quality, and Amazon's search engine makes it nigh impossible to filter them out except via the "best match" algorithm which often doesn't return the highest-rated products.
If Google's offering concentrates on meeting the needs of the buyer, rather than the seller, I could see it becoming very successful.
Not quite. It's a problem that isn't solvable by some government design, but only solvable through some very strict control of actions (road rules and we all love those).
There are several ways to improve street traffic via government design.
Bury sensors underneath the road at intersection, so left turn lights and in some cases straight-through lights are only triggered when there are actually cars which want to go in that direction.
Time the signal lights between different intersections, so drivers don't get a green light just to be caught by a red light at the next intersection. This is easier in streets laid out as a grid, but also possible on roads which are main thoroughfares (de-prioritize lights for side streets).
If you don't want to or can't time the lights, mount cameras next to the lights which look at oncoming traffic. That way the light doesn't remain green when there's no oncoming traffic and waiting cross-traffic, and doesn't turn red just as cars are approaching.
Anyhow, the premise in the summary (I haven't RTFA) is just wrong. Of course traffic apps increase traffic backups in the spatial dimension. But they decrease backups in the time dimension. That's the whole point - to minimize average travel time by spreading traffic around to fully utilize available road space to reduce the average duration of congestion. Because we generally care more about how much time it takes to travel between two points than we do the distance of the route.
VI's interface design was crafted to overcome the lack of a mouse and arrow keys on most computers back in the 1970s and 1980s. It's totally unnecessary today, and having to artificially switch between edit mode and cursor movement mode needlessly slows you down.
Emacs is way too reliant on the ctrl key. In grad school where we were encouraged to use emacs, I developed RSI in my left hand (pinky) from having to hold down ctrl so often when typing. Eventually I developed a habit of jamming it down with the side of my palm, but it's far from ideal. Alt is a better modifier key for repeated use IMHO, as your thumb on the space bar already sits next to it and you only have to curl the thumb a little to press it. I suspect the choice of ctrl as the modifier dates back to the 1970s when ctrl was typically positioned where caps lock is today.
These are the sorts of long-embedded user interface design choices I wish designers would challenge. Not useless and harmful stuff like flat UIs which slow you down and replacing informative text menus with a generic hamburger icon which simply adds an extra click to every menu navigation operation.
No they don't depend on being hot. Diesels generate spontaneous fuel-air combustion through adiabatic compression. The only requirement is that the ambient air temperature be high enough that the temperature and pressure after compression is sufficient for the fuel and air to spontaneously combust. If the ambient air temperature is too low (typically a little above freezing), then it initially needs the assistance of glow plugs to offset the loss of heat through the engine walls and help get the temp in the cylinders high enough for spontaneous combustion.
Gasoline engines match up well with a hybrid drivetrain because each covers for the other's weaknesses. Electric motors can produce their full torque output at zero RPM but start to get large and unwieldy for high power output. Gasoline engines produce little torque at low RPM, and don't hit their torque and HP peaks (roughly corresponds to most efficient operation - i.e. maximum mechanical energy generated per unit of fuel) until high RPM. Consequently, when you use a gasoline gar at low speed (stop and go traffic) or while cruising on the highway (typically only needs about 25 HP), you are nowhere near the engine's efficiency peak and you're burning a lot more fuel for the amount of work the engine is doing. If you pair the gas engine with an electric motor, you can use the electric motor for stop and go traffic and highway cruise. You only fire up the gasoline engine now and then, run it at its peak power efficiency (usually around 3500 RPM) long enough to top off the batteries to keep the electric motor going. Only when during high acceleration (high power output) do you need the gas engine + electric motor output.
Diesels power output doesn't match up well with this type of work distribution. Diesels already produce a lot of torque at low RPM, and their efficiency peak is at a lower RPM (usually around 1500-2000 RPM). Consequently, they're already operating pretty efficiently in stop-and-go traffic and at highway cruise speeds (I can hit almost 40 MPG at 65 MPH on the highway with my diesel SUV). Where diesels need help is adding more power at the high end - e.g. revving up the engine when you're on the highway to pass a slow car. This is why big rig trucks take so long to accelerate to highway speeds on an uphill on-ramp, or when they try to pass someone at freeway speeds. The engine is already close to its power peak at cruise, and there isn't much extra power at the high-end for these tasks. There's very little an electric-hybrid drivetrain can add to a diesel, and usually not enough to offset the penalty of the extra weight of the batteries and electric motors.
All this stems from the late 1990s when CARB first threatened to implement a zero emissions vehicle requirement (that a certain percentage of each manufacturer's sales had to be zero emissions if they wanted to be allowed to sell ICE vehicles, with the percentage increasing every year going forward). At the time, nobody knew which technology would pan out. GM and the Japanese automakers bet on battery-powered electric vehicles. Ford and Chrysler bet on hydrogen fuel cells. I don't think the European car makers cared, because their share of the California market was pretty small and there was a provision which if they failed to meet their quota they were allowedto purchase credits from other companies which exceeded their quota.
As we know now with 20/20 hindsight, electric turned out to be the easier solution to implement. Ford and Chrysler have been behind ever since, and many (all?) of the hybrid vehicles Ford put out were actually designs licensed from Japanese car companies.
Its worth noting that by 1999, none of the car companies except GM had a viable ZEV in production. GM's EV-1 (an EV powered by a lead-acid battery) looked to be the sole qualifying ZEV and GM was standing on a gold mine of licensing fees. The other car companies lobbied CARB saying the technology simply wasn't ready yet for a viable ZEV, and that hybrid-electric vehicles were the best choice with the technology of the time. CARB relented and rescinded the ZEV requirement, and GM's gold mine (and payoff for the billion dollars they'd invested in R&D on the EV-1) evaporated overnight. Consequently they recalled all the EV-1s and destroyed them and buried the research in retaliation for CARB's double-cross. If California wasn't going to let them profit from the EV-1, they weren't going to let California profit from it either.
Environmentalists initially hated hybrids because they still derived 100% of their energy from gasoline. I ate a lot of downvotes for trying to point out that regenerative braking and an electric motor's huge torque production at low RPM (where gasoline ICE engines are typically weak and inefficient) meant they could save a lot of energy when paired with a gas engine. It took them several years to come around to embrace hybrids.
CARB re-implemented the ZEV mandate (around 2013 I think?). The ZEV credit purchasing system was a big incentive for Musk to found Tesla. Since they produce only EVs, they always have excess ZEV credits which they can sell to other automakers. So mathematically they can actually sell each EV at a loss and still make money. It's been a rocky start thus far though. The "growth" in EV sales volume you see isn't organic - it's mandated by CARB's ZEV mandate (the formula is a bit complex, but currently a little more than 2% of each automaker's sales have to be ZEVs). If natural EV sales are short of CARB's target, the automakers simply offer discounts and incentives on their EVs towards the end of the year to try to spur sales so they can meet their target. That's why California always has the best deals on EV sales and leases - CARB only counts EVs purchased in California towards the ZEV quota.
Essentially, if consumer interest in EVs doesn't rise to match CARB's percentage goal, EVs get subsidized by ICE sales until it does. So the bigger the discounts and incentives you see on EVs in California towards the end of the year, the greater the mismatch between actual consumer interest and CARB's goals. 2015 was really bad, and there were some crazy EV deals that year (I almost pulled the trigger on a no money down, $79/mo 3-year lease for a VW eGolf). But 2016 and 2017 EV sales/leases seem to have been closer to target.
Don't compare against nonexistent perfectly safe solution. We use plastic bottles because they were much less dangerous than glass bottles. If you're worried about the dangers of plastic in the containers leaching into the water, you have to compare to the dangers of the next best alternative - glass bottles and the cuts they could give you when they broke.
Nothing in the world is 100% safe. You try to find what offers the best combination of utility and safety, and live with the inherent risks with that solution satisfied that you've minimized the risk..
And this is why all of the people telling us the market will find a solution are utterly full of shit.
What free market? Verizon has a government-granted monopoly in these areas.
The free-market is always going to be filled with players who will do anything to get an advantage. They'll outright lie to you or manipulate the game to their advantage.
Exactly right. Which is why in a free market, you have competition. You allow multiple players to offer products and services. If one regularly rips off customers, word gets around and customers stop buying from them, and they end up putting themselves out of business.
Except in this case the government decided to get involved, and prohibited anyone from competing with Verizon. Yeah Verizon are lying, cheating bastards. But it's not the free market which put them in the position of dominance they enjoy. The government did that. The market kills off bad businesses like Verizon. The government keeps Verizon alive.
There are lots of areas where the market has problems finding optimal solutions. But this isn't one of them. This turd is 100% on poor government regulation.
Actually, the business keeping Amazon afloat is their cloud services (AWS).Yes their e-commerce division pulls in the bulk of their revenue, but most of that is pass-through. Amazon only gets to keep a small portion of it as profit (net operating income). And globally their e-commerce actually operated at a loss last year. Meanwhile roughly a quarter of the revenue AWS pulls in is profit.
Amazon's North America e-commerce business delivered fourth quarter operating income of $1.69 billion on revenue of $37.3 billion.
International e-commerce sales were $18.04 billion with a operating loss of $919 million.
AWS had operating income of $1.35 billion for the fourth quarter with sales of $5.11 billion.
For the year, Amazon's international e-commerce operating losses eclipsed the company's North American operating profit. AWS had 2017 operating income of $4.33 billion on sales of $17.46 billion.
That's the pairing that should scare any retailer/distributor. Amazon is driving the profit margin for those businesses to zero, and paying for it with profit from their cloud services. A luxury most retailers and distributors don't have.
That's not entirely surprising: the more massive the galaxy, the faster the rotation at any given distance, but also the more distant the outer rim.
That was considered decades ago. The problem with thinking of galaxies in the classic 1/r^2 gravitational sense is that it implies that the spiral structure we observe in so many galaxies should have destroyed itself within a few billion years (a few rotations). There's something else going on with how they rotate that we don't understand. This apparent constant spin rate of outer edge is another piece of the puzzle. Dark matter may be another.
This mystery is similar to the spokes in Saturn's rings. Again, by classic gravitational theory, these spokes should tear apart relatively quickly as the inner structure rotates faster than the outer structure. But that's not what we see happening - somehow they appear to rotate at the same angular rate regardless of distance from Saturn. As if the rings were a solid disk instead of individual particles.
I biked to school in Southern California in the 1980s. This was back before the air pollution controls had gotten really strict, so every day there was a thick blanket of brown smog which rolled from Los Angeles eastward. Most days you couldn't even see the mountains 15-20 miles away. Ozone was a big component of this smog, and it makes your lungs hurt when you breathe deeply. Often after biking home, I'd be unable to do any strenuous physical activity simply because I couldn't breathe deeply enough. I recall reading an article in the L.A. Times about a study which concluded that going for a morning jog in the smog was worse for your health than being a couch potato and not jogging.
Have you had much luck with that? I've tried to fix a few grammar errors as well as obvious factual errors (e.g. cardinal direction between two cities). They've been reverted by overzealous editors who "own" the article.
The real problem is that tech companies want to pay programmers blue collar wages.. This is why their push for minorities to learn programming is no more than an attempt to saturate the market with skilled programmers to depress wages.
That's a bit far-fetched, considering that the average wage for a programmer is 3x the average blue collar wage. What a job is worth depends on how much productivity it generates. An employer will be willing to pay up to slightly less than the productivity generated by a job. If programming is more productive than blue collar work, then employers will always be willing to pay more for it.
It's the employer's job to try to reduce costs (wages). It's the employee's job to try to raise them (by asking for pay increases, and jumping ship to another company offering better pay if they feel they're underpaid). Where they meet in the middle is usually a pretty good indicator of what the job is actually worth (how much productivity it generates). Eliminate either and the entire market economy breaks down. You can't fault employers for doing what they're supposed to do in a functional market. (For the same reason, I never fault employees for asking for a pay raise (unless they just got one). I expect them to be keeping tabs on how much their job pays at other companies, and I expect them to ask for more if we're falling below what other companies pay.)
Whether more people decide to go into programming isn't up to the employers. It's a function of how much such jobs are paying vs how many (how few) people have the ability to perform those jobs. Criticism of pushes to encourage more minorities and women to take up programming is hypocritical - nothing more than an attempt to artificially constrain the labor market to boost wages. Ideally, everyone should be exposed to all possible career choices, so they can decide for themselves what they're best at. We keep art and music programs around for the same reason. While those fields typically result in low-paying jobs, you still want to expose kids to them so you can detect a fledgling Picasso or Mozart, and correctly guide them into that field.
If you're still not convinced, consider this: A blue collar worker who learns programming skills and gets a programming job lowers programming salaries slightly. But they also increase blue collar salaries slightly. You can't increase the supply of programmers in a vacuum. That additional supply has to come from somewhere. If it's coming from a lower-paying field like blue collar jobs, then it's a net win for everyone. The total productivity of the economy has gone up (programming job is more productive than the blue collar job so employers make more). And the total wages received has also gone up (person now earns more as a programmer than as a blue collar worker). Win-win.
I suspect the vast majority of problems go unreported even when management is responsive.
Why are we basing this on number of complaints? Anyone can file a complaint. One person with an overactive imagination can file multiple complaints without merit. The stat is already skewed far in favor of exaggerating the scope of the problem, and you're proposing skewing it even more. I thought one of the basic premises of our society was innocent until proven guilty?
If you truly want to gauge the scope of the problem, the number you should be looking at is the number of unique persons who were investigated and found to have committed sexual harassment. That eliminates the complaints found to be without merit. And it eliminates multiple complaints against a single individual. So the problem is likely much smaller than one complaint per 521 employees.
Even if 100% of the complaints are legit, the actual problem is probably on the order of one individual being complained against per 2000-5000 employees. If you base it on the number of accusers Bill Cosby has, the problem ends up being one individual in 15,000. Meaning of Microsoft's employee count of 124,000, there are probably only 8 individuals guilty of sexual harassment. And if 9 out of 10 people experiencing harassment don't report it, that still means it's being perpetrated by fewer than 100 individuals. The other 99.9% are innocent.
Allowing only upvotes means once an idea passes a certain threshold number of supporters, then it can be highlighted. But if you allow downvotes, then ideas popular only with the minority get downvoted back into oblivion by the majority. Resulting in a system where only the majority's ideas survive to be highlighted.
It seems like such a minor thing, but it creates a vastly different environment. One toxic to the premise of democracy - that new, disruptive and unpopular ideas can gradually build up support until they supplant the old majority viewpoint. That can't happen if the majority essentially has veto power over any new ideas contrary to their long-held beliefs.
As bad as hate speech is, if your underlying premise is that on balance people are good and that if given all sides of an argument they will usually make the right choice, then the proper way to fight hate speech isn't to ban it. It's to counter it by informing people of the contrary arguments. Otherwise you throw the speck which will grow into a baby out with the bathwater.
I can confirm what the others are saying - disabling the Windows Update service doesn't work to stop this. I've disabled it for months due to an incompatibility with the graphics drivers it kept trying to install. A few weeks ago, the same time I started getting messages saying I *had* to update to 1709, the Update service started re-enabling itself without my authorization.
I couldn't use 1709 because for some reason it broke about a third of my installed apps, but the damn thing kept installing itself. It caused me all kinds of headaches, including almost causing me to go over my ISP's quota because it kept downloading the 1709 update again (5 GB) every time I rolled it back so I could use my computer. Eventually I gave up, blocked off an evening so I could reinstall all the apps it broke. That's when I discovered not only were the apps broken, they couldn't be uninstalled nor reinstalled. I ended up having to roll back the 1709 update, uninstall the affected apps (which uninstalled fine in the previous version), do the 1709 update, then reinstall the broken (now uninstalled) apps. Didn't finish until well past 3 am.
I'm seriously thinking of going to go back to Windows 7.
is expected to have a resolution of some 5,500 by 3,000,
Current VR displays cover about a 200 degree field of view. 20/20 vision is defined as the ability to distinguish a line pair spaced 1 arc-minute apart, so 2 pixels per arc-minute. So this corresponds to (200 degrees) * (60 arc-minutes/degree) * (2 pixels/arc-minute) = 24,000 pixels. You need a display that's 24,000 pixels wide for it to display a 200 degree field of view and have the individual pixels not be discernible to the eye. So this display will be a little more than 1/5 of the way there.
Put another way, the angular resolution of this new VR headset will be (5500 pixels) / (200 degrees) = 27.5 pixels per degree. That's about the angular resolution of a 50" 1080p HDTV viewed from 31" away. Or a 24" 1080p monitor viewed from 15" away. The pixels will still be painfully obvious.
Slashdot Mirror
Where do school shootings rank? There have been about 250 deaths in school shootings over 18 years, or about 14 per year.(and K-12, not just ages 15-19). Since there are approximately 51 million K-12 students in the U.S., a student's chances of being killed in a non-gang, non-suicide school shooting in any given year are about 1 in 3.6 million. You are roughly 3x more likely to be struck by lightning (1 in 1.08 million).
Like airliner crashes, school shootings are one of these extremely rare, statistically insignificant events whose emotional impact creates a large amount of social interest. This causes a disproportionate amount of press coverage, leading people to wildly overestimate the actual danger. If you really want to save high schoolers' lives, teach them to: drive safely and buckle their seat belts, not to abuse drugs, seek counseling for depression, stay out of gangs, use sunscreen, eat healthy and exercise, get the flu shot, don't smoke, don't eat too many sweets, and avoid teen pregnancy. Each of these will save many more lives than all the hand-wringing over school shootings, some (like suicide-prevention) around a hundred times more.
If this were an Apple product, Cook would be telling you that you're touching it wrong.
The huge deviation of DC from the national average (29.3% of DC gun owners have a gun stolen vs 0.18% for the nation) makes me think gun theft is primarily a problem in urban areas, not rural. Further supporting this hypothesis is that Alaska, Montana, Idaho, and North and South Dakota all have gun ownership rates over 50%, but their rate of guns stolen is below the national average.
Also, the low rate for gun theft nationwide (0.18% per owner per year, vs 0.47% burglaries and 1.75% larceny) makes me think outside of certain cities, gun theft is not really a serious issue, and is more incidental property theft rather than targeted, and for the most part gun owners do a pretty good job keeping their guns safe from theft.
Which sort order you want isn't what's important. What's important is that different people will want different sort orders. The correct solution to this problem is to let user pick which sort order they want applied to their feed. But the people making the UI for these things seem to be on an ego trip, and enjoy playing god by deciding how millions of users must use an app, so won't give their users the decency of choosing what they want.
You see the same problem in Apple's walled garden, cellular carriers making certain apps on your phone undeletable, and printer manufacturers trying to restrict which ink cartridges you're allowed to use. VLC had the same problem - I and lots of others wanted to use the mouse wheel to seek (FF/RW) through the video. The main programmer thought it should control the volume, and refused to allow users to use it to seek. It was years before he finally relented and allowed an option to change the wheel's function to seek.
The real problem here is lack of respect for the user. Treating them like cattle instead of customers, just because you can.
LIDAR (and radar, and sonar) is one of those things which sounds great when you consider the car in isolation by itself. But isn't so great once you have multiple cars on the road all using it. I'm already noticing the problem with the sonar-based parking sensors. Despite using CHIRP (sonar frequency which varies over time), there's still enough random overlap from nearby cars that the parking sensor will occasionally trigger due to other cars which also have parking sensors. Usually it's at a red light when a lot of cars are bunched up. But I've had it trigger during normal driving and even a couple times on the freeway.
A passive system, like generating a 3D model of the surroundings based on the slight parallax between two eyeballs, does not interfere with other similar or identical systems sharing the road. So unless someone can come up with a bulletproof way of preventing two laser / radar / sonar systems from interfering with each other, the end goal should be an entirely passive environmental mapping system.
Actually, I think conflating the two is very relevant to the debate. Gun control is strongly favored in cities. Gun ownership is strongly favored in rural areas. I don't think this is a coincidence or due to any of the silly reasons proffered by both sides (that rural people are ignorant hicks, or urban people are sheep). Police response times are shorter in densely packed cities, so urban residents are more comfortable with the idea of calling 911 and waiting for police to show up. Police response times are a lot longer in rural areas, so rural residents will feel more comfortable having some means to defend themselves rather than wait 20-45 minutes for police to arrive.
Other means of defending yourself rely your physical strength exceeding that of an attacker. You have to be able to physically fight off an assailant(s). Guns are an equalizer - they remove physical strength from the equation. They're also fairly effective at equalizing an imbalance in numbers.
How does a school which charges $40,000/yr in tuition end up cash-strapped?
That's from an editorial in Mother Jones about 5 years back which made a big splash on the Internet, but didn't provide much evidence (mainly it just asserted the correlation was strong enough to be irrefutable).
A perusal of recent research on the topic turns up this recent JAMA paper which concluded that after controlling for childhood socioeconomic status, "Findings failed to support a dose-response association between BLL and consequential criminal offending." That would suggest that it's growing up in poverty which leads to future criminal behavior. And indeed if you look at the historical poverty rate, it dropped substantially right around 1970, around the time leaded gasoline began being phased out. And if you compare poverty rate by race, you find that the two races with the highest crime rates (black and hispanic) also have the highest childhood poverty rates.
This study which states "The consequences of lead exposure for later crime are theoretically compelling, but direct evidence from representative, longitudinal samples is sparse," reaches pretty much the same conclusion, but may be more useful as it provides direct links to other Google Scholar papers on the topic.
This isn't to say lead is safe. It's known to depress IQ, and though the link with future criminal behavior is weak, it is more strongly linked to antisocial behavior and delinquency. Just that the "irrefutable" link between leaded gasoline and crime presented in the Mother Jones article may in fact just have been a random correlation, not causation.
The bigger problem is that electricity pricing is done locally, while production is done nationally (or even internationally in the case of the U.S. and Canada). That is, your electricity rates depend on the cost of the plants which are nearby you. But the power grid is national. Shortages in one locale are made up by diverting power from a different location. The net result is that if there's a marginal increase in the consumption of electricity, the price the country pays for it is the marginal price for whatever power plant ends up generating that extra electricity. Hydro (and solar and wind) are supply-limited, so can never provide this marginal increase in electricity production. Nuclear plants are slow to ramp up or down, so are usually run at full capacity 24/7.
So the extra electricity used by crytocurrency mining comes entirely from fossil fuel plants. Even if they're located in an area which gets its electricity from hydro, their extra power consumption means there's less hydro power available to send to neighboring locales. That neighboring locale has to make up that electricity shortfall somehow, so a coal or gas plant near them ends up burning more fuel to generate it.
This is why it's pointless building your company next to a renewable power source just so you can advertise that your company is being green. Unless that renewable plant was built specifically to generate power for you (i.e. it wouldn't have been built otherwise), all you're doing is depriving someone else of renewable energy that they would've gotten if you hadn't built your company there. You haven't reduced the country's fossil fuel consumption, you've just pushed your fossil fuel consumption onto someone else just so you can claim the bragging rights of being green when in fact you're having zero net effect on the nation's pollution generation.
It's also why you should try to conserve electricity even if you live in an area with cheap electricity rates (e.g. Pacific Northwest, home of U.S. hydro power). Every kWh of hydro power you don't use is a kWh which gets transmitted to another part of the country, meaning some coal or gas plant somewhere has to generate a kWh less energy, and the air is that much cleaner for it.
Real reduction in fossil fuel emissions comes only two ways - reducing the country's overall power consumption, and increasing the percentage of power generated by nuclear and renewables. Cryptocurrency mining violates the first, so is just bad for the country regardless of where you do it. (Though there is an exception if you can do it during winter in an area which would've used electricity for heating anyway. It doesn't matter if the heat comes from an electric radiator or from a massive bank of GPUs. Both are electricity-in, heat-out at 100% conversion efficiency.)
Amazon could be dominating (more than they already are), except their search sucks. It's a large part of the reason I still buy most of my computer components from Newegg. Newegg has really useful and easy to use topical include/exclude options for pretty much every search I do, and it's easy to narrow down the results list to a handful of products which are exactly what I want. Amazon's searches seem to be fuzzy - even the include/exclude options seem to be polluted by vendors misrepresenting their products. Their "best match" algorithm seems to work best, except you can't sort it by ratings. If you try, you end up with a bunch of products which are seemingly only vaguely related to your search at the top of the list, or only have 1-2 ratings which are probably paid for. You often have to drill down 3-5 pages before you find a highly-rated product that you're actually searching for.
A lot of times I actually find it quicker to search for a product on Google, then follow the Amazon link in the search results. I mean I do that too for other sites (e.g. Best Buy, Staples) because they're intolerably slow. But I do it for Amazon simply because their search engine plain sucks. I'm pretty sure it all stems from Amazon trying to satisfy both sides - buyers and sellers. Buyers want highly-rated products that lots of other users have bought and reviewed. Sellers want to be able to break into a market with a new product. So Amazon feels compelled to return search results with few reviews even if that's not what customers want - to encourage more customers to try out new products instead of sticking with the safe choice. The problem is, many product markets are flooded with hundreds of cheap Chinese knockoffs of dubious quality, and Amazon's search engine makes it nigh impossible to filter them out except via the "best match" algorithm which often doesn't return the highest-rated products.
If Google's offering concentrates on meeting the needs of the buyer, rather than the seller, I could see it becoming very successful.
There are several ways to improve street traffic via government design.
Anyhow, the premise in the summary (I haven't RTFA) is just wrong. Of course traffic apps increase traffic backups in the spatial dimension. But they decrease backups in the time dimension. That's the whole point - to minimize average travel time by spreading traffic around to fully utilize available road space to reduce the average duration of congestion. Because we generally care more about how much time it takes to travel between two points than we do the distance of the route.
VI's interface design was crafted to overcome the lack of a mouse and arrow keys on most computers back in the 1970s and 1980s. It's totally unnecessary today, and having to artificially switch between edit mode and cursor movement mode needlessly slows you down.
Emacs is way too reliant on the ctrl key. In grad school where we were encouraged to use emacs, I developed RSI in my left hand (pinky) from having to hold down ctrl so often when typing. Eventually I developed a habit of jamming it down with the side of my palm, but it's far from ideal. Alt is a better modifier key for repeated use IMHO, as your thumb on the space bar already sits next to it and you only have to curl the thumb a little to press it. I suspect the choice of ctrl as the modifier dates back to the 1970s when ctrl was typically positioned where caps lock is today.
These are the sorts of long-embedded user interface design choices I wish designers would challenge. Not useless and harmful stuff like flat UIs which slow you down and replacing informative text menus with a generic hamburger icon which simply adds an extra click to every menu navigation operation.
No they don't depend on being hot. Diesels generate spontaneous fuel-air combustion through adiabatic compression. The only requirement is that the ambient air temperature be high enough that the temperature and pressure after compression is sufficient for the fuel and air to spontaneously combust. If the ambient air temperature is too low (typically a little above freezing), then it initially needs the assistance of glow plugs to offset the loss of heat through the engine walls and help get the temp in the cylinders high enough for spontaneous combustion.
Gasoline engines match up well with a hybrid drivetrain because each covers for the other's weaknesses. Electric motors can produce their full torque output at zero RPM but start to get large and unwieldy for high power output. Gasoline engines produce little torque at low RPM, and don't hit their torque and HP peaks (roughly corresponds to most efficient operation - i.e. maximum mechanical energy generated per unit of fuel) until high RPM. Consequently, when you use a gasoline gar at low speed (stop and go traffic) or while cruising on the highway (typically only needs about 25 HP), you are nowhere near the engine's efficiency peak and you're burning a lot more fuel for the amount of work the engine is doing. If you pair the gas engine with an electric motor, you can use the electric motor for stop and go traffic and highway cruise. You only fire up the gasoline engine now and then, run it at its peak power efficiency (usually around 3500 RPM) long enough to top off the batteries to keep the electric motor going. Only when during high acceleration (high power output) do you need the gas engine + electric motor output.
Diesels power output doesn't match up well with this type of work distribution. Diesels already produce a lot of torque at low RPM, and their efficiency peak is at a lower RPM (usually around 1500-2000 RPM). Consequently, they're already operating pretty efficiently in stop-and-go traffic and at highway cruise speeds (I can hit almost 40 MPG at 65 MPH on the highway with my diesel SUV). Where diesels need help is adding more power at the high end - e.g. revving up the engine when you're on the highway to pass a slow car. This is why big rig trucks take so long to accelerate to highway speeds on an uphill on-ramp, or when they try to pass someone at freeway speeds. The engine is already close to its power peak at cruise, and there isn't much extra power at the high-end for these tasks. There's very little an electric-hybrid drivetrain can add to a diesel, and usually not enough to offset the penalty of the extra weight of the batteries and electric motors.
All this stems from the late 1990s when CARB first threatened to implement a zero emissions vehicle requirement (that a certain percentage of each manufacturer's sales had to be zero emissions if they wanted to be allowed to sell ICE vehicles, with the percentage increasing every year going forward). At the time, nobody knew which technology would pan out. GM and the Japanese automakers bet on battery-powered electric vehicles. Ford and Chrysler bet on hydrogen fuel cells. I don't think the European car makers cared, because their share of the California market was pretty small and there was a provision which if they failed to meet their quota they were allowedto purchase credits from other companies which exceeded their quota.
As we know now with 20/20 hindsight, electric turned out to be the easier solution to implement. Ford and Chrysler have been behind ever since, and many (all?) of the hybrid vehicles Ford put out were actually designs licensed from Japanese car companies.
Its worth noting that by 1999, none of the car companies except GM had a viable ZEV in production. GM's EV-1 (an EV powered by a lead-acid battery) looked to be the sole qualifying ZEV and GM was standing on a gold mine of licensing fees. The other car companies lobbied CARB saying the technology simply wasn't ready yet for a viable ZEV, and that hybrid-electric vehicles were the best choice with the technology of the time. CARB relented and rescinded the ZEV requirement, and GM's gold mine (and payoff for the billion dollars they'd invested in R&D on the EV-1) evaporated overnight. Consequently they recalled all the EV-1s and destroyed them and buried the research in retaliation for CARB's double-cross. If California wasn't going to let them profit from the EV-1, they weren't going to let California profit from it either.
Environmentalists initially hated hybrids because they still derived 100% of their energy from gasoline. I ate a lot of downvotes for trying to point out that regenerative braking and an electric motor's huge torque production at low RPM (where gasoline ICE engines are typically weak and inefficient) meant they could save a lot of energy when paired with a gas engine. It took them several years to come around to embrace hybrids.
CARB re-implemented the ZEV mandate (around 2013 I think?). The ZEV credit purchasing system was a big incentive for Musk to found Tesla. Since they produce only EVs, they always have excess ZEV credits which they can sell to other automakers. So mathematically they can actually sell each EV at a loss and still make money. It's been a rocky start thus far though. The "growth" in EV sales volume you see isn't organic - it's mandated by CARB's ZEV mandate (the formula is a bit complex, but currently a little more than 2% of each automaker's sales have to be ZEVs). If natural EV sales are short of CARB's target, the automakers simply offer discounts and incentives on their EVs towards the end of the year to try to spur sales so they can meet their target. That's why California always has the best deals on EV sales and leases - CARB only counts EVs purchased in California towards the ZEV quota.
Essentially, if consumer interest in EVs doesn't rise to match CARB's percentage goal, EVs get subsidized by ICE sales until it does. So the bigger the discounts and incentives you see on EVs in California towards the end of the year, the greater the mismatch between actual consumer interest and CARB's goals. 2015 was really bad, and there were some crazy EV deals that year (I almost pulled the trigger on a no money down, $79/mo 3-year lease for a VW eGolf). But 2016 and 2017 EV sales/leases seem to have been closer to target.
Don't compare against nonexistent perfectly safe solution. We use plastic bottles because they were much less dangerous than glass bottles. If you're worried about the dangers of plastic in the containers leaching into the water, you have to compare to the dangers of the next best alternative - glass bottles and the cuts they could give you when they broke.
Nothing in the world is 100% safe. You try to find what offers the best combination of utility and safety, and live with the inherent risks with that solution satisfied that you've minimized the risk..
What free market? Verizon has a government-granted monopoly in these areas.
Exactly right. Which is why in a free market, you have competition. You allow multiple players to offer products and services. If one regularly rips off customers, word gets around and customers stop buying from them, and they end up putting themselves out of business.
Except in this case the government decided to get involved, and prohibited anyone from competing with Verizon. Yeah Verizon are lying, cheating bastards. But it's not the free market which put them in the position of dominance they enjoy. The government did that. The market kills off bad businesses like Verizon. The government keeps Verizon alive.
There are lots of areas where the market has problems finding optimal solutions. But this isn't one of them. This turd is 100% on poor government regulation.
That's the pairing that should scare any retailer/distributor. Amazon is driving the profit margin for those businesses to zero, and paying for it with profit from their cloud services. A luxury most retailers and distributors don't have.
That was considered decades ago. The problem with thinking of galaxies in the classic 1/r^2 gravitational sense is that it implies that the spiral structure we observe in so many galaxies should have destroyed itself within a few billion years (a few rotations). There's something else going on with how they rotate that we don't understand. This apparent constant spin rate of outer edge is another piece of the puzzle. Dark matter may be another.
This mystery is similar to the spokes in Saturn's rings. Again, by classic gravitational theory, these spokes should tear apart relatively quickly as the inner structure rotates faster than the outer structure. But that's not what we see happening - somehow they appear to rotate at the same angular rate regardless of distance from Saturn. As if the rings were a solid disk instead of individual particles.
I biked to school in Southern California in the 1980s. This was back before the air pollution controls had gotten really strict, so every day there was a thick blanket of brown smog which rolled from Los Angeles eastward. Most days you couldn't even see the mountains 15-20 miles away. Ozone was a big component of this smog, and it makes your lungs hurt when you breathe deeply. Often after biking home, I'd be unable to do any strenuous physical activity simply because I couldn't breathe deeply enough. I recall reading an article in the L.A. Times about a study which concluded that going for a morning jog in the smog was worse for your health than being a couch potato and not jogging.
Have you had much luck with that? I've tried to fix a few grammar errors as well as obvious factual errors (e.g. cardinal direction between two cities). They've been reverted by overzealous editors who "own" the article.
That's a bit far-fetched, considering that the average wage for a programmer is 3x the average blue collar wage. What a job is worth depends on how much productivity it generates. An employer will be willing to pay up to slightly less than the productivity generated by a job. If programming is more productive than blue collar work, then employers will always be willing to pay more for it.
It's the employer's job to try to reduce costs (wages). It's the employee's job to try to raise them (by asking for pay increases, and jumping ship to another company offering better pay if they feel they're underpaid). Where they meet in the middle is usually a pretty good indicator of what the job is actually worth (how much productivity it generates). Eliminate either and the entire market economy breaks down. You can't fault employers for doing what they're supposed to do in a functional market. (For the same reason, I never fault employees for asking for a pay raise (unless they just got one). I expect them to be keeping tabs on how much their job pays at other companies, and I expect them to ask for more if we're falling below what other companies pay.)
Whether more people decide to go into programming isn't up to the employers. It's a function of how much such jobs are paying vs how many (how few) people have the ability to perform those jobs. Criticism of pushes to encourage more minorities and women to take up programming is hypocritical - nothing more than an attempt to artificially constrain the labor market to boost wages. Ideally, everyone should be exposed to all possible career choices, so they can decide for themselves what they're best at. We keep art and music programs around for the same reason. While those fields typically result in low-paying jobs, you still want to expose kids to them so you can detect a fledgling Picasso or Mozart, and correctly guide them into that field.
If you're still not convinced, consider this: A blue collar worker who learns programming skills and gets a programming job lowers programming salaries slightly. But they also increase blue collar salaries slightly. You can't increase the supply of programmers in a vacuum. That additional supply has to come from somewhere. If it's coming from a lower-paying field like blue collar jobs, then it's a net win for everyone. The total productivity of the economy has gone up (programming job is more productive than the blue collar job so employers make more). And the total wages received has also gone up (person now earns more as a programmer than as a blue collar worker). Win-win.
Why are we basing this on number of complaints? Anyone can file a complaint. One person with an overactive imagination can file multiple complaints without merit. The stat is already skewed far in favor of exaggerating the scope of the problem, and you're proposing skewing it even more. I thought one of the basic premises of our society was innocent until proven guilty?
If you truly want to gauge the scope of the problem, the number you should be looking at is the number of unique persons who were investigated and found to have committed sexual harassment. That eliminates the complaints found to be without merit. And it eliminates multiple complaints against a single individual. So the problem is likely much smaller than one complaint per 521 employees.
Even if 100% of the complaints are legit, the actual problem is probably on the order of one individual being complained against per 2000-5000 employees. If you base it on the number of accusers Bill Cosby has, the problem ends up being one individual in 15,000. Meaning of Microsoft's employee count of 124,000, there are probably only 8 individuals guilty of sexual harassment. And if 9 out of 10 people experiencing harassment don't report it, that still means it's being perpetrated by fewer than 100 individuals. The other 99.9% are innocent.
That's the problem with allowing downvotes.
Allowing only upvotes means once an idea passes a certain threshold number of supporters, then it can be highlighted. But if you allow downvotes, then ideas popular only with the minority get downvoted back into oblivion by the majority. Resulting in a system where only the majority's ideas survive to be highlighted.
It seems like such a minor thing, but it creates a vastly different environment. One toxic to the premise of democracy - that new, disruptive and unpopular ideas can gradually build up support until they supplant the old majority viewpoint. That can't happen if the majority essentially has veto power over any new ideas contrary to their long-held beliefs.
As bad as hate speech is, if your underlying premise is that on balance people are good and that if given all sides of an argument they will usually make the right choice, then the proper way to fight hate speech isn't to ban it. It's to counter it by informing people of the contrary arguments. Otherwise you throw the speck which will grow into a baby out with the bathwater.
I can confirm what the others are saying - disabling the Windows Update service doesn't work to stop this. I've disabled it for months due to an incompatibility with the graphics drivers it kept trying to install. A few weeks ago, the same time I started getting messages saying I *had* to update to 1709, the Update service started re-enabling itself without my authorization.
I couldn't use 1709 because for some reason it broke about a third of my installed apps, but the damn thing kept installing itself. It caused me all kinds of headaches, including almost causing me to go over my ISP's quota because it kept downloading the 1709 update again (5 GB) every time I rolled it back so I could use my computer. Eventually I gave up, blocked off an evening so I could reinstall all the apps it broke. That's when I discovered not only were the apps broken, they couldn't be uninstalled nor reinstalled. I ended up having to roll back the 1709 update, uninstall the affected apps (which uninstalled fine in the previous version), do the 1709 update, then reinstall the broken (now uninstalled) apps. Didn't finish until well past 3 am.
I'm seriously thinking of going to go back to Windows 7.
Current VR displays cover about a 200 degree field of view. 20/20 vision is defined as the ability to distinguish a line pair spaced 1 arc-minute apart, so 2 pixels per arc-minute. So this corresponds to (200 degrees) * (60 arc-minutes/degree) * (2 pixels/arc-minute) = 24,000 pixels. You need a display that's 24,000 pixels wide for it to display a 200 degree field of view and have the individual pixels not be discernible to the eye. So this display will be a little more than 1/5 of the way there.
Put another way, the angular resolution of this new VR headset will be (5500 pixels) / (200 degrees) = 27.5 pixels per degree. That's about the angular resolution of a 50" 1080p HDTV viewed from 31" away. Or a 24" 1080p monitor viewed from 15" away. The pixels will still be painfully obvious.