i pointed this out before, but google's policy of forcing people to give their *real* names is incredibly dangerous.
I know you young people might find this hard to believe, but being forced to use your *real* name on the Internet was the norm until the mid-1990s. Don't believe me? Go to Google Groups (where old USENET posts are archived) and browse anything from the early 1990s or before. It has everyone's real names, and *gasp* sometimes even their contact info. School, company, and government sysadmins voluntarily enforced an unwritten rule that you could not be anonymous on the Internet. When a method of doing something anonymously on the Internet was discovered, it was reported as a bug, and quashed at the earliest opportunity.
What brought anonymity to the Internet was, ironically, AOL joining USENET in 1993. See, AOL required you to use your real name when signing up (so they could bill your credit card). But they also allowed you to make up to 5 sub-accounts for free, ostensibly so your family members could use AOL services under their own name. Of course people immediately took advantage of this to create alter-egos which could make USENET posts anonymously.
So while I do think anonymity is better for the Internet (not that we could do anything about it if it were bad - that horse has long since fled the stable), don't make up stuff like real names being "incredibly dangerous." The Internet worked just fine for ~2 decades with everyone using their real names.
I thought the point of schools was to provide children with an education. When did babysitting children's online behavior become part of their responsibilities?
So, you have no issues with people recording you when you don't know about it?
You think is OK for some Glasshole to walk into a restaurant where you are enjoying a public yet private dinner with a friend, record it and put it up on the Intertubes? You are OK with that. I mean, it is a "public" place, right?
You're already being recorded at your public yet private dinner with a friend. Nearly every restaurant has had a security camera system recording 24/7 for a couple decades, apparently without you knowing about it. And yes the recordings sometimes get posted on the internet.
The problem here isn't Google Glass. The problem is a disconnect between reality and your perception of it. All Google Glass is guilty of is educating you that your perception is wrong.
Apparently, this product has real stability issues potentially related to the upgrade to KitKat, or related to only 1 gb of RAM (which seems odd because there are a multiple of tablets (see iPad) that function well with only 1 gb of RAM).
Is this an Android on Intel problem?
Until last year, I used an old Galaxy S (yes, the first one) with just 512 MB of RAM. I upgraded it to Jelly Bean, then Kit Kat via Cyanogenmod. Never had any stability issues. The only problem was that about 400 MB of the RAM gets used by the OS, so Android starts playing whack-a-mole killing background tasks after you use more than about 3 apps. (Android kills task to free up memory, task says "wait I need to be running" and restarts itself, Android kills different task, that task says "wait I need to be running" and restarts itself, repeat. and everything bogs down.)
The votes in Kyllo were all over the political map.
Majority: Scalia (right), joined by Souter (center left), Thomas (right), Ginsburg (left), Breyer (center left)
Dissent: Stevens (center-left), Rehnquist (right), O'Connor (center right), Kennedy (center)
If you read through the decision, nearly every justice had their own reasons for voting the way they did. That's usually a sign of a bad decision (bad as in likely to be overturned in the future). The good decisions tend to have strong central themes defining the votes. Those usually stand the test of time (unless/until the country's feelings about those themes shifts). While I was pleased with the outcome, I expect this issue will become a SCotUS case again simply because the decision was so schizophrenic it provided little guidance for lower courts trying to pigeonhole similar cases into the framework the SCotUS set up.
Current traffic monitoring systems use either CC video analysis, ramp meters, magnetic loop, or blue tooth detection. I've heard of systems to pick up tire pressure indicator signals also, but I haven't seen them first hand.
Er, no. I was playing around with the traffic feature in Google Maps (which shows you how bad traffic congestion is). I had always assumed they were getting their data from CalTrans cameras along the freeways or something. But then I zoomed in and noticed they had traffic info for the major streets. I zoomed in some more and I noticed it showed traffic levels for the street in front of my house, where I know there aren't any cameras or magnetic loop road sensors.
That's when it hit me. Google is using the GPS-measured movements of people in cars with Android phones in their pockets to generate real-time traffic maps. That's what it means when you click OK to sending anonymized location info from your phone back to Google. (Apple did a similar thing to generate their geographical wifi hotspot map as a backup for GPS. Google actually did that The Right Way - by sending employee-driven cars all around the world recording wifi network SSIDs (and a little else by accident), and the EU crucified them for it. Apple did it by pulling GPS and wifi data from private individuals' iPhones, and no regulatory agency raised a fuss about it.)
The ability to get a warrant "without providing specific details" and the person doesn't have to be within the court's jurisdiction.
I believe that's a mischaracterization of what's going on here. It's not that the FBI wants the warrant without having to reveal specific details that they know, it's that they don't know the specific details. Only that there's criminal activity going at that node. Likewise, it's not that they want to get a warrant for a target known to be outside the court's jurisdiction, it's that they don't know what jurisdiction the target falls within. And they don't want the warrant to retroactively be declared invalid because after serving it they discover the target was outside the court's jurisdiction.
This is a subtle distinction I see a lot of people missing in a lot of things. Logic doesn't resolve down to everything being binary in state - true or false. It resolves down to three states - true, false, unknown/cannot be determined. (Math is the same way - that's why x/0 is undefined, not infinity.) If you assume binary logic and absolutely disallow warrants which are outside the court's jurisdiction, then yes you've eliminated the abuse of granting warrants outside a known jurisdiction. But you've also made crimes where the jurisdiction is unknown immune to warrants.
3.4GHz Intel i7-4770HQ
Intel Iris Pro Graphics 5200
375 x 244 x 22mm 2.0Kg
14 x 9.6 x 0.86" 4.4lbs
48 Wh lithium polymer battery
Up to 8 hours usage
That battery life is a pipe dream. The Macbook Pro 15 (which is much better optimized for battery life than Windows) w/o discrete graphics gets 8 hours under light use on the same CPU using a 95 Wh battery. This thing is more likely to get 4 hours best case, probably closer to 2-3 hours since most open source software won't be optimized for power savings on this exact hardware. (Yes I've tested this, when I put together my NAS/VM server. I plugged it into a Kill-a-Watt and measured power draw from a variety of OSes. Windows came in best at 30 Watts idle. The best default install of a Linux distro was 35 Watts idle. The worst 55 Watts idle. All were right around 105 Watts under load.)
Most of the Windows laptops with an quad core i7 (without Iris Pro graphics) managing 4 hours under light use have a 60+ Wh battery. The two with 52/54 Wh batteries (Lenovo Y50, MSI GS60) come in at 3-4 hours battery life in reviews. An 8 hour battery life in this thing is going to be attainable only in the useless "I leave the laptop sitting there powered on, but doing nothing" case (where BTW the MBP 15 hits 14 hours due to its gargantuan battery, and the 60+ Wh Windows laptops manage about 8 hours).
Which brings us to the weight. Given the short battery life, why not increase the weight to put in a bigger battery? Obviously they're trying to match the Macbook Pro 15. But if you can't match it, sacrificing battery size to keep the weight low is probably the worst compromise you can make. As it is, this thing is going to be an super-light (for a 15" notebook) ultra-portable laptop that has to sit on the desk plugged into AC power most of the time. People who buy ultra-portable laptops buy them so they can take it with them and use it away from the desk and power outlet. People who don't mind short battery life don't mind it because their laptop usually sits on a desk plugged into AC power, and thus weight doesn't matter as much. Pick one or the other.
This is incorrect. The process is called deconvolution.
It's limited by your knowledge of the len's point spread function (or in the case of blurred corners and edges, how the PSF changes as you get further from the center), and the sensor's ability to accurately capture the resulting image. And you can't deconvolve close to the edges, where you're missing image data. But mathematically the process is straightforward, if processing-intensive. Technically the light field cameras are capturing a completely blurry image, which is selectively brought into focus by deconvolution.
Also, it's easier for a tiny smartphone camera to generate images with sharp corners than a DSLR. Their lenses are so tiny you can easily (and cheaply) mold and/or grind them into aspherical shapes which reduce distortions at the edges and corners. DSLR lenses OTOH are much bigger, and thus more difficult to design and a lot more expensive to grind into the requisite shape.
If the charger can see your phone, it's not in your pocket or purse.
Obviously this is just a cover story for research into a laser designed to melt people's phones while in use. e.g. if they don't put them away in a theater when the movie starts, or kids who won't put their phone away while in class.
If that were a major factor at play, you'd expect the teams of only men to do as well as teams of only women, and teams with mixed men and women performing worst (because they'd be short on brainpower due to the women not speaking up).
I think what's going on is that this test is pretty limited in its scope. In the real world, women tend to be more risk-adverse than men. They tend to stick with the tried and true instead of striking out into the unknown. If you limit the group task to something which involves little or no risk, then women end up doing best. If you limit the group task to things where significant gains can be made from risky decisions, then the men end up doing best (with a lot of casualties along the way; e.g. the Darwin Award winners are mostly men). Combine the two and you end up with social groupings which can function well in both low-risk and high-risk situations.
If you live in the USofA, you are in more danger of being killed by someone in your family than by a terrorist.
If you live in the USA, you are in more danger of being killed by your TV or furniture falling on you than by a terrorist. (24 fatalities per year vs 17 per year).
When we were at war with Nazi Germany and Imperial Japan, I could see temporarily giving up some of our freedoms for security. But to give them up in perpetuity for something as trivial as terrorism? If you believe that's a worthwhile tradeoff, then the terrorists have truly won. The whole reason terrorists use terrorist tactics is because they don't have enough manpower or firepower to mount a head-on attack, or even a guerrilla attack. They deliberately choose a tactic which has maximum impact on public sentiment (i.e. terrorizes them, hence the name) for minimal effort. 9/11 was a fluke, a statistical outlier, whose tactic was already rendered impossible to replicate by the time the people on the 4th plane realized what was happening.
In the grand scheme of things, terrorists are nothing, less than a roundoff error in our vehicle accident fatality statistics. Would you accept government monitoring of your everyday driving behavior for added safety? Why not? Nearly 2500x as many Americans are killed in car accidents each year than by terrorism. Shouldn't that be making your "would rather not die" alarm go off like crazy?
I'll start taking all this gender equality stuff being reported seriously when I see at least half as many articles complaining about the latter as I see about the former. If one is a "problem", so is the other. Otherwise I'll take it there's an implicit assumption that women like to teach (or are better teachers) than men. And likewise men like STEM (or are better at STEM) than women.
I'd add another caveat - that once the error is pointed out, the caller ceases any attempts to market anything. Otherwise it's no different than them calling you and putting you on the "not interested" list.
Also, this problem is somewhat unique to the U.S. Cellular subscribers in the U.S. pay for incoming calls. In most countries, it's the caller who pays, so they have a built-in incentive to avoid mobile phone numbers. (Long story short is that U.S. landline phone plans switched over to fixed monthly rates before cell phones became popular. That is, you pay $30/mo, and all you could make unlimited local calls (and later, long distance calls). When cell phones arrived on the scene, this necessitated the cell phone owner pay cellular charges for incoming calls, in order to avoid breaking these fixed rate plans.)
This technique has been used since the 16th century and led to discoveries of all Planets, Planetoids, various Asteroids, Comets, and Plutoids ever since without the need of direct imaging; just some very cool math...
I don't think orbital dynamics has ever been used to discover an asteroid or comet or trans-Neptunian object. Certainly it's used to confirm their orbits (I've done that myself, freezing my ass off overnight taking a glass photographic plate, then measuring how much a small dot moved night to night). But asteroids have too little mass to to appreciably change the orbits of the larger planets. Ceres (along with a lot of other asteroids in the asteroid belt) in particular was discovered by blind luck by people searching almost at random for another planet between Mars and Jupiter. So to for that matter was Pluto - people were chasing what turned out to be an error in Neptune's calculated mass, and Pluto just happened to be near the spot that error predicted at the time they were looking.
Comets are discovered by (obsessed) people scanning the sky every night for a fuzzy dot that shouldn't be there. It's actually the same process as for asteroids (except now you have a computer do the observation instead of freezing your ass off like I did), and if the orbital calculations say it's a highly elliptical orbit instead of circular, you have a comet. The gas jets from vaporizing material as they approach the sun (which gives them their "tail) are pointed in random directions, and perturbs their orbit enough to make precise orbital calculations useless. Only general calculations like Halley's Comet returning every 86 years work.
Orbital calculations work well for (A) objects which are relatively close together since gravity decreases as the inverse square of distance, and (B) have relatively short orbital periods since this means they move faster and thus generate a larger measurable motion against the background stars. Neither of these hold true for trans-Neptunan objects.
If you subscribe to the theory that the solar system started out as a cloud of matter, and a slightly larger lump somewhere happened to coalesce into the sun by gravity, then it makes sense that the further you go out, the more material there is simply because of geometry. The volume of space (restricted to near the plane of the solar system) goes up as the square of the distance from the sun. While the length of the orbit only goes up proportional to the radius. So there must be more stuff in the outer solar system than in the inner. It's just spread out more.
Unfortunately, you and the video are focusing on the wrong metric. Things like child mortality, starvation, access to clean water, housing, etc. can all be artificially skewed by foreign aid.
The one true metric that matters is productivity per person. If each person (on average) is producing barely enough economic output to feed himself, then the country is a 3rd world developing country. Most European nations were in this state in the Middle Ages, where people had to work in the fields all day to barely keep themselves fed. And a single bad season or plague sidelining workers meant mass starvation.
If an average person is producing enough to easily feed himself and still have plenty left over to fritter away on extravagances like going to the movies, a high-end GPU for the latest Call of Duty, and the latest iPhone, then the country is a developed nation.
Corruption correlates fairly strongly with productivity per person. The more corrupt you are, the lower your GDP per capita. It's particularly revealing when you look at countries like South Korea, which by all accounts is a modernized country, yet its worker productivity has stalled at about 20%-30% lower than those of Europe and North America. Then you look at the level of corruption and it makes sense. Money which should be going from productive person to productive person thus increasing productivity even more, is instead being diverted into the pockets of corrupt non-productive people. Resulting in a lower amount of productivity per capita even though all the modern infrastructure for a thriving economy is there.
Giving people in developing countries medical care, food, clean water, and modern conveniences is pointless if they're going to continue to be dependent on foreign charity for those things in perpetuity. The primary goal of foreign assistance should always be domestic economic development (secondary being education to help staff those new domestic jobs being developed). Once these people have been set up with a functional economy where they can generate the maximum amount of productive work they're capable of, they will build their own hospitals and train their own doctors, plant their own farms, drill for or desalinate their own water, and build their own utilities and communications infrastructure.
Foreign aid like medical care, food, clean water may make the donor feel better, but its net effect doesn't really help the people in the country. And in some cases it even hurts (e.g. food donated as foreign aid depresses food prices and kills the economic viability of local farms). It should be reserved for times of calamity and bad luck, e.g. when a country which was just barely getting by gets hit by a natural disaster and crosses the threshold into regressing.
The problem with so few women in IT is that one has to ask is there something that is preventing women from getting jobs in IT. It's a fair question. In our society, there should be nothing that stops someone from getting a job - equal access is important. The problem is that no one is asking what sucks about IT.
Actually the first question that comes to my mind is, why isn't there a similar crusade to stamp out gender bias in public education? 76% of public school teachers are female. That's actually the exact same ratio as STEM, where 76% of STEM employees are male. Why do STEM jobs get all the press while public education gets none? You would think with them being public school teachers, it would be a lot easier problem to address it first since they all effectively have the same employer.
Contrary to all the hype on the news whenever it happens, the odds of your child being abducted by a non-family member are exceedingly slim. It happens about 58,000 times a year, which on a per child basis is 0.08% (72 million children in 1999 when the study took place). 99.9% of abductions end with the child being recovered or returned, 90% of them within 24 hours.
Only 115 of those cases were classified as a stereotypical kidnapping (child moved more than 50 miles, held at least one night, held for ransom or intent to permanently keep the child, or the child killed). 57% of kidnappings end with the child recovered, 40% end in the child's death, 3% with the child never found. So as a cause of death, your child is roughly 3x more likely to be killed in a firearms accident, 8x more likely to be killed in a fire, 18x more likely to be poisoned, 22x more likely to drown, and nearly 100x more likely to die in a car accident.
If you ever wanted an example of the stereotype of a government agency preying on people's fears about their safety to amass more power for itself, CPS is it.
And also, why redundancy is not backup. If your backup is plugged in and/or mounted, it's not a backup any more.
That's not strictly true anymore. True, a RAID user would've been screwed in this scenario. But newer redundant filesystems like ZFS and btrfs support snapshots, which would in fact have made this situation recoverable. Provided you were generating snapshots at regular intervals.
Some other potentially embarrassing English-to-English translation problems I've encountered:
"Hi, I'm Randy." Randy = popular male first name in American English, horny in British English.
"Knock me up in the morning." Knock up = wake up in British English, get pregnant in American English,
rubber = eraser in British English, condom in American English
"Blow me" = expression of surprise in British English, insult where a male insinuates requesting a blowjob in American English.
pissed = annoyed in American English, drunk in British English.
first floor in British English = second floor to Americans, first floor in American English = ground floor to Brits.
And one German-to-English one when the weather is hot:
"I am hot" literally translated is "Ich bin heiss" or I am horny. Germans say "Es ist heiss", or "it is hot". Either "I am hot" or "it is hot" works in American English, but the former is common enough that it's assumed to refer to the weather first, a brag about how good looking or lucky you are second, and a statement that you are horny a distant third.
Modularity sounds like a good idea, but in practice, in cellphones, I don't think it'll work. In objects of that size every millimeter counts, and modularity takes up quite a bit of space at that scale, because each part needs to be enclosed, securely attach to the others, etc. The trade-offs will mean you'll be able to pick one or two things (e.g. speed, battery life, extra features, etc.) but not all at the same time.
A lot of people think phones are already too thin. I had to buy a case for mine just to thicken it up so I wouldn't drop it as easily. I dropped my previous phone 3 times in 2 years. I dropped my new one 3 times in 2 weeks before I got the case.
I think it'll flop, but for a different reason. 4-5 years ago it would've been a huge success. Back then, the advances were coming in quickly and steadily. 2G, 3G, 4G. Single core, dual core, quad core. 512MB, 1GB, 2GB, 4GB. 800x480, 1024x600, 1280x720, 1920x1080, 2560x1440. Today, not so much. Smartphones are pretty close to the point where you can buy one and use it until it breaks. A lot people I know who aren't obsessed with having the latest and greatest have the Galaxy S3 (released 2.7 years ago) and have zero reason to upgrade. It does everything they need, and could potentially need from a phone for the foreseeable future.
I find it hard to believe the DK2 or the new WXQGA AMOLED screens on the next gen are still too low rez?
20/20 vision is defined as the ability to distinguish a line pair with 1 arc-minute separation. Or 2 pixels per arc-minute. At a viewing distance of 2 feet, this works out to just about 300 DPI. Which is where the 300 DPI spec for printers and 227 PPI spec for Apple's Retina laptop displays come from. You cannot distinguish the pixels at 2-3 feet if you have 20/20 vision.
Half of this spec - 150 DPI on printers, ~100 PPI on monitors - is considered "good enough" for most purposes. A 24" 1080p monitor is 91 PPI, so "good enough" for viewing from about 3 feet away.
A WXQGA screen (2560x1600) breaks down to 1280x1600 per eye. For the pixels to be small enough to be invisible to the eye, the image has to be smaller than 10.7 degrees by 13.3 degrees. That's a tiny image - a bit bigger than your fist at arm's length. For a "good enough" image, the image has to be smaller than 21 degrees by 27 degrees. Which is still tiny (roughly 2x2.5 fists at arms length). If the image is any bigger than that, you easily see the pixels.
So there's a lot of progress which needs to be made, not just in display resolution but in GPUs to drive those higher resolutions, before these VR units will stop being "too low rez." 4k and 8k displays are going to be much more important for VR than for big screen TVs.
(One possible workaround on the GPU side is for a camera to monitor where you're looking at, and only render that spot in high resolution. The image your eyes send to the brain is really crappy, with just a tiny spot in high resolution. GPUs have to render the entire scene in that high resolution because they don't know where your eyes are looking. But if you can track where the eye is looking, you can eliminate a lot of the GPU's workload. On a monitor this doesn't really work because other people can't watch the screen at the same time you are. But on a head-mounted display, there is no problem.)
1 - Make "franchise agreements" in cities, towns and states ILLEGAL. Paying a kickback to the government to keep out competition and to just do business is wrong. time to smack the hands of all these scumbag politicians.
The franchise agreement in the previous city I lived in was a straight kickback deal. The city asked ISPs to bid on how much they'd pay the city per house which subscribed to Internet, and the highest bidder won a monopoly. The city I lived in before gave the monopoly in exchange for a guarantee to hook up 99% of the city, including low income areas.
I'm still not convinced that granting a monopoly is ever a good idea. But there were bad franchise agreements and good franchise agreements.
2 - Government funded and OWNED fiber everywhere. Dont let AT&T own it or Comcast. It's all government ownd so that a company can come into town and set up shop as an ISP without having to spend millions to run fibers right next to all the other competition. Plus this allows you to force regulate ISP's from being dicks and only offering their service to the rich parts of town.
Heck no. During the California budget crunch, Caltrans pretty much stopped maintaining the roads. Only the larger potholes got fixed. The smaller ones and cracks in the road were left, and several highways I regularly drove on went 1-3 years beyond the point where they would've been repaired or repaved in the past. All this despite fuel taxes being unchanged. You don't want fiber to be owned by an entity which can decide money paid ostensibly to maintain the fiber should instead go to the general fund because they've got a budget shortfall.
The company which owns the fiber should do nothing but own the fiber, maintain it, roll out new installations, and occasionally upgrade it. They shouldn't sell internet service. They shouldn't have their budget tied to other government functions.
Do it like the other utilities - a private company owns and maintains the pipes/wires, but they're prohibited from selling anything carried over the pipes. Because they're a monopoly the transport prices they charge are regulated by a Public Utilities Commission, which has to approve any price hikes. That's how my gas and electricity work. I pick one of dozens of gas and electricity providers. Their bill includes a surcharge for a transport fee, which is the amount of my check that goes to the company which owns the pipes/wires.
3 - FORCE HONEST PRICING make Service Contracts ILLEGAL.. you can not find anywhere on comcasts website the real prices of internet, only their special sale prices that go up from 100 to 600% at a later date. No more of this bullshit, honest prices prominently displayed. no Contracts allowed in any way for any reason.
Contracts are necessary for stability because the lines are intentionally oversubscribed to lower cost. The accountants need to know that next month's revenue will be pretty close to this month's revenue in order to properly buy upstream bandwidth and set prices. (And don't say they shouldn't oversubscribe. A dedicated OC3 is 155 Mbps (149 Mbps usable) and costs about $30,000/mo, or about $200/mo per Mbps. The only way you're able to get 25 Mbps for $50/mo is because of oversubscription and sharing bandwidth with other customers.)
I do agree the price should not be allowed to change within the contract period. That's just a time-shifted version of giving with one hand while stealing with the other.
Slashdot Mirror
I know you young people might find this hard to believe, but being forced to use your *real* name on the Internet was the norm until the mid-1990s. Don't believe me? Go to Google Groups (where old USENET posts are archived) and browse anything from the early 1990s or before. It has everyone's real names, and *gasp* sometimes even their contact info. School, company, and government sysadmins voluntarily enforced an unwritten rule that you could not be anonymous on the Internet. When a method of doing something anonymously on the Internet was discovered, it was reported as a bug, and quashed at the earliest opportunity.
What brought anonymity to the Internet was, ironically, AOL joining USENET in 1993. See, AOL required you to use your real name when signing up (so they could bill your credit card). But they also allowed you to make up to 5 sub-accounts for free, ostensibly so your family members could use AOL services under their own name. Of course people immediately took advantage of this to create alter-egos which could make USENET posts anonymously.
So while I do think anonymity is better for the Internet (not that we could do anything about it if it were bad - that horse has long since fled the stable), don't make up stuff like real names being "incredibly dangerous." The Internet worked just fine for ~2 decades with everyone using their real names.
I thought the point of schools was to provide children with an education. When did babysitting children's online behavior become part of their responsibilities?
You're already being recorded at your public yet private dinner with a friend. Nearly every restaurant has had a security camera system recording 24/7 for a couple decades, apparently without you knowing about it. And yes the recordings sometimes get posted on the internet.
The problem here isn't Google Glass. The problem is a disconnect between reality and your perception of it. All Google Glass is guilty of is educating you that your perception is wrong.
Until last year, I used an old Galaxy S (yes, the first one) with just 512 MB of RAM. I upgraded it to Jelly Bean, then Kit Kat via Cyanogenmod. Never had any stability issues. The only problem was that about 400 MB of the RAM gets used by the OS, so Android starts playing whack-a-mole killing background tasks after you use more than about 3 apps. (Android kills task to free up memory, task says "wait I need to be running" and restarts itself, Android kills different task, that task says "wait I need to be running" and restarts itself, repeat. and everything bogs down.)
The Asus MeMo Pad 7 has very similar specs to the Dell Venue 7 (Intel CPU, 1 GB RAM, Android 4.4), and I don't see reviewers complaining about stability. So maybe a Dell problem?
The votes in Kyllo were all over the political map.
Majority: Scalia (right), joined by Souter (center left), Thomas (right), Ginsburg (left), Breyer (center left)
Dissent: Stevens (center-left), Rehnquist (right), O'Connor (center right), Kennedy (center)
If you read through the decision, nearly every justice had their own reasons for voting the way they did. That's usually a sign of a bad decision (bad as in likely to be overturned in the future). The good decisions tend to have strong central themes defining the votes. Those usually stand the test of time (unless/until the country's feelings about those themes shifts). While I was pleased with the outcome, I expect this issue will become a SCotUS case again simply because the decision was so schizophrenic it provided little guidance for lower courts trying to pigeonhole similar cases into the framework the SCotUS set up.
Er, no. I was playing around with the traffic feature in Google Maps (which shows you how bad traffic congestion is). I had always assumed they were getting their data from CalTrans cameras along the freeways or something. But then I zoomed in and noticed they had traffic info for the major streets. I zoomed in some more and I noticed it showed traffic levels for the street in front of my house, where I know there aren't any cameras or magnetic loop road sensors.
That's when it hit me. Google is using the GPS-measured movements of people in cars with Android phones in their pockets to generate real-time traffic maps. That's what it means when you click OK to sending anonymized location info from your phone back to Google. (Apple did a similar thing to generate their geographical wifi hotspot map as a backup for GPS. Google actually did that The Right Way - by sending employee-driven cars all around the world recording wifi network SSIDs (and a little else by accident), and the EU crucified them for it. Apple did it by pulling GPS and wifi data from private individuals' iPhones, and no regulatory agency raised a fuss about it.)
I believe that's a mischaracterization of what's going on here. It's not that the FBI wants the warrant without having to reveal specific details that they know, it's that they don't know the specific details. Only that there's criminal activity going at that node. Likewise, it's not that they want to get a warrant for a target known to be outside the court's jurisdiction, it's that they don't know what jurisdiction the target falls within. And they don't want the warrant to retroactively be declared invalid because after serving it they discover the target was outside the court's jurisdiction.
This is a subtle distinction I see a lot of people missing in a lot of things. Logic doesn't resolve down to everything being binary in state - true or false. It resolves down to three states - true, false, unknown/cannot be determined. (Math is the same way - that's why x/0 is undefined, not infinity.) If you assume binary logic and absolutely disallow warrants which are outside the court's jurisdiction, then yes you've eliminated the abuse of granting warrants outside a known jurisdiction. But you've also made crimes where the jurisdiction is unknown immune to warrants.
Some key specs on this thing:
3.4GHz Intel i7-4770HQ
Intel Iris Pro Graphics 5200
375 x 244 x 22mm 2.0Kg
14 x 9.6 x 0.86" 4.4lbs
48 Wh lithium polymer battery
Up to 8 hours usage
That battery life is a pipe dream. The Macbook Pro 15 (which is much better optimized for battery life than Windows) w/o discrete graphics gets 8 hours under light use on the same CPU using a 95 Wh battery. This thing is more likely to get 4 hours best case, probably closer to 2-3 hours since most open source software won't be optimized for power savings on this exact hardware. (Yes I've tested this, when I put together my NAS/VM server. I plugged it into a Kill-a-Watt and measured power draw from a variety of OSes. Windows came in best at 30 Watts idle. The best default install of a Linux distro was 35 Watts idle. The worst 55 Watts idle. All were right around 105 Watts under load.)
Most of the Windows laptops with an quad core i7 (without Iris Pro graphics) managing 4 hours under light use have a 60+ Wh battery. The two with 52/54 Wh batteries (Lenovo Y50, MSI GS60) come in at 3-4 hours battery life in reviews. An 8 hour battery life in this thing is going to be attainable only in the useless "I leave the laptop sitting there powered on, but doing nothing" case (where BTW the MBP 15 hits 14 hours due to its gargantuan battery, and the 60+ Wh Windows laptops manage about 8 hours).
Which brings us to the weight. Given the short battery life, why not increase the weight to put in a bigger battery? Obviously they're trying to match the Macbook Pro 15. But if you can't match it, sacrificing battery size to keep the weight low is probably the worst compromise you can make. As it is, this thing is going to be an super-light (for a 15" notebook) ultra-portable laptop that has to sit on the desk plugged into AC power most of the time. People who buy ultra-portable laptops buy them so they can take it with them and use it away from the desk and power outlet. People who don't mind short battery life don't mind it because their laptop usually sits on a desk plugged into AC power, and thus weight doesn't matter as much. Pick one or the other.
This is incorrect. The process is called deconvolution.
It's limited by your knowledge of the len's point spread function (or in the case of blurred corners and edges, how the PSF changes as you get further from the center), and the sensor's ability to accurately capture the resulting image. And you can't deconvolve close to the edges, where you're missing image data. But mathematically the process is straightforward, if processing-intensive. Technically the light field cameras are capturing a completely blurry image, which is selectively brought into focus by deconvolution.
Also, it's easier for a tiny smartphone camera to generate images with sharp corners than a DSLR. Their lenses are so tiny you can easily (and cheaply) mold and/or grind them into aspherical shapes which reduce distortions at the edges and corners. DSLR lenses OTOH are much bigger, and thus more difficult to design and a lot more expensive to grind into the requisite shape.
I've wondered, what's to stop them from collect that data even if you're not a subscriber?
Obviously this is just a cover story for research into a laser designed to melt people's phones while in use. e.g. if they don't put them away in a theater when the movie starts, or kids who won't put their phone away while in class.
If that were a major factor at play, you'd expect the teams of only men to do as well as teams of only women, and teams with mixed men and women performing worst (because they'd be short on brainpower due to the women not speaking up).
I think what's going on is that this test is pretty limited in its scope. In the real world, women tend to be more risk-adverse than men. They tend to stick with the tried and true instead of striking out into the unknown. If you limit the group task to something which involves little or no risk, then women end up doing best. If you limit the group task to things where significant gains can be made from risky decisions, then the men end up doing best (with a lot of casualties along the way; e.g. the Darwin Award winners are mostly men). Combine the two and you end up with social groupings which can function well in both low-risk and high-risk situations.
If you live in the USA, you are in more danger of being killed by your TV or furniture falling on you than by a terrorist. (24 fatalities per year vs 17 per year).
When we were at war with Nazi Germany and Imperial Japan, I could see temporarily giving up some of our freedoms for security. But to give them up in perpetuity for something as trivial as terrorism? If you believe that's a worthwhile tradeoff, then the terrorists have truly won. The whole reason terrorists use terrorist tactics is because they don't have enough manpower or firepower to mount a head-on attack, or even a guerrilla attack. They deliberately choose a tactic which has maximum impact on public sentiment (i.e. terrorizes them, hence the name) for minimal effort. 9/11 was a fluke, a statistical outlier, whose tactic was already rendered impossible to replicate by the time the people on the 4th plane realized what was happening.
In the grand scheme of things, terrorists are nothing, less than a roundoff error in our vehicle accident fatality statistics. Would you accept government monitoring of your everyday driving behavior for added safety? Why not? Nearly 2500x as many Americans are killed in car accidents each year than by terrorism. Shouldn't that be making your "would rather not die" alarm go off like crazy?
Ran across this interesting tidbit while looking up some stats for myself last time one of these articles got posted here.
Of the 7.6 million STEM workers, only 24% are women.
Of the 3.7 million public schoolteachers, only 24% are men.
I'll start taking all this gender equality stuff being reported seriously when I see at least half as many articles complaining about the latter as I see about the former. If one is a "problem", so is the other. Otherwise I'll take it there's an implicit assumption that women like to teach (or are better teachers) than men. And likewise men like STEM (or are better at STEM) than women.
I'd add another caveat - that once the error is pointed out, the caller ceases any attempts to market anything. Otherwise it's no different than them calling you and putting you on the "not interested" list.
Also, this problem is somewhat unique to the U.S. Cellular subscribers in the U.S. pay for incoming calls. In most countries, it's the caller who pays, so they have a built-in incentive to avoid mobile phone numbers. (Long story short is that U.S. landline phone plans switched over to fixed monthly rates before cell phones became popular. That is, you pay $30/mo, and all you could make unlimited local calls (and later, long distance calls). When cell phones arrived on the scene, this necessitated the cell phone owner pay cellular charges for incoming calls, in order to avoid breaking these fixed rate plans.)
I don't think orbital dynamics has ever been used to discover an asteroid or comet or trans-Neptunian object. Certainly it's used to confirm their orbits (I've done that myself, freezing my ass off overnight taking a glass photographic plate, then measuring how much a small dot moved night to night). But asteroids have too little mass to to appreciably change the orbits of the larger planets. Ceres (along with a lot of other asteroids in the asteroid belt) in particular was discovered by blind luck by people searching almost at random for another planet between Mars and Jupiter. So to for that matter was Pluto - people were chasing what turned out to be an error in Neptune's calculated mass, and Pluto just happened to be near the spot that error predicted at the time they were looking.
Comets are discovered by (obsessed) people scanning the sky every night for a fuzzy dot that shouldn't be there. It's actually the same process as for asteroids (except now you have a computer do the observation instead of freezing your ass off like I did), and if the orbital calculations say it's a highly elliptical orbit instead of circular, you have a comet. The gas jets from vaporizing material as they approach the sun (which gives them their "tail) are pointed in random directions, and perturbs their orbit enough to make precise orbital calculations useless. Only general calculations like Halley's Comet returning every 86 years work.
Orbital calculations work well for (A) objects which are relatively close together since gravity decreases as the inverse square of distance, and (B) have relatively short orbital periods since this means they move faster and thus generate a larger measurable motion against the background stars. Neither of these hold true for trans-Neptunan objects.
If you subscribe to the theory that the solar system started out as a cloud of matter, and a slightly larger lump somewhere happened to coalesce into the sun by gravity, then it makes sense that the further you go out, the more material there is simply because of geometry. The volume of space (restricted to near the plane of the solar system) goes up as the square of the distance from the sun. While the length of the orbit only goes up proportional to the radius. So there must be more stuff in the outer solar system than in the inner. It's just spread out more.
Unfortunately, you and the video are focusing on the wrong metric. Things like child mortality, starvation, access to clean water, housing, etc. can all be artificially skewed by foreign aid.
The one true metric that matters is productivity per person. If each person (on average) is producing barely enough economic output to feed himself, then the country is a 3rd world developing country. Most European nations were in this state in the Middle Ages, where people had to work in the fields all day to barely keep themselves fed. And a single bad season or plague sidelining workers meant mass starvation.
If an average person is producing enough to easily feed himself and still have plenty left over to fritter away on extravagances like going to the movies, a high-end GPU for the latest Call of Duty, and the latest iPhone, then the country is a developed nation.
Corruption correlates fairly strongly with productivity per person. The more corrupt you are, the lower your GDP per capita. It's particularly revealing when you look at countries like South Korea, which by all accounts is a modernized country, yet its worker productivity has stalled at about 20%-30% lower than those of Europe and North America. Then you look at the level of corruption and it makes sense. Money which should be going from productive person to productive person thus increasing productivity even more, is instead being diverted into the pockets of corrupt non-productive people. Resulting in a lower amount of productivity per capita even though all the modern infrastructure for a thriving economy is there.
Giving people in developing countries medical care, food, clean water, and modern conveniences is pointless if they're going to continue to be dependent on foreign charity for those things in perpetuity. The primary goal of foreign assistance should always be domestic economic development (secondary being education to help staff those new domestic jobs being developed). Once these people have been set up with a functional economy where they can generate the maximum amount of productive work they're capable of, they will build their own hospitals and train their own doctors, plant their own farms, drill for or desalinate their own water, and build their own utilities and communications infrastructure.
Foreign aid like medical care, food, clean water may make the donor feel better, but its net effect doesn't really help the people in the country. And in some cases it even hurts (e.g. food donated as foreign aid depresses food prices and kills the economic viability of local farms). It should be reserved for times of calamity and bad luck, e.g. when a country which was just barely getting by gets hit by a natural disaster and crosses the threshold into regressing.
Actually the first question that comes to my mind is, why isn't there a similar crusade to stamp out gender bias in public education? 76% of public school teachers are female. That's actually the exact same ratio as STEM, where 76% of STEM employees are male. Why do STEM jobs get all the press while public education gets none? You would think with them being public school teachers, it would be a lot easier problem to address it first since they all effectively have the same employer.
Contrary to all the hype on the news whenever it happens, the odds of your child being abducted by a non-family member are exceedingly slim. It happens about 58,000 times a year, which on a per child basis is 0.08% (72 million children in 1999 when the study took place). 99.9% of abductions end with the child being recovered or returned, 90% of them within 24 hours.
Only 115 of those cases were classified as a stereotypical kidnapping (child moved more than 50 miles, held at least one night, held for ransom or intent to permanently keep the child, or the child killed). 57% of kidnappings end with the child recovered, 40% end in the child's death, 3% with the child never found. So as a cause of death, your child is roughly 3x more likely to be killed in a firearms accident, 8x more likely to be killed in a fire, 18x more likely to be poisoned, 22x more likely to drown, and nearly 100x more likely to die in a car accident.
If you ever wanted an example of the stereotype of a government agency preying on people's fears about their safety to amass more power for itself, CPS is it.
That's not strictly true anymore. True, a RAID user would've been screwed in this scenario. But newer redundant filesystems like ZFS and btrfs support snapshots, which would in fact have made this situation recoverable. Provided you were generating snapshots at regular intervals.
Some other potentially embarrassing English-to-English translation problems I've encountered:
"Hi, I'm Randy." Randy = popular male first name in American English, horny in British English.
"Knock me up in the morning." Knock up = wake up in British English, get pregnant in American English,
rubber = eraser in British English, condom in American English
"Blow me" = expression of surprise in British English, insult where a male insinuates requesting a blowjob in American English.
pissed = annoyed in American English, drunk in British English.
first floor in British English = second floor to Americans, first floor in American English = ground floor to Brits.
And one German-to-English one when the weather is hot:
"I am hot" literally translated is "Ich bin heiss" or I am horny. Germans say "Es ist heiss", or "it is hot". Either "I am hot" or "it is hot" works in American English, but the former is common enough that it's assumed to refer to the weather first, a brag about how good looking or lucky you are second, and a statement that you are horny a distant third.
Reminds me of the backronym someone came up with for the old PCMCIA slots on laptops. People Can't Memorize Computer Industry Acronyms.
A lot of people think phones are already too thin. I had to buy a case for mine just to thicken it up so I wouldn't drop it as easily. I dropped my previous phone 3 times in 2 years. I dropped my new one 3 times in 2 weeks before I got the case.
I think it'll flop, but for a different reason. 4-5 years ago it would've been a huge success. Back then, the advances were coming in quickly and steadily. 2G, 3G, 4G. Single core, dual core, quad core. 512MB, 1GB, 2GB, 4GB. 800x480, 1024x600, 1280x720, 1920x1080, 2560x1440. Today, not so much. Smartphones are pretty close to the point where you can buy one and use it until it breaks. A lot people I know who aren't obsessed with having the latest and greatest have the Galaxy S3 (released 2.7 years ago) and have zero reason to upgrade. It does everything they need, and could potentially need from a phone for the foreseeable future.
20/20 vision is defined as the ability to distinguish a line pair with 1 arc-minute separation. Or 2 pixels per arc-minute. At a viewing distance of 2 feet, this works out to just about 300 DPI. Which is where the 300 DPI spec for printers and 227 PPI spec for Apple's Retina laptop displays come from. You cannot distinguish the pixels at 2-3 feet if you have 20/20 vision.
Half of this spec - 150 DPI on printers, ~100 PPI on monitors - is considered "good enough" for most purposes. A 24" 1080p monitor is 91 PPI, so "good enough" for viewing from about 3 feet away.
A WXQGA screen (2560x1600) breaks down to 1280x1600 per eye. For the pixels to be small enough to be invisible to the eye, the image has to be smaller than 10.7 degrees by 13.3 degrees. That's a tiny image - a bit bigger than your fist at arm's length. For a "good enough" image, the image has to be smaller than 21 degrees by 27 degrees. Which is still tiny (roughly 2x2.5 fists at arms length). If the image is any bigger than that, you easily see the pixels.
So there's a lot of progress which needs to be made, not just in display resolution but in GPUs to drive those higher resolutions, before these VR units will stop being "too low rez." 4k and 8k displays are going to be much more important for VR than for big screen TVs.
(One possible workaround on the GPU side is for a camera to monitor where you're looking at, and only render that spot in high resolution. The image your eyes send to the brain is really crappy, with just a tiny spot in high resolution. GPUs have to render the entire scene in that high resolution because they don't know where your eyes are looking. But if you can track where the eye is looking, you can eliminate a lot of the GPU's workload. On a monitor this doesn't really work because other people can't watch the screen at the same time you are. But on a head-mounted display, there is no problem.)
The franchise agreement in the previous city I lived in was a straight kickback deal. The city asked ISPs to bid on how much they'd pay the city per house which subscribed to Internet, and the highest bidder won a monopoly. The city I lived in before gave the monopoly in exchange for a guarantee to hook up 99% of the city, including low income areas.
I'm still not convinced that granting a monopoly is ever a good idea. But there were bad franchise agreements and good franchise agreements.
Heck no. During the California budget crunch, Caltrans pretty much stopped maintaining the roads. Only the larger potholes got fixed. The smaller ones and cracks in the road were left, and several highways I regularly drove on went 1-3 years beyond the point where they would've been repaired or repaved in the past. All this despite fuel taxes being unchanged. You don't want fiber to be owned by an entity which can decide money paid ostensibly to maintain the fiber should instead go to the general fund because they've got a budget shortfall.
The company which owns the fiber should do nothing but own the fiber, maintain it, roll out new installations, and occasionally upgrade it. They shouldn't sell internet service. They shouldn't have their budget tied to other government functions.
Do it like the other utilities - a private company owns and maintains the pipes/wires, but they're prohibited from selling anything carried over the pipes. Because they're a monopoly the transport prices they charge are regulated by a Public Utilities Commission, which has to approve any price hikes. That's how my gas and electricity work. I pick one of dozens of gas and electricity providers. Their bill includes a surcharge for a transport fee, which is the amount of my check that goes to the company which owns the pipes/wires.
Contracts are necessary for stability because the lines are intentionally oversubscribed to lower cost. The accountants need to know that next month's revenue will be pretty close to this month's revenue in order to properly buy upstream bandwidth and set prices. (And don't say they shouldn't oversubscribe. A dedicated OC3 is 155 Mbps (149 Mbps usable) and costs about $30,000/mo, or about $200/mo per Mbps. The only way you're able to get 25 Mbps for $50/mo is because of oversubscription and sharing bandwidth with other customers.)
I do agree the price should not be allowed to change within the contract period. That's just a time-shifted version of giving with one hand while stealing with the other.