They do. Some data center operators get a deal on their electricity if they run their backup generators during peaks, reducing the load on the grid. As a bonus, it's a cheap readiness test of their backup power systems.
To be fair, they've improved data storage reliability as a result of incidents like these. Now, when you provision a storage account under Azure, there's a check box (checked by default) that turns on "global backup". When on, any data written to storage is fanned out across two geographically diverse data centers.
Say what you will about their technical mistakes, Microsoft does learn from them. As far as their decisions when it comes to product development and marketing, though, they don't appear to have learned a damn thing in the last decade.
That's certainly a different way of looking at it. Primitive people had it good because they were unaware of how bad they had it. To them it was natural to bear four children in hopes that one might mature to adulthood - although a fifth child was unlikely, as the mother's chances of surviving that many childbirths were not high. People didn't die from the nasty age-related diseases like cancer or heart disease because they died much earlier due to injury, hardship, or violence. So no, I don't buy the "non-material primitive society had it as good as us, but in their own way" position for a second. On average, people in today's western societies work much less and live far longer than our forebears, and have our basic needs well and truly exceeded. (Are we happier today? Different question. Today we have so much idle time that people seem to invent problems just to have something to worry about.)
Beyond that, though, I don't see us as being on different sides. I think a basic income is a great idea. Our society certainly is rich enough to afford it.
I would advise caution, though, because it will breed a dual economy. There will be the rich, who will continue to have the money, toys, power, and authority, and the poor who will be given their bread and circuses.
Crime would stay low: most criminals start from a position of desperation, but once they realize the profitability of crime, they expand. Remove desperation, and fewer people will start. That's huge in terms of safety.
In the best case, all it would take to get out of the basic income bracket is to expend enough effort to better yourself. Do something to earn more money, and you've improved your life by that much. Rewards are a powerful incentive.
But the rich historically have not been welcoming of additions to their ranks. Sharing power is unlikely. Sharing their wealth is equally unlikely, even if it would make society an overall better and safer place for them. Could a self-starter really rise to join the elite? Instead, I'd guess that not only would we have a dual economy, we'd have a dual currency, far more specific than the difference between "old money" and "new money." Your Basic Income dollars would buy you food and shelter, and there'd be a thriving barter economy of swap meets and marketplaces. But it would take the new ThalliumVisa to buy a car, or a Versace handbag, and for some reason the ThalliumVisa bank wouldn't accept trade in BI$. Class differences would become unbridgeable. And those differences would breed strife.
Shifting slightly, something I've noticed about people is that people hold different concepts of fairness. Some think that fairness means "you must earn everything yourself, and we should all have an equal chance to earn." To those people, welfare is bad because the recipients didn't earn it, or it's bad because there are "welfare queens" who abuse the system. They see examples of abuse and think the system is a failure because abuse happened. I believe instead that fairness is socialism: have a safety net for everyone. We have so many resources that we should be appalled that there are any hungry children in this society, or that someone should be denied health care or shelter because they can't afford it. I understand there will be abuses in that system because they're not preventable, and I'm willing to live with a few of them while we continually fix the system.
Anyway, yeah, bring on the dual economy. Let's try that for a while.
Stonewashed blue jeans are tumbled in a slurry of abrasive pellets. The breakdown of these pellets is a grit that is trapped in the pockets, and which scratches the heck out of anything stored in them.
No, the solution has to be government interference of some kind. There is no world government (thankfully), but the downside is that nobody mandates fair trade or labor practices between countries. That's up to the countries themselves.
Consider an example under way right now in the solar panel industry. Domestic solar panel makers build factories that cost $10 million dollars to produce panels that cost $5,000 each. A Chinese factory with a leftover billion dollar chip factory (they upgraded to a $10 billion dollar factory to make state of the art chips) now starts making solar panels much more efficiently and with much cheaper labor, for $1,000 per panel. Domestic makers can't compete and will soon go out of business. Chinese factory will raise prices to $10,000 per panel. Domestic consumers will then be paying twice the price of domestic solar panels for foreign imports. Everyone in the U.S. knows that if a new domestic maker tries to enter the market to sell $4,000 panels, the Chinese will lower the price to $1,000 until the domestic factory is out of business, then raise them back to $10,000.
That's monopolistic behavior, and would be illegal under the Sherman Act. Under current trade practices, it's perfectly legal for the Chinese to do it.
There is no way to compete in a global market when domestic producers have to follow domestic laws that do not apply outside our borders. And those domestic laws are there to prevent domestic companies from doing exactly the same thing, which we already agree is bad for all consumers.
The answer is punitive tariffs, (aka government interference.) We shouldn't try to balance everything out and derive a "fair price" for solar panels, and tax the Chinese solar panels until they're competitive with domestic solar panels. We should look at the behavior in this case, and say "you profited at $10,000 panels, you drove an American company out of business with $1,000 panels, the tax on all imported panels is now $10,000 and the tax on all solar panels from China is now punitively set at $20,000." If we do this in every case where industries practice dumping on us, the Chinese will get very mad at the factory owners who keep trying to screw the U.S., and will take care of the problem internally. Meanwhile, the $10,000 tax will prevent gray market panels from China making an end-run through other countries. And in the U.S., domestic companies will be free to compete with each other for our business, selling panels at fair domestic prices of $5,000 or $4,000.
We're never going to take care of our own people until we protect them.
No, advertisers have no business value in taking this seriously, and every incentive to find ways around it. Advertisers know that targeted advertising is both more effective and less offensive than random advertising. And today, in order to provide targeted advertising, tracking provides an overall picture of an individual's tastes and habits.
Targeted advertising will continue, but in a different fashion. Instead of tracking you at an individual level, I predict they'll find ways to identify surfers at an aggregate level. Perhaps cookies will contain your advertising preferences instead of your identity. If you visit a few car forums every so often, the advertiser might store the CAR=2 value in your cookie, if you visit Etsy a lot they'll store CRAFT=5, and if you visit a political campaign donation site, they'll store SUCKER=9. Then, as you surf the web, if you're identified as a SUCKER>3, the ads will feature Vote Rombama. They're not tracking you personally or identifying your habits, but they're still targeting you.
The bigger problem, and why I think Google was the big holdout, is Google Analytics. They don't use the tracking data just for advertising. They sell information about shopping habits to marketers. If you search for "stereo reviews", then click the links to stereo-reviews.com and hifiworld.com, then buy a Coby home stereo for $299 from shopping.com, they can provide that data to stereo retailers around the net: people who buy overpriced crap surf believe the information on stereo-reviews.com and hifiworld.com. Killing tracking kills that intelligence business.
This would be apt if Google did not already provide a way for users to universally opt out of doubleclick, and a chrome plugin that controls your opt out at key internet advertising associations.
In other words, "Google already provides a combination of non-intuitive and user-download-required ad hoc ways to do one fourth of something that's being standardized elsewhere, so we should think it's OK for them to ignore the standardizations under way."
Hmm... for a minute there I thought you were shilling for Google. But now I realize you're just cognitively impaired.
There's a big difference. Almost all of the "serfs" today (at least the ones living in America) have shelter, food, clothing, running water, heat, electricity, lighting, TV sets, refrigeration, cell phones, cheap and fast transportation, medical services, borders secure from invasion, and on a level far surpassing the living standards of even the kings of the middle ages.
Sure there's inequity. The balance of wealth distribution is today skewed beyond understanding. I don't have a four-Lamborghini garage, or a stable of race horses; I have a ten-year-old Ford truck and a couple of dogs. But when you start worrying about how bad we have it in comparison to the 1%, or whine about money spent on taxes, try to also compare yourself to the 99% from 150 or more years ago. We live better today than every single human ever prior to 1850.
What would make an audio codec something worth using that would make you switch?
For the listener, there doesn't have to be a "switch", because it's not an "either-or" proposition. An audio playback system already has a library of codecs at its disposal, and chooses the correct one based on the encoding. Some players like Windows Media Player will dynamically download new decoders on demand as required. There is no burden on the consumer, as long as their devices support the new codec. And sure, other players (like iPods) may require some kind of user-initiated update, and those could mean a loss of audience if no update is available. There will always be those who are disenfranchised for a few years.
But the real cost of encoding is borne by the distributor of the audio. Those are the people who have money reasons to switch. Consider the guy running a web service who is pumping out music streams to thousands of listeners. Adding a new codec that's 10% more efficient at compressing audio means he can fit 10% more listeners in the same bandwidth, which gives him 10% more ad revenue. If only half the users have the new codec, well, he's still gaining 5% more capacity. If I'm making a greeting card playback chip, perhaps I can buy smaller (and therefore cheaper) ROM chips to hold the same amount of sound. If I'm building a telephony switch, perhaps I can increase the number of channels I support simultaneously. If I'm building a battery powered music player, a more efficient decoder can mean longer battery life, more song storage capacity, or could make more CPU available to other applications making it a faster overall device. And to anyone paying an annual license for their current encoding software, a truly free codec could mean lower costs.
Out of everybody involved, the merchant is almost the least at fault. In order to accept a few dollars worth of a transaction, the merchant is forced to handle these things called "account numbers" and "credit cards" that represent tremendous potential value, even though the merchant might be a dollar store with transactions never worth more than a few dollars. Imagine a two-buck shrimp shack on the beach, where half the customers pay with thousand dollar bills, and they each expect $998 in change. That's pretty close to the current situation.
The customers are not really to blame, either, except that they have proven themselves incapable of coping with increased security measures. They demand convenience and speed, perhaps unreasonable convenience given the value their cards represent. They take little responsibility for their own security, because the law says they're liable for only $50 worth of loss.
The banks are slightly more to blame. They are supposed to be responsible for securely handling all my money, but if they screw up they can send it all out the doors without my approval.
The real culprits are the card payment networks. The entire credit system is insecure from start to end, from its inception through today. The card networks are the ones who established the insecure systems and the insecure standards, they approved the insecure protocols, and they way they were designed back in the 1980s they didn't care about anything except making sure they got their slice of the money as it passed through their networks. When the fraud problems got too high, instead of redesigning their systems to be secure, they took a giant stick to the retailers and said "make your systems carry our insecure data more securely or we will fine you and kick you out of the credit card club." They didn't even tell the retailers what to do, or give them secure protocols, or provide tools for them to do it, they just said "you figure out a way to make it secure, or else." So now thanks to those very expensive non-standardized regulations, there are ill-defined and haphazardly implemented patchworks of security here and there, and little wonder that there are data breaches reported almost daily.
If you could get everyone (or at least a very significant number of people) in the country to switch to cash, then maybe prices would go down. Otherwise, me switching to cash isn't going to reduce my costs one bit. All it's going to do is stop earning me cash back and sign up bonuses.
If everyone switched to cash, prices would likely go up, not down. For large merchants, I know the cost of handling cash is substantially higher than the cost of handling credit transactions.
Credit transactions: A cash register has to be on a network, and have a PIN pad attached. A reader has to read a card, and some bits take a few milliseconds to flow through a wire. An occasional piece of paper has to be printed, signed, and collected. Visa and the bank take their cut on the back end. Occasionally, a bad credit transaction will be charged back to the retailer. All in all, the process is very automated from start to finish, with a few special hardware requirements like PIN pads, and some secure handling activities go on, like changing certificates.
Cash transactions: A store keeps a large, expensive safe full of money in a secured back room. Two people open the safe, take out the day's money, and run it through another expensive machine to count it. They put it in bags, and someone carries those bags to the cash registers. Cashiers take those bags, count the money, and put it in their specially designed armored money drawers (tills). Customer comes, and says "I'd like to pay with cash." Cashier takes time to manually count the customer's money, places it in the till, then counts out and returns the correct change to the customer. At the end of the day, the cash is collected from the tills, brought to the back room, counted in the expensive machine, and is then stored in the expensive safe. The next day, an armored car service is contracted to drive up and pick up the contents of the safe, which they then drive to a bank. All that security equipment, safes, cash counters, cash register tills, money room security systems, alarms, all cost a lot of money to buy and maintain. The handling of the money so many times throughout the day costs a lot in labor. And the cost of an armored car service takes their cut as well.
Finally, cash has a lot of unique risks. There's a risk that people who are trusted to handle it might actually steal it; there's a risk that someone will pass you forged money; and it's a deliciously tempting robbery target, meaning all those people who handle it throughout the day are at risk of being shot by a robber. Nobody ever gets shot for a stack of credit slips.
For small retailers, the expenses are likely much lower. The safe might be right under the single cash register. The manager would likely have free time at the start and end of day to do some of the cash management activities. The money is probably deposited in the bank by the shop owner, and not driven in an armored car. But for the big stores, accepting cash is a huge expense.
surely these scanners can record serial numbers since they scan the bill for denomination anyway.
Surely they don't.
Bill acceptors rely on several attributes to test a piece of paper to ensure it really is money. They shine various lights at it and through it, they run it over a magnetic ink detector, they check the thickness, they measure the dimensions, they match the images to a database of known images, but they do not record the images of the bills they accept. They just keep track of the amount.
Now, the cameras over the cash registers are taking plenty of photos of you. It would be possible to go back through a day's footage and see who paid in cash throughout the day, and figure out which image had you spending the $100 bill. But even that's just a picture, and it still doesn't tell me what your name is.
Any data in a QR code that is invalid should only be marked as invalid and the bill sorted aside for later, manual investigation. No "action" with the data itself is required. It shouldn't matter if the data is a URL or an IP address or "echo y|format C:/q". There should be nothing processed but an ack that the data doesn't correspond to correct ranges.
Nobody here is arguing that they should do anything else. But people are instead arguing that the reality is different, and that stupider things have happened to many institutions that should have known better.
Consider this scenario: John's First National Bank and Laundromat decides they need verifier software. They outsource the writing to a cheap software contractor shop who doesn't care much other than they deliver on time. Developer at the shop says "I would be done faster if I download a QR reader library from the intarwebs." Developer downloads a copy of zxing and wires it into the app. Unbeknownst to everyone involved, he included the portion of the library that has keyword recognition built in, and will auto-launch a URL. Nobody thinks to test it with a URL, because their specs don't say anything about dollar bill QR codes with URLs encoded in them.
It's not like every bank needs to have this happen. If even one bill verifier package in one bank contains this flaw, a tampered QR bill will flow through it at some point.
Implausible? Spend enough time in Corporate America and you'll find it's not even improbable.
Who said that the QR code will encode an URL? This is not written in the engadget article, and that's the main erroneous assumption of the Slasdot poster (planetzuda).
It's not an erroneous assumption at all. The banks wouldn't print a URL in their QR code, but we're talking about an attacker modifying a bill, not the bank.
QR codes encode data, and that data can include ways to automate the processing of it. Smartphone QR reader applications know how to parse many recognized QR keywords, which include sms:, smsto:, mms:, mailto:, tel:, geo:, market:, youtube:, and of course http: and https:. While it's highly unlikely, it's not impossible that the bank's software reading the QR codes might interpret one of the known keywords and take an action.
An attacker would encode a malicious URL and print it on the bill, with the expectation that the bank would scan the code while in the normal course of verifying the currency. They'd be betting on the off-chance that there's something wrong with the security of the QR reader software in the bank's system, in hopes that their bill verifier would become corrupted in some way.
Think about it: for an attacker, having a zombie machine inside a bank's firewall would be like downloading a gold bar. And the risk? Virtually zero. Print up a bunch of nano-sized QR dots, stick them on your banknotes, and go spend them. Nobody will be able to trace that cash back to you. At some point in its future, that note will flow through a bank's verifier, and then you might get lucky.
It's no different than someone printing a QR code containing a URL that points to some malware site and sticking it on a movie poster, hoping that some idiot with their phone will read it and get infected.
Of course, a good tactic for the bank would be to have the U.S. Secret Service follow up on any illicit URLs they found embedded in the cash. They'd be able to proactively ID the malware server before even connecting to it, catching the bad guys unawares.
The 10 minute wait is an interesting compromise. Most security measures work by slowing down a determined attacker, not by absolutely preventing the loss, so that's a good idea. But in the CCTV footage I saw where a couple guys stole a BMW in minutes, they were a bit jumpy but didn't seem to mind that it took that long.
The balance might come in a combination of time and physical effort. Perhaps the lost key compromise could require lifting the car, then using a specialized, bulky tool (which is functionally not much different than a master lock and key) to access something from beneath the center of the car, combined with a 10 or 30 minute wait. Or perhaps the suspension has travel limit sensors that could be used - all four wheels have to be off the ground for 10 minutes. Or maybe something else that requires a tremendous amount of energy - compressing a large spring to trigger a contained switch. Make it require a set of things that are too difficult for the average thief to carry - a jack and jack stands, a long specialized tool, at least something more than a guy can carry in his pocket.
If it's going to remain purely digital, have it use digital certificates and require on-line network access to reset it. The BMW factory would then be the gatekeepers of who would be authorized to access the "blank key reset service". The on-board computer would log all key reset activities on a ROM, which they could provide access logs to police for investigations. A crooked dealer would quickly be identified, and immediately revoked by the service.
It still won't stop a tow truck or a mobile garage, but they're not stopped today.
Do you find there is still demand for mid/side stereo encoding? I'm not in the audio business, but it seems to be a holdover from ancient technologies like FM stereo. Since virtually everything is digitally processed these days, do people really need the pre-combined streams? When does it provide a realistic advantage?
I know that a mid-channel track can be processed on its own and played back as a mono source. You don't have to burn the energy to decode a second track. But that's its only advantage. And it is an advantage only in a very limited energy environment, where the tiny draw of a chip processing a second audio track would cost you battery life. But when I look at modern uses of audio, there are no cases where mono playback intersects with tightly limited energy:
Stereo: personal music (headphones); car audio; home entertainment systems; commercial television; video playback.
Pure mono source: telephone; web videos.
Mid-channel mono only: public address; ambient background music.
For all cases of stereo, you have to decode two streams anyway. There is no saving in having a mid/side encoding, as you have the mid channel in one track and the side channel in another. Two tracks == twice the energy, regardless of encoding technology.
For all cases of pure mono, you don't need mid/side encoding, as you have only one channel of input. Energy saved automatically.
For the cases where you would downconvert a stereo mid/side channel encoding to a mono playback system, the typical environments are fixed installations, where wall power is readily available, and the tiny energy consumption of the second track is not an issue.
So from my limited point of view, I don't see any rational reason to continue to support such an arcane format. What rationale did you have for including it?
So I was trying to figure out if SILK would permit tying streams together, or if it provides any kind of synchronization with an external time source. All I can really find is a section on clock drift, and using tools like OPUS_GET_PITCH() to figure out how far the drift is. And there's a section on stuffing multiple frames (up to 48) into a CBR Code 3 packet. But it doesn't indicate anything about how those frames would be related.
Which is too bad, because it really does limit the applications. Without sync, how could Opus streams be used for the audio tracks on a video?
All I can think is that I'm really missing something fundamental here. They had to have considered these other applications, didn't they?
They also left out n-channel support. You get mono or stereo, but that's it. No 7.1 surround encoding. That would have made it the one true codec for everything. That and lossless encoding. The two true codecs for everything.
Oh, and support in the iPhone. That would have made three true codecs. Among the many true codecs... oh bugger, I'll start again.
At some point, I do trust that at least some of the people who have these hard jobs still do them for honest and noble reasons.
I think it's really hard for guys like street cops to maintain objectivity when they're facing lying criminals 8 hours out of every day; and these are awful people, like guys who beat up their wives. They're usually faced with scum who clearly deserve more punishment than a night in lockup will give them, but they can do nothing within the bounds of the law. But the lab guys don't have these same stressors. Most just want to find the truth.
RTFA. They addressed this very topic. They said the FBI lab was run this way, but the investigator from Philadelphia said that he never felt pressure to opine one way or another.
The FBI lab was forced to undergo serious changes as a result.
I think he was talking about unit testing, not system or integration or even feature testing. Effective unit tests do not rely on external dependencies. You don't need or even want a poorly configured or missing "test environment" to mess up your ability to run the tests. Instead, you provide fakes that yield exactly the expected values to trigger your program's behavior. You can have fakes that drive you in the direction of the happy path, fakes that steer the code down the sad paths, and still other fakes that take you down exception paths. Mock objects can provide everything from simulated database data to simulated service responses. And the values are all coded into the tests, where they provide documentation to anyone who can read code: if you call the Add(2,3) method you should expect a result of 5, if you call the Divide(5,0) method you should expect a DivideByZeroException will be thrown, etc.
And if you have code that is not unit testable today, you can certainly refactor it until it is testable. Once you learn a few patterns and anti-patterns, it's not even that hard to do. The real question is whether or not your boss thinks it's worth it to go back into a working million line project and add unit tests until you achieve 100% test coverage. (Sadly, the reply is generally "you want to spend HOW MUCH on testing code that already works? Get out of here, you psycho!") But is it worth it to add unit tests around any changes you make going forward? Absolutely.
NASA, Boeing and Lockheed are unlikely to ever build a new launcher especially one with a price tag running to $40 billion dollars. They have completely lost the engineering capacity, the fire in the belly and the desire.
For the Biggest Science projects, NASA only does what they're told - the politicians have to say "send a colony to Mars", it's not a choice that NASA makes on their own. And Lockheed and Boeing are contractors. They don't choose what NASA does, either. Remember, NASA didn't stand in front of America and say "we will send a man to the moon by the end of the decade; not because it is easy, but because it is hard." That was Kennedy.
I don't think they've lost the engineering capabilities. But what they have done is added so much process in terms of safety engineering that they probably couldn't get another Apollo mission together without passing several centuries worth of review boards. Trust me, that kind of bureaucracy will kill any engineer's spirit to create or innovate. (That's why I admire Boeing engineers more than any others: they are still able to create new airliners in an environment that requires a dedicated production line at 3M just to create all the red tape they see. Boeing may not roll out a new model every year, but they persevere until they ship it. That takes serious dedication.)
Slashdot Mirror
They do. Some data center operators get a deal on their electricity if they run their backup generators during peaks, reducing the load on the grid. As a bonus, it's a cheap readiness test of their backup power systems.
To be fair, they've improved data storage reliability as a result of incidents like these. Now, when you provision a storage account under Azure, there's a check box (checked by default) that turns on "global backup". When on, any data written to storage is fanned out across two geographically diverse data centers.
Say what you will about their technical mistakes, Microsoft does learn from them. As far as their decisions when it comes to product development and marketing, though, they don't appear to have learned a damn thing in the last decade.
That's certainly a different way of looking at it. Primitive people had it good because they were unaware of how bad they had it. To them it was natural to bear four children in hopes that one might mature to adulthood - although a fifth child was unlikely, as the mother's chances of surviving that many childbirths were not high. People didn't die from the nasty age-related diseases like cancer or heart disease because they died much earlier due to injury, hardship, or violence. So no, I don't buy the "non-material primitive society had it as good as us, but in their own way" position for a second. On average, people in today's western societies work much less and live far longer than our forebears, and have our basic needs well and truly exceeded. (Are we happier today? Different question. Today we have so much idle time that people seem to invent problems just to have something to worry about.)
Beyond that, though, I don't see us as being on different sides. I think a basic income is a great idea. Our society certainly is rich enough to afford it.
I would advise caution, though, because it will breed a dual economy. There will be the rich, who will continue to have the money, toys, power, and authority, and the poor who will be given their bread and circuses.
Crime would stay low: most criminals start from a position of desperation, but once they realize the profitability of crime, they expand. Remove desperation, and fewer people will start. That's huge in terms of safety.
In the best case, all it would take to get out of the basic income bracket is to expend enough effort to better yourself. Do something to earn more money, and you've improved your life by that much. Rewards are a powerful incentive.
But the rich historically have not been welcoming of additions to their ranks. Sharing power is unlikely. Sharing their wealth is equally unlikely, even if it would make society an overall better and safer place for them. Could a self-starter really rise to join the elite? Instead, I'd guess that not only would we have a dual economy, we'd have a dual currency, far more specific than the difference between "old money" and "new money." Your Basic Income dollars would buy you food and shelter, and there'd be a thriving barter economy of swap meets and marketplaces. But it would take the new ThalliumVisa to buy a car, or a Versace handbag, and for some reason the ThalliumVisa bank wouldn't accept trade in BI$. Class differences would become unbridgeable. And those differences would breed strife.
Shifting slightly, something I've noticed about people is that people hold different concepts of fairness. Some think that fairness means "you must earn everything yourself, and we should all have an equal chance to earn." To those people, welfare is bad because the recipients didn't earn it, or it's bad because there are "welfare queens" who abuse the system. They see examples of abuse and think the system is a failure because abuse happened. I believe instead that fairness is socialism: have a safety net for everyone. We have so many resources that we should be appalled that there are any hungry children in this society, or that someone should be denied health care or shelter because they can't afford it. I understand there will be abuses in that system because they're not preventable, and I'm willing to live with a few of them while we continually fix the system.
Anyway, yeah, bring on the dual economy. Let's try that for a while.
Stonewashed blue jeans are tumbled in a slurry of abrasive pellets. The breakdown of these pellets is a grit that is trapped in the pockets, and which scratches the heck out of anything stored in them.
No, the solution has to be government interference of some kind. There is no world government (thankfully), but the downside is that nobody mandates fair trade or labor practices between countries. That's up to the countries themselves.
Consider an example under way right now in the solar panel industry. Domestic solar panel makers build factories that cost $10 million dollars to produce panels that cost $5,000 each. A Chinese factory with a leftover billion dollar chip factory (they upgraded to a $10 billion dollar factory to make state of the art chips) now starts making solar panels much more efficiently and with much cheaper labor, for $1,000 per panel. Domestic makers can't compete and will soon go out of business. Chinese factory will raise prices to $10,000 per panel. Domestic consumers will then be paying twice the price of domestic solar panels for foreign imports. Everyone in the U.S. knows that if a new domestic maker tries to enter the market to sell $4,000 panels, the Chinese will lower the price to $1,000 until the domestic factory is out of business, then raise them back to $10,000.
That's monopolistic behavior, and would be illegal under the Sherman Act. Under current trade practices, it's perfectly legal for the Chinese to do it.
There is no way to compete in a global market when domestic producers have to follow domestic laws that do not apply outside our borders. And those domestic laws are there to prevent domestic companies from doing exactly the same thing, which we already agree is bad for all consumers.
The answer is punitive tariffs, (aka government interference.) We shouldn't try to balance everything out and derive a "fair price" for solar panels, and tax the Chinese solar panels until they're competitive with domestic solar panels. We should look at the behavior in this case, and say "you profited at $10,000 panels, you drove an American company out of business with $1,000 panels, the tax on all imported panels is now $10,000 and the tax on all solar panels from China is now punitively set at $20,000." If we do this in every case where industries practice dumping on us, the Chinese will get very mad at the factory owners who keep trying to screw the U.S., and will take care of the problem internally. Meanwhile, the $10,000 tax will prevent gray market panels from China making an end-run through other countries. And in the U.S., domestic companies will be free to compete with each other for our business, selling panels at fair domestic prices of $5,000 or $4,000.
We're never going to take care of our own people until we protect them.
No, advertisers have no business value in taking this seriously, and every incentive to find ways around it. Advertisers know that targeted advertising is both more effective and less offensive than random advertising. And today, in order to provide targeted advertising, tracking provides an overall picture of an individual's tastes and habits.
Targeted advertising will continue, but in a different fashion. Instead of tracking you at an individual level, I predict they'll find ways to identify surfers at an aggregate level. Perhaps cookies will contain your advertising preferences instead of your identity. If you visit a few car forums every so often, the advertiser might store the CAR=2 value in your cookie, if you visit Etsy a lot they'll store CRAFT=5, and if you visit a political campaign donation site, they'll store SUCKER=9. Then, as you surf the web, if you're identified as a SUCKER>3, the ads will feature Vote Rombama. They're not tracking you personally or identifying your habits, but they're still targeting you.
The bigger problem, and why I think Google was the big holdout, is Google Analytics. They don't use the tracking data just for advertising. They sell information about shopping habits to marketers. If you search for "stereo reviews", then click the links to stereo-reviews.com and hifiworld.com, then buy a Coby home stereo for $299 from shopping.com, they can provide that data to stereo retailers around the net: people who buy overpriced crap surf believe the information on stereo-reviews.com and hifiworld.com. Killing tracking kills that intelligence business.
This would be apt if Google did not already provide a way for users to universally opt out of doubleclick, and a chrome plugin that controls your opt out at key internet advertising associations.
In other words, "Google already provides a combination of non-intuitive and user-download-required ad hoc ways to do one fourth of something that's being standardized elsewhere, so we should think it's OK for them to ignore the standardizations under way."
Hmm... for a minute there I thought you were shilling for Google. But now I realize you're just cognitively impaired.
There's a big difference. Almost all of the "serfs" today (at least the ones living in America) have shelter, food, clothing, running water, heat, electricity, lighting, TV sets, refrigeration, cell phones, cheap and fast transportation, medical services, borders secure from invasion, and on a level far surpassing the living standards of even the kings of the middle ages.
Sure there's inequity. The balance of wealth distribution is today skewed beyond understanding. I don't have a four-Lamborghini garage, or a stable of race horses; I have a ten-year-old Ford truck and a couple of dogs. But when you start worrying about how bad we have it in comparison to the 1%, or whine about money spent on taxes, try to also compare yourself to the 99% from 150 or more years ago. We live better today than every single human ever prior to 1850.
Don't we already have enough people on the internet? Why do we keep encouraging more? :-)
Note: to all you humor-impaired people, the smiley face indicates this is a JOKE.
What would make an audio codec something worth using that would make you switch?
For the listener, there doesn't have to be a "switch", because it's not an "either-or" proposition. An audio playback system already has a library of codecs at its disposal, and chooses the correct one based on the encoding. Some players like Windows Media Player will dynamically download new decoders on demand as required. There is no burden on the consumer, as long as their devices support the new codec. And sure, other players (like iPods) may require some kind of user-initiated update, and those could mean a loss of audience if no update is available. There will always be those who are disenfranchised for a few years.
But the real cost of encoding is borne by the distributor of the audio. Those are the people who have money reasons to switch. Consider the guy running a web service who is pumping out music streams to thousands of listeners. Adding a new codec that's 10% more efficient at compressing audio means he can fit 10% more listeners in the same bandwidth, which gives him 10% more ad revenue. If only half the users have the new codec, well, he's still gaining 5% more capacity. If I'm making a greeting card playback chip, perhaps I can buy smaller (and therefore cheaper) ROM chips to hold the same amount of sound. If I'm building a telephony switch, perhaps I can increase the number of channels I support simultaneously. If I'm building a battery powered music player, a more efficient decoder can mean longer battery life, more song storage capacity, or could make more CPU available to other applications making it a faster overall device. And to anyone paying an annual license for their current encoding software, a truly free codec could mean lower costs.
Out of everybody involved, the merchant is almost the least at fault. In order to accept a few dollars worth of a transaction, the merchant is forced to handle these things called "account numbers" and "credit cards" that represent tremendous potential value, even though the merchant might be a dollar store with transactions never worth more than a few dollars. Imagine a two-buck shrimp shack on the beach, where half the customers pay with thousand dollar bills, and they each expect $998 in change. That's pretty close to the current situation.
The customers are not really to blame, either, except that they have proven themselves incapable of coping with increased security measures. They demand convenience and speed, perhaps unreasonable convenience given the value their cards represent. They take little responsibility for their own security, because the law says they're liable for only $50 worth of loss.
The banks are slightly more to blame. They are supposed to be responsible for securely handling all my money, but if they screw up they can send it all out the doors without my approval.
The real culprits are the card payment networks. The entire credit system is insecure from start to end, from its inception through today. The card networks are the ones who established the insecure systems and the insecure standards, they approved the insecure protocols, and they way they were designed back in the 1980s they didn't care about anything except making sure they got their slice of the money as it passed through their networks. When the fraud problems got too high, instead of redesigning their systems to be secure, they took a giant stick to the retailers and said "make your systems carry our insecure data more securely or we will fine you and kick you out of the credit card club." They didn't even tell the retailers what to do, or give them secure protocols, or provide tools for them to do it, they just said "you figure out a way to make it secure, or else." So now thanks to those very expensive non-standardized regulations, there are ill-defined and haphazardly implemented patchworks of security here and there, and little wonder that there are data breaches reported almost daily.
If you could get everyone (or at least a very significant number of people) in the country to switch to cash, then maybe prices would go down. Otherwise, me switching to cash isn't going to reduce my costs one bit. All it's going to do is stop earning me cash back and sign up bonuses.
If everyone switched to cash, prices would likely go up, not down. For large merchants, I know the cost of handling cash is substantially higher than the cost of handling credit transactions.
Credit transactions: A cash register has to be on a network, and have a PIN pad attached. A reader has to read a card, and some bits take a few milliseconds to flow through a wire. An occasional piece of paper has to be printed, signed, and collected. Visa and the bank take their cut on the back end. Occasionally, a bad credit transaction will be charged back to the retailer. All in all, the process is very automated from start to finish, with a few special hardware requirements like PIN pads, and some secure handling activities go on, like changing certificates.
Cash transactions: A store keeps a large, expensive safe full of money in a secured back room. Two people open the safe, take out the day's money, and run it through another expensive machine to count it. They put it in bags, and someone carries those bags to the cash registers. Cashiers take those bags, count the money, and put it in their specially designed armored money drawers (tills). Customer comes, and says "I'd like to pay with cash." Cashier takes time to manually count the customer's money, places it in the till, then counts out and returns the correct change to the customer. At the end of the day, the cash is collected from the tills, brought to the back room, counted in the expensive machine, and is then stored in the expensive safe. The next day, an armored car service is contracted to drive up and pick up the contents of the safe, which they then drive to a bank. All that security equipment, safes, cash counters, cash register tills, money room security systems, alarms, all cost a lot of money to buy and maintain. The handling of the money so many times throughout the day costs a lot in labor. And the cost of an armored car service takes their cut as well.
Finally, cash has a lot of unique risks. There's a risk that people who are trusted to handle it might actually steal it; there's a risk that someone will pass you forged money; and it's a deliciously tempting robbery target, meaning all those people who handle it throughout the day are at risk of being shot by a robber. Nobody ever gets shot for a stack of credit slips.
For small retailers, the expenses are likely much lower. The safe might be right under the single cash register. The manager would likely have free time at the start and end of day to do some of the cash management activities. The money is probably deposited in the bank by the shop owner, and not driven in an armored car. But for the big stores, accepting cash is a huge expense.
surely these scanners can record serial numbers since they scan the bill for denomination anyway.
Surely they don't.
Bill acceptors rely on several attributes to test a piece of paper to ensure it really is money. They shine various lights at it and through it, they run it over a magnetic ink detector, they check the thickness, they measure the dimensions, they match the images to a database of known images, but they do not record the images of the bills they accept. They just keep track of the amount.
Now, the cameras over the cash registers are taking plenty of photos of you. It would be possible to go back through a day's footage and see who paid in cash throughout the day, and figure out which image had you spending the $100 bill. But even that's just a picture, and it still doesn't tell me what your name is.
Any data in a QR code that is invalid should only be marked as invalid and the bill sorted aside for later, manual investigation. No "action" with the data itself is required. It shouldn't matter if the data is a URL or an IP address or "echo y|format C: /q". There should be nothing processed but an ack that the data doesn't correspond to correct ranges.
Nobody here is arguing that they should do anything else. But people are instead arguing that the reality is different, and that stupider things have happened to many institutions that should have known better.
Consider this scenario: John's First National Bank and Laundromat decides they need verifier software. They outsource the writing to a cheap software contractor shop who doesn't care much other than they deliver on time. Developer at the shop says "I would be done faster if I download a QR reader library from the intarwebs." Developer downloads a copy of zxing and wires it into the app. Unbeknownst to everyone involved, he included the portion of the library that has keyword recognition built in, and will auto-launch a URL. Nobody thinks to test it with a URL, because their specs don't say anything about dollar bill QR codes with URLs encoded in them.
It's not like every bank needs to have this happen. If even one bill verifier package in one bank contains this flaw, a tampered QR bill will flow through it at some point.
Implausible? Spend enough time in Corporate America and you'll find it's not even improbable.
Who said that the QR code will encode an URL?
This is not written in the engadget article, and that's the main erroneous assumption of the Slasdot poster (planetzuda).
It's not an erroneous assumption at all. The banks wouldn't print a URL in their QR code, but we're talking about an attacker modifying a bill, not the bank.
QR codes encode data, and that data can include ways to automate the processing of it. Smartphone QR reader applications know how to parse many recognized QR keywords, which include sms:, smsto:, mms:, mailto:, tel:, geo:, market:, youtube:, and of course http: and https:. While it's highly unlikely, it's not impossible that the bank's software reading the QR codes might interpret one of the known keywords and take an action.
An attacker would encode a malicious URL and print it on the bill, with the expectation that the bank would scan the code while in the normal course of verifying the currency. They'd be betting on the off-chance that there's something wrong with the security of the QR reader software in the bank's system, in hopes that their bill verifier would become corrupted in some way.
Think about it: for an attacker, having a zombie machine inside a bank's firewall would be like downloading a gold bar. And the risk? Virtually zero. Print up a bunch of nano-sized QR dots, stick them on your banknotes, and go spend them. Nobody will be able to trace that cash back to you. At some point in its future, that note will flow through a bank's verifier, and then you might get lucky.
It's no different than someone printing a QR code containing a URL that points to some malware site and sticking it on a movie poster, hoping that some idiot with their phone will read it and get infected.
Of course, a good tactic for the bank would be to have the U.S. Secret Service follow up on any illicit URLs they found embedded in the cash. They'd be able to proactively ID the malware server before even connecting to it, catching the bad guys unawares.
The 10 minute wait is an interesting compromise. Most security measures work by slowing down a determined attacker, not by absolutely preventing the loss, so that's a good idea. But in the CCTV footage I saw where a couple guys stole a BMW in minutes, they were a bit jumpy but didn't seem to mind that it took that long.
The balance might come in a combination of time and physical effort. Perhaps the lost key compromise could require lifting the car, then using a specialized, bulky tool (which is functionally not much different than a master lock and key) to access something from beneath the center of the car, combined with a 10 or 30 minute wait. Or perhaps the suspension has travel limit sensors that could be used - all four wheels have to be off the ground for 10 minutes. Or maybe something else that requires a tremendous amount of energy - compressing a large spring to trigger a contained switch. Make it require a set of things that are too difficult for the average thief to carry - a jack and jack stands, a long specialized tool, at least something more than a guy can carry in his pocket.
If it's going to remain purely digital, have it use digital certificates and require on-line network access to reset it. The BMW factory would then be the gatekeepers of who would be authorized to access the "blank key reset service". The on-board computer would log all key reset activities on a ROM, which they could provide access logs to police for investigations. A crooked dealer would quickly be identified, and immediately revoked by the service.
It still won't stop a tow truck or a mobile garage, but they're not stopped today.
Do you find there is still demand for mid/side stereo encoding? I'm not in the audio business, but it seems to be a holdover from ancient technologies like FM stereo. Since virtually everything is digitally processed these days, do people really need the pre-combined streams? When does it provide a realistic advantage?
I know that a mid-channel track can be processed on its own and played back as a mono source. You don't have to burn the energy to decode a second track. But that's its only advantage. And it is an advantage only in a very limited energy environment, where the tiny draw of a chip processing a second audio track would cost you battery life. But when I look at modern uses of audio, there are no cases where mono playback intersects with tightly limited energy:
Stereo: personal music (headphones); car audio; home entertainment systems; commercial television; video playback.
Pure mono source: telephone; web videos.
Mid-channel mono only: public address; ambient background music.
For all cases of stereo, you have to decode two streams anyway. There is no saving in having a mid/side encoding, as you have the mid channel in one track and the side channel in another. Two tracks == twice the energy, regardless of encoding technology.
For all cases of pure mono, you don't need mid/side encoding, as you have only one channel of input. Energy saved automatically.
For the cases where you would downconvert a stereo mid/side channel encoding to a mono playback system, the typical environments are fixed installations, where wall power is readily available, and the tiny energy consumption of the second track is not an issue.
So from my limited point of view, I don't see any rational reason to continue to support such an arcane format. What rationale did you have for including it?
So I was trying to figure out if SILK would permit tying streams together, or if it provides any kind of synchronization with an external time source. All I can really find is a section on clock drift, and using tools like OPUS_GET_PITCH() to figure out how far the drift is. And there's a section on stuffing multiple frames (up to 48) into a CBR Code 3 packet. But it doesn't indicate anything about how those frames would be related.
Which is too bad, because it really does limit the applications. Without sync, how could Opus streams be used for the audio tracks on a video?
All I can think is that I'm really missing something fundamental here. They had to have considered these other applications, didn't they?
They also left out n-channel support. You get mono or stereo, but that's it. No 7.1 surround encoding. That would have made it the one true codec for everything. That and lossless encoding. The two true codecs for everything.
Oh, and support in the iPhone. That would have made three true codecs. Among the many true codecs... oh bugger, I'll start again.
That's the nice thing about standards. There are so many to choose from!
At some point, I do trust that at least some of the people who have these hard jobs still do them for honest and noble reasons.
I think it's really hard for guys like street cops to maintain objectivity when they're facing lying criminals 8 hours out of every day; and these are awful people, like guys who beat up their wives. They're usually faced with scum who clearly deserve more punishment than a night in lockup will give them, but they can do nothing within the bounds of the law. But the lab guys don't have these same stressors. Most just want to find the truth.
At least the optimist in me still believes that.
What if the test consists in building an AI that can pass the test?
Then they have built a test that will accurately detect the arrival of the Singularity.
RTFA. They addressed this very topic. They said the FBI lab was run this way, but the investigator from Philadelphia said that he never felt pressure to opine one way or another.
The FBI lab was forced to undergo serious changes as a result.
I think he was talking about unit testing, not system or integration or even feature testing. Effective unit tests do not rely on external dependencies. You don't need or even want a poorly configured or missing "test environment" to mess up your ability to run the tests. Instead, you provide fakes that yield exactly the expected values to trigger your program's behavior. You can have fakes that drive you in the direction of the happy path, fakes that steer the code down the sad paths, and still other fakes that take you down exception paths. Mock objects can provide everything from simulated database data to simulated service responses. And the values are all coded into the tests, where they provide documentation to anyone who can read code: if you call the Add(2,3) method you should expect a result of 5, if you call the Divide(5,0) method you should expect a DivideByZeroException will be thrown, etc.
And if you have code that is not unit testable today, you can certainly refactor it until it is testable. Once you learn a few patterns and anti-patterns, it's not even that hard to do. The real question is whether or not your boss thinks it's worth it to go back into a working million line project and add unit tests until you achieve 100% test coverage. (Sadly, the reply is generally "you want to spend HOW MUCH on testing code that already works? Get out of here, you psycho!") But is it worth it to add unit tests around any changes you make going forward? Absolutely.
NASA, Boeing and Lockheed are unlikely to ever build a new launcher especially one with a price tag running to $40 billion dollars. They have completely lost the engineering capacity, the fire in the belly and the desire.
For the Biggest Science projects, NASA only does what they're told - the politicians have to say "send a colony to Mars", it's not a choice that NASA makes on their own. And Lockheed and Boeing are contractors. They don't choose what NASA does, either. Remember, NASA didn't stand in front of America and say "we will send a man to the moon by the end of the decade; not because it is easy, but because it is hard." That was Kennedy.
I don't think they've lost the engineering capabilities. But what they have done is added so much process in terms of safety engineering that they probably couldn't get another Apollo mission together without passing several centuries worth of review boards. Trust me, that kind of bureaucracy will kill any engineer's spirit to create or innovate. (That's why I admire Boeing engineers more than any others: they are still able to create new airliners in an environment that requires a dedicated production line at 3M just to create all the red tape they see. Boeing may not roll out a new model every year, but they persevere until they ship it. That takes serious dedication.)