Exactly. As much as I love my Linuxy goodness, I'm fully aware that practically every driver I use is community-written, community-packaged, and community-maintained. Yes, there's a few vendors out there releasing officially-supported drivers, but they're still rare.
Most warranties come with a disclaimer voiding them in the event of misuse. Every FOSS driver is effectively misuse from the perspective of a warranty, because the driver could do something to the device the manufacturer didn't approve. Yes, you could look into the source and find out what the driver's actually doing, but that's prohibitively expensive.
And if she'd rolled with the hit, she wouldn't even have problems with her wrist. You can run into a solid brick wall at 10mph without sustaining any injury unless you stiff-arm it, then you're gonna break something.
She did roll with it. Specifically, a barrel roll. The force of impact was enough to shatter her wrist, raise her center of mass two feet, and spin her upside. She may have also fractured her skull, but the doctor did not do an X-ray on it because it showed no signs of being severe enough to require special treatment.
The assertion that her arm spared her femur came from the attending doctor, not me.
At the risk of hearing people crying about "blaming the victim", I'll just suggest that your wife doesn't pay any attention to anything around her if that's what happened.
I'll just suggest that you're an ass. My wife was already halfway across the crosswalk as the car approached the intersection. Had the car stopped and proceeded slowly through the intersection as required by law, there would have been ample time to finish crossing. The driver didn't stop, didn't proceed slowly, and didn't account for the reduced visibility.
My wife is in this category, and now has bone problems in her wrist. If she weren't a rugby player at the time and followed her instinct to stiff-arm her vehicular opponent, she'd have likely ended up in a wheelchair for a few months.
The driver just didn't see her. It was nighttime, in a pretty empty part of a small town, and my wife was wearing dark green on the far side of an intersection at the bottom of a hill. The driver rolled through the stop and accelerated immediately, right into my wife.
Agreed. Interestingly, a lot of processing work in medicine is working toward eliminating animal testing, by better simulating physics and correlating similar chemicals with known results, ultimately reducing the amount of animal testing that's necessary to produce a safe drug. Animals (including the human kind) are still involved, though, and will be for a very long time.
As for finance, I don't actually see any ethical problems here. Some bankers did things that sounded good on paper and matched all known criteria for something where gain outweighed risk, but those formulas didn't anticipate the combined effect of everybody doing the same thing at once. Was it a mistake? Yes. Did some bankers realize they were doing something wrong, and do nothing? Indeed they did. Does that make the entire industry a wretched hive of scum and villainy? Not really, despite Slashdot's vocal opinion of those "evil bastard bankers".
As the saying goes, you've got to break a few eggs to make an omelette. Every industry has its distasteful activities, and every industry is trying to eliminate them, albeit in a large-scale, disorganized, and very slow way. Why not help them move forward to a world of more ethical practices, rather than just hiding from the problem and assuming that anyone will care about your silent and ineffectual protest? As another saying goes, all that is necessary for evil to triumph is for good men to do nothing.
Furthermore, population growth and AIDS are larger problems than the African civil wars. Rwanda's population is already larger that what is was before the genocide there.
Population growth and AIDS are partly due to the civil wars. If an area's unsafe, volunteer educators and doctors are far more rare, so STDs spread rampantly. With uncertainty about the future and high rates of child mortality, people reproduce as much as they can, trying to ensure that their family/tribe/group will endure, and even grow large enough to eventually win whatever the current conflict is. These civil wars have grown from centuries of tribal conflict, so the battle plans are laid out on a scale of generations, with parents expecting that their children will some day fight for their tribe in glorious battle, if only those damned Westerners would get out of their way and stop saving whoever's losing the war this decade...
Supporting peaceful endeavors (including "mining companies that won't kill each other") and education is the best way I know of to solve all three problems. With education comes a better economy, sanitation, a more stable future, lower birth rate, which finally leads to better education.
Exactly. The mechanism is not the loophole, but the abuse is.
Transferring your money to a different part of the same company? That shouldn't be taxed.
Your company's expenses were the same as its income, so you had no profits? That shouldn't be taxed.
Most of your profits are made in a country with low tax rates? That should be taxed, but only at the low rate.
Put them together with a hefty helping of accounting mayonnaise, and you have a Double Irish and Dutch Sandwich arrangement, a fully-legal loophole. Properly pulling it off requires at least four companies in three nations, so it's not something the average person can do in its entirety.
I personally, however, have made use of several of the provisions that make it work, so I won't claim I'm against any single part. I've transferred money to (and from) a business of my own, being happy not to face taxes on every transfer. I've moved money to a country with practically no taxes, because I was living there.
But there's realistically no way to know that from an attacker's perspective. The password could be words, or it could be a string of random letters. If the system allows symbols, symbols will need to be included in the search space.
This tool calculates the brute-force time on a character basis. It says that dictionary attacks still work and should be mitigated by policy and practice.
The reason Google, Netflix, and the like don't already pay enormous amounts of taxes is because old tax laws have been riddled with loopholes. Legislators try to fix this by adding new taxes, because it's easier to make new laws than revise old ones.
As mentioned elsewhere, there's not much better than having Real Engineers go on sales calls, too, to answer the technical questions. You can teach salesmen all you want, but they won't be able to fake the insight gained through experience.
All salesmen should have some familiarity with the industry they're marketing to, though. They should have an understanding of how a programmer's mind works, and how your product makes the customers' lives better. For that, I recommend BASIC more than anything else. Not VB, mind you, but good ol' BASIC:
It's (usually) plain English. There are few abbreviations, and most structures read as a straightforward sentence. That helps to keep focus on general structures and concepts rather than syntax details.
No overhead. There is no boilerplate necessary to just make something that runs. That means that your first lessons can cover things like "the program runs one step at a time, in order," which is a lesson often missed in many introductory courses, and not obvious to many non-programmer folks.
Most structures (depending on version), in simple form. No, you likely won't find multithreading, but you can show a function call, loops, conditionals, variables, objects, and most other programming elements just fine, and without needing much other syntax to make a demonstration program. Pick a flavor of BASIC that includes features supported by your product, for illustration.
No practical application. This is a bit of a lie that really should be told to all students. Make it clear from the start that they should never attempt to write a "real" program in BASIC, not because it's impossible, but because there are far better languages out there. Toward the end of the lessons, start introducing them (especially C/C++, since it's what your customers use). Use that as a leaping-off point to show that all languages are functionally similar.
Once the run-through with BASIC is complete, you can expect the salesmen to understand how to read a simple (and commented!) program, and work out what it does. Show them equivalent programs written in C, C++, and BASIC. Be sure to point out how your product makes life easier, and show how a competitor (or Notepad) doesn't, tying in the lesson with the ultimate goal of making better salesmen.
You definitely won't be producing any great programmers, but you'll give them a glimpse of the mental juggling we do. They'll be able to recognize common use among customers, and possibly even impress a few with their knowledge. That's enough to significantly improve their relationship with the potential customer.
Here's an idea that could definitely save these retailers.
People need entertainment, right? And they want something they like, right?
We'll hire a bunch of college kids, call them the "Media Squad" and have them review movies, video games, and music, and recommend them based on each customer's particular tastes. By sheer coincidence, they'll recommend things most that are brand new, and stuck with a high markup. To prevent the kids from getting any silly notions like recommending Hulu or other things we don't make profit on, we'll give them sales quotas.
A friend of mine (whom we shall call "Art", for he is now an art teacher) was once involved in a city-wide scavenger hunt, where one item to find was "a police officer's badge number". Art and his team found a patrol car stopped in front of them at an intersection. Doing what all (un)reasonable young adults do in their young adult years, they chose the most straightforward method to get the officer's attention: they rammed his car.
More or less... They pulled up behind the car, then inched forward while the cop gave them quizzical looks in his mirror. Moving as slowly as he could manage, Art bumped the officer's vehicle. The officer got out, asked what was going on, understood the situation, laughed about it, gave them his business card (in lieu of his badge number, which is apparently against city policy to give out), and let them off with a warning to avoid hitting patrol cars in the future.
Despite Slashdot's popular opinion, most police officers are decent people. Treat them as such, with respect for the fact that they're trying to do their job, and they'll usually treat you decently as well.
"Let's find out," the officer cheerfully replies, as he picks up the radio and calls the dispatcher. "Can I get some assistance at the corner of 3rd and main? There's some wacko here who says he comes from the Internet on a quest for knowledge and troll skins. Could be a mental patient."
Cops could (in theory, with the right legal framework in place, and the right IT support, and funding, etc.) use Google's data and analysis as strong indicators of suspicion. That could be useful, but it's not nearly enough to warrant an in-person police response.
An analogy would be for me to run up to a random cop on the street and ask him how long it'd take to get reinforcements to the area. It's not the kind of activity that normally happens, so I've probably earned a bit of surveillance and a few funny looks, but it's no reason to be arrested on the spot.
No, they wouldn't. Actual background checks take too long and return far too much information. Today, your visit is recorded in a database, correlated with the fact that your plane had held a suspected terrorist in its last flight, so the conclusion is that you must be the recipient of a secret package hidden inside your seat cushion. That's enough to get a GPS tracker on your car and addition to the no-fly list.
Cue the hundreds of Slashdotters who proclaim "Oh, they're reinvented obsolete databases" and "Just wait until they need ACID, then they'll be fucked", the NoSQL blind-faith followers who harp about pure scalability and clustering, and at least a dozen references to an animated video of a retarded strawman saying "webscale" repeatedly.
Somewhere in the depths of poorly-researched comments will be some guy who thinks that NoSQL is a tool that really just might be useful for particular use cases, and should be used where appropriate, and nowhere else. Sadly, his post will be missed because everyone's too busy talking about how everything can be done just as easily on a $500,000 server farm running Oracle's latest and greatest turd.
You know, you don't HAVE to make the call. There are people who can make calls for you.
And heck, there are wonderful people known as "ham radio operators" who have a backup radio network as well
Privacy? We don't need no steenkin' privacy!
We'll try another round for getting the point across: All these systems have problems, and "there are cases where a particular radio communication system fails" is a common one. Sure, you can work around it by using a different network, and try to shoehorn the existing system in ("Let's shout through the storm drain system to a ham operator who can radio the hospital so they can call MedicAlert and find out if this guy's allergic to anything"), or you can use something else ("I'll take a picture of this code, and when we get cell service en route to the hospital, we'll know more").
Once again, my main point is that having options is a good thing. Just because you can cobble together a system with existing parts that's still functional in contrived circumstances is no reason to dismiss an innovative idea.
Microsoft has Bing, which supposedly uses user histories to judge what kind of results they want. They already have the reputation for being evil, though, so nobody really expects them to honor DNT. They can gather all the data they want, and laugh for a while until Google launches its next product to embarrassingly point out Microsoft's lack of innovation.
It does seem that you are doing a little fear mongering of your own to promote a QR code based system over the current system.
Absolutely not. That's just sarcasm. I'm in favor of both systems, and RFID chips, and carrying card in wallets, and if somebody could work out a system where a skywriter writes medical information about people in the street, there's a good chance I'm in favor of that, too. Paramedics need warnings of "this will make things worse" as soon as possible, and I'm in favor of supplying that information through whatever mechanism is convenient, fastest, and most accurate at the time it's needed. One paramedic may work better with the QR code system, and one may prefer to have a human to talk to. If both are options, the paramedic can pick.
I wouldn't call the GGP suggestion of printing all the information on a bracelet as "fear mongering" just impractical.
This is the fear mongering I object to:
2) Person is in a reception dead zone. (Soon to literally be a dead zone.)
3) Disaster scenario: What happens to mobile phone reception?
Both of those are current problems with the MedicAlert system (since it requires a phone call, as well). To suggest they're unique to the QR system is ignorant, at best. They also affect (to a lesser degree) a paramedic's access to information the hospital already has. If you're being pulled from a car crash in a tunnel, they're not going to have your information, and that's just too bad.
I think you meant availability which is not equivalent to accuracy.
Very true. I worked with medical data a while back, and the single biggest problem I encountered was missing information. Fortunately, there are efforts underway to remedy this. Healthcare providers and pharmacies are increasingly using protocols like HL7 to share data,and (through data use agreements) it's legal for this project and MedicAlert to share information to ensure they always have accurate information available. Whether they're doing so is another matter, but that's still no reason to reject the idea outright. Conditions that are severe enough to drastically affect paramedics' treatment is unlikely to change often enough to matter, anyway.
I used to work in the healthcare field, and I can assure you that this is half-right. Saying 95% of people are terrible with their recordkeeping is being optimistic about humanity. I personally carry an emergency contact card in my wallet, where the first four of eight numbers don't work anymore. I'll update it someday (or maybe just put a sticker over them...)
That said, it is possible for this to work. HIPAA doesn't outlaw access to "protected health information" but only outlaws unauthorized access. That means your receptionists' monitors must be hidden from view, data transfer must be encrypted, no lookups without a professional need, and so on. It's possible (and common) for third parties to receive medical information if they have a contractual agreement and operational need for it. That's how insurance providers can get treatment details, and how pharmacies can notice possible interactions.
Many healthcare providers are already transferring information via HL7 and other protocols. Adding this system to an existing (and ideally-configured) HL7 transport takes about an hour, and all relevant information will be passed on.
MedicAlert does have all of my father's information. MedicAlert also has the reception and time problems (since it involves a phone call) that the GGP is using as a fearmongering argument against a QR code system, but that's not what he wanted:
Just write the dam thing in English on the bracelet and all you need to do is be able to read English
This system is prone to many failures, some of which are similar to ones the bracelets have. Again, the main point is that redundancy offers reliability.
I present a counterexample, illustrating necessity: my father. Among other problems he has diabetes, a congenital heart condition (aggravated by a later trauma), a deformed left hip, stunted growth in his right leg, skin cancer, an allergy to iodine-based surgical preparations (because of an "inert" additive), a severe allergy to all seafood (the root cause of the iodine issue), and high blood pressure. He's also an organ donor, and has requested DNR.
That's all useful for paramedics to know, and it's a heck of a lot for a bracelet.
I'm surprised they aren't using a proprietary 2D barcode format, but that would mean hiring 3x as many developers...
1) EMT doesn't have the app.
I'm actually starting a project of my own, and looked at many barcode formats recently. QR is the obvious choice, for the main reason that it's already so popular and reliable. The summary implies the QR code is just a website link (and/or a password) to your medical records, which are hosted elsewhere. That means that any old QR code reader would work, so there's practically widespread deployment already.
2) Person is in a reception dead zone. (Soon to literally be a dead zone.)
So nothing changes from now, and the paramedics know nothing, until the ambulance starts moving. If you live in the boondocks with many areas of no reception, get a bracelet and a sticker. My father has his bracelet and carries a folding card in his wallet (on the off chance a paramedic will see it) detailing his problems.
3) Disaster scenario: What happens to mobile phone reception?
Same as above. No new risk, possibly some good. If the lookup doesn't work, paramedics are no worse off than they are right now. This may be a shocking surprise, but paramedics have brains. There are very few conditions that will lead to death if the paramedic doesn't know about them. My father's iodine reaction was discovered thirty minutes into open-heart surgery. They gave him a few more drugs to suppress the reaction, finished the surgery, then noted that they'd need a different preparation for the next round of surgery.
4) Paramedics have time to surfe the web while trying to save lives?
Yes, they sometimes do. My most recent encounter with paramedics was after my wife was hit by a car. Three paramedics arrived. One used a flashlight to direct traffic (since it was late evening), one treated my wife, and one held the ambulance door open. I suspect the doorstop could have taken a few moments to scan a code and checked for allergies.
5) LifeSquare's web site is down. Whoops, guess I'll die of a reaction to penicillin then.
No, you'll have a reaction to penicillin, which the paramedics will notice, and take care of. It's more work for them, and makes the hospital trip more complicated, but it's how things work now.
Sure, there are privacy issues, but that seems to be the least of the faults with the system. Just write the dam thing in English on the bracelet and all you need to do is be able to read English. Low-tech solution is the right one.
What if your bracelet falls off, or gets dirty, or is tarnished, or if it's upside down and the paramedics can't move your arm to read it? You could die of a reaction to penicillin because the bracelet didn't work!
Like any important system, redundancy across different methods is the most reliable.
That doesn't work. Unless people are embarrassed they won't lift a finger, they'll just ignore it. Besides, the atrocities at Abu Ghraib were committed with the full knowledge and consent of the US Government.
Embarrassment works, eh? So now we have no more war crimes in Iraq, politicians are clean, and the world is a happy place now that WikiLeaks has embarrassed everyone into behaving, right?
Wrong. Utterly wrong. Using the fear of embarrassment as a motivator is a quick and easy way for any random person with information to feel powerful, but it's no more effective than, say, appealing to someone's judgement. Very often, all that's needed to correct an immoral action is to convince someone to take a fresh look at what they're doing. That can best be done with care, not fear.
It's funny that you should use Abu Ghraib as an example, since an internal investigation (started through those official channels) reported abuse to the media in 2004, two years before WikiLeaks was launched. In fact, practically all of the events (including trials) involving Abu Ghraib were concluded by 2007, well before the mass release in 2010. Boy, that sure is effective embarrassment, being three years too late and accomplishing nothing.
It happens in other fields, too. Try secretly reporting a security bug to Microsoft and see if they fix it. Compare to how long they take to fix published bugs.
That's what responsible disclosure is for: Go through the official channels, then at a later point inform the general public. In the worst case, the malware developers get a bit more time to run their exploits, and a few days after release the malware will be analyzed and the bug becomes public knowledge anyway.
As for telling journalists: It wasn't Julian Assange who released the cables, everything he released was redacted to avoid the problems you describe. People only got full access to the original documents when a journalist published the decryption key to Julian's "insurance file".
Note also that the "insurance file" itself was released to the public by a defector, Daniel Domscheit-Berg, who obtained it while working on an internal server. The file had previously been given to the journalist Leigh, with instructions to make it public. This raises three important questions:
1) Why put unredacted information in a file with potential to be released, anyway, if not to cause harm?
2) Why leave the file on any server that anyone but Assange could access after it's been copied elsewhere?
Assange's "insurance" was an outright extortion scheme. Either the government gave in to his demands of amnesty, or he'd risk people's lives. We can see how well that fear tactic is working out for him...
Slashdot Mirror
Exactly. As much as I love my Linuxy goodness, I'm fully aware that practically every driver I use is community-written, community-packaged, and community-maintained. Yes, there's a few vendors out there releasing officially-supported drivers, but they're still rare.
Most warranties come with a disclaimer voiding them in the event of misuse. Every FOSS driver is effectively misuse from the perspective of a warranty, because the driver could do something to the device the manufacturer didn't approve. Yes, you could look into the source and find out what the driver's actually doing, but that's prohibitively expensive.
And if she'd rolled with the hit, she wouldn't even have problems with her wrist. You can run into a solid brick wall at 10mph without sustaining any injury unless you stiff-arm it, then you're gonna break something.
She did roll with it. Specifically, a barrel roll. The force of impact was enough to shatter her wrist, raise her center of mass two feet, and spin her upside. She may have also fractured her skull, but the doctor did not do an X-ray on it because it showed no signs of being severe enough to require special treatment.
The assertion that her arm spared her femur came from the attending doctor, not me.
At the risk of hearing people crying about "blaming the victim", I'll just suggest that your wife doesn't pay any attention to anything around her if that's what happened.
I'll just suggest that you're an ass. My wife was already halfway across the crosswalk as the car approached the intersection. Had the car stopped and proceeded slowly through the intersection as required by law, there would have been ample time to finish crossing. The driver didn't stop, didn't proceed slowly, and didn't account for the reduced visibility.
My wife is in this category, and now has bone problems in her wrist. If she weren't a rugby player at the time and followed her instinct to stiff-arm her vehicular opponent, she'd have likely ended up in a wheelchair for a few months.
The driver just didn't see her. It was nighttime, in a pretty empty part of a small town, and my wife was wearing dark green on the far side of an intersection at the bottom of a hill. The driver rolled through the stop and accelerated immediately, right into my wife.
Agreed. Interestingly, a lot of processing work in medicine is working toward eliminating animal testing, by better simulating physics and correlating similar chemicals with known results, ultimately reducing the amount of animal testing that's necessary to produce a safe drug. Animals (including the human kind) are still involved, though, and will be for a very long time.
As for finance, I don't actually see any ethical problems here. Some bankers did things that sounded good on paper and matched all known criteria for something where gain outweighed risk, but those formulas didn't anticipate the combined effect of everybody doing the same thing at once. Was it a mistake? Yes. Did some bankers realize they were doing something wrong, and do nothing? Indeed they did. Does that make the entire industry a wretched hive of scum and villainy? Not really, despite Slashdot's vocal opinion of those "evil bastard bankers".
As the saying goes, you've got to break a few eggs to make an omelette. Every industry has its distasteful activities, and every industry is trying to eliminate them, albeit in a large-scale, disorganized, and very slow way. Why not help them move forward to a world of more ethical practices, rather than just hiding from the problem and assuming that anyone will care about your silent and ineffectual protest? As another saying goes, all that is necessary for evil to triumph is for good men to do nothing.
Furthermore, population growth and AIDS are larger problems than the African civil wars. Rwanda's population is already larger that what is was before the genocide there.
Population growth and AIDS are partly due to the civil wars. If an area's unsafe, volunteer educators and doctors are far more rare, so STDs spread rampantly. With uncertainty about the future and high rates of child mortality, people reproduce as much as they can, trying to ensure that their family/tribe/group will endure, and even grow large enough to eventually win whatever the current conflict is. These civil wars have grown from centuries of tribal conflict, so the battle plans are laid out on a scale of generations, with parents expecting that their children will some day fight for their tribe in glorious battle, if only those damned Westerners would get out of their way and stop saving whoever's losing the war this decade...
Supporting peaceful endeavors (including "mining companies that won't kill each other") and education is the best way I know of to solve all three problems. With education comes a better economy, sanitation, a more stable future, lower birth rate, which finally leads to better education.
Exactly. The mechanism is not the loophole, but the abuse is.
Transferring your money to a different part of the same company? That shouldn't be taxed.
Your company's expenses were the same as its income, so you had no profits? That shouldn't be taxed.
Most of your profits are made in a country with low tax rates? That should be taxed, but only at the low rate.
Put them together with a hefty helping of accounting mayonnaise, and you have a Double Irish and Dutch Sandwich arrangement, a fully-legal loophole. Properly pulling it off requires at least four companies in three nations, so it's not something the average person can do in its entirety.
I personally, however, have made use of several of the provisions that make it work, so I won't claim I'm against any single part. I've transferred money to (and from) a business of my own, being happy not to face taxes on every transfer. I've moved money to a country with practically no taxes, because I was living there.
This tool calculates the brute-force time on a character basis. It says that dictionary attacks still work and should be mitigated by policy and practice.
The reason Google, Netflix, and the like don't already pay enormous amounts of taxes is because old tax laws have been riddled with loopholes. Legislators try to fix this by adding new taxes, because it's easier to make new laws than revise old ones.
Link in link form.
As mentioned elsewhere, there's not much better than having Real Engineers go on sales calls, too, to answer the technical questions. You can teach salesmen all you want, but they won't be able to fake the insight gained through experience.
All salesmen should have some familiarity with the industry they're marketing to, though. They should have an understanding of how a programmer's mind works, and how your product makes the customers' lives better. For that, I recommend BASIC more than anything else. Not VB, mind you, but good ol' BASIC:
Once the run-through with BASIC is complete, you can expect the salesmen to understand how to read a simple (and commented!) program, and work out what it does. Show them equivalent programs written in C, C++, and BASIC. Be sure to point out how your product makes life easier, and show how a competitor (or Notepad) doesn't, tying in the lesson with the ultimate goal of making better salesmen.
You definitely won't be producing any great programmers, but you'll give them a glimpse of the mental juggling we do. They'll be able to recognize common use among customers, and possibly even impress a few with their knowledge. That's enough to significantly improve their relationship with the potential customer.
Here's an idea that could definitely save these retailers.
People need entertainment, right? And they want something they like, right?
We'll hire a bunch of college kids, call them the "Media Squad" and have them review movies, video games, and music, and recommend them based on each customer's particular tastes. By sheer coincidence, they'll recommend things most that are brand new, and stuck with a high markup. To prevent the kids from getting any silly notions like recommending Hulu or other things we don't make profit on, we'll give them sales quotas.
What could possibly go wrong?
A friend of mine (whom we shall call "Art", for he is now an art teacher) was once involved in a city-wide scavenger hunt, where one item to find was "a police officer's badge number". Art and his team found a patrol car stopped in front of them at an intersection. Doing what all (un)reasonable young adults do in their young adult years, they chose the most straightforward method to get the officer's attention: they rammed his car.
More or less... They pulled up behind the car, then inched forward while the cop gave them quizzical looks in his mirror. Moving as slowly as he could manage, Art bumped the officer's vehicle. The officer got out, asked what was going on, understood the situation, laughed about it, gave them his business card (in lieu of his badge number, which is apparently against city policy to give out), and let them off with a warning to avoid hitting patrol cars in the future.
Despite Slashdot's popular opinion, most police officers are decent people. Treat them as such, with respect for the fact that they're trying to do their job, and they'll usually treat you decently as well.
"Let's find out," the officer cheerfully replies, as he picks up the radio and calls the dispatcher. "Can I get some assistance at the corner of 3rd and main? There's some wacko here who says he comes from the Internet on a quest for knowledge and troll skins. Could be a mental patient."
No.
Cops could (in theory, with the right legal framework in place, and the right IT support, and funding, etc.) use Google's data and analysis as strong indicators of suspicion. That could be useful, but it's not nearly enough to warrant an in-person police response.
An analogy would be for me to run up to a random cop on the street and ask him how long it'd take to get reinforcements to the area. It's not the kind of activity that normally happens, so I've probably earned a bit of surveillance and a few funny looks, but it's no reason to be arrested on the spot.
No, they wouldn't. Actual background checks take too long and return far too much information. Today, your visit is recorded in a database, correlated with the fact that your plane had held a suspected terrorist in its last flight, so the conclusion is that you must be the recipient of a secret package hidden inside your seat cushion. That's enough to get a GPS tracker on your car and addition to the no-fly list.
Oh, look, it's a NoSQL article.
Cue the hundreds of Slashdotters who proclaim "Oh, they're reinvented obsolete databases" and "Just wait until they need ACID, then they'll be fucked", the NoSQL blind-faith followers who harp about pure scalability and clustering, and at least a dozen references to an animated video of a retarded strawman saying "webscale" repeatedly.
Somewhere in the depths of poorly-researched comments will be some guy who thinks that NoSQL is a tool that really just might be useful for particular use cases, and should be used where appropriate, and nowhere else. Sadly, his post will be missed because everyone's too busy talking about how everything can be done just as easily on a $500,000 server farm running Oracle's latest and greatest turd.
My next project: A network with the SSID "ThisRouterIsUpsideDown", with some specific settings applied.
Oh, give him a break... he's an IE user.
You know, you don't HAVE to make the call. There are people who can make calls for you.
And heck, there are wonderful people known as "ham radio operators" who have a backup radio network as well
Privacy? We don't need no steenkin' privacy!
We'll try another round for getting the point across: All these systems have problems, and "there are cases where a particular radio communication system fails" is a common one. Sure, you can work around it by using a different network, and try to shoehorn the existing system in ("Let's shout through the storm drain system to a ham operator who can radio the hospital so they can call MedicAlert and find out if this guy's allergic to anything"), or you can use something else ("I'll take a picture of this code, and when we get cell service en route to the hospital, we'll know more").
Once again, my main point is that having options is a good thing. Just because you can cobble together a system with existing parts that's still functional in contrived circumstances is no reason to dismiss an innovative idea.
Microsoft has Bing, which supposedly uses user histories to judge what kind of results they want. They already have the reputation for being evil, though, so nobody really expects them to honor DNT. They can gather all the data they want, and laugh for a while until Google launches its next product to embarrassingly point out Microsoft's lack of innovation.
Let's get one thing out of the way real quick:
It does seem that you are doing a little fear mongering of your own to promote a QR code based system over the current system.
Absolutely not. That's just sarcasm. I'm in favor of both systems, and RFID chips, and carrying card in wallets, and if somebody could work out a system where a skywriter writes medical information about people in the street, there's a good chance I'm in favor of that, too. Paramedics need warnings of "this will make things worse" as soon as possible, and I'm in favor of supplying that information through whatever mechanism is convenient, fastest, and most accurate at the time it's needed. One paramedic may work better with the QR code system, and one may prefer to have a human to talk to. If both are options, the paramedic can pick.
I wouldn't call the GGP suggestion of printing all the information on a bracelet as "fear mongering" just impractical.
This is the fear mongering I object to:
2) Person is in a reception dead zone. (Soon to literally be a dead zone.)
3) Disaster scenario: What happens to mobile phone reception?
Both of those are current problems with the MedicAlert system (since it requires a phone call, as well). To suggest they're unique to the QR system is ignorant, at best. They also affect (to a lesser degree) a paramedic's access to information the hospital already has. If you're being pulled from a car crash in a tunnel, they're not going to have your information, and that's just too bad.
I think you meant availability which is not equivalent to accuracy.
Very true. I worked with medical data a while back, and the single biggest problem I encountered was missing information. Fortunately, there are efforts underway to remedy this. Healthcare providers and pharmacies are increasingly using protocols like HL7 to share data,and (through data use agreements) it's legal for this project and MedicAlert to share information to ensure they always have accurate information available. Whether they're doing so is another matter, but that's still no reason to reject the idea outright. Conditions that are severe enough to drastically affect paramedics' treatment is unlikely to change often enough to matter, anyway.
I used to work in the healthcare field, and I can assure you that this is half-right. Saying 95% of people are terrible with their recordkeeping is being optimistic about humanity. I personally carry an emergency contact card in my wallet, where the first four of eight numbers don't work anymore. I'll update it someday (or maybe just put a sticker over them...)
That said, it is possible for this to work. HIPAA doesn't outlaw access to "protected health information" but only outlaws unauthorized access. That means your receptionists' monitors must be hidden from view, data transfer must be encrypted, no lookups without a professional need, and so on. It's possible (and common) for third parties to receive medical information if they have a contractual agreement and operational need for it. That's how insurance providers can get treatment details, and how pharmacies can notice possible interactions.
Many healthcare providers are already transferring information via HL7 and other protocols. Adding this system to an existing (and ideally-configured) HL7 transport takes about an hour, and all relevant information will be passed on.
MedicAlert does have all of my father's information. MedicAlert also has the reception and time problems (since it involves a phone call) that the GGP is using as a fearmongering argument against a QR code system, but that's not what he wanted:
Just write the dam thing in English on the bracelet and all you need to do is be able to read English
This system is prone to many failures, some of which are similar to ones the bracelets have. Again, the main point is that redundancy offers reliability.
I present a counterexample, illustrating necessity: my father. Among other problems he has diabetes, a congenital heart condition (aggravated by a later trauma), a deformed left hip, stunted growth in his right leg, skin cancer, an allergy to iodine-based surgical preparations (because of an "inert" additive), a severe allergy to all seafood (the root cause of the iodine issue), and high blood pressure. He's also an organ donor, and has requested DNR.
That's all useful for paramedics to know, and it's a heck of a lot for a bracelet.
I'm surprised they aren't using a proprietary 2D barcode format, but that would mean hiring 3x as many developers...
1) EMT doesn't have the app.
I'm actually starting a project of my own, and looked at many barcode formats recently. QR is the obvious choice, for the main reason that it's already so popular and reliable. The summary implies the QR code is just a website link (and/or a password) to your medical records, which are hosted elsewhere. That means that any old QR code reader would work, so there's practically widespread deployment already.
2) Person is in a reception dead zone. (Soon to literally be a dead zone.)
So nothing changes from now, and the paramedics know nothing, until the ambulance starts moving. If you live in the boondocks with many areas of no reception, get a bracelet and a sticker. My father has his bracelet and carries a folding card in his wallet (on the off chance a paramedic will see it) detailing his problems.
3) Disaster scenario: What happens to mobile phone reception?
Same as above. No new risk, possibly some good. If the lookup doesn't work, paramedics are no worse off than they are right now. This may be a shocking surprise, but paramedics have brains. There are very few conditions that will lead to death if the paramedic doesn't know about them. My father's iodine reaction was discovered thirty minutes into open-heart surgery. They gave him a few more drugs to suppress the reaction, finished the surgery, then noted that they'd need a different preparation for the next round of surgery.
4) Paramedics have time to surfe the web while trying to save lives?
Yes, they sometimes do. My most recent encounter with paramedics was after my wife was hit by a car. Three paramedics arrived. One used a flashlight to direct traffic (since it was late evening), one treated my wife, and one held the ambulance door open. I suspect the doorstop could have taken a few moments to scan a code and checked for allergies.
5) LifeSquare's web site is down. Whoops, guess I'll die of a reaction to penicillin then.
No, you'll have a reaction to penicillin, which the paramedics will notice, and take care of. It's more work for them, and makes the hospital trip more complicated, but it's how things work now.
Sure, there are privacy issues, but that seems to be the least of the faults with the system. Just write the dam thing in English on the bracelet and all you need to do is be able to read English. Low-tech solution is the right one.
What if your bracelet falls off, or gets dirty, or is tarnished, or if it's upside down and the paramedics can't move your arm to read it? You could die of a reaction to penicillin because the bracelet didn't work!
Like any important system, redundancy across different methods is the most reliable.
That doesn't work. Unless people are embarrassed they won't lift a finger, they'll just ignore it. Besides, the atrocities at Abu Ghraib were committed with the full knowledge and consent of the US Government.
Embarrassment works, eh? So now we have no more war crimes in Iraq, politicians are clean, and the world is a happy place now that WikiLeaks has embarrassed everyone into behaving, right?
Wrong. Utterly wrong. Using the fear of embarrassment as a motivator is a quick and easy way for any random person with information to feel powerful, but it's no more effective than, say, appealing to someone's judgement. Very often, all that's needed to correct an immoral action is to convince someone to take a fresh look at what they're doing. That can best be done with care, not fear.
It's funny that you should use Abu Ghraib as an example, since an internal investigation (started through those official channels) reported abuse to the media in 2004, two years before WikiLeaks was launched. In fact, practically all of the events (including trials) involving Abu Ghraib were concluded by 2007, well before the mass release in 2010. Boy, that sure is effective embarrassment, being three years too late and accomplishing nothing.
It happens in other fields, too. Try secretly reporting a security bug to Microsoft and see if they fix it. Compare to how long they take to fix published bugs.
That's what responsible disclosure is for: Go through the official channels, then at a later point inform the general public. In the worst case, the malware developers get a bit more time to run their exploits, and a few days after release the malware will be analyzed and the bug becomes public knowledge anyway.
As for telling journalists: It wasn't Julian Assange who released the cables, everything he released was redacted to avoid the problems you describe. People only got full access to the original documents when a journalist published the decryption key to Julian's "insurance file".
Note also that the "insurance file" itself was released to the public by a defector, Daniel Domscheit-Berg, who obtained it while working on an internal server. The file had previously been given to the journalist Leigh, with instructions to make it public. This raises three important questions:
1) Why put unredacted information in a file with potential to be released, anyway, if not to cause harm?
2) Why leave the file on any server that anyone but Assange could access after it's been copied elsewhere?
Assange's "insurance" was an outright extortion scheme. Either the government gave in to his demands of amnesty, or he'd risk people's lives. We can see how well that fear tactic is working out for him...