Well, then what about the cost? The component difference to embed what is basically a laptop PC is a lot more cost than using the chips found in smartphones. _Especially_ 7 years ago.
I don't think you'd be so blithely confident if it were you facing exploding drone attacks. Would you feel better about being shot at with live ammo if you were wearing a level IV vest + helmet and thus have a good chance of not dying? It only takes one bullet to find a hole in your defense and you die. Defense inherently is also harder to do the offense because the attacker can choose the time and place to attack. Pick a spot where the nets are decayed or not installed. Use an overwhelming number of drones. That kind of thing.
You know those shit retail jobs that pay like crap but require minimal qualifications? Working for Walmart and other similar stores? Well, an Amazon worker, since robots help bring him the stuff and a computer micromanages his every action, is probably 5 times as effective at moving product as a retail stocker is. So +20k seasonal jobs = -100k shitty retail jobs elsewhere.
Ok, thinking over my post, that isn't quite true. If consumer incomes were still growing - if people were consuming more - then if consumption of various shit from Amazon were 5x what it was before Amazon, then the number of jobs would be the same.
And how would you deal with the power consumption? 7 years ago, even low power PCs sucked down 40+ watts for the CPU+GPU. SSDs were just starting to even become a thing. You need a GPU to have smoothly scrolling navigation maps and basic animations, especially since you'd be forced to use a very low power, low performance processor.
The tech to do this at scale is just a few years old. The key piece that was missing is you need cheap sensors and cheap processors able to accurately recognize people. Probably an IR sensor, a camera, and machine vision on FPGAs/custom chips running neural nets.
That's 2015 technology. That is, before 2015, it existed but sucked ass and was too expensive.
And so maybe in 2025 the military procurement process will have the first acceptable prototypes. But yeah it's a great idea, I'm imagining some kind of containerized weapon. You essentially fire it by connecting a control device to a port on the side. You specify the GPS coordinates of the target area. You expect it to just kill everyone in the target area, so you need a blue force tracking system to make sure no friendly troops are there.
Realistically I can't think of a reliable means to tell the difference between friendly troops, enemy troops, and civilians - well ok I think you could maybe train the recognition to PREFER enemy troops but it would still sometimes screw up so the target area would have to be clear of friendly troops.
Anyways, the containerized weapon makes a terrible racket as all the drones insides start up (would sound like bees) and then they all escape through an opening in the top.
I dunno man. Helicopters/attack drones have immense aircraft performance, driven by gas turbine engines. Battery powered drones don't have anywhere close to the power/weight and they also are always about to run out of battery. Maybe a hovering helicopter could be hit but attack drones orbit at 10,000+ feet and you're not going to hit a moving helicopter.
Also, 2 lbs worth of warhead is marginal for destroying an attack helicopter. You could do it maybe but only if the detonation were in exactly the right spot and it was a shaped charge.
If they can get access to a drone, they can either :
1. Order the right chemicals for high grade explosives and synthesize them. You know that ANFO is high explosive, right? Any idiot with the 2 ingredients can make some. I _think_ the fertilizer has gotten hard to come by but I can't say for sure.
2. Extract high grade plastic explosive from other munitions in a war zone. Artillery shells, mines, RPG warheads - whatever they can get their hands on.
1 lb is a hand grenade. And it could probably be molded into a claymore style shaped charge aimed downwards so that it makes more efficient use of the blast.
Now, now. Don't exaggerate. That (probably) won't blow a Samsung phone. You'd have to share a link to a power hungry game so the victim's phone draws enough current to set off the battery.
The high end models have autopilots, cameras, and laser/ultrasonic sensors to avoid collisions. So it's within the state of the art to build a custom drone that uses off the shelf face recognition to pursue a specific person.
Well, ish. Yeah I know, in reality the person would turn away from the drone and it would lose lock, from a distance the camera wouldn't have enough resolution, reflections off window glass or rain would mess it up, etc etc etc. I'm trying to say that a movie like seeking quadcopter with a bomb isn't far off. And today you could make one that homes on GPS coordinates, giving it an accuracy that would sometimes be pinpoint, able to hit a specific window in a building or a specific spot on the ground.
And sometimes it would be terrible, since GPS is spotty and depends on a number of factors.
The attackers could send several drones from different angles, and program in a series of flight waypoints that take them between structures in the target area as cover so they can't be engaged until it's too late.
The attackers could hide the drones on rooftops or other unattended places in an urban area before the attack. Maybe inside a disguised container with a solar panel on it or something. (the solar panel would keep the drone's battery and a cell phone used for communication charged)
Sort of darkly ironic is that while high end consumer drones can basically do this (you'd need a high end one with LIDAR so it doesn't collide with walls on the way to the target - not sure if any models you can buy have LIDAR, but it's totally doable if you use a solid state sensor), this kind of attack pattern is what a military using cruise missiles would do against a hardened target.
Maybe there will have to be automated anti-drone defenses. Some kind of net round that won't be lethal if it misses the drone and hits a person. Or maybe important figures will have to spend all their time in hardened buildings and armored vehicles for fear of getting droned.
Heh. I picked up a moto G4 for $130 with sales tax from that deal that Amazon is offering. It came with ads/bloatware but I was able to remove them with a readily available tool.
It technically does have a 5.5" 1920x1080 screen, a large battery, 2 gigs of RAM, and a quad core processor. It's also technically slow as a snail compared to the phones you listed, equivalent to a good phone probably at least 2 years old....But it works fine. You can browse the web. Use maps. Send texts. Answer phone calls. None of these basic tasks, save loading the more bloated web pages, are particular slow...
So from my perspective, spending another $270, or triple what I spent, for a phone that's only distinguishing advantage is snappier app opening, the ability to play demanding mobile games I don't play, and better camera performance at night is questionable...
They should be excluded from the final ballot because sometimes third party candidates actually suck enough votes away that democracy fails to even pick the candidate the majority of voters want, if they were not distracted by irrelevant choices. The method to do this is commonly known as a "runoff" election or ranked voting or a dozen other possible fixes.
It's still zero. For 240 years, a third party has never won and never will win for the next million years as long as the rules for elections remain the same.
What sometimes happens is a party will die, and a party that WAS a third party replaces them. That might eventually happen to the Republicans, that does seem to be where the trend is heading.
Prosecuting attorneys can agree to binding agreements in return for cooperation. That's what an immunity deal is - the state is promising to not prosecute, even if they have the evidence they need, in return for something. The deal can include terms like "you agree to destroy certain evidence and any copies used you agree to not use in any future investigation".
Rather disturbingly, the defense can agree to similar terms such as letting the prosecution destroy evidence that may be exculpatory in return for a reduced prison sentence.
Oh. Sure. I mean it's not totally out there. Obviously you can do it, easy, with a few bits and bobs to make a radiation lens and a wee little nuclear bomb to get the party started. Works great. So doing the same thing, just on a smaller scale with a tiny puff of fusion fuel gas, has always seemed to be right around the corner. If a bomb that weighs under 100 kg can do it on a large scale, why can't you make it happen on a small scale with 100 tons or so of apparatus, including lots of superconducting magnets and capacitors and lasers and whatnot?
Certainly there's been many attempts and there ARE tiny fusors that work, they just have pathetically low fusion rates.
How does end to end cryptographic security relate to this in any way?
For "Something you know" security - end to end encryption does nothing to stop a hidden camera or hidden electronic device that can detect the actual buttons pressed on the keypad.
For "something you are" security - a tampered with biometric sensor can have it's readings copied, for example you can tamper with a fingerprint scanner so the sensor copies any prints it detects to another embedded device that stores a copy of the print images. This is a little harder because then the thieves have to make a replica that will fool a scanner (some of the better ones have capacitance and heat sensors so it's harder to fool with a fake rubber fingerprint) but still doable.
For 'something you have' - if what you "have" is a fixed magnetic strip, that's easy to steal and copy. If it's a chip with a secret internal private key or OTP, forget it. The thief is going to have to steal the physical card.
In NONE of these cases does end to end encryption do shit for you.
Are you trolling? Of course fusion would be nice, but there's very little reason to think that lightweight, controlled fusion is imminent. It might happen but I wouldn't bet on it.
That's not at all what the articles say on it. It says it was essentially a flight ready technology. A few chips can either be solved by better QC or just determining the probability of failure due to the chips. 0% chance of failure is never possible, any vehicle humans has ever invented has a nonzero chance of failing spontaneously and killing you in use.
Well, first of all, those returns are generally done via person, which means you'd use the chip feature of the card.
Second, obviously the credit card issuer would authorize canceling a transaction or a credit but not a debit without the CV2 number. You didn't read the article, in actuality the credit card still has a number, it's the authentication code on the back that changes.
It can be made impossible if the seed is complex enough. What you're saying is, "I know what a number in a sequence is, every hour. I know the algorithm the sequence uses, I just don't know the seed. How many numbers in series do I need to see to calculate the seed".
If they did it right, you need more than 26,280 numbers to determine the seed. But honestly, if you think about it, if a thief steals your card and starts watching the numbers flick by. As long as it takes more than a few numbers to determine the seed, that's plenty. After all, the whole reason to figure out the seed by crook is so you can then return a card to someone and clone their card and they won't report it stolen. If this takes more than a few hours, best case, the person is going to notice and report it.
You know, when I see this argument, there's a critical flaw in it. Yes, 2 factor is generally better than 1- factor, and so forth.
However, all factors are not equal. "Something you know" is the worst case because humans have limited memory so there will always be easy ways to steal anything a user knows. This is why passwords are so shitty. "Something you are"(fingerprint, iris scanners) unfortunately devolves back to "something you know" in that it's a fixed biometric signature that you can spoof.
Something you have, if it's a chip or other method that uses a secret OTP or private key, is by far the strongest. There's no fixed code you can steal of any sort. Different every time. Yes, someone can steal the physical card - but that still reduces fraud enormously because if a card is a missing, the victim can report it and cause it to be invalidated. It's imperfect - you are correct that chip and pin is better than chip and signature - but it's still a huge reduction in fraud. It also makes it much easier to catch the thief - if someone you know the name of steals your credit card, you can give their name to the police. The police can obtain security tapes, and if the person on camera resembles the person who's name you gave, boom, they got them.
That's a crime that is vastly more likely to be solved than "someone I never saw or met stole my card # somehow and used it in India"
Slashdot Mirror
Well, then what about the cost? The component difference to embed what is basically a laptop PC is a lot more cost than using the chips found in smartphones. _Especially_ 7 years ago.
I don't think you'd be so blithely confident if it were you facing exploding drone attacks. Would you feel better about being shot at with live ammo if you were wearing a level IV vest + helmet and thus have a good chance of not dying? It only takes one bullet to find a hole in your defense and you die. Defense inherently is also harder to do the offense because the attacker can choose the time and place to attack. Pick a spot where the nets are decayed or not installed. Use an overwhelming number of drones. That kind of thing.
You know those shit retail jobs that pay like crap but require minimal qualifications? Working for Walmart and other similar stores? Well, an Amazon worker, since robots help bring him the stuff and a computer micromanages his every action, is probably 5 times as effective at moving product as a retail stocker is. So +20k seasonal jobs = -100k shitty retail jobs elsewhere.
Ok, thinking over my post, that isn't quite true. If consumer incomes were still growing - if people were consuming more - then if consumption of various shit from Amazon were 5x what it was before Amazon, then the number of jobs would be the same.
But they aren't.
And how would you deal with the power consumption? 7 years ago, even low power PCs sucked down 40+ watts for the CPU+GPU. SSDs were just starting to even become a thing. You need a GPU to have smoothly scrolling navigation maps and basic animations, especially since you'd be forced to use a very low power, low performance processor.
Can I ask what you had in mind for "standard" PC hardware? What kind of processor?
The tech to do this at scale is just a few years old. The key piece that was missing is you need cheap sensors and cheap processors able to accurately recognize people. Probably an IR sensor, a camera, and machine vision on FPGAs/custom chips running neural nets.
That's 2015 technology. That is, before 2015, it existed but sucked ass and was too expensive.
And so maybe in 2025 the military procurement process will have the first acceptable prototypes. But yeah it's a great idea, I'm imagining some kind of containerized weapon. You essentially fire it by connecting a control device to a port on the side. You specify the GPS coordinates of the target area. You expect it to just kill everyone in the target area, so you need a blue force tracking system to make sure no friendly troops are there.
Realistically I can't think of a reliable means to tell the difference between friendly troops, enemy troops, and civilians - well ok I think you could maybe train the recognition to PREFER enemy troops but it would still sometimes screw up so the target area would have to be clear of friendly troops.
Anyways, the containerized weapon makes a terrible racket as all the drones insides start up (would sound like bees) and then they all escape through an opening in the top.
I dunno man. Helicopters/attack drones have immense aircraft performance, driven by gas turbine engines. Battery powered drones don't have anywhere close to the power/weight and they also are always about to run out of battery. Maybe a hovering helicopter could be hit but attack drones orbit at 10,000+ feet and you're not going to hit a moving helicopter.
Also, 2 lbs worth of warhead is marginal for destroying an attack helicopter. You could do it maybe but only if the detonation were in exactly the right spot and it was a shaped charge.
If they can get access to a drone, they can either :
1. Order the right chemicals for high grade explosives and synthesize them. You know that ANFO is high explosive, right? Any idiot with the 2 ingredients can make some. I _think_ the fertilizer has gotten hard to come by but I can't say for sure.
2. Extract high grade plastic explosive from other munitions in a war zone. Artillery shells, mines, RPG warheads - whatever they can get their hands on.
1 lb is a hand grenade. And it could probably be molded into a claymore style shaped charge aimed downwards so that it makes more efficient use of the blast.
Now, now. Don't exaggerate. That (probably) won't blow a Samsung phone. You'd have to share a link to a power hungry game so the victim's phone draws enough current to set off the battery.
The high end models have autopilots, cameras, and laser/ultrasonic sensors to avoid collisions. So it's within the state of the art to build a custom drone that uses off the shelf face recognition to pursue a specific person.
Well, ish. Yeah I know, in reality the person would turn away from the drone and it would lose lock, from a distance the camera wouldn't have enough resolution, reflections off window glass or rain would mess it up, etc etc etc. I'm trying to say that a movie like seeking quadcopter with a bomb isn't far off. And today you could make one that homes on GPS coordinates, giving it an accuracy that would sometimes be pinpoint, able to hit a specific window in a building or a specific spot on the ground.
And sometimes it would be terrible, since GPS is spotty and depends on a number of factors.
The attackers could send several drones from different angles, and program in a series of flight waypoints that take them between structures in the target area as cover so they can't be engaged until it's too late.
The attackers could hide the drones on rooftops or other unattended places in an urban area before the attack. Maybe inside a disguised container with a solar panel on it or something. (the solar panel would keep the drone's battery and a cell phone used for communication charged)
Sort of darkly ironic is that while high end consumer drones can basically do this (you'd need a high end one with LIDAR so it doesn't collide with walls on the way to the target - not sure if any models you can buy have LIDAR, but it's totally doable if you use a solid state sensor), this kind of attack pattern is what a military using cruise missiles would do against a hardened target.
Maybe there will have to be automated anti-drone defenses. Some kind of net round that won't be lethal if it misses the drone and hits a person. Or maybe important figures will have to spend all their time in hardened buildings and armored vehicles for fear of getting droned.
There still can only be two. And it takes a shift over generations to replace one political party with another.
Heh. I picked up a moto G4 for $130 with sales tax from that deal that Amazon is offering. It came with ads/bloatware but I was able to remove them with a readily available tool.
It technically does have a 5.5" 1920x1080 screen, a large battery, 2 gigs of RAM, and a quad core processor. It's also technically slow as a snail compared to the phones you listed, equivalent to a good phone probably at least 2 years old. ...But it works fine. You can browse the web. Use maps. Send texts. Answer phone calls. None of these basic tasks, save loading the more bloated web pages, are particular slow...
So from my perspective, spending another $270, or triple what I spent, for a phone that's only distinguishing advantage is snappier app opening, the ability to play demanding mobile games I don't play, and better camera performance at night is questionable...
They should be excluded from the final ballot because sometimes third party candidates actually suck enough votes away that democracy fails to even pick the candidate the majority of voters want, if they were not distracted by irrelevant choices. The method to do this is commonly known as a "runoff" election or ranked voting or a dozen other possible fixes.
It's still zero. For 240 years, a third party has never won and never will win for the next million years as long as the rules for elections remain the same.
What sometimes happens is a party will die, and a party that WAS a third party replaces them. That might eventually happen to the Republicans, that does seem to be where the trend is heading.
Prosecuting attorneys can agree to binding agreements in return for cooperation. That's what an immunity deal is - the state is promising to not prosecute, even if they have the evidence they need, in return for something. The deal can include terms like "you agree to destroy certain evidence and any copies used you agree to not use in any future investigation".
Rather disturbingly, the defense can agree to similar terms such as letting the prosecution destroy evidence that may be exculpatory in return for a reduced prison sentence.
There's still precisely a 0.0 repeating chance that anyone but Trump or Hillary becomes president. Do you understand this mathematical fact?
It doesn't change. Only the code at the back. RTFA.
Oh. Sure. I mean it's not totally out there. Obviously you can do it, easy, with a few bits and bobs to make a radiation lens and a wee little nuclear bomb to get the party started. Works great. So doing the same thing, just on a smaller scale with a tiny puff of fusion fuel gas, has always seemed to be right around the corner. If a bomb that weighs under 100 kg can do it on a large scale, why can't you make it happen on a small scale with 100 tons or so of apparatus, including lots of superconducting magnets and capacitors and lasers and whatnot?
Certainly there's been many attempts and there ARE tiny fusors that work, they just have pathetically low fusion rates.
How does end to end cryptographic security relate to this in any way?
For "Something you know" security - end to end encryption does nothing to stop a hidden camera or hidden electronic device that can detect the actual buttons pressed on the keypad.
For "something you are" security - a tampered with biometric sensor can have it's readings copied, for example you can tamper with a fingerprint scanner so the sensor copies any prints it detects to another embedded device that stores a copy of the print images. This is a little harder because then the thieves have to make a replica that will fool a scanner (some of the better ones have capacitance and heat sensors so it's harder to fool with a fake rubber fingerprint) but still doable.
For 'something you have' - if what you "have" is a fixed magnetic strip, that's easy to steal and copy. If it's a chip with a secret internal private key or OTP, forget it. The thief is going to have to steal the physical card.
In NONE of these cases does end to end encryption do shit for you.
Are you trolling? Of course fusion would be nice, but there's very little reason to think that lightweight, controlled fusion is imminent. It might happen but I wouldn't bet on it.
That's not at all what the articles say on it. It says it was essentially a flight ready technology. A few chips can either be solved by better QC or just determining the probability of failure due to the chips. 0% chance of failure is never possible, any vehicle humans has ever invented has a nonzero chance of failing spontaneously and killing you in use.
Well, first of all, those returns are generally done via person, which means you'd use the chip feature of the card.
Second, obviously the credit card issuer would authorize canceling a transaction or a credit but not a debit without the CV2 number. You didn't read the article, in actuality the credit card still has a number, it's the authentication code on the back that changes.
It can be made impossible if the seed is complex enough. What you're saying is, "I know what a number in a sequence is, every hour. I know the algorithm the sequence uses, I just don't know the seed. How many numbers in series do I need to see to calculate the seed".
If they did it right, you need more than 26,280 numbers to determine the seed. But honestly, if you think about it, if a thief steals your card and starts watching the numbers flick by. As long as it takes more than a few numbers to determine the seed, that's plenty. After all, the whole reason to figure out the seed by crook is so you can then return a card to someone and clone their card and they won't report it stolen. If this takes more than a few hours, best case, the person is going to notice and report it.
You know, when I see this argument, there's a critical flaw in it. Yes, 2 factor is generally better than 1- factor, and so forth.
However, all factors are not equal. "Something you know" is the worst case because humans have limited memory so there will always be easy ways to steal anything a user knows. This is why passwords are so shitty. "Something you are"(fingerprint, iris scanners) unfortunately devolves back to "something you know" in that it's a fixed biometric signature that you can spoof.
Something you have, if it's a chip or other method that uses a secret OTP or private key, is by far the strongest. There's no fixed code you can steal of any sort. Different every time. Yes, someone can steal the physical card - but that still reduces fraud enormously because if a card is a missing, the victim can report it and cause it to be invalidated. It's imperfect - you are correct that chip and pin is better than chip and signature - but it's still a huge reduction in fraud. It also makes it much easier to catch the thief - if someone you know the name of steals your credit card, you can give their name to the police. The police can obtain security tapes, and if the person on camera resembles the person who's name you gave, boom, they got them.
That's a crime that is vastly more likely to be solved than "someone I never saw or met stole my card # somehow and used it in India"