"It doesn't need to "reliably" block unauthorized shooters. If it gives even a 50% chance that a kid who steals dad's gun won't be able to shoot up their school, or that a young child won't be able to accidentally shoot themselves, then surely that's worth something."
Yes, of course. But you also have to weigh that against whether it DOES work, when it's supposed to. If you're a police officer or military, for example, and your buddies get killed because your gun wouldn't go off, they're just as dead as that young child you mention.
As I wrote above: it would take an awful lot of convincing to get me to believe this thing has the right mix of reliability. I'm not about to say it's impossible, but so far I haven't seen technology that even comes close.
"Gun manufacturers are not responsible for the safety of their devices. They convinced / paid off congress to pass laws that allows them to be immune to even gun safety lawsuits. Take your red herring anti-tort arguments elsewhere."
It's not a red-herring argument. You just thought he meant something else. I think.
"The first lawsuit" would be by somebody who tried to use it for self-defense, and it didn't work.
Gun manufacturers may not be responsible for misuse of their product, but they sure as hell ARE responsible for products that don't actually perform their primary function, i.e., shoot when you want them to shoot.
All manufacturer's disclaimers aside, most states today have "implied warranty" laws, and there have been some successful suits over that in the past.
"Never a failure to feed, never a failure to fire, ever. Damm reliable gun."
I don't dispute that. But there are trade-offs. The Stoner design (at least today) though have an accuracy goal that the AK just doesn't match. You noted that yourself.
On the other hand, you can at least to some degree get the best of both worlds. Some of the newer piston-operated variations are way less prone to fouling and jamming. Not all of them, but some. Since there are so many companies making them now, there has been a lot of experimentation and a lot of improvements.
Excuse me? Please explain how I "moved the goalposts".
The problem -- as I already explained -- is that if you make these accurate enough to reliably allow the authorized (true positive), then you have to keep the parameters so loose as to also allow some of the unauthorized. At least with today's technology.
How is that "moving the goalposts"? The stated PURPOSE of these things is to prevent unauthorized access, while allowing reliable authorized access. If they can't do that (and I argue that they can't, at least reliably) then the whole thing is a waste of time.
It's fairly simple math. If they have a technology that is so far advanced that it can actually do this, then I would be very surprised and maybe actually pleased. But I am far from convinced, because it would have to involve a technology I've never even heard of. Not impossible but not very likely.
"Guns kept in the home for "self protection" are alarmingly likely to be used against their owners, either by burglars who find them first, children by accident, or the owner himself for suicide (not that this tech would prevent that)."
I was going to argue with you -- your first paragraph about false positives and negatives just doesn't apply in this case; the statistical problems are pretty well known and those ain't them.
But this second paragraph made me stop. I'm not going to argue with you because you're just arguing nonsense. The statistics do NOT show anything of this sort. I know because I have studied this very subject for years.
But I'm not going to argue or try to persuade, or go out of my way to present evidence. I just don't have time for this kind of nonsense.
"Wiretap warrants require a lot more than just reasonable suspicion of a crime, though. "
Absolutely. They require probable cause, which means real evidence. Of course, then there are the secret rooms the government built into some telco offices that simply siphon off data without anybody's knowledge or consent. Those are established fact... they are the whole reason Congress had to give telcos "immunity" for passing on the information. But as far as I know, there still isn't a law that allows the government to do it legally or constitutionally.
"Wiretap laws were written to fit the idea that phone companies were simple carriers who would respect the integrity of customer's conversations, and since they didn't provide services themselves, people had a reasonable expectation of privacy."
It's not that they didn't provide services. They didn't provide content. As the courts have ruled: there is a lesser standard of evidence needed for telephone records (who called who, and when, for example) than there is for the content of the telephone conversation (wiretap).
But this brings up a good point. Telcos were (FCC Regulations) classified as Title II "Common Carriers". I.e., they provide the call service, but are strictly forbidden from intercepting or interfering with the content (conversation) without a warrant.
It is quite possible to classify and regulate Cable companies and other ISPs as Title II Common Carriers. In fact, the FCC has wanted to do it for decades. But lobbyists got Congress to pass a law specifically excluding ISPs from Common Carrier status. That was one of the biggest mistakes of the last few decades.
The solution: get Congress to remove the exclusion from ISPs. Then the vast majority of your privacy concerns go away, virtually overnight: it will then be prohibited for ISPs (or anybody, including usage trackers) from monitoring your activities without a warrant. Most of the major privacy and security concerns surrounding the Internet simply disappear.
Sure, there will still be a few criminals doing it now and then. But criminals tapped (probably still tap) telephones, too. But the big problem -- government and corporations -- will be forced to leave it alone.
"The police walk into your telephone switch room with a warrant, you let them listen. That's much much older than CALEA, that's only 20 years old."
That's pretty irrelevant, though, because with telephones, tapping is pretty darned easy. But with other technologies it has NEVER been possible to "just listen in"... it just wasn't built in.
That's not "refusal", it's simply not building something in a way that expressly caters to the police. And I don't give a damn. The police don't have a right to run the tech world.
"How about 'I know how to write quality code, but I'm no longer interested in spending the necessary cycles to learn every new faddish tech. that comes down the pipe'?"
Seconded.
I get so tired of hearing, "Woman, you need to learn NodeJS if you want to get ahead!" "No, you need to learn MongoDB!" "No, you need 'responsive' web design!" Etc. Etc.
I did not learn NodeJS, and never once have I felt a lack because of it. MongoDB is not what it's cracked up to be. Actual responsive design has some pretty decent ideas, but it's hardly the only way to achieve them. And so on.
Functional programming is great for some applications. For others, it's not the right tool for the job. And so on.
I spend a lot of time looking at the "latest and greatest". And I decide not to bother with a lot of them. I know other people who almost always jump on the bandwagon... and spend months or even more struggling.
I'm pretty happy with the tools I'm currently using. I can change them and have when I felt a genuine need.
When I say "near 100% reliable", I'm not joking. 99.9% just isn't good enough for something I'd trust my life to. But if it approached 99.99%, then it's getting near the reliability of the gun itself, and may be good enough. That's approximately 1 error in 1000 rounds. Even that is pushing what I view as acceptable limits.
And even just given that it's battery-powered, it probably will never reach that goal in the foreseeable future.
As for its intended purpose (blocking unauthorized users), I have no doubt that it would work some of the time. But how often, given that it has to be that accurate for the authorized? I'm not confident that it would be that good at its job. It's a very difficult balancing act, and I would need a lot of convincing.
"When I pull the trigger, I want the gun to fire. I doubt this will be reliable enough to depend upon."
More to the point: if you want it to be reliable, then the fingerprint technology has to be loose enough to be UNreliable. We already know this. With today's technology, if you want to allow access with fingerprints reliably, you have to make your parameters loose enough that false positives slip in too easily.
Which means that in order to be near 100% reliable for an "authorized" shooter, this thing provably can't do what it's intended to do: reliably block the UNauthorized.
"How does using a strong password prevent password re-use?"
The key issue here is not so much strong passwords, but "auditing".
If this were the beginning of the month, I would suspect an April Fools joke. They want to improve your use of passwords by monitoring your password usage!!!
Gee, what could possibly be wrong with this idea? [sarcasm]
"All completely irrelevant, because foreigners are less likely to come from university recognized as "top-ranked" by Americans, or have any ties to US university to get a Ph.D there while working, or be involved in getting a US patent even if invention is mostlly his work."
This is a straw-man argument. Not only are those things not irrelevant, they are among the top reasons corporations claimed they needed H1-B workers: the claim has been that those workers are among the top in those metrics.
"There is a significant band in the middle of people who will pirate if it's easy and buy if it's not."
Multiple studies have shown that the most prolific "illegal" downloaders are the same people who spend the most on media: music, movies in theaters, and DVDs. Yet another study just a month or so ago (discussed here on/.) had the same conclusion.
"My focus here is the aspect of how DRM protects the rights of content creators (aka, artists) and helps to prevent people freely distributing their works and with no compensation."
This is an assumption that is not borne out by the actual data.
Study after study of various aspects of DRM, in regard to software and published works anyway, belie this assumption.
People who "illegally" download movies and music also happen to be the people who spend the most on music and movies (both in-theater and DVDs).
The fact is that products that are solidly locked up under DRM tend not to sell very well. Look at the latest rebellion against Electronic Arts and Ubisoft over DRM. EA has been laying off employees.
This is not to say it might not be useful under some circumstances. But by and large, it has tended to make products less attractive to consumers.
"When you introduce commercial aspects to OS, it becomes a completely different beast"
That was my point. If they want regular, professionally-provided support, they won't get it for most Free-and-Open-Source products. They should probably opt for commercial instead.
"Don't think that a difference in "averages" says anything about people in the top percentiles."
The whole point was that relatively few of the H1-B workers were IN the top percentiles.
Companies have been claiming that we need more H1-B workers because the U.S. has not been creating enough of them. These studies show that excuse to be nothing but bullshit.
For advanced students of literature or writing, Jack Vance and Barry Longyear should be requirements.
Though some people have found Vance hard to read, his English prose is impeccable.
Longyear never uses the "he said," "she retorted," "he quipped" kind of lazy and awkward sentence construction that has come to be almost universal today. Studying how he gets around it while making it seem natural is very educational. (He did publish one short story in which he did that, but it was intended as satire of that very thing.)
"Why do you doubt that? If you have a camera at the scene of a crime and refuse to turn over the footage, a judge would happily grant a subpoena for that."
I should have qualified my statement.
If it's AT a crime scene (cameras right there), then I can certainly see a warrant being issued. But as for cameras that just happen to be in the same general area of the city, it would (and should) be a lot harder to obtain a warrant because a warrant has to be based on probable cause.
"Because it was within 300 meters and might have accidentally caught the perpetrators on video" is not probable cause.
Slashdot Mirror
Wow. I hadn't thought about that angle, but it's a very good point.
"It doesn't need to "reliably" block unauthorized shooters. If it gives even a 50% chance that a kid who steals dad's gun won't be able to shoot up their school, or that a young child won't be able to accidentally shoot themselves, then surely that's worth something."
Yes, of course. But you also have to weigh that against whether it DOES work, when it's supposed to. If you're a police officer or military, for example, and your buddies get killed because your gun wouldn't go off, they're just as dead as that young child you mention.
As I wrote above: it would take an awful lot of convincing to get me to believe this thing has the right mix of reliability. I'm not about to say it's impossible, but so far I haven't seen technology that even comes close.
"Gun manufacturers are not responsible for the safety of their devices. They convinced / paid off congress to pass laws that allows them to be immune to even gun safety lawsuits. Take your red herring anti-tort arguments elsewhere."
It's not a red-herring argument. You just thought he meant something else. I think.
"The first lawsuit" would be by somebody who tried to use it for self-defense, and it didn't work.
Gun manufacturers may not be responsible for misuse of their product, but they sure as hell ARE responsible for products that don't actually perform their primary function, i.e., shoot when you want them to shoot.
All manufacturer's disclaimers aside, most states today have "implied warranty" laws, and there have been some successful suits over that in the past.
"Never a failure to feed, never a failure to fire, ever. Damm reliable gun."
I don't dispute that. But there are trade-offs. The Stoner design (at least today) though have an accuracy goal that the AK just doesn't match. You noted that yourself.
On the other hand, you can at least to some degree get the best of both worlds. Some of the newer piston-operated variations are way less prone to fouling and jamming. Not all of them, but some. Since there are so many companies making them now, there has been a lot of experimentation and a lot of improvements.
"It's not 10,000 rounds. It's 10,000 grips. If it recognizes the authorized user, then it would not re-run the check until the user changes grip."
That's actually a good point. I stated it wrong. Let's say 10,000 TIMES. I'm good with that.
Excuse me? Please explain how I "moved the goalposts".
The problem -- as I already explained -- is that if you make these accurate enough to reliably allow the authorized (true positive), then you have to keep the parameters so loose as to also allow some of the unauthorized. At least with today's technology.
How is that "moving the goalposts"? The stated PURPOSE of these things is to prevent unauthorized access, while allowing reliable authorized access. If they can't do that (and I argue that they can't, at least reliably) then the whole thing is a waste of time.
It's fairly simple math. If they have a technology that is so far advanced that it can actually do this, then I would be very surprised and maybe actually pleased. But I am far from convinced, because it would have to involve a technology I've never even heard of. Not impossible but not very likely.
"Guns kept in the home for "self protection" are alarmingly likely to be used against their owners, either by burglars who find them first, children by accident, or the owner himself for suicide (not that this tech would prevent that)."
I was going to argue with you -- your first paragraph about false positives and negatives just doesn't apply in this case; the statistical problems are pretty well known and those ain't them.
But this second paragraph made me stop. I'm not going to argue with you because you're just arguing nonsense. The statistics do NOT show anything of this sort. I know because I have studied this very subject for years.
But I'm not going to argue or try to persuade, or go out of my way to present evidence. I just don't have time for this kind of nonsense.
"Wiretap warrants require a lot more than just reasonable suspicion of a crime, though. "
Absolutely. They require probable cause, which means real evidence. Of course, then there are the secret rooms the government built into some telco offices that simply siphon off data without anybody's knowledge or consent. Those are established fact... they are the whole reason Congress had to give telcos "immunity" for passing on the information. But as far as I know, there still isn't a law that allows the government to do it legally or constitutionally.
"Wiretap laws were written to fit the idea that phone companies were simple carriers who would respect the integrity of customer's conversations, and since they didn't provide services themselves, people had a reasonable expectation of privacy."
It's not that they didn't provide services. They didn't provide content. As the courts have ruled: there is a lesser standard of evidence needed for telephone records (who called who, and when, for example) than there is for the content of the telephone conversation (wiretap).
But this brings up a good point. Telcos were (FCC Regulations) classified as Title II "Common Carriers". I.e., they provide the call service, but are strictly forbidden from intercepting or interfering with the content (conversation) without a warrant.
It is quite possible to classify and regulate Cable companies and other ISPs as Title II Common Carriers. In fact, the FCC has wanted to do it for decades. But lobbyists got Congress to pass a law specifically excluding ISPs from Common Carrier status. That was one of the biggest mistakes of the last few decades.
The solution: get Congress to remove the exclusion from ISPs. Then the vast majority of your privacy concerns go away, virtually overnight: it will then be prohibited for ISPs (or anybody, including usage trackers) from monitoring your activities without a warrant. Most of the major privacy and security concerns surrounding the Internet simply disappear.
Sure, there will still be a few criminals doing it now and then. But criminals tapped (probably still tap) telephones, too. But the big problem -- government and corporations -- will be forced to leave it alone.
"The police walk into your telephone switch room with a warrant, you let them listen. That's much much older than CALEA, that's only 20 years old."
That's pretty irrelevant, though, because with telephones, tapping is pretty darned easy. But with other technologies it has NEVER been possible to "just listen in"... it just wasn't built in.
That's not "refusal", it's simply not building something in a way that expressly caters to the police. And I don't give a damn. The police don't have a right to run the tech world.
If they can't keep up, tough shit.
"How about 'I know how to write quality code, but I'm no longer interested in spending the necessary cycles to learn every new faddish tech. that comes down the pipe'?"
Seconded.
I get so tired of hearing, "Woman, you need to learn NodeJS if you want to get ahead!" "No, you need to learn MongoDB!" "No, you need 'responsive' web design!" Etc. Etc.
I did not learn NodeJS, and never once have I felt a lack because of it. MongoDB is not what it's cracked up to be. Actual responsive design has some pretty decent ideas, but it's hardly the only way to achieve them. And so on.
Functional programming is great for some applications. For others, it's not the right tool for the job. And so on.
I spend a lot of time looking at the "latest and greatest". And I decide not to bother with a lot of them. I know other people who almost always jump on the bandwagon... and spend months or even more struggling.
I'm pretty happy with the tools I'm currently using. I can change them and have when I felt a genuine need.
Meh. Bad math. That should be "1 error per 10000 rounds".
Qualifiers:
When I say "near 100% reliable", I'm not joking. 99.9% just isn't good enough for something I'd trust my life to. But if it approached 99.99%, then it's getting near the reliability of the gun itself, and may be good enough. That's approximately 1 error in 1000 rounds. Even that is pushing what I view as acceptable limits.
And even just given that it's battery-powered, it probably will never reach that goal in the foreseeable future.
As for its intended purpose (blocking unauthorized users), I have no doubt that it would work some of the time. But how often, given that it has to be that accurate for the authorized? I'm not confident that it would be that good at its job. It's a very difficult balancing act, and I would need a lot of convincing.
"When I pull the trigger, I want the gun to fire. I doubt this will be reliable enough to depend upon."
More to the point: if you want it to be reliable, then the fingerprint technology has to be loose enough to be UNreliable. We already know this. With today's technology, if you want to allow access with fingerprints reliably, you have to make your parameters loose enough that false positives slip in too easily.
Which means that in order to be near 100% reliable for an "authorized" shooter, this thing provably can't do what it's intended to do: reliably block the UNauthorized.
"I missed Mr. Longyear's books while growing up. I'm currently enjoying Infinity hold. Thank you!"
Hah! It has been rare that anyone has thanked me on Slashdot. You are most welcome.
"How does using a strong password prevent password re-use?"
The key issue here is not so much strong passwords, but "auditing".
If this were the beginning of the month, I would suspect an April Fools joke. They want to improve your use of passwords by monitoring your password usage!!!
Gee, what could possibly be wrong with this idea? [sarcasm]
"he "studies" you refer to found that teenagers are into music. Nothing more, nothing less."
Nonsense. And putting the word "studies" in quotes does not make them bullshit.
If you spend a few minutes on Google, you can find at least several legitimate studies that showed these things, exactly as I stated.
The first one I read about was back in the year 2000. And there have been quire a few since. Their findings have been very consistent.
"What you said and what I said are not incompatible. My comment explains the observation you've echoed."
I wasn't trying to argue with you, but just to expand a little on the point.
"All completely irrelevant, because foreigners are less likely to come from university recognized as "top-ranked" by Americans, or have any ties to US university to get a Ph.D there while working, or be involved in getting a US patent even if invention is mostlly his work."
This is a straw-man argument. Not only are those things not irrelevant, they are among the top reasons corporations claimed they needed H1-B workers: the claim has been that those workers are among the top in those metrics.
But the fact is, they are not.
"There is a significant band in the middle of people who will pirate if it's easy and buy if it's not."
Multiple studies have shown that the most prolific "illegal" downloaders are the same people who spend the most on media: music, movies in theaters, and DVDs. Yet another study just a month or so ago (discussed here on /.) had the same conclusion.
"My focus here is the aspect of how DRM protects the rights of content creators (aka, artists) and helps to prevent people freely distributing their works and with no compensation."
This is an assumption that is not borne out by the actual data.
Study after study of various aspects of DRM, in regard to software and published works anyway, belie this assumption.
People who "illegally" download movies and music also happen to be the people who spend the most on music and movies (both in-theater and DVDs).
The fact is that products that are solidly locked up under DRM tend not to sell very well. Look at the latest rebellion against Electronic Arts and Ubisoft over DRM. EA has been laying off employees.
This is not to say it might not be useful under some circumstances. But by and large, it has tended to make products less attractive to consumers.
"When you introduce commercial aspects to OS, it becomes a completely different beast"
That was my point. If they want regular, professionally-provided support, they won't get it for most Free-and-Open-Source products. They should probably opt for commercial instead.
s / enough of them / enough competent tech workers
"Don't think that a difference in "averages" says anything about people in the top percentiles."
The whole point was that relatively few of the H1-B workers were IN the top percentiles.
Companies have been claiming that we need more H1-B workers because the U.S. has not been creating enough of them. These studies show that excuse to be nothing but bullshit.
For advanced students of literature or writing, Jack Vance and Barry Longyear should be requirements.
Though some people have found Vance hard to read, his English prose is impeccable.
Longyear never uses the "he said," "she retorted," "he quipped" kind of lazy and awkward sentence construction that has come to be almost universal today. Studying how he gets around it while making it seem natural is very educational. (He did publish one short story in which he did that, but it was intended as satire of that very thing.)
"Why do you doubt that? If you have a camera at the scene of a crime and refuse to turn over the footage, a judge would happily grant a subpoena for that."
I should have qualified my statement.
If it's AT a crime scene (cameras right there), then I can certainly see a warrant being issued. But as for cameras that just happen to be in the same general area of the city, it would (and should) be a lot harder to obtain a warrant because a warrant has to be based on probable cause.
"Because it was within 300 meters and might have accidentally caught the perpetrators on video" is not probable cause.