Damn straight. I love these lines from his recent script of episode III:
EWAN MCGREGOR
It's over, Hayden. I've got the high
ground, just like Darth Maul did in
Episode 1 right before I killed him
successfully. Ignoring that, if you
jump over to me, I will cut your
shit off.
HAYDEN CHRISTENSEN
You underestimate my power to
decide not to jump to the low ground
in front of you where I will be able
to safely continue duelling, but to
instead try to jump all the way over
you and get my shit cut off!
...but on human subjects using fMRI. This research really isn't related to the matrix or DNI's directly, it's about seeing whether or not electrical signals from the brain contain enough information for a classifier (ironically, in our case, artificial neural networks) to distinguish between some subjective cognitive state.
Considering the progress we've made in distinguishing cognitive states (is this person looking at a face, a house, a squirrel, etc?) in human subjects using fMRI (an extremely noisy dataset), I'm not surprised that they found that there's enough information in a few neurons to perform classification.
Really, the best pop-sci term to describe this would be "mind reading" -- the high level goal is to have a function that transforms physical space to some sort of cognitive space. I guess you could say it's the "I" of the I/O DNI in the matrix.
Sam and Max 2 was just cancelled, so why don't they just take all that writing and whatnot they were working on an turn it into a movie? Instead of focusing on action games, some quality and humor would be nice...
...unless of course they let Uwe Boll work on it, in which case, I would have to kill myself.
Let me say first that though I'm not an expert, I am studying for a degree in psychology and neuroscience with a specific emphasis on connectionist modelling of the brain, so this research is very relevant to what I have some experience/interest in...
Anyways, before everyone gets excited about the brain's "cache", it's important to remember that computer processors and neural networks like our brain process information in entirely different ways. You get similar results some of the time, but for different reasons. The key difference is that our brain processes information in parallel, on a massive scale.
People talk about the computer-brain analogy being useful on a general level, but it's actually entirely wrong on any level. When it comes to memory, this is especially important. Our brains work by sloshing around activity through enormous numbers of neurons across interconnected layers; basically, this leads to two types of memory: active memory (patterns of activity that are actively maintained across time) and weight-based memory (adjusting the connections between neurons to influence the future processing of activity.) Usually such "short term" memory as that is being discussed in the article is referring to active memory.
Anyhow, the important bit to take away from all of this is that active memory in the brain is something that requires a lot of upkeep. It's not like computer memory that holds specific information that can be erased or retrieved--rather, it biases current processing based on a pattern of activity that resulted from past processing. Without going into too much detail, in the case of remembering dots positioned on a screen, you can imagine that seeing the dots spreads activity through the cortex, including both the spatial processing areas and some "active maintainer" area that is able to lock in patterns of activity. In the context of the test, the representation of the dots in the spatial layer activates another pattern of activity in the "active maintainer," which sort of "locks on" to the activity in the maintainer that corresponds to the the represenation of the dots in the spatial layer. When recall time comes, the active maintainer sends activation to the dots representation in the spatial layer--you can then visualize what you just saw a moment ago (literally activating the same neurons). This depends on the quality of the represenation in the active maintainer, of course, and is really oversimplified, but you can sort of get an idea of the complexity involved.
Anyways, there's already a lot of evidence that the prefrontal cortex is heavily involved in actively maintaining a set pattern of activity in the face of distraction, but since prettymuch all distinctions in the brain are gradual and not absolute anyways, it wouldn't be too surprising to find that another part of cortex could be more specifically involved in maintaing representations in the spatial processing part of the brain.
As for cognition and intelligence, there's no question that active memory is important for intelligence--if you don't have it (if you are lobotomized, removing the entire prefrontal cortex), you can't direct your thoughts to reflect anything that came before, and you become a vegetable. But as to the contribution of this specific brain area, that's clearly going to be speculation at this point.
Unfortunately, the typical lifetime of a spammer's website is around 2 hours.
Well, that may be true, but we're talking about the guys paying the spammers here. Spam directs people to companies's websites who are paying them to do so. Those are the guys this method goes after.
It also has the disadvantage of being susceptible to joe jobs and similar, someone maliciously making you or your software believe some innocent site is the culprit.
Yes, but, like I said, since you know at all times who you are attacking (the client tells you, it's open source and all) you can easily check for yourself.
if you can identify the spam so accurately and quickly why are you seeing any?
If you went to the URL, you'll see that the spam is identified by hand. Basically, you get spam, you send it to this guy, he then uses it to generate a template file that is then sent back to the distributed network as instructions for the next attack. Crappy system, yeah, but it's a pretty small project right now. In any case, that's beside the point--everyone can identify spam, I mean, that's not the problem we're trying to solve here, is it?
I keep posting about this, I've submitted a story about it, but nobody ever listens, and this strikes me as the only ORIGINAL idea that I've heard in a long time:
Everyone says that filtering all the spam in the world isn't going to help if we can't stop users from clicking on it. They're right. So if we can't stop them from clicking, why not do the reverse--flood the SPAMMER'S inbox with false positives of our own??
Basically UC is a little program that goes to companies that spam's websites and fills out their sign up forms with real looking but randomly generated info. At SOME point, there is an opportunity cost to checking up on these false positives. For example, if it costs $0.02 to check up on a false positive, and the companies make $10 for each order they sell from spamming, then we need is a distributed network to put in more than 500 false responses for each positive response they receive. If you've got a distributed network of 1000+ computers, and you put in a false positive every 30 seconds, then in 1 hr that's enough 120,000 false positives or enough to cover for 240 real responses. The beauty of this is that there is no longer any profit for the business using the spammer. It hits them where it hurts most.
But this method requires a large distributed network to work! It could, but nobody seems to know about it! Right now it's just some guy's pet project--if this thing got a serious team and some serious PR, it could really take the spamming world by storm! (Of course you'd have to watch out for abuses--targetting innocent businesses networks--but we already have large blacklists a la spamcop and under an open framework I think it'd be safe enough to use.)
For god's sake people, if we got a large enough network, it could really work!
Basically, you try to fill up spamming companies's inboxes with false responses using randomly generated yet realistic looking information. Theoretically, you get enough people doing this sort of thing, you could remove some of the profitability from spamming. At some point, the company's gotta spend a least a little effort trying to verify information. Too much time wasted investigating false responses, maybe the company's going to change its approach.
'Course, it's just some guy's pet project right now, but these sorts of approachs are kind of interesting.
Okay. The Harvard study that you are referring to was published in 1991 based on data taken from New York hospitlas in 1984. So already it's 20 years out of date, but let's ignore that for a moment.
Physicians reviewed approximately 7,743 charts. These were then classified into 1,278 "adverse events," which were essentially accidents or mishaps, and 306 of those were deemed to be due to negligence. Not necessarily malpractice--just an accident that could have been avoided, regardless of whether or not it is criminal. So yes--there are plenty of
people who are suffering from negligence.
But who is actually doing the suing?
Of 280 of the negligent claims that were studied, 47 actual malpractice suits resulted. But the board of physician reviewers concluded that in only 8 of these 47 events were the injuries caused by medical malpractice. Yet in 40 percent of these unwarranted cases, the jury awarded indemnities to the plaintiffs.
So what do we get from this study? It turns out that actually, the way the system is, people who deserve to be suing aren't, while those who don't are. Yet the study researchers, despite their own statistic that fewer than 20% of malpractice claims actually filed had an identifiable basis in medical negligence, claim that "the underlying assumption that too many groundless malpractice suits are initiated is unfounded." So it's now the most widely cited study to prove the "myth" of frivolous lawsuits.
Basically, the researchers conclude that, rather than simply trying to make sure only legitimate claims are filed, there should be more claims filed by everybody, since that too will increase the number of legitimate claims. They claimed, essentially, that since the legal system does an okay job of separating the false from the true, it's worth it.
Twenty years later, after their study has been used by everyone on all sides of the issue, I think we are seeing that it is NOT worth it. Not if you want to have a baby anytime soon--good luck finding someone who can either afford or is willing to take the risk to deliver it.
So I'm not trying to be insensitive or rude--it's just that you have to admit that there REALLY ARE frivolous lawsuits.
And these lawsuits DO result in big payoffs.
After all, if nobody's getting awards past $75k, then why is setting a limit to something like $300,000 such a big deal? This was proposed as part of a tort reform deal recently in MD--physicans even had a large rally in front of the governor in support of it--but it got shot down. Why? Because there are people making tons of money off this stuff, and in our case it just so happens that the head of the MD state senate is (well was) one of them.
20 years ago, things were different, ask any doctor struggling against the rising insurance rates today. The problem is that within the past few years it's gone crazy.
So I have read TFA, and if you want to read about the survey yourself just google "harvard malpractice study" and you'll get both sides of the issue.
That's the point, though. Whether or not those cases where people were actually wronged by criminally incompetent doctors give off huge payoffs still doesn't account for the fact that in ANY circumstance, for the other 99% of cases, there are huge LEGAL costs that largely fall on the doctor and the doctor's insure provider.
I completely agree that there are times where the plantiffs deserve all the money they get--usually, when they get money, they deserve it. But the other 99% of times, it just puts a huge burden on the doctor and his insurance provider. The payoffs aren't the problem--the problem is the number of lawsuits that DONT result in payoffs. Just getting sued is enough to cause a doctor's insurance payments to rise like a rocket, regardless of whether or not it was actually a valid claim and regardless of whether or not that doctor successfully proves their innocence.
And look, I don't believe that crap about only one percent of valid claims getting rewards. What is their standard for "valid"? A valid claim is one that was decided by the court to be valid. In a legal sense, a criminal, regardless of whatever somebody else may think after reviewing the case on his own, is innocent if that's what the court has decided. If you disagree you can go to court and appeal.
Finally, the way the system is now, it's totally dependent on the juries. While there are cases like your sister, there are also plenty of cases where the plaintiffs do receive millions of dollars.
Good point, but there's one thing you're missing.
1/100 of the cases that actually file suit actually result in an award, right?
So, who's paying for the other 99?
The patient's lawyers usually only get paid if they win...but what about the doctor's lawyers? Who's going to pay the ridiculous legal fees that a serious, competent law firm is going to charge to take the case? That's right--even an UNSUCCESSFUL lawsuit costs a TON of money in legal fees for the doctor and the doctor's insurance. Lawsuits don't happen for free.
I mean, honestly, if you had to pay insurance fees of $250,000 just to do your job--as some Ob/Gyn do in TX--wouldn't you be pretty angry too?
Look, yeah, obviously you don't want the "problem doctors" to continue to be practicing, but right now, insurance prices are driving the competent, caring doctors to simply close up shop. So not only is the current system preventing those incompetent or simply dishonest doctors from practicing, it's screwing honest doctors over as well.
If you have to view each patient as a potential enemy, what's the point in practicing medicine anymore?
As was pointed out in the article, the situation with spammers sucks right now. The only way it's going to change is if we can change the economics of the situation--this calls for novel ideas, such as Unsolicited Commando, which uses the idea of false positives to make it economically less profitable for spammers.
The idea is based around the fact that there are to places to attack the economics of spam: one, the sending (spammer) side, and, two, the response processing (employer of spammer) side. It's already been argued that making email cost money to send isn't really feasible, at least not in the future.
But you can increase the cost of the response processing: every time companies get a positive response to their spam, the company must put at least some amount of effort into validating the information and then processing it (such as a subscription, product placement, etc.) So, what if the company received lots of potentially valid fake responses (false positives) to spam, so many that the processing costs would actually outweight the benefits of advertising with spam? If companies could never figure out who their real customers were, it wouldn't be worth it.
That's the idea behind Unsolicited Commando, a small program that runs in the background on your PC and that receives "orders" from a central server essentially giving enough information for the program to go to a website and fill out a form with real-sounding but bogus info. If enough computers were doing it, bogus info would be coming from such a variety of internet addresses that there'd be no way for spam companies to filter it.
So far as I can see, this type of approach is our best bet.
Slashdot Mirror
Considering the progress we've made in distinguishing cognitive states (is this person looking at a face, a house, a squirrel, etc?) in human subjects using fMRI (an extremely noisy dataset), I'm not surprised that they found that there's enough information in a few neurons to perform classification.
Really, the best pop-sci term to describe this would be "mind reading" -- the high level goal is to have a function that transforms physical space to some sort of cognitive space. I guess you could say it's the "I" of the I/O DNI in the matrix.
Anyways, before everyone gets excited about the brain's "cache", it's important to remember that computer processors and neural networks like our brain process information in entirely different ways. You get similar results some of the time, but for different reasons. The key difference is that our brain processes information in parallel, on a massive scale.
People talk about the computer-brain analogy being useful on a general level, but it's actually entirely wrong on any level. When it comes to memory, this is especially important. Our brains work by sloshing around activity through enormous numbers of neurons across interconnected layers; basically, this leads to two types of memory: active memory (patterns of activity that are actively maintained across time) and weight-based memory (adjusting the connections between neurons to influence the future processing of activity.) Usually such "short term" memory as that is being discussed in the article is referring to active memory.
Anyhow, the important bit to take away from all of this is that active memory in the brain is something that requires a lot of upkeep. It's not like computer memory that holds specific information that can be erased or retrieved--rather, it biases current processing based on a pattern of activity that resulted from past processing. Without going into too much detail, in the case of remembering dots positioned on a screen, you can imagine that seeing the dots spreads activity through the cortex, including both the spatial processing areas and some "active maintainer" area that is able to lock in patterns of activity. In the context of the test, the representation of the dots in the spatial layer activates another pattern of activity in the "active maintainer," which sort of "locks on" to the activity in the maintainer that corresponds to the the represenation of the dots in the spatial layer. When recall time comes, the active maintainer sends activation to the dots representation in the spatial layer--you can then visualize what you just saw a moment ago (literally activating the same neurons). This depends on the quality of the represenation in the active maintainer, of course, and is really oversimplified, but you can sort of get an idea of the complexity involved.
Anyways, there's already a lot of evidence that the prefrontal cortex is heavily involved in actively maintaining a set pattern of activity in the face of distraction, but since prettymuch all distinctions in the brain are gradual and not absolute anyways, it wouldn't be too surprising to find that another part of cortex could be more specifically involved in maintaing representations in the spatial processing part of the brain.
As for cognition and intelligence, there's no question that active memory is important for intelligence--if you don't have it (if you are lobotomized, removing the entire prefrontal cortex), you can't direct your thoughts to reflect anything that came before, and you become a vegetable. But as to the contribution of this specific brain area, that's clearly going to be speculation at this point.
Well, that may be true, but we're talking about the guys paying the spammers here. Spam directs people to companies's websites who are paying them to do so. Those are the guys this method goes after.
It also has the disadvantage of being susceptible to joe jobs and similar, someone maliciously making you or your software believe some innocent site is the culprit.
Yes, but, like I said, since you know at all times who you are attacking (the client tells you, it's open source and all) you can easily check for yourself.
if you can identify the spam so accurately and quickly why are you seeing any?
If you went to the URL, you'll see that the spam is identified by hand. Basically, you get spam, you send it to this guy, he then uses it to generate a template file that is then sent back to the distributed network as instructions for the next attack. Crappy system, yeah, but it's a pretty small project right now. In any case, that's beside the point--everyone can identify spam, I mean, that's not the problem we're trying to solve here, is it?
Unsolicited Commando
Everyone says that filtering all the spam in the world isn't going to help if we can't stop users from clicking on it. They're right. So if we can't stop them from clicking, why not do the reverse--flood the SPAMMER'S inbox with false positives of our own?? Basically UC is a little program that goes to companies that spam's websites and fills out their sign up forms with real looking but randomly generated info. At SOME point, there is an opportunity cost to checking up on these false positives. For example, if it costs $0.02 to check up on a false positive, and the companies make $10 for each order they sell from spamming, then we need is a distributed network to put in more than 500 false responses for each positive response they receive. If you've got a distributed network of 1000+ computers, and you put in a false positive every 30 seconds, then in 1 hr that's enough 120,000 false positives or enough to cover for 240 real responses. The beauty of this is that there is no longer any profit for the business using the spammer. It hits them where it hurts most.
But this method requires a large distributed network to work! It could, but nobody seems to know about it! Right now it's just some guy's pet project--if this thing got a serious team and some serious PR, it could really take the spamming world by storm! (Of course you'd have to watch out for abuses--targetting innocent businesses networks--but we already have large blacklists a la spamcop and under an open framework I think it'd be safe enough to use.)
For god's sake people, if we got a large enough network, it could really work!
Basically, you try to fill up spamming companies's inboxes with false responses using randomly generated yet realistic looking information. Theoretically, you get enough people doing this sort of thing, you could remove some of the profitability from spamming. At some point, the company's gotta spend a least a little effort trying to verify information. Too much time wasted investigating false responses, maybe the company's going to change its approach.
'Course, it's just some guy's pet project right now, but these sorts of approachs are kind of interesting.
That being said, it's important to keep in mind that it doesn't work if you're the only one laughing...
Okay. The Harvard study that you are referring to was published in 1991 based on data taken from New York hospitlas in 1984. So already it's 20 years out of date, but let's ignore that for a moment.
Physicians reviewed approximately 7,743 charts. These were then classified into 1,278 "adverse events," which were essentially accidents or mishaps, and 306 of those were deemed to be due to negligence. Not necessarily malpractice--just an accident that could have been avoided, regardless of whether or not it is criminal. So yes--there are plenty of people who are suffering from negligence.
But who is actually doing the suing?
Of 280 of the negligent claims that were studied, 47 actual malpractice suits resulted. But the board of physician reviewers concluded that in only 8 of these 47 events were the injuries caused by medical malpractice. Yet in 40 percent of these unwarranted cases, the jury awarded indemnities to the plaintiffs.
So what do we get from this study? It turns out that actually, the way the system is, people who deserve to be suing aren't, while those who don't are. Yet the study researchers, despite their own statistic that fewer than 20% of malpractice claims actually filed had an identifiable basis in medical negligence, claim that "the underlying assumption that too many groundless malpractice suits are initiated is unfounded." So it's now the most widely cited study to prove the "myth" of frivolous lawsuits.
Basically, the researchers conclude that, rather than simply trying to make sure only legitimate claims are filed, there should be more claims filed by everybody, since that too will increase the number of legitimate claims. They claimed, essentially, that since the legal system does an okay job of separating the false from the true, it's worth it.
Twenty years later, after their study has been used by everyone on all sides of the issue, I think we are seeing that it is NOT worth it. Not if you want to have a baby anytime soon--good luck finding someone who can either afford or is willing to take the risk to deliver it.
So I'm not trying to be insensitive or rude--it's just that you have to admit that there REALLY ARE frivolous lawsuits.
And these lawsuits DO result in big payoffs. After all, if nobody's getting awards past $75k, then why is setting a limit to something like $300,000 such a big deal? This was proposed as part of a tort reform deal recently in MD--physicans even had a large rally in front of the governor in support of it--but it got shot down. Why? Because there are people making tons of money off this stuff, and in our case it just so happens that the head of the MD state senate is (well was) one of them.
20 years ago, things were different, ask any doctor struggling against the rising insurance rates today. The problem is that within the past few years it's gone crazy.
So I have read TFA, and if you want to read about the survey yourself just google "harvard malpractice study" and you'll get both sides of the issue.
I completely agree that there are times where the plantiffs deserve all the money they get--usually, when they get money, they deserve it. But the other 99% of times, it just puts a huge burden on the doctor and his insurance provider. The payoffs aren't the problem--the problem is the number of lawsuits that DONT result in payoffs. Just getting sued is enough to cause a doctor's insurance payments to rise like a rocket, regardless of whether or not it was actually a valid claim and regardless of whether or not that doctor successfully proves their innocence.
And look, I don't believe that crap about only one percent of valid claims getting rewards. What is their standard for "valid"? A valid claim is one that was decided by the court to be valid. In a legal sense, a criminal, regardless of whatever somebody else may think after reviewing the case on his own, is innocent if that's what the court has decided. If you disagree you can go to court and appeal.
Finally, the way the system is now, it's totally dependent on the juries. While there are cases like your sister, there are also plenty of cases where the plaintiffs do receive millions of dollars.
So, who's paying for the other 99?
The patient's lawyers usually only get paid if they win...but what about the doctor's lawyers? Who's going to pay the ridiculous legal fees that a serious, competent law firm is going to charge to take the case? That's right--even an UNSUCCESSFUL lawsuit costs a TON of money in legal fees for the doctor and the doctor's insurance. Lawsuits don't happen for free.
I mean, honestly, if you had to pay insurance fees of $250,000 just to do your job--as some Ob/Gyn do in TX--wouldn't you be pretty angry too?
Look, yeah, obviously you don't want the "problem doctors" to continue to be practicing, but right now, insurance prices are driving the competent, caring doctors to simply close up shop. So not only is the current system preventing those incompetent or simply dishonest doctors from practicing, it's screwing honest doctors over as well.
If you have to view each patient as a potential enemy, what's the point in practicing medicine anymore?
Down with freedom in video games!!!
As was pointed out in the article, the situation with spammers sucks right now. The only way it's going to change is if we can change the economics of the situation--this calls for novel ideas, such as Unsolicited Commando, which uses the idea of false positives to make it economically less profitable for spammers.
The idea is based around the fact that there are to places to attack the economics of spam: one, the sending (spammer) side, and, two, the response processing (employer of spammer) side. It's already been argued that making email cost money to send isn't really feasible, at least not in the future.
But you can increase the cost of the response processing: every time companies get a positive response to their spam, the company must put at least some amount of effort into validating the information and then processing it (such as a subscription, product placement, etc.) So, what if the company received lots of potentially valid fake responses (false positives) to spam, so many that the processing costs would actually outweight the benefits of advertising with spam? If companies could never figure out who their real customers were, it wouldn't be worth it.
That's the idea behind Unsolicited Commando, a small program that runs in the background on your PC and that receives "orders" from a central server essentially giving enough information for the program to go to a website and fill out a form with real-sounding but bogus info. If enough computers were doing it, bogus info would be coming from such a variety of internet addresses that there'd be no way for spam companies to filter it.
So far as I can see, this type of approach is our best bet.