... since knowing there's something there provides additional information that can be used to calibrate the extraction routine.
but not visible until new knowledge could see the picture in a fresh light.
This says it all. In fact, you could create a much simpler extraction technique consisting of a black box around the known item that meets this same standard. Can the new extraction technique do more than this? That, apparently, remains to be seen.
This is the just as likely to add burden as to remove it.
I can't count the number of times I've seen attempts to 'standardize' data, or even just notation, in a given field. It all works very well for data to that point, but then the field expands or changes, or new assumptions become important, and the whole thing becomes either unwieldy or obsolete. This is one reason why every different field, it seems, has their own standards in their literature.
Speaking of the literature, most of these proposals are quickly followed by a 'let's just ask authors to conform to this now' approach to adopting these things. Papers get rewritten (or rejected), key points get lost, and the community gets weaker, all so that some standard with a half life of 12 months can be implemented.
This might be different. I applaud people trying to solve hard problems, and this is certainly one. I do think that more of the burden should be on demonstrating that the standradization is applicable for 12 months or more AFTER final development in a given field, never mind several.
Generally, though we shouldn't fear context. We should embrace it.
Ouch... Isn't part of a backup strategy to sometimes attempt a recovery from a backup, on a test system?
Yes. He addresses this and acknowledges he did not either deliberately fail his system or conduct extensive tests to ensure his backup scheme was adequate.
He acknowledges this was one of many 'lessons learned' (aka huge mistakes made).
You will find some formulas in mathematical notation but some are really hard to read since they are printed in a font size of about 4 while the text is written in 10
Yikes. How'd that ever happen? Is this common with Manning's Equations? (Hah!)
No. Modern cryptographic systems are far more versatile than your 2 scenario viewpoint suggests. Every time you use the internet, you tap into some of this. 2-of-3 authorization is so trivial, I'm surprised I even have to mention it.
Here's how you do it with a face: instead of using your own face, you a photo of Brad Pitt on your Iphone or related device.
When they brute force that, you switch to a picture of Jennifer Anniston.
You can change your 'biometric-based' password just as easily as they can brute force it. Just don't limit yourself to your own biometrics.
Once they have your password, you choose another one and that's it. I'd like to see you do that with your face.
I take your point, but I don't understand the either/or philosophy of security. Besides, in most cases that matter, once they have your 'password', they have you. Period.
To me, security is all about layering anyways. Adding a biometric layer that works well for the user (i.e. effortless) and typically involves a brute force attack to defeat? Why not?
It was a bit of a joke. But I don't think your scenario would work anyways given their need to adjust lighting conditions as they mentioned.
More to the point, you could use something like an Iphone with a DB of randomly generated photos until it cracked. This is what the researchers here did. This is the real vulnerability. But it's brute force attack, and on any proper 'secured' system it would have to be one of several.
He says the laptop makers should remove the facial biometrics feature from their products because the vulnerability of this technology can't be fixed.
If that's the standard, all security features should be removed. Everything is somewhat vulnerable, and a determined intruder with infinite resource will almost always find a way in. The object is to make this unreasonably hard for most applications.
If you get your laptop lifted at the coffee shop, they better lift your wallet too I guess.
My first thought upon reading this article is that Vegas wants to get lots of people to think they can count cards and come on down to empty their pockets. Even (maybe especially?) in relatively desperate times.
Most won't dare use the IPhone in a Casino anyways, though they might kid themselves long enough to book a trip and hit the floor.
Some might decide to do it themselves once they figure out the mechanics of card counting. The casinos love people like this.
It's basically a big commercial for Vegas, Blackjack, and deluding yourself that you can maybe win too. Just like everything else these days, it seems.
Canada's system of government is (and probably needs to be) Parliamentary system, Westminster-style. The nonsense with the Newfoundland delegates being allowed to vote for their constituent's interest, and the whole debacle about the coalition government a few months back illustrate how far we've gotten from this -- most people are apparently not even aware of what constitutes a government in Canada.
CRTC's biggest challenge is in mirroring this representation domestically, not branding at all. Anyone who watches Hockey Night in Canada knows that even with the amount of regional protection rules we have enforced now, we still get what seems like 80-90% coverage of a terrible Toronto-based team preferentially over much better teams in playoff races. Contrary to what LEaf fans think, this is not because it's 'what the people want to see'.
With all that said, I don't think any Canadian content as envision by the parent is currently needed or really, technologically viable.
I DO think that networks, websites, and shows that want to have regional IP blocking and a 'Canadian viewers go here' set up (as many now do) may reasonably be required to have a mandatory amount of Canadian content available though. Since they are the ones enforcing the distinction, it would be both reasonable and feasible for the CRTC to concern itself with 'airwaves' targeted purely at Canadians.
I don't know where to start, so let's work on this:
It's absurd that we argue over whether certain works were written by shakespeare as if it would make them better plays if they had been and take some undiscovered painting by a classical master to be a great work but dismiss it if we discover it was truly modern.
and
We don't go back and read Newton and neither does it make sense to look at the originals for anything but historical purpose.
Clearly you're not a scientist, or at least not a very good one. We do precisely that. There's as much value in rehashing Newton or Fermat (did he know? did he make a mistake) or Descartes or Shakespeare to understand the MAN and the history as there is to 'discover' something new. And new-old discoveries happen on a regular basis too.
Since these are atmospheric releases, they're almost certainly are items already 'tagged'. Presumably this would be a destructive test in that they need to check areas that are -in- the paint.
\
This is grade school stuff. Correlation is not causation.
Which means if you're approaching a region you haven't sampled, then you can't understand what's going to happen because you've thrown away your interest in 'why [something] does what it does, [because] it just does it.'
If you're only using models as correlations or proxies, what are you using models for anyways? There's nothing 'increasingly' true about that.
To all the "do your own work" flamers out here: More often than not, PhD projects are based on vague ideas or old projects that were left lying around by the professor because it is too risky to put an expensive PostDoc on it.
Complete bullshit. If you're going to say that, cite a reference. I can find 5000 that say the opposite, but if pressed I'll list 10.
That's why I have a Ph.D. and you don't even know how to abbreviate it properly.
Get some good journal papers and go from there. Please, please take 'articles' with a grain of salt, no matter where they come from, but particularly online articles and to a lesser extent 'proceedings' (I note Stafford lists her journal papers with her conference ones, and this can be problematic)
Agreed. Get thee to a library my friend. If you can't cite the 10 most important and 10 most recent journal articles on the question, what the hell are you doing asking for help on Slashdot? Who's your supervisor?
Well, guesstimating 0.5% of 300mill in about 40million in that age group, that's about 4%. If you're willing to assume that a disproportionate amount of those enlisted (and targeted by Nintendo) are young men, then it shoots up to close to 10%. As the recession gets deeper and casualties hopefully get lower, the enrollment rate is likely to go up.
All kinds of things. We probably both know them ad nauseum, but suffice it to say a weak dollar is not in and of itself driving the recession.
which started with mortgage market collapsing
Nope. It started well before that. When fast food jobs started becoming listed as manufacturing jobs (back in 2002-2003?) in Michigan that was a sign, for instance. Increasingly tight immigration policies that are hurting the available workforce is another. Credit crunch is certainly another, but that's been going on for 30 years.
It's also a weak MARKET issue. As a recession hits, the market for toys tends to dwindle. As a larger portion of 18 to 30 year olds serve in foreign engagements, the market for toys for that age group tends to dwindle too.
Nobody spent their economic incentive checks. That's pretty much all anybody needs to know.
Since surely this is at least in part teaching issue, and seems to be commonplace in most of western society (not just nations with the NCLBA or equiv.) doesn't this really suggest a drop in teaching quality/ability? Are we handing out 'licenses' to the wrong people, or too easily (to people with integrity issues who bend to NCLB standards) or some other flawed way? Should we be handing out 'licenses' at all? Can teaching ever be taught, really? Are teaching unions a help or a hindrance to the education product (and not to the teachers standard of living which I agree is generally too low)?
I'm sure this'll kill my karma, but I'd be genuinely interested in hearing thoughts along those lines since pinning it all on some amorphous 'system' or act or generation seems like a bit of a cop out, to say the least.
Slashdot Mirror
... since knowing there's something there provides additional information that can be used to calibrate the extraction routine.
but not visible until new knowledge could see the picture in a fresh light.
This says it all. In fact, you could create a much simpler extraction technique consisting of a black box around the known item that meets this same standard. Can the new extraction technique do more than this? That, apparently, remains to be seen.
This is the just as likely to add burden as to remove it.
I can't count the number of times I've seen attempts to 'standardize' data, or even just notation, in a given field. It all works very well for data to that point, but then the field expands or changes, or new assumptions become important, and the whole thing becomes either unwieldy or obsolete. This is one reason why every different field, it seems, has their own standards in their literature.
Speaking of the literature, most of these proposals are quickly followed by a 'let's just ask authors to conform to this now' approach to adopting these things. Papers get rewritten (or rejected), key points get lost, and the community gets weaker, all so that some standard with a half life of 12 months can be implemented.
This might be different. I applaud people trying to solve hard problems, and this is certainly one. I do think that more of the burden should be on demonstrating that the standradization is applicable for 12 months or more AFTER final development in a given field, never mind several.
Generally, though we shouldn't fear context. We should embrace it.
Ouch... Isn't part of a backup strategy to sometimes attempt a recovery from a backup, on a test system?
Yes. He addresses this and acknowledges he did not either deliberately fail his system or conduct extensive tests to ensure his backup scheme was adequate.
He acknowledges this was one of many 'lessons learned' (aka huge mistakes made).
You will find some formulas in mathematical notation but some are really hard to read since they are printed in a font size of about 4 while the text is written in 10
Yikes. How'd that ever happen? Is this common with Manning's Equations? (Hah!)
Seriously though, what a nightmare.
There are 2 ways to use biometrics.
No. Modern cryptographic systems are far more versatile than your 2 scenario viewpoint suggests. Every time you use the internet, you tap into some of this. 2-of-3 authorization is so trivial, I'm surprised I even have to mention it.
Here's how you do it with a face: instead of using your own face, you a photo of Brad Pitt on your Iphone or related device. When they brute force that, you switch to a picture of Jennifer Anniston. You can change your 'biometric-based' password just as easily as they can brute force it. Just don't limit yourself to your own biometrics.
Once they have your password, you choose another one and that's it. I'd like to see you do that with your face.
I take your point, but I don't understand the either/or philosophy of security. Besides, in most cases that matter, once they have your 'password', they have you. Period.
To me, security is all about layering anyways. Adding a biometric layer that works well for the user (i.e. effortless) and typically involves a brute force attack to defeat? Why not?
It was a bit of a joke. But I don't think your scenario would work anyways given their need to adjust lighting conditions as they mentioned.
More to the point, you could use something like an Iphone with a DB of randomly generated photos until it cracked. This is what the researchers here did. This is the real vulnerability. But it's brute force attack, and on any proper 'secured' system it would have to be one of several.
He says the laptop makers should remove the facial biometrics feature from their products because the vulnerability of this technology can't be fixed.
If that's the standard, all security features should be removed. Everything is somewhat vulnerable, and a determined intruder with infinite resource will almost always find a way in. The object is to make this unreasonably hard for most applications.
If you get your laptop lifted at the coffee shop, they better lift your wallet too I guess.
My first thought upon reading this article is that Vegas wants to get lots of people to think they can count cards and come on down to empty their pockets. Even (maybe especially?) in relatively desperate times.
Most won't dare use the IPhone in a Casino anyways, though they might kid themselves long enough to book a trip and hit the floor.
Some might decide to do it themselves once they figure out the mechanics of card counting. The casinos love people like this.
It's basically a big commercial for Vegas, Blackjack, and deluding yourself that you can maybe win too. Just like everything else these days, it seems.
I don't really agree.
Canada's system of government is (and probably needs to be) Parliamentary system, Westminster-style. The nonsense with the Newfoundland delegates being allowed to vote for their constituent's interest, and the whole debacle about the coalition government a few months back illustrate how far we've gotten from this -- most people are apparently not even aware of what constitutes a government in Canada.
CRTC's biggest challenge is in mirroring this representation domestically, not branding at all. Anyone who watches Hockey Night in Canada knows that even with the amount of regional protection rules we have enforced now, we still get what seems like 80-90% coverage of a terrible Toronto-based team preferentially over much better teams in playoff races. Contrary to what LEaf fans think, this is not because it's 'what the people want to see'.
With all that said, I don't think any Canadian content as envision by the parent is currently needed or really, technologically viable.
I DO think that networks, websites, and shows that want to have regional IP blocking and a 'Canadian viewers go here' set up (as many now do) may reasonably be required to have a mandatory amount of Canadian content available though. Since they are the ones enforcing the distinction, it would be both reasonable and feasible for the CRTC to concern itself with 'airwaves' targeted purely at Canadians.
Exactly. It's called spectroscopy, and really you need to check for both decays and their peaks. I worked on the 'detectors' a few years back.
I don't know where to start, so let's work on this:
It's absurd that we argue over whether certain works were written by shakespeare as if it would make them better plays if they had been and take some undiscovered painting by a classical master to be a great work but dismiss it if we discover it was truly modern.
and
We don't go back and read Newton and neither does it make sense to look at the originals for anything but historical purpose.
Clearly you're not a scientist, or at least not a very good one. We do precisely that. There's as much value in rehashing Newton or Fermat (did he know? did he make a mistake) or Descartes or Shakespeare to understand the MAN and the history as there is to 'discover' something new. And new-old discoveries happen on a regular basis too.
they're? Really? Sheesh. I need a coffee.
Since these are atmospheric releases, they're almost certainly are items already 'tagged'. Presumably this would be a destructive test in that they need to check areas that are -in- the paint. \
This is grade school stuff. Correlation is not causation.
Which means if you're approaching a region you haven't sampled, then you can't understand what's going to happen because you've thrown away your interest in 'why [something] does what it does, [because] it just does it.'
If you're only using models as correlations or proxies, what are you using models for anyways? There's nothing 'increasingly' true about that.
That's gonna be one hell of a party man.
One. Hell. Of a Party.
To all the "do your own work" flamers out here: More often than not, PhD projects are based on vague ideas or old projects that were left lying around by the professor because it is too risky to put an expensive PostDoc on it.
Complete bullshit. If you're going to say that, cite a reference. I can find 5000 that say the opposite, but if pressed I'll list 10.
That's why I have a Ph.D. and you don't even know how to abbreviate it properly.
That is not a peer-reviewed article is it?
Get some good journal papers and go from there. Please, please take 'articles' with a grain of salt, no matter where they come from, but particularly online articles and to a lesser extent 'proceedings' (I note Stafford lists her journal papers with her conference ones, and this can be problematic)
Just my 2cents.
Agreed. Get thee to a library my friend. If you can't cite the 10 most important and 10 most recent journal articles on the question, what the hell are you doing asking for help on Slashdot? Who's your supervisor?
Just wondering if the USACEWP (the Bircher Group) and the AFCC (Lord's group) have ever gamed off against one another, or do so routinely?
If not, why the heck not?
If so, who wins most?
Well, guesstimating 0.5% of 300mill in about 40million in that age group, that's about 4%. If you're willing to assume that a disproportionate amount of those enlisted (and targeted by Nintendo) are young men, then it shoots up to close to 10%. As the recession gets deeper and casualties hopefully get lower, the enrollment rate is likely to go up.
What do you think is driving the recession?
All kinds of things. We probably both know them ad nauseum, but suffice it to say a weak dollar is not in and of itself driving the recession.
which started with mortgage market collapsing
Nope. It started well before that. When fast food jobs started becoming listed as manufacturing jobs (back in 2002-2003?) in Michigan that was a sign, for instance. Increasingly tight immigration policies that are hurting the available workforce is another. Credit crunch is certainly another, but that's been going on for 30 years.
It's also a weak MARKET issue. As a recession hits, the market for toys tends to dwindle. As a larger portion of 18 to 30 year olds serve in foreign engagements, the market for toys for that age group tends to dwindle too.
Nobody spent their economic incentive checks. That's pretty much all anybody needs to know.
I'm a licensed teacher
Since surely this is at least in part teaching issue, and seems to be commonplace in most of western society (not just nations with the NCLBA or equiv.) doesn't this really suggest a drop in teaching quality/ability? Are we handing out 'licenses' to the wrong people, or too easily (to people with integrity issues who bend to NCLB standards) or some other flawed way? Should we be handing out 'licenses' at all? Can teaching ever be taught, really? Are teaching unions a help or a hindrance to the education product (and not to the teachers standard of living which I agree is generally too low)?
I'm sure this'll kill my karma, but I'd be genuinely interested in hearing thoughts along those lines since pinning it all on some amorphous 'system' or act or generation seems like a bit of a cop out, to say the least.