There is some hope that this whole surveillance may start crumbling because the French are much more likely to take it more seriously, fish out that old, trusted and may be a little rusted Guillotine to behead Hollande.
I have to disappoint you: I currently live in France, and the revelation about the French state spying on its citizens get hardly any airtime. No one cares here. They are too busy with good-old-fashioned politics with faces attached to issues.
Indeed, I'd have modded you up if I'd have mod points.
This study is yet another one showing these effects, but is by far not the first. The effects of video-game playing, in particular action video-game playing, on various part of the decision making process have been studied extensively. The whole research was kicked off by the publication of
The French would disagree with this. They have single handedly foisted on the world ever longer copyrights since the 19th century. I don't know why the French are this way, but given that they have invented croissants, mayonaisse and champagne, I'm inclined to believe them.
You shouldn't love them too much for their croissants, as those were blatantly copied from the Austrians, who created the croissant's predecessor, known as the 'Kipferl', which - after being introduced by a Viennese into the Parisian society - was copied by French 'viennoiseries' and became the croissant. Luckily - for them - intellectual property protection wasn't going havoc in the 1830s, yet.
You'd think they'd have gotten the police involved instead of trying to scoop a story...
From the article:
Wie so viele Kontonummern illegal in Umlauf gelangen konnten, muss in den nächsten Wochen die Staatsanwaltschaft Düsseldorf klären. Die WirtschaftsWoche übergab den Ermittlern am vergangenen Donnerstag die CD mit den 1,2 Millionen Datensätzen und Kontonummern.
which roughly means:
How that many account number reached circulation illegally is to be clarified over the next weeks by the prosecuting authorities of Düsseldorf. Reporters of the WirtschaftsWoche handed the CD with the 1.2 million data sets and account numbers to the investigators last Thursday.
So, they firstly contacted the responsible branch of jurisdiction and after that published the article.
People post things like this to archiv all the time. It doesn't mean it is correct or deep.
Anton Zeilinger is one of the leading researchers in experimental quantum physics and generally seems to do good stuff. I can't imagine him putting his name on a paper that has utterly wrong claims.
There is a correlation between performance in visual tasks and the amount of time people have been playing action video games. The initial study has shown that action-video-game (AVG, e.g. Unreal Tournament, other ego-shooters) players perform significantly better in a range of visual attention tasks than non-AVG players. In later studies it has been shown that this increased performance is not observed for people who do play games that are not of the AVG-genre (e.g. The Sims), and also that 50h of game playing of AVG games is sufficient to observe a significant performance increase in visual tasks. Currently, the same lab is investigating whether this effect is also observed in the elderly, with positive initial results. For more information, just have a look at the lab's list of publications (disclaimer: I'm in the same department as that lab, though not member of that lab).
In relation to the article, they seem to recommend the people to play games of the non-AVG type. For this reason I have my doubts that these games will significantly improve performance in visual tasks. On the other hand, it might support other tasks that are required while driving, but that remains to be shown.
You don't even have to read the article to understand that the 25 minutes are the amount of music that can be kept in RAM, not on the HD. WTF do you get an 'Insightful' for?
..as they don't search the state space exhaustively. Going through all possible combinations of parameters would be brute force, but in this case, as the parameters are real-valued, this is even impossible (if ignoring the possibility of quantisation)
Evolutionary Algorithms provide informed search as they perform competition among the individuals (each representing one possible solution) in the population. Their performance is way above exhaustive search techniques (which _are_ brute force) but below classical search techniques. In this case, however, such classical techniques cannot be applied as the problem space is not well-defined.
Ugh. The magic doesn't come from vectors. Vectors are just how you throw the numbers around. The reason the classification apparently works well is their choice of representation of the document: a word histogram -- the occurance count for each word.
...and the word histogram is represented using a vector. 'Histogram' is proably not the best analogy as the words don't have a numeric value assigned to them (I guess they use word hash tables in this method).
To measure the distance between two histograms, you usually use the chi-squared test.
AFAIK, normalised cross-correlation between two vectors, each representing a document, is faster and gives better results in this case [1].
So, forget all about "vectors", the real work horse is the histogram. And, we can discuss about "clustering", but it's just as imporant to know how you're measuring the distance from one document to another.
This distance is given by the correlation coefficient.
[...] But, the hard part about the clustering is getting them into a space in which they actually, nicely cluster.
That's what Singular Value Decomposition was used for. I'm sure that this is not the only thing they've used to get rid of the noise and somewhere later in the article he mentions a well-performing tokeniser, maybe also doing dictionary lookups to get rid of closed class words (function words, without semantics but only to provide the grammatical framework of the sentence).
I had to stop reading the article because it was so clearly written by someone who had no comfort with the mathematical concepts or techniques. (Sorry, but seriously, it's the blind leading the blind.)
I don't know if you have noticed, but the article was written for the lay reader. Did you expect a scientific publication embedded in an interview? It does not give away all technical details (they would be stupid to do so anyway) but with a bit of imagination and background knowledge it describes well an outline of the technology used. The described method is pretty standard in the Information Extraction community anyway, but the magic usually lies in the details.
[1] H Schuetze, "Automatic Word Sense Discrimination", Computational Linguistics, 24 (1), pp97-124, 1998.
It clearly states that the keys won't be used if evesdropping was detected. Hence, no part of the message will be sent when evesdropping was detected and so the evesdropper won't have anything except for a worthless key.
You've never heard of a standard companie's callback procedure, right?
Products are only called back for repair if the probability that the costs of customers complaining and hurting the image of the compay (also related to income) is higher than the costs of the recall. Ethics is way behind this simple calculation.
"Their sales are generally too small to justify catering for region four. This reduces competition to the advantage of US studios," he said.
By the end of 1999, there were 720 DVDs available in region four, but more than 5000 in the US.
Australian Consumers Association spokesman Charles Britton said yesterday the zoning system imposed a "severe restriction of choice".
So this is not just about the small Aussie outfits but also about the choice of DVDs they get in Australia. As mentioned in the article the region four doesn't seem to be om much importance to the US studios.
Slashdot Mirror
There is some hope that this whole surveillance may start crumbling because the French are much more likely to take it more seriously, fish out that old, trusted and may be a little rusted Guillotine to behead Hollande.
I have to disappoint you: I currently live in France, and the revelation about the French state spying on its citizens get hardly any airtime. No one cares here. They are too busy with good-old-fashioned politics with faces attached to issues.
There's been other similar prior work. For example, there's evidence that gamers can quickly allocate their attention in an efficient fashion. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2680769/ and that gamers have faster reaction times for a large variety of tasks http://cdp.sagepub.com/content/18/6/321.short.
Indeed, I'd have modded you up if I'd have mod points.
This study is yet another one showing these effects, but is by far not the first. The effects of video-game playing, in particular action video-game playing, on various part of the decision making process have been studied extensively. The whole research was kicked off by the publication of
Green, C.S. & Bavelier, D. (2003). Action video games modify visual selective attention. Nature, 423, 534-537
with more publications related to that topic available on the lab page of Daphne Bavelier.
Disclaimer: I was working in the same department as the above-mentioned lab some years ago
http://xkcd.com/932/
The French would disagree with this. They have single handedly foisted on the world ever longer copyrights since the 19th century. I don't know why the French are this way, but given that they have invented croissants, mayonaisse and champagne, I'm inclined to believe them.
You shouldn't love them too much for their croissants, as those were blatantly copied from the Austrians, who created the croissant's predecessor, known as the 'Kipferl', which - after being introduced by a Viennese into the Parisian society - was copied by French 'viennoiseries' and became the croissant. Luckily - for them - intellectual property protection wasn't going havoc in the 1830s, yet.
.
You'd think they'd have gotten the police involved instead of trying to scoop a story...
From the article:
Wie so viele Kontonummern illegal in Umlauf gelangen konnten, muss in den nächsten Wochen die Staatsanwaltschaft Düsseldorf klären. Die WirtschaftsWoche übergab den Ermittlern am vergangenen Donnerstag die CD mit den 1,2 Millionen Datensätzen und Kontonummern.
which roughly means:
How that many account number reached circulation illegally is to be clarified over the next weeks by the prosecuting authorities of Düsseldorf. Reporters of the WirtschaftsWoche handed the CD with the 1.2 million data sets and account numbers to the investigators last Thursday.
So, they firstly contacted the responsible branch of jurisdiction and after that published the article.
People post things like this to archiv all the time. It doesn't mean it is correct or deep.
Anton Zeilinger is one of the leading researchers in experimental quantum physics and generally seems to do good stuff. I can't imagine him putting his name on a paper that has utterly wrong claims.
There is a correlation between performance in visual tasks and the amount of time people have been playing action video games. The initial study has shown that action-video-game (AVG, e.g. Unreal Tournament, other ego-shooters) players perform significantly better in a range of visual attention tasks than non-AVG players. In later studies it has been shown that this increased performance is not observed for people who do play games that are not of the AVG-genre (e.g. The Sims), and also that 50h of game playing of AVG games is sufficient to observe a significant performance increase in visual tasks. Currently, the same lab is investigating whether this effect is also observed in the elderly, with positive initial results. For more information, just have a look at the lab's list of publications (disclaimer: I'm in the same department as that lab, though not member of that lab).
In relation to the article, they seem to recommend the people to play games of the non-AVG type. For this reason I have my doubts that these games will significantly improve performance in visual tasks. On the other hand, it might support other tasks that are required while driving, but that remains to be shown.
There's this assumption that what is good for Disney is what's good for America, but that's an oversimplification.
There's also what's good for Yahoo and Google.
Is that supposed to imply that what's good for the economy is good for Ameria? Where does that leave its citizens?
You found out what 10% of the brain does (the sensory/motor areas)? The other 90% must not be used for anything.
This old myth actually never had its origin in science, but was created and then spread through popular media. Please don't help it survive - it's time to let it die.
Now it is official: Firefox is the new EMACS.
If only Firefox had a good web browser.
You don't even have to read the article to understand that the 25 minutes are the amount of music that can be kept in RAM, not on the HD. WTF do you get an 'Insightful' for?
..as they don't search the state space exhaustively. Going through all possible combinations of parameters would be brute force, but in this case, as the parameters are real-valued, this is even impossible (if ignoring the possibility of quantisation)
Evolutionary Algorithms provide informed search as they perform competition among the individuals (each representing one possible solution) in the population. Their performance is way above exhaustive search techniques (which _are_ brute force) but below classical search techniques. In this case, however, such classical techniques cannot be applied as the problem space is not well-defined.
Ugh. The magic doesn't come from vectors. Vectors are just how you throw the numbers around. The reason the classification apparently works well is their choice of representation of the document: a word histogram -- the occurance count for each word.
...and the word histogram is represented using a vector. 'Histogram' is proably not the best analogy as the words don't have a numeric value assigned to them (I guess they use word hash tables in this method).
To measure the distance between two histograms, you usually use the chi-squared test.
AFAIK, normalised cross-correlation between two vectors, each representing a document, is faster and gives better results in this case [1].
So, forget all about "vectors", the real work horse is the histogram. And, we can discuss about "clustering", but it's just as imporant to know how you're measuring the distance from one document to another.
This distance is given by the correlation coefficient.
[...] But, the hard part about the clustering is getting them into a space in which they actually, nicely cluster.
That's what Singular Value Decomposition was used for. I'm sure that this is not the only thing they've used to get rid of the noise and somewhere later in the article he mentions a well-performing tokeniser, maybe also doing dictionary lookups to get rid of closed class words (function words, without semantics but only to provide the grammatical framework of the sentence).
I had to stop reading the article because it was so clearly written by someone who had no comfort with the mathematical concepts or techniques. (Sorry, but seriously, it's the blind leading the blind.)
I don't know if you have noticed, but the article was written for the lay reader. Did you expect a scientific publication embedded in an interview? It does not give away all technical details (they would be stupid to do so anyway) but with a bit of imagination and background knowledge it describes well an outline of the technology used. The described method is pretty standard in the Information Extraction community anyway, but the magic usually lies in the details.
[1] H Schuetze, "Automatic Word Sense Discrimination", Computational Linguistics, 24 (1), pp97-124, 1998.
Did you RTFA?
It clearly states that the keys won't be used if evesdropping was detected. Hence, no part of the message will be sent when evesdropping was detected and so the evesdropper won't have anything except for a worthless key.
I'd say that the cars don't even have a position before you look at them.
You have IIS?
Holy shit, your fucked!
You've never heard of a standard companie's callback procedure, right?
Products are only called back for repair if the probability that the costs of customers complaining and hurting the image of the compay (also related to income) is higher than the costs of the recall. Ethics is way behind this simple calculation.
Uhm, just had a Fight Club flashback, sorry.
"Their sales are generally too small to justify catering for region four. This reduces competition to the advantage of US studios," he said.
By the end of 1999, there were 720 DVDs available in region four, but more than 5000 in the US.
Australian Consumers Association spokesman Charles Britton said yesterday the zoning system imposed a "severe restriction of choice".
So this is not just about the small Aussie outfits but also about the choice of DVDs they get in Australia. As mentioned in the article the region four doesn't seem to be om much importance to the US studios.