From a New Scientist article covering the research mentioned here - http://www.newscientist.com/article/mg20827801.300-mental-muscle-six-ways-to-boost-your-brain.html (sorry, subscription required for full text, but you should get one anyway): "Several studies have shown that simply exposing people to light improves performance on many cognitive tasks." and "In another study, volunteers had their brains scanned as they performed a short-term memory task while exposed to either violet, blue or green light. The scans revealed that after just a few seconds of light exposure an area of the brain stem known to play a role in alertness became more active (PLoS One, vol 2, p 1247). Blue light was the most potent. Similarly, in simple reaction tasks, exposure to blue light is more effective in sustaining cognitive performance than green light (Sleep, vol 29, p 161)."
Some of us not only tried, but succeeded. Among the APL systems on which I've worked, one was used by five traders who accounted for 1% of the volume of the NYSE and made a lot of money for the firm. Another was an engineering design system that far surpassed anything that was commercially available for years after it was created in APL.
Anyone who doesn't know what he's talking about, slam APL - which had features in the 1960s that are still in advance of contemporary languages. I'm sorry that your little brain can't deal with it. It remains a tremendously powerful tool - there are still four or five commercial vendors who sell a version of the language. It's a pity that so many programmers still act as if they are paid by the hour and choose large cumbersome tools when there are so many elegant and powerful ones available - not just APL, but this is the most frequently maligned by ignoramuses.
A more important issue is one of scapegoating. I have no doubt that he willfully did something wrong, thereby incurring unnecessary expense to the city and he should be punished. But how did his supervisors let him get in that situation? They, of course, have no threat of prosecution even though they allowed him to get in position to commit the abuse by failing to have policies in place to prevent it.
It's a bit like a rogue trader taking full blame when his management failed to adequately monitor and limit his positions.
Even without knowing these idiosyncrasies, it's unlikely that the code is runnable without the expensive, unique infrastructure for which it's been customized.
As far as "gaming the algorithm" - it's doing real-time processing between multiple sources. You're unlikely to discover anything by reading the code other than that they do some seemingly-odd things in the interests of efficiency. In fact, you can probably guess the most important algorithm: if the same thing is priced differently at two different exchanges, buy the cheaper one and sell the more expensive one.
> Because it's pretty easy to get people to agree to spend the necessary money, if it might save their, or their children's, lives
Not if it's their own money. Seatbelts and airbags (probably other things as well) have reduced the annual death toll from cars in the US: down from 50,000/year when the population was about 200e6 (back when I took Driver's Ed) to about 40,000/year now for a population of about 300e6. People wear seatbelts and have airbags not because of clamorous consumer demand but because of laws and regulations forcing them to save their own lives and the lives of their children.
Some of us have submitted programs in APL, or its younger sibling J, to the shootout (see http://www.mail-archive.com/general@jsoftware.com/msg02859.html). However, since the rules of the shootout specify the algorithm you have to use, you end up writing C programs in APL or J which is no way to take advantage of the power and expressiveness of these languages.
It's like taking part in a poetry contest where you're only allowed baby-talk.
However, about the claim of APL being faster than C, this is only true if you're talking about development time (which may be the most important metric). An interpreted language like APL or J might approach the execution speed of C for operations on large arrays but the interpretive overhead will always impose a performance penalty.
Basically, it says that the ethanol lobbyists are fighting back against the EPA attempting to do its job by actually measuring the effects of ethanol as fuel.
I work down here where this happened. When someone walked into our meeting today to tell us we should be prepared to evacuate if there was an alarm because of this, and that they had already evacuated the higher floors in our building, we continued our meeting after a few comments.
I imagine most people did the same. Note that at least the people who evacuated our building did so because they had been told to, not because they panicked. Also, the evacuation was only for the very highest floors, not the whole building.
This is mostly good advice, right up to the point where the poster takes seriously "Elliott Wave Theory" - technical methods are more math-like than they are serious financial math.
"The trouble is, baroque complexity of financial instruments and transactions was the primary concealment tool that allowed all the lying in the first place"
Is it lying if you don't even understand that you're doing it?
I see a lot of attempts like these to shift the onus of the housing crisis back to Clinton: "Heavy-handed incetives to take risky loans that were first implemented 20 some-odd years ago, but greatly ramped up by the Clinton administration"
But the facts don't support it. According to a good, recent Economist article (http://www.economist.com/finance/PrinterFriendly.cfm?story_id=13491933), the subprime percentage of the total American mortgage market increased from 7% in 2001 to over 20% in 2006. Also, a recent study out of UNC Chapel Hill, also quoted in the article, finds that subprime mortgages made under the Community Advantage Programs (CAP) were only one-quarter as likely to default as non-CAP subprimes.
Undoubtedly Clinton does share some of the blame, as do all recent administrations, in continuing to subsidize the inefficient and costly suburban lifestyle, but that's a separate issue.
> All that stuff about a "jury of your peers", and being judged by people who are well versed in the law (but not necessarily the material issue at hand), doesn't work in modern society.... > Because society has become too complicated for us to have just judges, or just lawyers, or outdated concepts like a "jury of our peers"... Who's only qualifications were that they registered to vote in your county.
But we let people post (well-reasoned) diatribes on Slashdot even when they don't know the difference between "Who's" and "Whose".
Since we tend to be technocrats here, we often overlook the essentially political nature of "a jury of your peers". A good part of this is that a jury of "average citizens" might refuse to find you guilty if you broke an unpopular law.
The bad part of a "jury of your peers" is when this gets interpreted narrowly to be "people just like you", say "rich male land-owners". Since I have the good fortune to have taken up residence in the state of New York, I've seen one redress to this: no one gets an automatic exemption for jury duty.
Here, the governor has been called to jury duty, as has the mayor of New York City. That they did not serve is not the point: they were called just like everyone else. In fact, they may have been more well-versed "in the law", but that isn't the point either.
Most cases have little to do with esoteric material issues but a lot to do with what seems plausible and what seems right. A friend of mine was on a jury for a murder trial and they let the guy off completely. The murder victim was a well-known bully who maybe got what was coming to him. The defendant may have pulled the trigger but the jury decided, all things considered, that he was not guilty.
That's what it's about - a shared perception of fairness.
People are never "proven innocent" in a U.S. court - they are assumed to be innocent. They can be judged "not guilty", however. The distinction is important as it addresses the question of the background assumptions.
I was reading in my comfortable chair, three feet away from where I'm now typing this.
Am I the only one who still finds it more comfortable to curl up with a book than to read a screen?
I really, really like modern digital stuff as much as any slashdotter out there but a book, or magazine, is still a superior technology in many ways: it needs no power, it's durable, I can stuff it into a pocket and take it with me, I can read anywhere there's enough light, from any position I find comfortable; if I lose it or drop it in the bathtub, no big whoop.
Some of these advantages would go away if I had one of these new-fangled readers, I suppose, rather than the laptop I mostly use but dead trees are still more "user-friendly".
Slashdot Mirror
From a New Scientist article covering the research mentioned here - http://www.newscientist.com/article/mg20827801.300-mental-muscle-six-ways-to-boost-your-brain.html (sorry, subscription required for full text, but you should get one anyway):
"Several studies have shown that simply exposing people to light improves performance on many cognitive tasks."
and
"In another study, volunteers had their brains scanned as they performed a short-term memory task while exposed to either violet, blue or green light. The scans revealed that after just a few seconds of light exposure an area of the brain stem known to play a role in alertness became more active (PLoS One, vol 2, p 1247). Blue light was the most potent. Similarly, in simple reaction tasks, exposure to blue light is more effective in sustaining cognitive performance than green light (Sleep, vol 29, p 161)."
Some of us not only tried, but succeeded. Among the APL systems on which I've worked, one was used by five traders who accounted for 1% of the volume of the NYSE and made a lot of money for the firm. Another was an engineering design system that far surpassed anything that was commercially available for years after it was created in APL.
Anyone who doesn't know what he's talking about, slam APL - which had features in the 1960s that are still in advance of contemporary languages. I'm sorry that your little brain can't deal with it. It remains a tremendously powerful tool - there are still four or five commercial vendors who sell a version of the language. It's a pity that so many programmers still act as if they are paid by the hour and choose large cumbersome tools when there are so many elegant and powerful ones available - not just APL, but this is the most frequently maligned by ignoramuses.
A more important issue is one of scapegoating. I have no doubt that he willfully did something wrong, thereby incurring unnecessary expense to the city and he should be punished. But how did his supervisors let him get in that situation? They, of course, have no threat of prosecution even though they allowed him to get in position to commit the abuse by failing to have policies in place to prevent it.
It's a bit like a rogue trader taking full blame when his management failed to adequately monitor and limit his positions.
The Mayans were right - their world did end.
Since I don't own a car, the Bugatti Veyron is the car I've chosen not to own.
This new model is great news for me, as now I'm saving over $2 million by not owning it, as opposed to the mere $1.5 million savings I used to get.
Even without knowing these idiosyncrasies, it's unlikely that the code is runnable without the expensive, unique infrastructure for which it's been customized.
As far as "gaming the algorithm" - it's doing real-time processing between multiple sources. You're unlikely to discover anything by reading the code other than that they do some seemingly-odd things in the interests of efficiency. In fact, you can probably guess the most important algorithm: if the same thing is priced differently at two different exchanges, buy the cheaper one and sell the more expensive one.
Only if they're Hot Tuna.
> Because it's pretty easy to get people to agree to spend the necessary money, if it might save their, or their children's, lives
Not if it's their own money. Seatbelts and airbags (probably other things as well) have reduced the annual death toll from cars in the US: down from 50,000/year when the population was about 200e6 (back when I took Driver's Ed) to about 40,000/year now for a population of about 300e6. People wear seatbelts and have airbags not because of clamorous consumer demand but because of laws and regulations forcing them to save their own lives and the lives of their children.
Counter-intuitive, but true.
Some of us have submitted programs in APL, or its younger sibling J, to the shootout (see http://www.mail-archive.com/general@jsoftware.com/msg02859.html). However, since the rules of the shootout specify the algorithm you have to use, you end up writing C programs in APL or J which is no way to take advantage of the power and expressiveness of these languages.
It's like taking part in a poetry contest where you're only allowed baby-talk.
One language that might be terser is J: see jsoftware.org.
For instance, Newton's method is written
N=: 1 : '- u % u d. 1'
(see http://www.jsoftware.com/jwiki/Essays/Newton's Method)
However, about the claim of APL being faster than C, this is only true if you're talking about development time (which may be the most important metric). An interpreted language like APL or J might approach the execution speed of C for operations on large arrays but the interpretive overhead will always impose a performance penalty.
..if this NY Times editorial is a sign of the times: http://www.nytimes.com/2009/05/24/opinion/24sun2.htm .
Basically, it says that the ethanol lobbyists are fighting back against the EPA attempting to do its job by actually measuring the effects of ethanol as fuel.
Or do they translate "strike" as "coup" as in "coup de grÃce"? Which would make it "one strike" and you're out.
I work down here where this happened. When someone walked into our meeting today to tell us we should be prepared to evacuate if there was an alarm because of this, and that they had already evacuated the higher floors in our building, we continued our meeting after a few comments.
I imagine most people did the same. Note that at least the people who evacuated our building did so because they had been told to, not because they panicked. Also, the evacuation was only for the very highest floors, not the whole building.
This is mostly good advice, right up to the point where the poster takes seriously "Elliott Wave Theory" - technical methods are more math-like than they are serious financial math.
"The trouble is, baroque complexity of financial instruments and transactions was the primary concealment tool that allowed all the lying in the first place"
Is it lying if you don't even understand that you're doing it?
I see a lot of attempts like these to shift the onus of the housing crisis back to Clinton:
"Heavy-handed incetives to take risky loans that were first implemented 20 some-odd years ago, but greatly ramped up by the Clinton administration"
But the facts don't support it. According to a good, recent Economist article (http://www.economist.com/finance/PrinterFriendly.cfm?story_id=13491933), the subprime percentage of the total American mortgage market increased from 7% in 2001 to over 20% in 2006. Also, a recent study out of UNC Chapel Hill, also quoted in the article, finds that subprime mortgages made under the Community Advantage Programs (CAP) were only one-quarter as likely to default as non-CAP subprimes.
Undoubtedly Clinton does share some of the blame, as do all recent administrations, in continuing to subsidize the inefficient and costly suburban lifestyle, but that's a separate issue.
The map only lets you see where you are, not where you're going - you can't move away from a view centered on your current location.
Just in case we need more proof that the initial cost of a bad decision is the same as for a good one.
> All that stuff about a "jury of your peers", and being judged by people who are well versed in the law (but not necessarily the material issue at hand), doesn't work in modern society. ...
> Because society has become too complicated for us to have just judges, or just lawyers, or outdated concepts like a "jury of our peers"... Who's only qualifications were that they registered to vote in your county.
But we let people post (well-reasoned) diatribes on Slashdot even when they don't know the difference between "Who's" and "Whose".
Since we tend to be technocrats here, we often overlook the essentially political nature of "a jury of your peers". A good part of this is that a jury of "average citizens" might refuse to find you guilty if you broke an unpopular law.
The bad part of a "jury of your peers" is when this gets interpreted narrowly to be "people just like you", say "rich male land-owners". Since I have the good fortune to have taken up residence in the state of New York, I've seen one redress to this: no one gets an automatic exemption for jury duty.
Here, the governor has been called to jury duty, as has the mayor of New York City. That they did not serve is not the point: they were called just like everyone else. In fact, they may have been more well-versed "in the law", but that isn't the point either.
Most cases have little to do with esoteric material issues but a lot to do with what seems plausible and what seems right. A friend of mine was on a jury for a murder trial and they let the guy off completely. The murder victim was a well-known bully who maybe got what was coming to him. The defendant may have pulled the trigger but the jury decided, all things considered, that he was not guilty.
That's what it's about - a shared perception of fairness.
> Frankly, I think anyone who calls their kid John should be guilty of child abuse.
It's worse if they call him "Dick".
People are never "proven innocent" in a U.S. court - they are assumed to be innocent. They can be judged "not guilty", however. The distinction is important as it addresses the question of the background assumptions.
I was reading in my comfortable chair, three feet away from where I'm now typing this.
Am I the only one who still finds it more comfortable to curl up with a book than to read a screen?
I really, really like modern digital stuff as much as any slashdotter out there but a book, or magazine, is still a superior technology in many ways: it needs no power, it's durable, I can stuff it into a pocket and take it with me, I can read anywhere there's enough light, from any position I find comfortable; if I lose it or drop it in the bathtub, no big whoop.
Some of these advantages would go away if I had one of these new-fangled readers, I suppose, rather than the laptop I mostly use but dead trees are still more "user-friendly".
To quote someone else's sig:
If you watch TV news,
you know less about the world
than if you just drank gin
straight from the bottle.
- mtDNA (123855)
The digital signal I get is a little flaky but they are supposed to boost the signal after the switch.
You're evidently not familiar with the American motto: give'm a fair trial, then string'em up!