And yet there is greater socioeconomic mobility in the US than in other places, such as Europe.
It would be hard to be more wrong — the US is on par with the UK, and both are well behind most of Western Europe. See, for example, this Wikipedia entry.
Much like BlackPignouf, it took me a month of daily practice to learn how to do the standard three cascade, and having been through that, I've found it very easy to teach others how to do it. (Probably, familiarity with a whole suite of failure modes helps.)
I persisted with juggling, and had a lot of fun with more complicated patterns, pass juggling, and so on. But it is all a case of directed persistance overcoming a complete lack of any natural ability.
[...] using real numbers (real, not floating-point) would give a trans-Turing capability.
Given that almost every real number encodes an uncountable number of bits of information, I guess this isn't especially surprising in retrospect. The result though should make us suspicious of the assumption that the physical constants and properties in our physical theories can indeed take any real number value.
Let me second this recommendation. I used to bring back supplies from Japan, or inveigle friends to bring them back from work trips to Singapore, but now they are sold locally (Australia)! I use them for everything, but particularly value the ease with which they write, and the fine width which allows the precise writing of both mathematical expressions and complicated characters.
One downside is that they can vary a little in consistency: if you have the opportunity, try a number of them and pick the best one. Also, they do not work well when used with a ruler — they quickly wear out the nib.
An alternative, not quite as good, is the Muji-brand ball-point pen. They periodically need to have the barrel tightened (simply done with the fingers), but otherwise can give a good, fine line.
There is no design for a "working commercial thorium reactors". It's all just bits and pieces of theory, and experimental reactors that only answered some of the questions.
This is not true — CANDU reactors can burn thorium in a number of fuel configurations, and they have been around for decades. That none are operating commercially on a thorium fuel cycle is, I believe, primarily due to a combination of regulation and infrastructure considerations.
Next generation thorium reactors will be great, but we already have the technology to use thorium. We just don't.
Top-level elite athletes are already genetic outliers who have also benefitted from good fortune in early training and nutrition and, typically, tens or hundreds of thousands of dollars worth of targetted training.
It's not just a matter of will power or clever training schedules. It matters not how strong be my willpower, or how dedicated my training: I will never be an olympic-class athlete.
Bring on the drugs, and the treatments. It would make elite sport more equitable, and further, the medical risks taken by those with the burning desire to compete at any cost will allow the greater majority of people to benefit from enhancements with more safety.
It's a 'good-enough' approximation to an optimal selection process.
The probability of someone clicking on option A, B or C is unknown, but is expected to be constant when averaged over the population. Given the ratio of clicks versus views on any given option, the posterior distribution of that probability can be modelled as a Beta distribution. The experimental question is then: given the current estimates, which option should be presented to maximise the utility of the test?
For simply ranking the options, the utility may be the Shannon information. In this case though, the utility also has to incorporate the expected benefit of a click-through. One could set up a utility function which is weighted between the two outcomes, possibly varying over time.
In practice though, Beta distributions with different means tend to converge to separate peaks quite quickly, so taking a possible 10% hit on the current best estimate click-through outcome seems an entirely plausible approximation. Bayesian experimental design though could also tell you when to stop testing and stick with the winner.
On a counterpoint, the engine of a recent simulation project of ours is written in C++ and hand-optimized for speed. It took much more work to get VC 2008 to produce tight code than it did for g++ 4.5, and even then, the g++ compiled code ran a bit faster. (The code has a mix of fp and integer work, and the core of it and working set are small enough to fit comfortably in L2 cache; there are some unpredictable branches in tight loops too, for good measure.)
'Hundreds of millions'?
Qantas has 33000 or so employees. Even a $5000 p.a. increase in salary to every employee wouldn't make it to the hundreds of millions. On the other hand, Qantas did just double its yearly net profits, to a total of $250 million.
So, don't the other domestic Australian airlines employ people belonging to these unions? Meanwhile Qantas doubles its profits, spends 10 million dollars on a re-branding exercise, and gives a 1.5 million dollar raise to its CEO. Now this current suspension is estimated to be costing them $20 million per day.
If I had to choose a side based on the available evidence, it would not be Qantas' management.
... at least, this matches my experience at an Australian game development company. At least we didn't have to suffer this for seven years before shipping, though.
Sure enough, after shipping, the company lost 70% of their coders and they were reduced to producing shallow clones of their original (good) game.
Imagine if we could have a reset button.
Imagine a world where, physically at least, most decisions could be undone, and the only lasting consequences of a poor choice were social.
No more: crippled by that car accident; freak fall that took out your friend's eye; paraplegic after the top rope snapped; blinded after the pressure cooker exploded.
We should, as a society, be aiming to conquer death and injury. Technology can be our saviour. Physical misery through misfortune, or through the sheer arbitrariness of our biology and environment, may one day be only a historical curiosity, but we need to dedicate real resources and real time to the project.
Do not accept the status quo! The only reason why we have so much death, is that we haven't yet figured out how to fix it.
... if you're not in the US, and you don't use a Mac or Windows PC.
Amazon does not sell MP3s to Australians; iTunes is Mac/Windows only. Personally, I've had most success with emusic.com, but as the querier has noted, their range is limited.
I really miss being able to walk into a store with a large classical range, have a listen, chat with knowledgeable staff, and have the chance of a serendipitous discovery. The web can theoretically provide the equivalent and more besides, but is hobbled by overly restrictive domains and copyright paranoia.
Thanks to Microsoft, the typical computer user believes that sporadic crashing is unavoidable, machines and programs must be restarted periodically if they are to maintain efficiency, and that the threat of viruses is the price paid for the convenience of email. It has come to the point that recently, when trying to explain that it was important for long-running (scientific number-crunching) code to be careful about memory management, the people I was talking to refused to believe it was possible that a program could run for over a week without slowing down. Trying to convince people that the overhead of ECC in cost and speed for computers destined for number crunching is worthwhile is hard when they believe crashes and instability are as manageable and predictable as bad weather.
Remember the days of breathless warnings about emails, which if read, would destroy your computer? And how Microsoft made the dream come true?
I should not be surprised at the gall of Microsoft to suggest that this world-wide problem, born from their neglect and short-sightedness, should be addressed with public money.
The article makes a strong case for the i3-530 and the i5-750, but
unlike the comparable AMD processors, they have no support for ECC.
If you're using a computer just for game playing and email, that's
fine. On the other hand, if you are doing anything which requires
reliability — both in terms of machine stability and
the consistency of results and data — ECC is a must. The premium
that Intel charge for what should be a standard feature prices
them out of the value computing market.
Personally, I'm very much hoping to be hale and hearty well past 2050. It would be nice to enjoy a world that is not suffering global upheaval resulting from say, anthropogenic climate change.
fixed, AKA 6x13, or more formally, -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso10646-1.
The one true programming font. No other font better manages the compromise between legibility and compactness,
and being a well-crafted bitmap font, it is crisper and clearer than ever on modern LCD screens.
X11 got it right 25 odd years ago, and now with near-full Unicode support, it's only gotten better.
Now I can't say what policies he might have violated;...
Someone on the comment thread attached to the FA gave an actual link
to the school's actual policies.
There's nothing there about bringing in an electronics project,
though I guess there was always the possibility that he was so enamored
with it that he engaged in a "public display of affection".
The Tesla c1060 processor boards sound like a very efficient way of packing in compute power, but unless they're neglecting to mention it, the 4GB of GDDR3 RAM each has on board has no error correction. Given the rates of correctable errors observed e.g. here, I could never recommend using it for computing simulations that matter. A flipped bit in a floating point number can have a disproportionate affect on the outcome of calculations that rely upon it, and short of running the whole simulation a second or third time, one couldn't be confident that such an error did not occur.
Large compute-intensive simulations can take weeks, and are used to justify engineering and business decisions that involve the disposition of large amounts of money and other resources — it is important that the computational part of the process can be relied upon.
I, too, would love to learn of some low-power fanless ECC-backed small servers.
The closest I found was an experimental MV78100 (ARM) based board that serves
LinuxArmOrg, which
according to a
recent blog entry is using ECC RAM.
A later post by the same author points out that Atom just doesn't support ECC
at all, making it a poor choice for server applications.
Slashdot Mirror
It would be hard to be more wrong — the US is on par with the UK, and both are well behind most of Western Europe. See, for example, this Wikipedia entry.
Much like BlackPignouf, it took me a month of daily practice to learn how to do the standard three cascade, and having been through that, I've found it very easy to teach others how to do it. (Probably, familiarity with a whole suite of failure modes helps.)
I persisted with juggling, and had a lot of fun with more complicated patterns, pass juggling, and so on. But it is all a case of directed persistance overcoming a complete lack of any natural ability.
Indeed you are right.
Given that almost every real number encodes an uncountable number of bits of information, I guess this isn't especially surprising in retrospect. The result though should make us suspicious of the assumption that the physical constants and properties in our physical theories can indeed take any real number value.
This, too, might be accommodated in a space of negative curvature: hyperbolic space admits a tiling by regular octagons.
Let me second this recommendation. I used to bring back supplies from Japan, or inveigle friends to bring them back from work trips to Singapore, but now they are sold locally (Australia)! I use them for everything, but particularly value the ease with which they write, and the fine width which allows the precise writing of both mathematical expressions and complicated characters.
One downside is that they can vary a little in consistency: if you have the opportunity, try a number of them and pick the best one. Also, they do not work well when used with a ruler — they quickly wear out the nib.
An alternative, not quite as good, is the Muji-brand ball-point pen. They periodically need to have the barrel tightened (simply done with the fingers), but otherwise can give a good, fine line.
There is no design for a "working commercial thorium reactors". It's all just bits and pieces of theory, and experimental reactors that only answered some of the questions.
This is not true — CANDU reactors can burn thorium in a number of fuel configurations, and they have been around for decades. That none are operating commercially on a thorium fuel cycle is, I believe, primarily due to a combination of regulation and infrastructure considerations.
Next generation thorium reactors will be great, but we already have the technology to use thorium. We just don't.
Top-level elite athletes are already genetic outliers who have also benefitted from good fortune in early training and nutrition and, typically, tens or hundreds of thousands of dollars worth of targetted training.
It's not just a matter of will power or clever training schedules. It matters not how strong be my willpower, or how dedicated my training: I will never be an olympic-class athlete.
Bring on the drugs, and the treatments. It would make elite sport more equitable, and further, the medical risks taken by those with the burning desire to compete at any cost will allow the greater majority of people to benefit from enhancements with more safety.
I am honestly curious: why should Bayesian experimental design not be used for serious work?
It's a 'good-enough' approximation to an optimal selection process.
The probability of someone clicking on option A, B or C is unknown, but is expected to be constant when averaged over the population. Given the ratio of clicks versus views on any given option, the posterior distribution of that probability can be modelled as a Beta distribution. The experimental question is then: given the current estimates, which option should be presented to maximise the utility of the test?
For simply ranking the options, the utility may be the Shannon information. In this case though, the utility also has to incorporate the expected benefit of a click-through. One could set up a utility function which is weighted between the two outcomes, possibly varying over time.
In practice though, Beta distributions with different means tend to converge to separate peaks quite quickly, so taking a possible 10% hit on the current best estimate click-through outcome seems an entirely plausible approximation. Bayesian experimental design though could also tell you when to stop testing and stick with the winner.
On a counterpoint, the engine of a recent simulation project of ours is written in C++ and hand-optimized for speed. It took much more work to get VC 2008 to produce tight code than it did for g++ 4.5, and even then, the g++ compiled code ran a bit faster. (The code has a mix of fp and integer work, and the core of it and working set are small enough to fit comfortably in L2 cache; there are some unpredictable branches in tight loops too, for good measure.)
'Hundreds of millions'? Qantas has 33000 or so employees. Even a $5000 p.a. increase in salary to every employee wouldn't make it to the hundreds of millions. On the other hand, Qantas did just double its yearly net profits, to a total of $250 million.
So, don't the other domestic Australian airlines employ people belonging to these unions? Meanwhile Qantas doubles its profits, spends 10 million dollars on a re-branding exercise, and gives a 1.5 million dollar raise to its CEO. Now this current suspension is estimated to be costing them $20 million per day.
If I had to choose a side based on the available evidence, it would not be Qantas' management.
... at least, this matches my experience at an Australian game development company. At least we didn't have to suffer this for seven years before shipping, though.
Sure enough, after shipping, the company lost 70% of their coders and they were reduced to producing shallow clones of their original (good) game.
The game industry is, basically, sick.
Imagine if we could have a reset button.
Imagine a world where, physically at least, most decisions could be undone, and the only lasting consequences of a poor choice were social.
No more: crippled by that car accident; freak fall that took out your friend's eye; paraplegic after the top rope snapped; blinded after the pressure cooker exploded.
We should, as a society, be aiming to conquer death and injury. Technology can be our saviour. Physical misery through misfortune, or through the sheer arbitrariness of our biology and environment, may one day be only a historical curiosity, but we need to dedicate real resources and real time to the project.
Do not accept the status quo! The only reason why we have so much death, is that we haven't yet figured out how to fix it.
... if you're not in the US, and you don't use a Mac or Windows PC. Amazon does not sell MP3s to Australians; iTunes is Mac/Windows only. Personally, I've had most success with emusic.com, but as the querier has noted, their range is limited.
I really miss being able to walk into a store with a large classical range, have a listen, chat with knowledgeable staff, and have the chance of a serendipitous discovery. The web can theoretically provide the equivalent and more besides, but is hobbled by overly restrictive domains and copyright paranoia.
I must have a different calendar. When is Holiday? Is that after or before Septemberary?
Those of you who do not work hard but succeed regardless also pay, while those who work hard but are subject to some misfortune also receive.
Thanks to Microsoft, the typical computer user believes that sporadic crashing is unavoidable, machines and programs must be restarted periodically if they are to maintain efficiency, and that the threat of viruses is the price paid for the convenience of email. It has come to the point that recently, when trying to explain that it was important for long-running (scientific number-crunching) code to be careful about memory management, the people I was talking to refused to believe it was possible that a program could run for over a week without slowing down. Trying to convince people that the overhead of ECC in cost and speed for computers destined for number crunching is worthwhile is hard when they believe crashes and instability are as manageable and predictable as bad weather.
Remember the days of breathless warnings about emails, which if read, would destroy your computer? And how Microsoft made the dream come true?
I should not be surprised at the gall of Microsoft to suggest that this world-wide problem, born from their neglect and short-sightedness, should be addressed with public money.
The article makes a strong case for the i3-530 and the i5-750, but unlike the comparable AMD processors, they have no support for ECC.
If you're using a computer just for game playing and email, that's fine. On the other hand, if you are doing anything which requires reliability — both in terms of machine stability and the consistency of results and data — ECC is a must. The premium that Intel charge for what should be a standard feature prices them out of the value computing market.
Personally, I'm very much hoping to be hale and hearty well past 2050. It would be nice to enjoy a world that is not suffering global upheaval resulting from say, anthropogenic climate change.
fixed, AKA 6x13, or more formally, -misc-fixed-medium-r-semicondensed--13-120-75-75-c-60-iso10646-1.
The one true programming font. No other font better manages the compromise between legibility and compactness, and being a well-crafted bitmap font, it is crisper and clearer than ever on modern LCD screens.
X11 got it right 25 odd years ago, and now with near-full Unicode support, it's only gotten better.
Now I can't say what policies he might have violated; ...
Someone on the comment thread attached to the FA gave an actual link to the school's actual policies.
There's nothing there about bringing in an electronics project, though I guess there was always the possibility that he was so enamored with it that he engaged in a "public display of affection".
The Tesla c1060 processor boards sound like a very efficient way of packing in compute power, but unless they're neglecting to mention it, the 4GB of GDDR3 RAM each has on board has no error correction. Given the rates of correctable errors observed e.g. here, I could never recommend using it for computing simulations that matter. A flipped bit in a floating point number can have a disproportionate affect on the outcome of calculations that rely upon it, and short of running the whole simulation a second or third time, one couldn't be confident that such an error did not occur.
Large compute-intensive simulations can take weeks, and are used to justify engineering and business decisions that involve the disposition of large amounts of money and other resources — it is important that the computational part of the process can be relied upon.
I, too, would love to learn of some low-power fanless ECC-backed small servers. The closest I found was an experimental MV78100 (ARM) based board that serves LinuxArmOrg, which according to a recent blog entry is using ECC RAM.
A later post by the same author points out that Atom just doesn't support ECC at all, making it a poor choice for server applications.