Except that if you look at a globe, you'll see that most of the large blue areas are connected. It might take a while, but what leaks into the gulf blue area will eventually end up in most of the other blue areas.
However the short-term outlook for the SE Asia fish farmers is very good... good enough to plan an early retirement.
Lacking mod points, I just want to say I like the voice and rhythm of your writing. It has the feel of some of the science fiction I like to read. I hope are able to do more with it than write interesting bits of stories on Slashdot.
You just explained in great detail why we should not be drilling for oil at these depths. We lack the technology and capability to control the inevitable problems like this.
Well, even if these machines aren't network ready, there's nothing to stop these valiant TSA workers from bringing their own cameras in and snapping shots of their favorite passengers.
Yes, you're right that there is only one way to represent each real number with a fixed number of bits, but it's a one-way hash... there are an infinite number of number of real numbers that can be represented by each digital representation. If you have 2 bits, 1.001, 1.0001, 1.00001 all have the same binary representation.
So if you started with 4 bits and then lower that to 2 bits (those other two are being used for something else), you used to have a space divided into 16 possible values that is now only divided into 4 possible values. This means you have to map each of the 16 possible values to one of the 4 possible values.
If I have a digital display with one digit and it shows 0 - 9 (this is like the lowest exponent you described above). Now I need it to show values up to 90, so I tape a "0" next to the digital display and rework the logic behind the display.
Where I used to be able to show 0,1,2,3,4,5,6,7,9, those discrete values can now only be shown as either 0 or 1.
I see how you're describing a variable scheme for allocating bits, but you can't go from a case where all the bits are used to represent integers to a some are the mantissa and some are the exponent and still be able to represent all of those original integers with full fidelity.
Hmm, a double has 64 bits, where do the last 12 go?
A 64 bit integer will lose precision when converted to a 64 bit double because the double has to use some of the bits for the exponent and the rest go the mantissa. If 12 bits go to the exponent, then your 64 bit integer now has to be expressed as a 52 bit integer, raised to some power. There are (I think) 2^11 distinct numbers that would have the same representation as a double.
Not in the eyes of the law. If I use a gun to shoot bullets at you but miss, there was no actual harm - only potential harm. Yet I would still most likely be prosecuted for attempted homicide.
Agreed. Asimov wrote in the forward of one of his robot books, "If knowledge poses a dangerous problem, I can't believe that ignorance is the solution." I think it applies aptly here.
Sure, some people will accidentally misuse the data, and others (hopefully fewer) will intentionally misuse the data, but for many, having that data available has a great potential for increasing the understanding we all have.
We already have printers, so the marginal cost of printing out a page for a meeting is less than 5 cents. On that page, I can write notes, draw figures, etc. and I don't have to worry about my battery running out.
For the $500 of an iPad, I can print out 10,000 pages, which probably covers all of my meetings for 5 years or more. I doubt the iPad will last that long.
I'm all about electronic storage and using computers where possible. But sometimes a piece of paper is simpler and better.
And how many meetings have you really been in? If everyone's sitting there with iPads, the one thing they won't be doing is paying attention in the meeting - they'll all be checking their email, updating Facebook, or posting on Slashdot.
Not everybody has to, or should, do things they way you do. An iPad works great for you. Congratulations. For many situations, though, it's not the best answer and people should be free to chose the options that work best.
I take it people at the school don't have to pay anything for those prints? If there aren't any charges or quotas, they'll print tremendous amounts and often leave it in the printers.
In the school lab I worked at, we had a quota system where every student was allocated a "free" 500 pages every term. Few used that much, but where a handful that given the chance would burn through cases of paper.
Another option is to bill per page, at least over a certain number of them. Do it through the student accounts office and they have to pay before they can register for the next term.
That's what lots of people are saying here. Yet the study's results indicate that if you are trying to gain wait, HFCS is a better choice than "table sugar". Either there is something wrong with the study, or there is something wrong with your understanding of the difference between the two.
If they are indeed the same, then in a well designed experiment, there should be no difference in outcome after taking into account statistical variations.
"Should be the same based on my/your/our understanding of the systems involved" is not usually equivalent to "Is the same".
Perhaps there was some evolutionary advantage to having teenagers available to patrol the early night hours.
It's been a while since I was a teenager, but I think I have it figured out. It's easier for teenagers to sneak off and have sex after the parents have already fallen asleep. That leads to babies of teenagers who stay awake longer than their parents, and thus an evolutionary advantage to the trait.
I can relate to what you're saying. I've ALWAYS been a night person. I have always had trouble getting up in the morning. However, I have no problem working from say noon to about 5 AM. In fact, I do that regularly and am consistently most productive from around 11:00pm to 4:00am or so.
People like to tell me I just need to start getting up in the morning and then I'll be tired and be "normal". It doesn't work. Even 4 years in the army didn't "fix" me... I was just tired all the time - and still awake at night.
I was happiest in grad school when all my classes were after 4:00pm and I worked on my homework all night long. The sun came up, I went to bed, and was up again in time to have lunch.
Even my current boss tried to get me to sleep normally but when I tried drugs (legal ones) I was in the office at 8:00am, but I was pretty much a zombie all day and couldn't get anything done. He has finally decided he prefers the very productive me. So I work in the evenings and night (great for working with my counterparts around the globe) and we just never schedule meetings before 11:00am. It works because I accomplish a lot, solve problems, and make him look really good to his bosses.
Even though I just got promoted, I realize overall it's career limiting because not many people are that understanding, so there aren't many places in the company I can go. But I prefer to do great things rather than show up 9-5 and be a drone.
I wish I could wake up at 6:00am and be as creative and productive as I am at night, but 30 years of studying and working have demonstrated to me that it's not doable. Sure, I can be at work at 8:00, but I'm worthless for anything but manual labor at that point.
The model merely jiggers with the numbers, but the statistics ultimately tells you the trend.
I suspect most climate models are actually very large systems of differential equations that express the relationships between various aspects of the climate. They would then be solved numerically in discrete time steps to predict the future state of the climate. The statistical analysis of observed data would then be used to set values of constants and to establish initial conditions.
Indeed, climate modeling uses statistics, but it's only one part of it and there's a lot going on that is not statistics.
And then it struck me - most of the research I had read had applied parametric statistical tests to their data - that it, the researchers made an assumption that the underlying distribution of results would fall on a normal curve.
It seems I heard a radio interview recently that discussed how many of the mathematical models used on Wall Street make the same error. That the probability of events don't actually follow a normal distribution, so high-impact anomalous events that would normally be in the very thin parts of the normal curve are actually in a thicker part of some other curve... and thus more likely to happen.
There are some things you should never be able to forge.... Do people forget basic definitions so easily?
Given a couple years with little contact with people who speak your native language, you'll actually begin to forget that very language you have lived speaking all your life. So it doesn't surprise me at all that people would forget basic definitions if they don't actually think about those definitions very often.
I figure if you can forget your native language then pretty much all bets are off for the stuff you've known for a lot less time and used a much smaller percentage of your thinking life.
So the apparent contradiction you brought up isn't really surprising.
When my father worked for a consulting company, that was actually what he was told to do. It didn't matter what the client company needed or wanted; it was all about making money for the consulting company. To make reach that end, he was to recommend whatever they weren't doing as the answer to all their problems... then proceed to rack up the billable hours. That's why he stopped working for them in pretty short order.
And that wasn't some po-dunk shop with a handful of employees - it's one of the largest engineering consulting firms you'll find.
So sadly, the apparent contradiction is real and main part of the business model - at least in some consulting companies.
If you go in and see that a company has a centralized structure, you try to sell them on decentralization. If they're decentralized, you try to sell them on centralization. If they out-source, preach in-sourcing; if they in-source, preach out-sourcing.
Oh, and in both cases, we have just the products and the consulting teams to help you achieve a synergistic paradigm shift to streamline your enterprise and facilitate a win-win situation with end-to-end empowerment for your core team.
Just another feel good appointment of an academic to a position where they can't really do anything
Yeah, I just don't see what good it would be to have someone who's known for being able to deal with large amounts of complex information and present it in easy-to-understand ways... especially an academic. I mean, just look at what horrible failure it was to have that academic Richard Feynman on the committee studying the Challenger explosion. Those ivory tower types just have no grasp on how things really work.
So, clearly you don't like Obama or Tufte... who do YOU recommend be put on this committee? And if you don't think the committee should exist, what do you suggest for better tracking and visibility of the stimulus funds?
Oh right, that whole natural process that led to the sequestration of vast amounts of CO2 over the billions of years of our planet's history? Totally forgot about that. Damn those plants and algae for making me look and feel dumb.
CO2 is risky because it has a half-life of over a century.
I ask this in earnest - what do you mean that CO2 has this half-life a century? What happens to it? I thought it was pretty stable and not likely to break down into other compounds (unless from strong UV in the upper atmosphere?) Or is it because it bleeds off into space?
I think you got it half-right. Talk to the boss. If the boss doesn't care, then stop worrying about it. If the other users complain to you, explain the situation and tell them there's nothing more you can do - that they need to talk to the boss about it.
It never pays to care about something more than the boss does. And if your boss doesn't care about the right things, then it's time to find a new boss.
I think what he means by "Pi is not predictable" is that not already knowing you're dealing with Pi, given a sequence of digits from the expansion of Pi, you could not easily predict the next digit.
For example, given the following: 7,8,4,8,8,9,1,0,1,5,9,8,6,0,3,0,9
without knowing that these are digits of pi in advance, you'd have trouble being able to predict that the next digit is 5.
Slashdot Mirror
Except that if you look at a globe, you'll see that most of the large blue areas are connected. It might take a while, but what leaks into the gulf blue area will eventually end up in most of the other blue areas.
However the short-term outlook for the SE Asia fish farmers is very good... good enough to plan an early retirement.
Lacking mod points, I just want to say I like the voice and rhythm of your writing. It has the feel of some of the science fiction I like to read. I hope are able to do more with it than write interesting bits of stories on Slashdot.
You just explained in great detail why we should not be drilling for oil at these depths. We lack the technology and capability to control the inevitable problems like this.
Well, even if these machines aren't network ready, there's nothing to stop these valiant TSA workers from bringing their own cameras in and snapping shots of their favorite passengers.
Yes, you're right that there is only one way to represent each real number with a fixed number of bits, but it's a one-way hash... there are an infinite number of number of real numbers that can be represented by each digital representation. If you have 2 bits, 1.001, 1.0001, 1.00001 all have the same binary representation.
So if you started with 4 bits and then lower that to 2 bits (those other two are being used for something else), you used to have a space divided into 16 possible values that is now only divided into 4 possible values. This means you have to map each of the 16 possible values to one of the 4 possible values.
If I have a digital display with one digit and it shows 0 - 9 (this is like the lowest exponent you described above). Now I need it to show values up to 90, so I tape a "0" next to the digital display and rework the logic behind the display.
Where I used to be able to show 0,1,2,3,4,5,6,7,9, those discrete values can now only be shown as either 0 or 1.
I see how you're describing a variable scheme for allocating bits, but you can't go from a case where all the bits are used to represent integers to a some are the mantissa and some are the exponent and still be able to represent all of those original integers with full fidelity.
Hmm, a double has 64 bits, where do the last 12 go?
A 64 bit integer will lose precision when converted to a 64 bit double because the double has to use some of the bits for the exponent and the rest go the mantissa. If 12 bits go to the exponent, then your 64 bit integer now has to be expressed as a 52 bit integer, raised to some power. There are (I think) 2^11 distinct numbers that would have the same representation as a double.
Potential harm is meaningless.
Not in the eyes of the law. If I use a gun to shoot bullets at you but miss, there was no actual harm - only potential harm. Yet I would still most likely be prosecuted for attempted homicide.
That's call Jury Nullification. It's often legal, but defense lawyers are typically not allowed to mention it as an option.
Agreed. Asimov wrote in the forward of one of his robot books, "If knowledge poses a dangerous problem, I can't believe that ignorance is the solution." I think it applies aptly here.
Sure, some people will accidentally misuse the data, and others (hopefully fewer) will intentionally misuse the data, but for many, having that data available has a great potential for increasing the understanding we all have.
We already have printers, so the marginal cost of printing out a page for a meeting is less than 5 cents. On that page, I can write notes, draw figures, etc. and I don't have to worry about my battery running out.
For the $500 of an iPad, I can print out 10,000 pages, which probably covers all of my meetings for 5 years or more. I doubt the iPad will last that long.
I'm all about electronic storage and using computers where possible. But sometimes a piece of paper is simpler and better.
And how many meetings have you really been in? If everyone's sitting there with iPads, the one thing they won't be doing is paying attention in the meeting - they'll all be checking their email, updating Facebook, or posting on Slashdot.
Not everybody has to, or should, do things they way you do. An iPad works great for you. Congratulations. For many situations, though, it's not the best answer and people should be free to chose the options that work best.
I take it people at the school don't have to pay anything for those prints? If there aren't any charges or quotas, they'll print tremendous amounts and often leave it in the printers.
In the school lab I worked at, we had a quota system where every student was allocated a "free" 500 pages every term. Few used that much, but where a handful that given the chance would burn through cases of paper.
Another option is to bill per page, at least over a certain number of them. Do it through the student accounts office and they have to pay before they can register for the next term.
It's like a mini tragedy of the commons.
That's what lots of people are saying here. Yet the study's results indicate that if you are trying to gain wait, HFCS is a better choice than "table sugar". Either there is something wrong with the study, or there is something wrong with your understanding of the difference between the two.
If they are indeed the same, then in a well designed experiment, there should be no difference in outcome after taking into account statistical variations.
"Should be the same based on my/your/our understanding of the systems involved" is not usually equivalent to "Is the same".
Perhaps there was some evolutionary advantage to having teenagers available to patrol the early night hours.
It's been a while since I was a teenager, but I think I have it figured out. It's easier for teenagers to sneak off and have sex after the parents have already fallen asleep. That leads to babies of teenagers who stay awake longer than their parents, and thus an evolutionary advantage to the trait.
I can relate to what you're saying. I've ALWAYS been a night person. I have always had trouble getting up in the morning. However, I have no problem working from say noon to about 5 AM. In fact, I do that regularly and am consistently most productive from around 11:00pm to 4:00am or so.
People like to tell me I just need to start getting up in the morning and then I'll be tired and be "normal". It doesn't work. Even 4 years in the army didn't "fix" me... I was just tired all the time - and still awake at night.
I was happiest in grad school when all my classes were after 4:00pm and I worked on my homework all night long. The sun came up, I went to bed, and was up again in time to have lunch.
Even my current boss tried to get me to sleep normally but when I tried drugs (legal ones) I was in the office at 8:00am, but I was pretty much a zombie all day and couldn't get anything done. He has finally decided he prefers the very productive me. So I work in the evenings and night (great for working with my counterparts around the globe) and we just never schedule meetings before 11:00am. It works because I accomplish a lot, solve problems, and make him look really good to his bosses.
Even though I just got promoted, I realize overall it's career limiting because not many people are that understanding, so there aren't many places in the company I can go. But I prefer to do great things rather than show up 9-5 and be a drone.
I wish I could wake up at 6:00am and be as creative and productive as I am at night, but 30 years of studying and working have demonstrated to me that it's not doable. Sure, I can be at work at 8:00, but I'm worthless for anything but manual labor at that point.
The model merely jiggers with the numbers, but the statistics ultimately tells you the trend.
I suspect most climate models are actually very large systems of differential equations that express the relationships between various aspects of the climate. They would then be solved numerically in discrete time steps to predict the future state of the climate. The statistical analysis of observed data would then be used to set values of constants and to establish initial conditions.
Indeed, climate modeling uses statistics, but it's only one part of it and there's a lot going on that is not statistics.
And then it struck me - most of the research I had read had applied parametric statistical tests to their data - that it, the researchers made an assumption that the underlying distribution of results would fall on a normal curve.
It seems I heard a radio interview recently that discussed how many of the mathematical models used on Wall Street make the same error. That the probability of events don't actually follow a normal distribution, so high-impact anomalous events that would normally be in the very thin parts of the normal curve are actually in a thicker part of some other curve... and thus more likely to happen.
I wish I could remember the interview.
There are some things you should never be able to forge.... Do people forget basic definitions so easily?
Given a couple years with little contact with people who speak your native language, you'll actually begin to forget that very language you have lived speaking all your life. So it doesn't surprise me at all that people would forget basic definitions if they don't actually think about those definitions very often.
I figure if you can forget your native language then pretty much all bets are off for the stuff you've known for a lot less time and used a much smaller percentage of your thinking life.
So the apparent contradiction you brought up isn't really surprising.
When my father worked for a consulting company, that was actually what he was told to do. It didn't matter what the client company needed or wanted; it was all about making money for the consulting company. To make reach that end, he was to recommend whatever they weren't doing as the answer to all their problems... then proceed to rack up the billable hours. That's why he stopped working for them in pretty short order.
And that wasn't some po-dunk shop with a handful of employees - it's one of the largest engineering consulting firms you'll find.
So sadly, the apparent contradiction is real and main part of the business model - at least in some consulting companies.
Learning that it won't be able to have ads for Viagra and Monostat, Twitter just got a lot more interesting to me.
Ever worked for a consulting company?
If you go in and see that a company has a centralized structure, you try to sell them on decentralization. If they're decentralized, you try to sell them on centralization. If they out-source, preach in-sourcing; if they in-source, preach out-sourcing.
Oh, and in both cases, we have just the products and the consulting teams to help you achieve a synergistic paradigm shift to streamline your enterprise and facilitate a win-win situation with end-to-end empowerment for your core team.
Just another feel good appointment of an academic to a position where they can't really do anything
Yeah, I just don't see what good it would be to have someone who's known for being able to deal with large amounts of complex information and present it in easy-to-understand ways... especially an academic. I mean, just look at what horrible failure it was to have that academic Richard Feynman on the committee studying the Challenger explosion. Those ivory tower types just have no grasp on how things really work.
So, clearly you don't like Obama or Tufte... who do YOU recommend be put on this committee? And if you don't think the committee should exist, what do you suggest for better tracking and visibility of the stimulus funds?
Oh right, that whole natural process that led to the sequestration of vast amounts of CO2 over the billions of years of our planet's history? Totally forgot about that. Damn those plants and algae for making me look and feel dumb.
CO2 is risky because it has a half-life of over a century.
I ask this in earnest - what do you mean that CO2 has this half-life a century? What happens to it? I thought it was pretty stable and not likely to break down into other compounds (unless from strong UV in the upper atmosphere?) Or is it because it bleeds off into space?
I think you got it half-right. Talk to the boss. If the boss doesn't care, then stop worrying about it. If the other users complain to you, explain the situation and tell them there's nothing more you can do - that they need to talk to the boss about it.
It never pays to care about something more than the boss does. And if your boss doesn't care about the right things, then it's time to find a new boss.
I think what he means by "Pi is not predictable" is that not already knowing you're dealing with Pi, given a sequence of digits from the expansion of Pi, you could not easily predict the next digit.
For example, given the following: 7,8,4,8,8,9,1,0,1,5,9,8,6,0,3,0,9
without knowing that these are digits of pi in advance, you'd have trouble being able to predict that the next digit is 5.