4 ms·

That's the nail on the head. Once you start getting into these sorts of assumptions about trends and correlations, you really have to have a good understanding

by netcan 11y ago

That's the nail on the head. Once you start getting into these sorts of assumptions about trends and correlations, you really have to have a good understanding of all these inputs. They are all suspect to different degrees across different time horizons and scenarios. What you really have when you put together these predictions is "Given A,B,C & D, this Y is likely to be whatever." That only works when you understand A,B,C & D quite well and keep them in your head. It's also bait for our biases and grand worldviews. I had this massive argument 10 years ago with a politically active friend on the left. He was/is a sort of modern socialist-environmetalist. He brought up exactly this energy-gdp correlation and was extrapolating a log way into the future. It also played into his general sense that GDP growth is an unbecoming central goal for a society and that economic progress is at odds with the environment. It's not a nonsensical perpective. That correlation is relevant. But, you need to get into the subtleties. Even the concept of "economic growth" is an abstract thing. GDP as a measure of economic output is becoming less and less useful metric. Wikipedia consumption contributes nothing to GDP. Some kid torrenting game of thrones contributes nothing to GDP. These things have value, but monetary exchanges are not good proxies for them. Even within the "energy intensive manufacturing" paradigm, an iphone's utility-to-energy cost ratio is a funky thing to try and fit in to the GDP concept which ultimately uses money to measure value. A 2015 iphone is worth more than a 2010 iphone. You can see this if you try to sell one and if you could go back to 2010 in time with a box of 2015 iphones you would see it as well. But beyond that, trying to do price adjustments (which are the basis of "keeping interest rate constant") on a rapidly developing technological product puts you in the weeds. This complexity is exactly why these correlations and trends are interesting They cut through the messiness. I think we should still do this. But… you need to treat these in the same way you'd treat Kurzweil's theories. Interesting, but built on a lot of unknowns. Too many to make you too confident.

DanielBMarkham 11y ago
Yep. Kurzweil's theories are very similar. Also the Drake Equation. Great for late-night BS sessions, and maybe a fine thing to put some faith in -- but it ain't science or logic. "...I had this massive argument 10 years ago with a politically active friend..." Reminds me of a similar story. Worked with a guy many years ago who was a double major, math and physics, I think. Brilliant guy. At the time, he assured me that now was the time to get in the market and bet on oil prices going higher. Peak oil was here! Shortages and chaos were upon us. I pointed out that peak oil had been called many times over the years. It was doubtful we were on that exact spot. In fact, some economists have even said that anything you can dig out of the ground is bound to go down in prices as extraction technology continues to mature. He countered by showing me a study from the U.S. Defense Department (I believe), it showed that yes, indeed, oil was going nowhere but up. I pointed out various examples of this type of thinking failing in the past. He pointed out that things can't continue on forever. Draw a trend line. There's no way that line can continue. Does this discussion sound familiar? We agreed to disagree, and I promised him that I would wait a few years and email him. We would revisit this discussion. Several years passed. Oil was not through the roof. In fact, IIRC, it was about in the same spot as it was before. So I emailed him. I asked him how he felt about his earlier assessment. I did not get a reply. Since then, of course, we have the advent of fracking, and the clear possibility that the U.S. is going to end up as a petroleum exporter, at least for a while. Faith is a fine thing. Heck, I'm a singularian, although I think we're looking at 300-500 years. But it's very important to be cognizant of those things you take on faith. Otherwise you end up having these long discussions in which your main defense is just repeating something along the lines of "science!" or "math!" over and over again.
- netcan 11y ago
  An interesting thing about Kurzweil's timeline is the 2045 singularity awakening date. It's conveniently scheduled around the end of his life, assuming medical technology extends life expectancy at his age to around 100 years. That said, I think what Kurzweil does is interesting and valuable. In fact, being so stridently futurist and speculative makes it a lot more honest than a typical "think tank" economic scenario predictions. You go into it with a much more assumption challenging type of good skepticism and engage with the unknowns. What makes you think 300-500 years? If we're talking 300-500 years, I'm lost. Compare 2015 to 1515. Consider accelerating rate of change, I think 300-500 years takes us outside of any kind of timeline we can map historically.
  - DanielBMarkham 11y ago
    I believe hard AI is going to be a lot more problematic to map out than we currently might think. I believe we're in for a push-back against overly-intrusive computer monitoring, which should slow down AI development. I also believe that "intelligent", "kind", "moral", and "useful" may also be much more mutually orthogonal than currently believed. Finally, mankind really needs to work through the period where mid-level AI could either exterminate us or love us into extinction. That's going to take some time to work out. But it's all just guesses, of course.
- alphydan 11y ago
  If you spoke to your friend 10 - 20 years ago, then he was right. Perhaps he imagined faster and more catastrophic dynamics, but inflation-adjusted Oil has been going up since 1998 [0]. If you had taken his advice and invested in oil futures in 2009, you could have become rather wealthy :) Of course, there have been impressive price fluctuation (One happening right now). But it's kind of expected in a highly complex system where technology, politics and financial crises interact in rather unpredictable ways. On average, I still think the trend is UP (and the "fracking miracle" will be a short lived spike). Peak Oil will happen (but predicting the date or dynamics inside a 20 year range is just as hard as predicting the timing and depth of a financial crisis). [0] http://inflationdata.com/articles/charts/inflation-adjusted-oil-prices-chart/ http://inflationdata.com/articles/charts/inflation-adjusted-...
danbruc 11y ago
What then is a faithful measure of economic growth or the size of an economy? This should provide further insight into possible limits.
- netcan 11y ago
  There just isn't one. Trying to measure the size of the economy "faithfully" is not possible. You can create abstractions like GDP that are useful where they are useful, but you can't treat them as solid concepts. When the "IQ" debate was hot and contentious, people often ended up with a tautological definition that measured IQ is whatever IQ measures. This is because "intelligence" is not a solid concept and the test is not a direct measure of this fuzzy concept anyway. That doesn't mean IQ is meaningless. It has predictive power. It can serve as a goal (unless you game it and make it useless). GDP is a measure of transactions, ultimately. That tends to correlate with the size of an economy and can be used for comparison, if everything is structurally similar. You can compare France to Sweden. If you are trying to compare the economic output of Monaco to Zimbabwe, GDP is not as good. If you are trying to compare the economy of today to the economy of 1836 or 2065, you're also in trouble. If you want to compare the economy of an alien civilization to ours, GDP won't work. Referencing GDP (or anything else, this is just an inherently hard problem) in these contexts is trying to squeeze some useful information out of a very limited and buggy tool. You need to understand the subtleties of it. What it does well and poorly. Most importantly, accept that our tools and therefore knowledge is limited.
  - dougabug 11y ago
    This would argue that economic growth is to a significant degree subjective, which would be consistent with the basic thrust of the original article. Economists once predicted that energy would be basically free, which turned out to be badly off base. Whether life can be made continuously better may be an open question, but the idea of essentially all major forms of scarcity disappearing due to never ending economic growth is cast into severe doubt.
    - chipsy 11y ago
Joeri 11y ago
The thing that makes GDP unsuitable in my view is that it measures destructive activity as being equal in value to constructive activity. You tear down a house, build an identical one in its place, and while nothing of value has been created, the GDP has received a sizable boost. Yes, people got paid for doing those things, but it would have been better had they been paid to produce something of value instead.
- nmrm2 11y ago
  > nothing of value has been created That's debatable, because no one tears down a house and rebuilds an identical one. * Was the old house livable? If not, then you haven't really wasted anything. * Is the new house up to more rigid environmental standards? * Does the new house take up more (less) space? * Is the new house wired/piped in a different way? etc.
  - netcan 11y ago
    It's not an invalid example. It shows that GDP has some weirdness. It's just not a scenario likely to be common enough to cause problems. But, it does demonstrate the leaky abstraction. There are also scenarios that break GDP but not homo economicus logic. What happens when a woman starts working and paying for childcare? Do you subtract the childcare she is no longer directly providing as negative GDP? Do you just add on this new service as growth, even though the kids are not getting and more hours of "childcare?" This is a real problem with data pre 1960s and post 1980s. GDP can't measure those. That's the economic equivalent of bulldozing a house and building a new one. There's a surplus, economic or otherwise but there is an unmeasured destruction too, parenting hours are reduced. We only count paid work so when one increases at the expense of another, the GDP abstraction leaks. What happens if important new products and services are free, like Wikipedia, Khan Academy, etc? What happens when you can buy a £500 android phone for £200? Do you do an inflation adjustment that counts the value of all £200 iphones produced @ £500? The real world mixes these up in ways that GDP can't really handle.
    - parasubvert 11y ago
roymurdock 11y ago
Here's a good (but long) resource for anyone who wants to understand how and why we have come to use the conventional measurements of economic growth that we do: http://www.oecd.org/std/41746710.pdf http://www.oecd.org/std/41746710.pdf Giovannini (former OECD Chief Statistician) has this to say about the future of economic accounting: Looking to the future, it is possible to identify certain trends in the demand for economic statistics that will presumably strengthen in the years to come. In the first place, demand will focus more and more on the services sector (which already accounts for over 70% of GDP in many developed countries), with new forms of production playing a more important part. The measurement of “intangible” activities will take on increased importance and this should bring about greater integration of economic statistics with social statistics: concepts such as those of human and social capital are already generating interest and becoming the subject of quantification, albeit still in embryonic form.
rayiner 11y ago
Does Wikipedia consumption have value if you don't use it to go and create something that has impact on GDP?
- andybak 11y ago
  This question goes very deep. If you pursue it then at some point you end up asking very deep questions about value, ethics and meaning. Unless you want to set a few boundaries to keep the debate about more prosaic matters then be prepared for a long night.
  - HiYaBarbie 11y ago
    The question is actually based on the false premise that value can be objectively measured/quantified. In fact, value is purely subjective. In other words, the value of 'Wikipedia consumption' is only what each individual places on it. For example, someone may find Wikipedia valuable in pursuing his quest for greater knowledge, whereas someone else is completely uninterested in 'consuming' it (and doesn't see Wikipedia as valuable). In a nutshell, the value of X is its subjectively perceived utility as a means towards an end. See http://mises.org http://mises.org for more.
- roymurdock 11y ago
  It adds to human capital, which we can't quantify as of now. I'm not sure if we'll ever find a good proxy for how much information is worth as it is an extremely heterogeneous, intangible input that goes through a complex, unstructured, and not-fully-understood process (the human brain) in order to create value in the world that we can measure using money as a proxy - a company's revenue, for instance.

by netcan 11y ago

DanielBMarkham 11y ago

Yep. Kurzweil's theories are very similar. Also the Drake Equation. Great for late-night BS sessions, and maybe a fine thing to put some faith in -- but it ain't science or logic. "...I had this massive argument 10 years ago with a politically active friend..." Reminds me of a similar story. Worked with a guy many years ago who was a double major, math and physics, I think. Brilliant guy. At the time, he assured me that now was the time to get in the market and bet on oil prices going higher. Peak oil was here! Shortages and chaos were upon us. I pointed out that peak oil had been called many times over the years. It was doubtful we were on that exact spot. In fact, some economists have even said that anything you can dig out of the ground is bound to go down in prices as extraction technology continues to mature. He countered by showing me a study from the U.S. Defense Department (I believe), it showed that yes, indeed, oil was going nowhere but up. I pointed out various examples of this type of thinking failing in the past. He pointed out that things can't continue on forever. Draw a trend line. There's no way that line can continue. Does this discussion sound familiar? We agreed to disagree, and I promised him that I would wait a few years and email him. We would revisit this discussion. Several years passed. Oil was not through the roof. In fact, IIRC, it was about in the same spot as it was before. So I emailed him. I asked him how he felt about his earlier assessment. I did not get a reply. Since then, of course, we have the advent of fracking, and the clear possibility that the U.S. is going to end up as a petroleum exporter, at least for a while. Faith is a fine thing. Heck, I'm a singularian, although I think we're looking at 300-500 years. But it's very important to be cognizant of those things you take on faith. Otherwise you end up having these long discussions in which your main defense is just repeating something along the lines of "science!" or "math!" over and over again.

netcan 11y ago
An interesting thing about Kurzweil's timeline is the 2045 singularity awakening date. It's conveniently scheduled around the end of his life, assuming medical technology extends life expectancy at his age to around 100 years. That said, I think what Kurzweil does is interesting and valuable. In fact, being so stridently futurist and speculative makes it a lot more honest than a typical "think tank" economic scenario predictions. You go into it with a much more assumption challenging type of good skepticism and engage with the unknowns. What makes you think 300-500 years? If we're talking 300-500 years, I'm lost. Compare 2015 to 1515. Consider accelerating rate of change, I think 300-500 years takes us outside of any kind of timeline we can map historically.
- DanielBMarkham 11y ago
  I believe hard AI is going to be a lot more problematic to map out than we currently might think. I believe we're in for a push-back against overly-intrusive computer monitoring, which should slow down AI development. I also believe that "intelligent", "kind", "moral", and "useful" may also be much more mutually orthogonal than currently believed. Finally, mankind really needs to work through the period where mid-level AI could either exterminate us or love us into extinction. That's going to take some time to work out. But it's all just guesses, of course.
alphydan 11y ago
If you spoke to your friend 10 - 20 years ago, then he was right. Perhaps he imagined faster and more catastrophic dynamics, but inflation-adjusted Oil has been going up since 1998 [0]. If you had taken his advice and invested in oil futures in 2009, you could have become rather wealthy :) Of course, there have been impressive price fluctuation (One happening right now). But it's kind of expected in a highly complex system where technology, politics and financial crises interact in rather unpredictable ways. On average, I still think the trend is UP (and the "fracking miracle" will be a short lived spike). Peak Oil will happen (but predicting the date or dynamics inside a 20 year range is just as hard as predicting the timing and depth of a financial crisis). [0] http://inflationdata.com/articles/charts/inflation-adjusted-oil-prices-chart/ http://inflationdata.com/articles/charts/inflation-adjusted-...

danbruc 11y ago

What then is a faithful measure of economic growth or the size of an economy? This should provide further insight into possible limits.

netcan 11y ago
There just isn't one. Trying to measure the size of the economy "faithfully" is not possible. You can create abstractions like GDP that are useful where they are useful, but you can't treat them as solid concepts. When the "IQ" debate was hot and contentious, people often ended up with a tautological definition that measured IQ is whatever IQ measures. This is because "intelligence" is not a solid concept and the test is not a direct measure of this fuzzy concept anyway. That doesn't mean IQ is meaningless. It has predictive power. It can serve as a goal (unless you game it and make it useless). GDP is a measure of transactions, ultimately. That tends to correlate with the size of an economy and can be used for comparison, if everything is structurally similar. You can compare France to Sweden. If you are trying to compare the economic output of Monaco to Zimbabwe, GDP is not as good. If you are trying to compare the economy of today to the economy of 1836 or 2065, you're also in trouble. If you want to compare the economy of an alien civilization to ours, GDP won't work. Referencing GDP (or anything else, this is just an inherently hard problem) in these contexts is trying to squeeze some useful information out of a very limited and buggy tool. You need to understand the subtleties of it. What it does well and poorly. Most importantly, accept that our tools and therefore knowledge is limited.
- dougabug 11y ago
  This would argue that economic growth is to a significant degree subjective, which would be consistent with the basic thrust of the original article. Economists once predicted that energy would be basically free, which turned out to be badly off base. Whether life can be made continuously better may be an open question, but the idea of essentially all major forms of scarcity disappearing due to never ending economic growth is cast into severe doubt.
  - chipsy 11y ago

Joeri 11y ago

The thing that makes GDP unsuitable in my view is that it measures destructive activity as being equal in value to constructive activity. You tear down a house, build an identical one in its place, and while nothing of value has been created, the GDP has received a sizable boost. Yes, people got paid for doing those things, but it would have been better had they been paid to produce something of value instead.

nmrm2 11y ago
> nothing of value has been created That's debatable, because no one tears down a house and rebuilds an identical one. * Was the old house livable? If not, then you haven't really wasted anything. * Is the new house up to more rigid environmental standards? * Does the new house take up more (less) space? * Is the new house wired/piped in a different way? etc.
- netcan 11y ago
  It's not an invalid example. It shows that GDP has some weirdness. It's just not a scenario likely to be common enough to cause problems. But, it does demonstrate the leaky abstraction. There are also scenarios that break GDP but not homo economicus logic. What happens when a woman starts working and paying for childcare? Do you subtract the childcare she is no longer directly providing as negative GDP? Do you just add on this new service as growth, even though the kids are not getting and more hours of "childcare?" This is a real problem with data pre 1960s and post 1980s. GDP can't measure those. That's the economic equivalent of bulldozing a house and building a new one. There's a surplus, economic or otherwise but there is an unmeasured destruction too, parenting hours are reduced. We only count paid work so when one increases at the expense of another, the GDP abstraction leaks. What happens if important new products and services are free, like Wikipedia, Khan Academy, etc? What happens when you can buy a £500 android phone for £200? Do you do an inflation adjustment that counts the value of all £200 iphones produced @ £500? The real world mixes these up in ways that GDP can't really handle.
  - parasubvert 11y ago

roymurdock 11y ago

Here's a good (but long) resource for anyone who wants to understand how and why we have come to use the conventional measurements of economic growth that we do: http://www.oecd.org/std/41746710.pdf http://www.oecd.org/std/41746710.pdf Giovannini (former OECD Chief Statistician) has this to say about the future of economic accounting: Looking to the future, it is possible to identify certain trends in the demand for economic statistics that will presumably strengthen in the years to come. In the first place, demand will focus more and more on the services sector (which already accounts for over 70% of GDP in many developed countries), with new forms of production playing a more important part. The measurement of “intangible” activities will take on increased importance and this should bring about greater integration of economic statistics with social statistics: concepts such as those of human and social capital are already generating interest and becoming the subject of quantification, albeit still in embryonic form.

rayiner 11y ago

Does Wikipedia consumption have value if you don't use it to go and create something that has impact on GDP?

andybak 11y ago
This question goes very deep. If you pursue it then at some point you end up asking very deep questions about value, ethics and meaning. Unless you want to set a few boundaries to keep the debate about more prosaic matters then be prepared for a long night.
- HiYaBarbie 11y ago
  The question is actually based on the false premise that value can be objectively measured/quantified. In fact, value is purely subjective. In other words, the value of 'Wikipedia consumption' is only what each individual places on it. For example, someone may find Wikipedia valuable in pursuing his quest for greater knowledge, whereas someone else is completely uninterested in 'consuming' it (and doesn't see Wikipedia as valuable). In a nutshell, the value of X is its subjectively perceived utility as a means towards an end. See http://mises.org http://mises.org for more.
roymurdock 11y ago
It adds to human capital, which we can't quantify as of now. I'm not sure if we'll ever find a good proxy for how much information is worth as it is an extremely heterogeneous, intangible input that goes through a complex, unstructured, and not-fully-understood process (the human brain) in order to create value in the world that we can measure using money as a proxy - a company's revenue, for instance.