9 ms·

He who refuses to do arithmetic is doomed to talk nonsense

51 points by p4bl0 15y ago

entangld 15y ago
Arithmetic is required to be logical? I doubt it.
- artmageddon 15y ago
  Logic can be expressed mathematically, and you will be in a lot of trouble if you don't understand how it works... especially in programming.
  - tiddchristopher 15y ago
    I would argue arithmetic is founded in logic, not the reverse. It's foolish to believe mathematics are anything more natural than an invention predicated on logical notions.
  - entangld 15y ago
    I'm saying mathematics might be sufficient to express logic, but it ain't necessary. Geez all the downvotes for a statement that isn't even controversial?
    - artmageddon 15y ago
      In cases like the article posted on the discussion of energy, it's hard to make cases without numbers. I admit that I hadn't read the article when I initially posted. I should also add that I didn't downvote your comment.
- Symmetry 15y ago
  Be logical? No. Make sense? Yes. You can make all the navel-gazing syllogism you want with pure logic, but if you want to make conclusions about the real world then in many cases (like the domain under discussion) you're going to need to involve math.
  - entangld 15y ago
    So what you're saying is you could listen to someone and agree that what they're saying is logical but it doesn't make sense? I'm not being a jerk, but could you please give me an example of something logical that doesn't make sense?
    - artmageddon 15y ago
      I'll try - to me, it's the difference between qualifying and quantifying. For instance, buying a new car when the one I own is facing serious problems is logical, but it may not make sense if I don't have the ability to pay for a new one. Math would be needed here to fully express that notion.
      - entangld 15y ago
        I don't want to get metaphysical here, but this is starting to meld into something a little too abstract. An argument doesn't need to make perfect sense in order to make sense. With math you might try to make it perfect, but even mathematical projections can be guesswork. With a certain level of uncertainty, it's not about the math. And when you're looking toward the future, you're mostly using theories. Maybe the theories don't become clear until they've proven themselves, but we can still understand them, because you can have a reasonable amount of uncertainty and still make sense.
    - gloob 15y ago
      All noses are frozzle. Some wiggles are noses. Therefore, some wiggles are frozzle.
      - entangld 15y ago
        I'm not advocating fuzzy logic, just like the people who believe math is needed to make sense aren't advocating bad statistics.
    - Symmetry 15y ago
      Example: "We want to survive. Survival requires sustainability by definition. Therefore, we must have sustainability". True, but rather useless without going into what sustainability means numerically.
dmm 15y ago
A major omission of his energy analysis is the concept of Energy-returned on Energy-invested or EROEI. EROEI simply means how much energy it takes to produce a give unit of energy. Taking this into account provides a much more meaningful analysis than just looking at simple reserve and consumption numbers. For example, it doesn't matter how many billions of barrels I might have in a field if each barrel takes 2 barrels to extract(except as perhaps a chemical feedstock). It's absolutely meaningless to take theoretical oil reserves and simply dividing them by currently yearly consumption and concluding that we have hundreds of years to go. The easy oil has been extracted and all that is left is the hard stuff. Even Saudi Arabia is drilling wells that sit under a mile of water. McCarthy mentions extracting uranium from seawater for christsake! That is really crazy. Also, it doesn't matter if all of our energy problems could be solved by building a million nuclear plants if the building of them takes so much capital no one can afford it. Again EROEI for nuclear is negative until the plants actually start producing electric. And it doesn't matter if we have tons of electric if all of our transportation infrastructure requires oil products. Converting all of that infrastructure is going to take lots of capital and energy. Check out theoildrum.com for lots of good energy articles.
- viggity 15y ago
  why does it take so much capital to build a nuke plant? Sounds like a regulatory problem.
  - brudgers 15y ago
    It's more of a public health issue - analogous to building sewer systems and clean water supplies.
  - grammaton 15y ago
    I really hope this was sarcasm. Plutonium is not cheap. Uranium is not cheap. Redundant safety systems are not cheap. Pressure proof concrete vessels are not cheap. Industrial grade power turbines, also not cheap. The list goes on and on.
    - 15y ago
viggity 15y ago
this should be read by every Malthusian environmentalist.
- grammaton 15y ago
  Why? Does it contain some sort of resounding rebuttal? Because I read it, and I didn't see one....
- btilly 15y ago
  His underlying assumption is that if we raise everyone's standard of living, population will stabilize permanently. It is true that raising people's standard of living drops population growth - typically to below the replacement level. But anyone who thinks that this effect will last for billions, millions, or even thousands of years has failed to understand evolution.
  - anamax 15y ago
    > But anyone who thinks that this effect will last for billions, millions, or even thousands of years has failed to understand evolution. Oh really? Evolution selects for survival, period. (Actually, survival selects for certain kinds of evolution.) Where is it written that survival implies over-population.
    - btilly 15y ago
      Evolution selects for survival, period. No. Evolution selects for genes that manage to leave lots of descendants. This is a local optimization that is not the same as long-term survival. In fact there are lots of examples of species going down evolutionary dead ends where eventually they get wiped out. Thanks to the miracle of exponential growth, a successful trait that is possessed by a small fraction of the population will spread like a cancer until it dominates the population. In the case of humans, any trait that leads to lots of kids despite economic pressures to have few of them is going to be evolutionarily successful. And it will be successful in a surprisingly short time. To take a simple model, suppose that we have a recessive allele that is 1% of the gene pool. (So 1.99% of the population carries the recessive, but only 0.01% of the population expresses it.) Suppose that people without it just replace themselves, while people with it expressed have twice as many kids. At first the trait will be utterly invisible. But in under 100 generations it will completely dominate the population. That is only a few thousand years.
rsheridan6 15y ago
>In particular, we argue that the whole world can reach and maintain American standards of living with a population of even 15 billion. That's one doubling away from the current population. There's no particular reason to believe population will never exceed 15 billion.
- jerf 15y ago
  Actually, the claim that it will is the one not currently supported by evidence. The UN currently estimates the peak of world population will be 9.22 billion in 2075, followed by a general decline, assuming current trends [1]. Now, I'd like to think we continue to extend lifespan, but birth rates are dropping. Against all (first-order) biology, when humans become wealthier, they have fewer children, and this result is very robust now, not just the aberration of one particular culture. [1]: http://www.un.org/esa/population/publications/longrange2/WorldPop2300final.pdf http://www.un.org/esa/population/publications/longrange2/Wor... , second paragraph of the executive summary. Oh, and I'm just quoting the 9.22; I think that's an absurd number of significant digits but one should quote sources as directly as possible.
  - rsheridan6 15y ago
    to last billions of years. You yourself think that projecting the tens of millions place 64 years in the future is an absurd number of significant digits. How many significant digits do you think is appropriate in a time frame 100M years longer? Even the UN itself only purports to project population trends until 2300, which is far short of a billion years, as hubristic as it is. Also, the UN projections have some unjustified assumptions. From page 7, under section C: >Over the long run, total fertility is projected to settle at 1.85 children per woman in each country—in between the current rate for Northern America (which is just below replacement level) and the rate for Europe (which is currently well below). This figure is somewhat arbitrary In the longer run they expect the growth rate of all countries to stabilize at 0: >The eventually similar growth between more developed and less developed regions is produced by similar long-range assumptions, especially the assumptions that fertility will fall in all countries below replacement (in the medium scenario) and rebound to replacement after a period largely similar across countries of a century or so. Starting their fertility declines later, less developed regions will reach low fertility levels after more developed regions but also stay at these levels until a later date. Of course population stabilizes at a rate not too much higher than the present rate if you assume fertility rates converge around the replacement level. Garbage in, garbage out.
grammaton 15y ago
This is a bunch of standard issue cornucopian gee-whizery. I can't help but notice that nowhere in this article does he cite so much as one, single, solitary number or statistic to back up any of his claims. Here is essentially how this entire page read, for me: "Q: Will we ever run out of resources?" "A: No, of course not, that's silly!" My personal favorite: If global warming is a problem, why, then, we can just reverse it! Of course! Why didn't we think of that before?!?
- nandemo 15y ago
  > Here is essentially how this entire page read, for me: "Q: Will we ever run out of resources?" "A: No, of course not, that's silly!" Well, then you need to read better. For instance, he estimates that nuclear and solar energy will last us 5 billion years. If he's wrong by a factor of 1000 that's still 5 million years which is not too bad. And there's a whole page of references: http://www-formal.stanford.edu/jmc/progress/references.html http://www-formal.stanford.edu/jmc/progress/references.html PS: I expect that some people will assume I'm in the same "camp" of the author, but I'm just pointing out that the criticism above is unfair.
  - brudgers 15y ago
    The estimates of 5 billion (or 5 million) years ignores the potential side effects of widespread distribution of the current generation of fissionable fuels and the level of political will required to develop an infrastructure for coping with these potentials. In other words, if one is inclined to ignore issues of this magnitude, why bother mentioning nuclear at all when no less questionable assumptions about relevant factors would allow a similar projection based on solar energy alone?
    - jerf 15y ago
      Well, we also have 5 million/billion years to deal with those problems. Our descendents, if they are still using nuclear power, will be no dumber and less capable than we are of solving problems, and probably a great deal more so. If nothing else, "blast it into space" is likely to be feasible within our lifetimes. (And not just "leave it in orbit", but drop it somewhere where it won't be a problem anymore. Space is full of such places; space arguably is nothing but such places. Space is big.) Besides, what massive problems are you worried about? On that timescale, radioactive waste just dissipates. It doesn't build up over millions of years. If it could build up over that time scale, it wouldn't be "radioactive" in the first place.
rlpb 15y ago
I refuse to do arithmetic in units such as the acre-foot.

51 points by p4bl0 15y ago

entangld 15y ago

Arithmetic is required to be logical? I doubt it.

artmageddon 15y ago
Logic can be expressed mathematically, and you will be in a lot of trouble if you don't understand how it works... especially in programming.
- tiddchristopher 15y ago
  I would argue arithmetic is founded in logic, not the reverse. It's foolish to believe mathematics are anything more natural than an invention predicated on logical notions.
- entangld 15y ago
  I'm saying mathematics might be sufficient to express logic, but it ain't necessary. Geez all the downvotes for a statement that isn't even controversial?
  - artmageddon 15y ago
    In cases like the article posted on the discussion of energy, it's hard to make cases without numbers. I admit that I hadn't read the article when I initially posted. I should also add that I didn't downvote your comment.
Symmetry 15y ago
Be logical? No. Make sense? Yes. You can make all the navel-gazing syllogism you want with pure logic, but if you want to make conclusions about the real world then in many cases (like the domain under discussion) you're going to need to involve math.
- entangld 15y ago
  So what you're saying is you could listen to someone and agree that what they're saying is logical but it doesn't make sense? I'm not being a jerk, but could you please give me an example of something logical that doesn't make sense?
  - artmageddon 15y ago
    I'll try - to me, it's the difference between qualifying and quantifying. For instance, buying a new car when the one I own is facing serious problems is logical, but it may not make sense if I don't have the ability to pay for a new one. Math would be needed here to fully express that notion.
    - entangld 15y ago
      I don't want to get metaphysical here, but this is starting to meld into something a little too abstract. An argument doesn't need to make perfect sense in order to make sense. With math you might try to make it perfect, but even mathematical projections can be guesswork. With a certain level of uncertainty, it's not about the math. And when you're looking toward the future, you're mostly using theories. Maybe the theories don't become clear until they've proven themselves, but we can still understand them, because you can have a reasonable amount of uncertainty and still make sense.
  - gloob 15y ago
    All noses are frozzle. Some wiggles are noses. Therefore, some wiggles are frozzle.
    - entangld 15y ago
      I'm not advocating fuzzy logic, just like the people who believe math is needed to make sense aren't advocating bad statistics.
  - Symmetry 15y ago
    Example: "We want to survive. Survival requires sustainability by definition. Therefore, we must have sustainability". True, but rather useless without going into what sustainability means numerically.

dmm 15y ago

A major omission of his energy analysis is the concept of Energy-returned on Energy-invested or EROEI. EROEI simply means how much energy it takes to produce a give unit of energy. Taking this into account provides a much more meaningful analysis than just looking at simple reserve and consumption numbers. For example, it doesn't matter how many billions of barrels I might have in a field if each barrel takes 2 barrels to extract(except as perhaps a chemical feedstock). It's absolutely meaningless to take theoretical oil reserves and simply dividing them by currently yearly consumption and concluding that we have hundreds of years to go. The easy oil has been extracted and all that is left is the hard stuff. Even Saudi Arabia is drilling wells that sit under a mile of water. McCarthy mentions extracting uranium from seawater for christsake! That is really crazy. Also, it doesn't matter if all of our energy problems could be solved by building a million nuclear plants if the building of them takes so much capital no one can afford it. Again EROEI for nuclear is negative until the plants actually start producing electric. And it doesn't matter if we have tons of electric if all of our transportation infrastructure requires oil products. Converting all of that infrastructure is going to take lots of capital and energy. Check out theoildrum.com for lots of good energy articles.

viggity 15y ago
why does it take so much capital to build a nuke plant? Sounds like a regulatory problem.
- brudgers 15y ago
  It's more of a public health issue - analogous to building sewer systems and clean water supplies.
- grammaton 15y ago
  I really hope this was sarcasm. Plutonium is not cheap. Uranium is not cheap. Redundant safety systems are not cheap. Pressure proof concrete vessels are not cheap. Industrial grade power turbines, also not cheap. The list goes on and on.
  - 15y ago

viggity 15y ago

this should be read by every Malthusian environmentalist.

grammaton 15y ago
Why? Does it contain some sort of resounding rebuttal? Because I read it, and I didn't see one....
btilly 15y ago
His underlying assumption is that if we raise everyone's standard of living, population will stabilize permanently. It is true that raising people's standard of living drops population growth - typically to below the replacement level. But anyone who thinks that this effect will last for billions, millions, or even thousands of years has failed to understand evolution.
- anamax 15y ago
  > But anyone who thinks that this effect will last for billions, millions, or even thousands of years has failed to understand evolution. Oh really? Evolution selects for survival, period. (Actually, survival selects for certain kinds of evolution.) Where is it written that survival implies over-population.
  - btilly 15y ago
    Evolution selects for survival, period. No. Evolution selects for genes that manage to leave lots of descendants. This is a local optimization that is not the same as long-term survival. In fact there are lots of examples of species going down evolutionary dead ends where eventually they get wiped out. Thanks to the miracle of exponential growth, a successful trait that is possessed by a small fraction of the population will spread like a cancer until it dominates the population. In the case of humans, any trait that leads to lots of kids despite economic pressures to have few of them is going to be evolutionarily successful. And it will be successful in a surprisingly short time. To take a simple model, suppose that we have a recessive allele that is 1% of the gene pool. (So 1.99% of the population carries the recessive, but only 0.01% of the population expresses it.) Suppose that people without it just replace themselves, while people with it expressed have twice as many kids. At first the trait will be utterly invisible. But in under 100 generations it will completely dominate the population. That is only a few thousand years.

rsheridan6 15y ago

>In particular, we argue that the whole world can reach and maintain American standards of living with a population of even 15 billion. That's one doubling away from the current population. There's no particular reason to believe population will never exceed 15 billion.

jerf 15y ago
Actually, the claim that it will is the one not currently supported by evidence. The UN currently estimates the peak of world population will be 9.22 billion in 2075, followed by a general decline, assuming current trends [1]. Now, I'd like to think we continue to extend lifespan, but birth rates are dropping. Against all (first-order) biology, when humans become wealthier, they have fewer children, and this result is very robust now, not just the aberration of one particular culture. [1]: http://www.un.org/esa/population/publications/longrange2/WorldPop2300final.pdf http://www.un.org/esa/population/publications/longrange2/Wor... , second paragraph of the executive summary. Oh, and I'm just quoting the 9.22; I think that's an absurd number of significant digits but one should quote sources as directly as possible.
- rsheridan6 15y ago
  to last billions of years. You yourself think that projecting the tens of millions place 64 years in the future is an absurd number of significant digits. How many significant digits do you think is appropriate in a time frame 100M years longer? Even the UN itself only purports to project population trends until 2300, which is far short of a billion years, as hubristic as it is. Also, the UN projections have some unjustified assumptions. From page 7, under section C: >Over the long run, total fertility is projected to settle at 1.85 children per woman in each country—in between the current rate for Northern America (which is just below replacement level) and the rate for Europe (which is currently well below). This figure is somewhat arbitrary In the longer run they expect the growth rate of all countries to stabilize at 0: >The eventually similar growth between more developed and less developed regions is produced by similar long-range assumptions, especially the assumptions that fertility will fall in all countries below replacement (in the medium scenario) and rebound to replacement after a period largely similar across countries of a century or so. Starting their fertility declines later, less developed regions will reach low fertility levels after more developed regions but also stay at these levels until a later date. Of course population stabilizes at a rate not too much higher than the present rate if you assume fertility rates converge around the replacement level. Garbage in, garbage out.

grammaton 15y ago

This is a bunch of standard issue cornucopian gee-whizery. I can't help but notice that nowhere in this article does he cite so much as one, single, solitary number or statistic to back up any of his claims. Here is essentially how this entire page read, for me: "Q: Will we ever run out of resources?" "A: No, of course not, that's silly!" My personal favorite: If global warming is a problem, why, then, we can just reverse it! Of course! Why didn't we think of that before?!?

nandemo 15y ago
> Here is essentially how this entire page read, for me: "Q: Will we ever run out of resources?" "A: No, of course not, that's silly!" Well, then you need to read better. For instance, he estimates that nuclear and solar energy will last us 5 billion years. If he's wrong by a factor of 1000 that's still 5 million years which is not too bad. And there's a whole page of references: http://www-formal.stanford.edu/jmc/progress/references.html http://www-formal.stanford.edu/jmc/progress/references.html PS: I expect that some people will assume I'm in the same "camp" of the author, but I'm just pointing out that the criticism above is unfair.
- brudgers 15y ago
  The estimates of 5 billion (or 5 million) years ignores the potential side effects of widespread distribution of the current generation of fissionable fuels and the level of political will required to develop an infrastructure for coping with these potentials. In other words, if one is inclined to ignore issues of this magnitude, why bother mentioning nuclear at all when no less questionable assumptions about relevant factors would allow a similar projection based on solar energy alone?
  - jerf 15y ago
    Well, we also have 5 million/billion years to deal with those problems. Our descendents, if they are still using nuclear power, will be no dumber and less capable than we are of solving problems, and probably a great deal more so. If nothing else, "blast it into space" is likely to be feasible within our lifetimes. (And not just "leave it in orbit", but drop it somewhere where it won't be a problem anymore. Space is full of such places; space arguably is nothing but such places. Space is big.) Besides, what massive problems are you worried about? On that timescale, radioactive waste just dissipates. It doesn't build up over millions of years. If it could build up over that time scale, it wouldn't be "radioactive" in the first place.

rlpb 15y ago

I refuse to do arithmetic in units such as the acre-foot.