Originally Posted by Copperfield

Poker players predict cards all the time. Sometimes they even get lucky.

The odds of calling a hand and then getting it? No greater than the odds of getting the hand in the first place. The call (a completely random guess) has zero influence over the deal (a physical event). Welcome to game theory.

Thank you.

Originally Posted by Mikebert

In the storm situation the key observation is the occurence of a special storm of probability p within a specific window of time chosen in advance during which there are N storms. p is such that the probability that any one of the N storms would be special (Np) is less than 0.05.

It is assumed that any window of time will contain a random sampling of storms, and so the window chosen will have such a random sample.

And if you get one special storm within a window that contains dozens and dozens of low energy storms such that the mean wind speed is actually significantly less than the control you still conclude that warming has increased storm strengths?

What if the mean wind speed during the window is two standard deviations above the control but your special storm doesn't occur? Do you conclude that warming has had no effect on storm strength?

Your search for singular events, no matter how rare, is a useless manner for measuring the treatment effect of warming because it allows for such ridiculous results.

Originally Posted by Vandal-72

And if you get one special storm within a window that contains dozens and dozens of low energy storms such that the mean wind speed is actually significantly less than the control you still conclude that warming has increased storm strengths?

Of course not! I would conclude that the many storms in that window come from a population that is different from the control population. It is different in two ways, (1) the mean value is lower; (2) the standard deviation is higher.

The hypothesis was: global warming make storms stronger (i.e. increase the mean), or more extreme storms more likely (i.e. increase the standard deviation) or both.

The null hypothesis is that global warming has no effect (i.e. neither the mean nor the standard deviation has changed).

The low-probability outcome p = NP < 0.05, (where P is the 4-sigma probability, and N is the number of storms in the window, i.e. opportunities for the 4-sigma event to happen) shows the population is different, the null hypothesis is rejected. That is, either the mean is higher OR the standard deviation is larger (or both).

The statistically significant lower mean shows that the mean is not increased. Therefore the standard deviation is higher (i.e. extreme storms are more likely, but average storms are NOT stronger). In fact, average storm strength is reduced.

So this data would support a conclusion that global warming makes storms weaker on average AND extreme storms more likely.

What if the mean wind speed during the window is two standard deviations above the control but your special storm doesn't occur? Do you conclude that warming has had no effect on storm strength?

If a t-test gave a sufficiently low probability, I would conclude that the many storms in that window came from a population that is different from the control population; the mean is higher.

This would support the hypothesis that global warming makes storms stronger. It says nothing about the standard deviation.

Originally Posted by Vandal-72

What if everyone starts predicting a royal flush with every hand? Do you conclude collusion when the two inevitably match up?

Then the prediction would be wrong many many times. This is the expected outcome. A successful prediction is necessarily made only once and then comes true. Otherwise it's a case of a broken clock being right twice aa day.

You are concluding collusion based on imagined evidence.

I would strongly suspect cheating and so would demand an investigation. You on the other hand, asserted that there was no difference between getting a successfully predicted royal flush and getting an unpredicted royal flush. In this you are claiming you would have no reason to be suspicious, which I found hard to believe.

I may may suspect collusion...

So now you've changed your mind.

but you have not presented any actual evidence so such.

Well your example of average storm strength that is signficantly higher is likewise not actual evidence. All the t-test does is establish that an unlikley coincidence has occurred: stronger storms happened to occur just when temperatures have risen. That is, stronger storms are correlated with rising temperature. But as you know, correlation does not necesarily mean causation.

In the poker example there is the coincidence of the predicition of the royal flush, and then it happening. Again correlation is not causation. In the poker case we can investigate, and perhaps find a cause.

Originally Posted by Vandal-72

Fine. Your treatment sample is not part of that population by definition, however.

The single storm is not a t-test. As you said a t-test asks are the means different. The null hypothesis asserts that the means are not different. It says nothing about the variance of the treatment sample. It can be different and the null hypothsis can hold.

The null hypothesis being tested is weaker than the t-test null hypothesis. The storm hypothesis allows for global warming that made extreme storms more likley but did not increase the mean. In such a case a t-test on all the storms in the window would not reject the null hypothesis that the mean is higher. It would still possible that standard deviation is larger.

The null hypothesis for the storms is rejected unless the treatment population mean AND standard deviation are the same. In this case you can use the control means and standard deviation to assess the probability of the special strom happening. And if it is low then you can reject the null hypothesis that both mean and standard deviation are the same.

There may be a test that you like better that tests a null hypothesis about standard deviations being the same. In that case you could run that test on the storms in the window and it would reject the null hypothesis that the treatment variance is the same.

But if you repeat the analysis without the special storm, and you now find a mean and standard deviation that are very similar to the control, then it will have been this single storm that made all the difference. You can't just throw out the outlier like a bad assay because the storm was undoubtly real. If this happens you can still conclude that soemthing has changed.

Storm strength are not going to be normally distribution. Transformed strengths could be normally distributed. For example perhaps the log of strengths is normally distributed. In this case one would say the probability distribution of storm strengths is lognormal. I specifically talked about the four sigma being defined with transformed strengths (so that they are normally distrubuted).

If a lognormal distrubtion is valid yu would take the log of a large set of control data and calculate the mean and standard deviation from that. A four sigma storm would be 4 standard deviations out from that mean. My use of the term "sigma" implied a control sample large enough that the t-distribution is not materially different from the normal distribution as so the control sample standard deviation s is very close to sigma.

You take the inverse log of this value and that storm strength would be the four-sigma storm strength

The t-tests you talked about would be done on the transformed data.

A drought is not so much an event as the absence of some normal events. To say that California is just not getting the rain that it usually does understates the case; San Francisco so far has the driest winter since records were first kept in 1849. California gets its rain almost entirely from winter storms. The usual westerly storm track has been shunted away.

If random, this is at least a one-in-150 year event.

http://www.wunderground.com/blog/wea...l?entrynum=233

Originally Posted by Mikebert

Then the prediction would be wrong many many times. This is the expected outcome. A successful prediction is necessarily made only once and then comes true. Otherwise it's a case of a broken clock being right twice aa day.

So when it is right, do you draw your gun and shoot the dealer?

I would strongly suspect cheating and so would demand an investigation. You on the other hand, asserted that there was no difference between getting a successfully predicted royal flush and getting an unpredicted royal flush. In this you are claiming you would have no reason to be suspicious, which I found hard to believe.

Nope. I stated that there is no statistical difference between being dealt the hand without claiming you will get it versus claiming you will. Stating you will get a certain hand has absolutely no effect on what hand you will be dealt.

You are putting words into my mouth.

So now you've changed your mind.

No. You have simply failed to see the difference between what you asked me and what you are claiming I said.

Well your example of average storm strength that is signficantly higher is likewise not actual evidence. All the t-test does is establish that an unlikley coincidence has occurred: stronger storms happened to occur just when temperatures have risen.

And your ridiculous test can't even do that.

BTW: Shall we address your failure to understand what a t-test actually does?

That is, stronger storms are correlated with rising temperature. But as you know, correlation does not necesarily mean causation.

Proper controls can differentiate between the two.

In the poker example there is the coincidence of the predicition of the royal flush, and then it happening. Again correlation is not causation. In the poker case we can investigate, and perhaps find a cause.

But, you claimed that the coincidence was evidence of wrong doing! Not the possibility of wrong doing, but positive evidence, like you are trying to claim the coincidence of a single large storm is supposed to be evidence of a relationship with global warming. Just as you are now rightly admitting that the poker event isn't really evidence, neither is your hypothetical storm event.

Originally Posted by Mikebert

The single storm is not a t-test.

Moving that goalpost? You linked to the t-test Wikipedia entry. You referenced the two sample t-test equation.

As you said a t-test asks are the means different. The null hypothesis asserts that the means are not different. It says nothing about the variance of the treatment sample. It can be different and the null hypothsis can hold.

The null hypothesis being tested is weaker than the t-test null hypothesis. The storm hypothesis allows for global warming that made extreme storms more likley but did not increase the mean. In such a case a t-test on all the storms in the window would not reject the null hypothesis that the mean is higher. It would still possible that standard deviation is larger.

Yeah, there are a whole bunch of statistical tests that could detect such a treatment effect. (Wilson rank sum, Mann-Whitney U, F-test) All of them require more than a single storm speed measurement. Your reliance on a single storm is still complete garbage.

The null hypothesis for the storms is rejected unless the treatment population mean AND standard deviation are the same. In this case you can use the control means and standard deviation to assess the probability of the special strom happening. And if it is low then you can reject the null hypothesis that both mean and standard deviation are the same.

No you can't.

There may be a test that you like better that tests a null hypothesis about standard deviations being the same. In that case you could run that test on the storms in the window and it would reject the null hypothesis that the treatment variance is the same.

Instead of one that I "like better", how about we use one that "actually works"?

But if you repeat the analysis without the special storm, and you now find a mean and standard deviation that are very similar to the control, then it will have been this single storm that made all the difference.

Not likely if you choose an appropriate sample size for analysis. That's the whole point. You are trying to rely on a sample size of one, which is just idiotic.

You can't just throw out the outlier like a bad assay because the storm was undoubtly real. If this happens you can still conclude that soemthing has changed.

1 - Use an appropriate sample size.

2 - Use a test statistic designed to deal with unequal variances.

3 - Done.

Storm strength are not going to be normally distribution. Transformed strengths could be normally distributed. For example perhaps the log of strengths is normally distributed.

Are they?

In this case one would say the probability distribution of storm strengths is lognormal. I specifically talked about the four sigma being defined with transformed strengths (so that they are normally distrubuted).

Are they?

If a lognormal distrubtion is valid yu would take the log of a large set of control data and calculate the mean and standard deviation from that. A four sigma storm would be 4 standard deviations out from that mean. My use of the term "sigma" implied a control sample large enough that the t-distribution is not materially different from the normal distribution as so the control sample standard deviation s is very close to sigma.

You take the inverse log of this value and that storm strength would be the four-sigma storm strength

The t-tests you talked about would be done on the transformed data.

So, you have arbitrarily chosen a single value to compare to the control population. That's a single sample t-test or you have simply calculated a probability value. A probability value for a single event can not be used to detect a treatment effect. A treatment effect is supposed to alter a probability value. You can only determine if the probability has been altered by measuring the distribution of all the storms in your window. The presence of your single storm alone can't tell you if the underlying probability has been altered just like a single poker hand alone can not tell you if the deal is truly random. There are real test statistics that can do that however.

Originally Posted by Mikebert

Of course not! I would conclude that the many storms in that window come from a population that is different from the control population. It is different in two ways, (1) the mean value is lower; (2) the standard deviation is higher.

The hypothesis was: global warming make storms stronger (i.e. increase the mean), or more extreme storms more likely (i.e. increase the standard deviation) or both.

Each of those options requires a test statistic specific for that parameter. You are trying to claim that you have a test statistic that can check all of them at the same time. The problem is that if you get a positive result you have absolutely no way of knowing which of the three options has occurred unless you analyze all the storms in your window.

The null hypothesis is that global warming has no effect (i.e. neither the mean nor the standard deviation has changed).

The low-probability outcome p = NP < 0.05, (where P is the 4-sigma probability, and N is the number of storms in the window, i.e. opportunities for the 4-sigma event to happen) shows the population is different, the null hypothesis is rejected.

1 - No it doesn't. All it shows is that a rare event happened. But rare events can happen without warming too. You can not conclude that the population is different just because a rare event happened.

2 - What is your rate of type II errors using your storm window test? How likely is it that warming has altered storms but you fail to detect it because your special storm did not show up? Hint: It will be huge.

Let's try this analogy. I know this is categorical versus continuous but it is just for illustrative purposes.

Let your four sigma storm be a 1/1000 chance occurrence based on the historical control data. We build a die that has 1000 sides to match the control population. A roll of a 1 on the die is equivalent to the occurrence of your special storm during your chosen window. So we wait the requisite window of time and roll the die. If we roll the 1, you are trying to claim that that is evidence that the die has been altered to include more 1's. You can't possibly know that after a single roll because an unaltered die is capable of rolling a 1 as well!

However, if we roll it ten times and get three 1's, then we have a much stronger claim of there being extra 1's on the die because it appears that the probability of a rolling a 1 is higher than it should be. But, we know that because we have multiple rolls for analysis.

That is, either the mean is higher OR the standard deviation is larger (or both).

The statistically significant lower mean shows that the mean is not increased. Therefore the standard deviation is higher (i.e. extreme storms are more likely, but average storms are NOT stronger). In fact, average storm strength is reduced.

So this data would support a conclusion that global warming makes storms weaker on average AND extreme storms more likely.

And your ridiculous one storm in a window test would fail to detect such a pattern. It's only the fact that all the storms in the window are included in the analysis that allows us to see what is really going on.

If a t-test gave a sufficiently low probability, I would conclude that the many storms in that window came from a population that is different from the control population; the mean is higher.

This would support the hypothesis that global warming makes storms stronger. It says nothing about the standard deviation.

Yes, we would use an appropriate test for variance (F-test) to answer that different question. Or, we could turn all the storm data into categorical (storm classes) and run a Mann-Whitney U to detect a different distribution pattern for storm strengths.

The point I can't seem to get you to grasp is that all of those tests require a sample size that is larger than one storm's presence or absence.

While the coins are flipping out, overhead is one of the craziest charts I have ever seen, one of multiple deranged frontal boundaries, and random pressure centers.

Originally Posted by Bad Dog

While the coins are flipping out, overhead is one of the craziest charts I have ever seen, one of multiple deranged frontal boundaries, and random pressure centers.

I think it's a subliminal message.

([F]-test) + Mann-Whitney [U ] = statistical output -> fuck you You know, garbage in, garbage out. Base metal coinage is garbage after all.

Originally Posted by Vandal-72

Let your four sigma storm be a 1/1000 chance occurrence based on the historical control data. We build a die that has 1000 sides to match the control population. A roll of a 1 on the die is equivalent to the occurrence of your special storm during your chosen window. So we wait the requisite window of time and roll the die.

This is a good analogy, but its not what I proposed. The die is rolled N times during the window, not once, because there are N storms, and each of them has a p chance of being special. The probability of getting a special during the window is Np. If Np is small then it is unlikely that this outcome would be seen.

There was an implied sample in the storm case. N was always there, I described early one that I had in mind a value of of N in the neighborhood of 500 . This is where the 4-sigma came from. I needed a single event with very low probability p so that Np < 0.05 even for large N.

Suppose warming changed the distribution pattern of storms so that very extreme events are much more common, but the distribution of non-extreme storms is unchanged. Example: suppose a 1 in 10000 event becomes a 1 in 200 event but the other 99.5% of events have the exact same distribution as the controls. How big of a smaple would you need to detect that this had happened?

The probability calculation I proposed is simply a quick way to determine whether an occurrence of an extreme event in a sample size N is of sufficiently low probability that its presence is signficant. It is very conservative. A more sophisticated test would work better, but my test has the advantage of being easy to understand. The reason why you don't see this as a regular formula is because it is so conservative, a test constructed by a statistician would have a smaller sample sze but the math involved in its construction would be over my head.

Statsitics is NOT the analysis of patterns. Statistics is a discipline that tries to maxmize the information that can be extracted from data, or a given amount a experiment effort, or at least that is what my stat prof said (super nice guy--he loved the conclusion to my beer taste test I did for my midterm project).

. My method is Np < 0.05. Solve for N: N > 0.05/p. In the problem above p = 0.0001, so N > 500. Minimum sample size is 500. If the 500-storm window fails to contain even one special storm the null hypothesis upheld. If the sample contains at least one extreme storm it is rejected.

If a storm 134 was an extreme storm you already know the null hypothesis will be rejected when you reach the end of the 500 storm sample and calculate Np. No matter what happens, the extreme storm cannot "unhappen". So you can conclude that result now.

Originally Posted by Mikebert

This is a good analogy, but its not what I proposed. The die is rolled N times during the window, not once, because there are N storms, and each of them has a p chance of being special. The probability of getting a special during the window is Np. If Np is small then it is unlikely that this outcome would be seen.

That may be what you think you are doing but it isn't. You just described a single roll for the window!

There was an implied sample in the storm case. N was always there, I described early one that I had in mind a value of of N in the neighborhood of 500 . This is where the 4-sigma came from. I needed a single event with very low probability p so that Np < 0.05 even for large N.

So each block of 500 storms is a roll.

Suppose warming changed the distribution pattern of storms so that very extreme events are much more common, but the distribution of non-extreme storms is unchanged. Example: suppose a 1 in 10000 event becomes a 1 in 200 event but the other 99.5% of events have the exact same distribution as the controls. How big of a smaple would you need to detect that this had happened?

Mann-Whitney U, F -test or you would need many 500 storm blocks to see if the special storms really are happening more often. A single event can not be a pattern of anything.

The probability calculation I proposed is simply a quick way to determine whether an occurrence of an extreme event in a sample size N is of sufficiently low probability that its presence is signficant.

No it isn't. No significance can be attached to the presence of a single storm.

It is very conservative. A more sophisticated test would work better, but my test has the advantage of being easy to understand.

The problem with your understanding is that it is just flat out wrong. You think you are doing statistical analysis but it really isn't.

The reason why you don't see this as a regular formula is because it is so conservative, a test constructed by a statistician would have a smaller sample sze but the math involved in its construction would be over my head.

The type II error probability of your "test" is so high that it is a pointless exercise. You can't conclude anything based on either a special storm happening or not happening.

Statistics is NOT the analysis of patterns.

Holy crap. Really?

Statistics is a discipline that tries to maxmize the information that can be extracted from data,

Like if there is or is not a pattern within the data.

or a given amount a experiment effort, or at least that is what my stat prof said (super nice guy--he loved the conclusion to my beer taste test I did for my midterm project).

. My method is Np < 0.05. Solve for N: N > 0.05/p. In the problem above p = 0.0001, so N > 500. Minimum sample size is 500. If the 500-storm window fails to contain even one special storm the null hypothesis upheld.

Nope. You have absolutely no idea what the likelihood of a type II error would be.

If the sample contains at least one extreme storm it is rejected.

Nope. Your 500 block storm is the trial, not each storm. Your are checking to see if a storm happens in the block so you get only one roll of the die.

If a storm 134 was an extreme storm you already know the null hypothesis will be rejected when you reach the end of the 500 storm sample and calculate Np.

Oh my god! Your understanding of statistics and experimental methods is even worse than I thought! You just described EXACTLY what I told you that New Agers do when they claim to find experimental evidence of ESP. It's called data peeking and it's a form of statistical fraud.

No matter what happens, the extreme storm cannot "unhappen". So you can conclude that result now.

You really, really don't understand how to run an experiment.

Originally Posted by Ragnarök_62

I think it's a subliminal message.

([F]-test) + Mann-Whitney [U ] = statistical output -> fuck you You know, garbage in, garbage out. Base metal coinage is garbage after all.

Unbelievably subliminal. I didn't even realize I had sent it.

I'm suing Bill O'Reilly for plagiarism!

Those who saw the segment of his show that John Stossel was on last (Tuesday) night will understand why.

Arctic warmth unprecedented in 44,000 years, reveals ancient moss

Jan 21, 2014

When the temperature rises on Baffin Island, in the Canadian high Arctic, ancient Polytrichum mosses, trapped beneath the ice for thousands of years, are exposed. Using radiocarbon dating, new research in Geophysical Research Letters has calculated the age of relic moss samples that have been exposed by modern Arctic warming. Since the moss samples would have been destroyed by erosion had they been previously exposed, the authors suggest that the temperatures in the Arctic are warmer than during any sustained period since the mosses were originally buried.

The authors collected 365 samples of recently exposed biological material from 110 different locations, cutting a 1000 kilometer long transect across Baffin Island. From their samples the authors obtained 145 viable measurements through radiocarbon dating. They found that most of their samples date from the past 5000 years, when a period of strong cooling overtook the Arctic. However, the authors also found older samples which were buried from 24,000 to 44,000 years ago.

The records suggest that in general, the eastern Canadian Arctic is warmer now than in any century in the past 5000 years, and in some places, modern temperatures are unprecedented in at least the past 44,000 years. The observations, the authors suggest, show that modern Arctic warming far exceeds the bounds of historical natural variability.

"The great time these plants have been entombed in ice, and their current exposure, is the first direct evidence that present summer warmth in the Eastern Canadian Arctic now exceeds the peak warmth there in the Early Holocene era", said Gifford Miller, from the University of Colorado. "Our findings add additional evidence to the growing consensus that anthropogenic emissions of greenhouse gases have now resulted in unprecedented recent summer warmth that is well outside the range of that attributable to natural climate variability."


Read more at: http://phys.org/news/2014-01-arctic-...veals.html#jCp

Two of the Humiliation Conga Line high pressure centers are passing just 50 NM to the north. The Polar Vortex has several more on the way.

Progressives respond to State of the Union

350.org Responds to State of the Union

350.org Executive Director May Boeve released the following response to tonight’s State of the Union from President Obama:

“President Obama says he recognizes the threat of climate change, but he sure doesn’t act like it. If he was serious, he’d reject the Keystone XL tar sands pipeline and stop promoting fossil fuels like natural gas. Fracking isn’t a solution, it’s a disaster for communities and the climate.

You can’t say you care about ending cancer and then go buy a carton of cigarettes–and you can’t say you care about the climate and then go dig up more fossil fuels. We need real leadership from this President, not more lip service. Rejecting Keystone XL would be the perfect place to start.”
350.org founder Bill McKibben added:

“An all of the above energy strategy is exactly as sensible as an all of the above foreign policy–I kept waiting for the part of the speech where he’d explain why North Korea and England should be treated the same. If he actually took climate change seriously, he’d understand that more oil means higher temperatures–that’s just how physics works.”

http://350.org/press-release/sotu/

Originally Posted by Deb C

Progressives respond to State of the Union

350.org Responds to State of the Union

350.org Executive Director May Boeve released the following response to tonight’s State of the Union from President Obama:

“President Obama says he recognizes the threat of climate change, but he sure doesn’t act like it. If he was serious, he’d reject the Keystone XL tar sands pipeline and stop promoting fossil fuels like natural gas. Fracking isn’t a solution, it’s a disaster for communities and the climate.

You can’t say you care about ending cancer and then go buy a carton of cigarettes–and you can’t say you care about the climate and then go dig up more fossil fuels. We need real leadership from this President, not more lip service. Rejecting Keystone XL would be the perfect place to start.”
350.org founder Bill McKibben added:

“An all of the above energy strategy is exactly as sensible as an all of the above foreign policy–I kept waiting for the part of the speech where he’d explain why North Korea and England should be treated the same. If he actually took climate change seriously, he’d understand that more oil means higher temperatures–that’s just how physics works.”

http://350.org/press-release/sotu/

Comments like this get ignored, or worse. Being counter productive is never a good plan. If you want to attack bad behavior, had viable good behavior to offer in its stead ... viable being the operable word.

Originally Posted by Marx & Lennon

Comments like this get ignored, or worse. Being counter productive is never a good plan. If you want to attack bad behavior, have viable good behavior to offer in its stead ... viable being the operable word.

That's the trouble with those of us on the left. Obama is about the best president the left will ever get, in this conservative country in which money speaks and money rules. We have to understand that, as much as we hate the all of the above strategy, the people will not support him now if he abandons it. He's got lots of fossil fuel rich states to worry about. People are brainwashed into thinking we need to burn more fossil fuels; that otherwise we'll run out of fuel and jobs will be lost. We don't at all need to use it, at least no more than we already are using until the switch is made to renewables; but the corporations are powerful, and the people think they still need all that oil and coal and gas. So I think he's doing pretty good, considering the people that he has to deal with-- namely us. All of us, left right and center.

But I don't disagree with 350.org giving voice to our concerns and pressuring him. Maybe it was too intemperate though, and when we talk without respect, I wonder if Obama and Co. can hear us. Maybe and maybe not.

(of course, I know and support 350.org)

Originally Posted by Marx & Lennon

Comments like this get ignored, or worse. Being counter productive is never a good plan. If you want to attack bad behavior, had viable good behavior to offer in its stead ... viable being the operable word.

Comments like these? The comments are from an organization that works for environmental justice. That, IMHO, is viable good behavior. I guess it was my bad assuming that you knew who the 350.org was. And, Bill McKibben is a progressive environmentalist who has worked tirelessly for educating and promoting legislation for lessening global warming. He has been an avid supporter of Obama, but he also knows that part of being a good citizen is looking at the underbelly of fancy words and rhetoric.

I have offered numerous positive ways to affect change. I didn't hear anything about those posts from you. I advocated many times about putting our time, energy and money into organizations that are working for a better America. I've even posted news and information from Yes magazine that talks about what we can do as citizens to make positive changes.

Originally Posted by Eric the Green

That's the trouble with those of us on the left. Obama is about the best president the left will ever get, in this conservative country in which money speaks and money rules. We have to understand that, as much as we hate the all of the above strategy, the people will not support him now if he abandons it. He's got lots of fossil fuel rich states to worry about. People are brainwashed into thinking we need to burn more fossil fuels; that otherwise we'll run out of fuel and jobs will be lost. We don't at all need to use it, at least no more than we already are using until the switch is made to renewables; but the corporations are powerful, and the people think they still need all that oil and coal and gas. So I think he's doing pretty good, considering the people that he has to deal with-- namely us. All of us, left right and center.

But I don't disagree with 350.org giving voice to our concerns and pressuring him. Maybe it was too intemperate though, and when we talk without respect, I wonder if Obama and Co. can hear us. Maybe and maybe not.

What did you read that was disrespectful? I didn't read anything disrespectful in the 350.org statement.

For those of you who do not know about the 350.org

http://350.org/about/what-we-do/

Originally Posted by Deb C

Comments like these? The comments are from an organization that works for environmental justice. That, IMHO, is viable good behavior. I guess it was my bad assuming that you knew who the 350.org was. And, Bill McKibben is a progressive environmentalist who has worked tirelessly for educating and promoting legislation for lessening global warming. He has been an avid supporter of Obama, but he also knows that part of being a good citizen is looking at the underbelly of fancy words and rhetoric.

I have offered numerous positive ways to affect change. I didn't hear anything about those posts from you. I advocated many times about putting our time, energy and money into organizations that are working for a better America. I've even posted news and information from Yes magazine that talks about what we can do as citizens to make positive changes.

You miss the point entirely. Demanding the impossible just gets eye-rolls from everyone, followed by your core point being either ignored or denigrated. That was a hard lesson I learned in the anti-war movement. Empowering your opposition is always a mistake.

In this case, McKibben demands an immediate end to fossil fuels. It can't be done, so no one takes anything else he says seriously. I certainly don't.

Originally Posted by Marx & Lennon

You miss the point entirely. Demanding the impossible just gets eye-rolls from everyone, followed by your core point being either ignored or denigrated. That was a hard lesson I learned in the anti-war movement. Empowering your opposition is always a mistake.

In this case, McKibben demands an immediate end to fossil fuels. It can't be done, so no one takes anything else he says seriously. I certainly don't.

Demanding the impossible is sometimes the only way to get things done. Ask the women in the suffrage movement or any of the crucial justice movements. Sure, it took awhile, but they never stopped demanding justice. The only time we really lose, is when we stop working for what's right. Just because it doesn't happen in our time frame, it doesn't mean that our ripples aren't having a positive affect for future generations.