It is an irreversible agonist that binds to nicotinic acetylcholine receptors and first activates then blocks them. At high doses it will paralyze muscles. At these low doses it would more likely act by interfering with cognition. Because it is irreversible, it likely has a cumulative effect.
No, the Biblical literalists are generally referred to as "young-earth" creationists. But it is still considered creationism if you believe that species were created in something very close to their current forms, as opposed to evolving from a common ancestor.
And substitution of 65:35 HFCS in place of sucrose (assuming equal sweetening and equal consumption of sweetened foods) would result in essentially no increase in fructose consumption, because 65:35 HFCS is about 39% sweeter than sucrose.
Cite some numbers here - if we continue at 0.1C/century warming, why should we assume any sort of catastrophe?
You seem to be experiencing a failure of reading comprehension. Which part of "None of the projections of climate science include what I would characterize as "catastrophic" global warming." did you fail to understand?
Oh, certainly there are physical mechanisms for natural climate change, but I don't have to model them in order for them to be true.
We seem to have another failure of comprehension here. Nobody is suggesting that the process of checking your work by doing the math magically causes your theory to be true. Rather, it is the way in which responsible scientists catch errors in their own ideas. Coming up with excuses for not doing the math to check his ideas is the mark of a crank, not a scientist.
Sure we've checked it mathematically - we've got indistinguishable periods of 50 years of warming, one of which is asserted to be "natural" and the other asserted to be due to human CO2. There is no discernible difference
It has been previously explained to you why this is nonsense. Try reading it again: "Surely, if you actually think this through, you will realize how ridiculous this is. If your living room is uncomfortably wrong, do you imagine that you can determine, solely by looking at the thermometer, whether it is too warm because it is a hot day--or because you left the heat on? No, you gather other information; you step outside to check the weather, you check the thermostat, you check whether you left the oven on." When there is "natural" climate change, there is other evidence of a change in "natural" mechanisms--a change in solar radiance, for example. Again, this is the sort of error that you fall into if you don't bother to "check your work" by mathematically modeling the physical mechanisms (whether you choose to call them "natural" or "unnatural").
Again, you seem to believe that a flawed, falsified model must be believed in the absence of another mathematical model. This isn't how science works - you don't need to prove something else right in order to prove a hypothesis wrong.
Considering how little you obviously know about science, your instruction to scientists regarding "how science works" carries little weight. In fact, science virtually always progresses by competing models (in the unlikely event that you actually wish to educate yourself, you might consider reading some Thomas Kuhn). Remember: "All models are wrong; that's why we call them models." So nitpicking somebody else's model does not make your model right. No model is perfect; you have to show that their model is wrong in a way that matters. How do you do that? You construct a better model, and you show that your improved model makes better predictions. And here is where the climate science "skeptics" have so miserably failed.
Of course, Robert Lustig also says that there's no significant difference between high fructose corn syrup and sugar.
He's right. It's pretty hard to come up with a plausible biochemical scenario in which such a modest difference in fructose content makes a big difference. Which is why the difference between HFCS and sucrose is not likely to be a basis of differences in health between Canadians and Americans. On the other hand, if Canadians consume substantially less sugary drinks overall than people in the U.S., that could be an explanation.
It's still pretty hard to imagine dramatically greater adverse effects with only 15% more fructose than sucrose. That would be an extraordinarily steep dose-response curve. And the actual increase in fructose consumption would be less than that, because fructose is 73% sweeter than sucrose, so with higher fructose content, less is needed to achieve a given amount of sweetness.
Seems unlikely; the commonly used form of HFCS contains about half glucose and half fructose. So does cane sugar. I suppose it could be an overdose of sugar in all forms, or of fructose whether derived from cane sugar or HFCS. Do Canadians consume as much sugary drinks as Americans?
Whoa, hold yer rhetoric there for a moment - we know that there is natural global warming, and natural global cooling. Fiddling with the magnitude of a feedback in a GCM is what gives rise to Catastrophic Anthropogenic Global Warming, or leaves things at simply Inconsequential Anthropogenic Global Warming. There is a difference.
Fiddling with the cloud feedback also does not eliminate the prediction of serious global warming that is likely to cause financial and human costs that far exceed the cost of mitigation. "Catastrophic" is a vague term. None of the projections of climate science include what I would characterize as "catastrophic" global warming. On the other hand, the greater uncertainties are at the upper end, particularly with respect to consequences. For example, biologists are not predicting that global warming will result in the emergence of a crop disease that will wipe out a major component of the world's food supply supply, resulting in the starvation of 10% of the world's population, but I expect that you will find few willing to assure you that it's beyond the realm of possibility.
Natural climate change isn't a model, it just *is*.
I see. So in your mind, there are no physical mechanisms responsible for "natural" climate change that could be studied and mathematically modeled. It has nothing to do with solar radiance, cloud feedback the earth's orbit, volcanic eruptions, or shifts in ocean currents. It's just....magic.
Remember, mathematical modeling is the major ways that scientists "check their work," to see if their ideas actually make sense, or if they are fooling themselves. So if somebody tells me that they believe that the modern warming is due to something other than CO2 ("cloud feedback" for example) and I ask them, "Have you checked your work mathematically? Is your theory consistent with known data?" and they answer, "No, I haven't bothered, because I'm really, really sure that I'm right, and besides, cloud feedback is natural, so I believe that I should be exempt from having to check my work," I'll conclude that the guy is not a real scientist, but just a crank, and give his views little credence.
Of course, it is true. Don't think for a moment that we have a model of physical processes that we can simply run backwards and recreate historical temperatures - we *started* with those simple models (and maintained the central conceit that CO2 was an overwhelming climate driver, instead of a second order function of temperature), and tweaked things until they gave a close match to historical records.
Well, that's all very interesting, but I'd be more convinced if you'd checked your work, and shown mathematically that it was possible to develop a physical model in which CO2 is not a strong climate driver, without losing consistency with the known climate data.
Your problem is this - modern temperature records aren't unprecedented on any scale. You have no way of excluding the null hypothesis of natural climate change.
So first we have a straw man--no scientist has claimed that modern temperature records are unprecedented, only that the natural causes that have produced high temperatures in the past are not present today. And then you fall back on your old "null hypothesis" mistake. Remember, the null hypothesis must be a zero statistical correlation hypothesis. So what are you supposing to have zero correlation? Temperature and time? Temperature and CO2?
And isn't that also true of your favored CO2 hypothesis? That what you claim is a trend could simply be natural background noise?
When examined over the period of time that statistical analysis indicates is necessary to be able to reliably detect the trend expected from global warming, it is highly significant. If you are asking, "But isn't it possible that global warming stopped for some unknown reason 9 years ago, or last week, or 10 seconds ago?" I will replay, "Absolutely. But absence of evidence is not evidence of absence."
Models are consistent with the known climate record because they've been curve fitted. Understand this isn't one of their strengths.
Once I get past the visceral repugnance I (like most scientists) feel for dishonesty, I can't help but admire how well-crafted some of these climate science talking points are. It is quite clear that they were crafted by people who themselves understood the science, and who carefully crafted a set of fallacious pseudoscientific arguments designed to deceive people with superficial understanding of science. Of course, they know that it is not possible to get global warming to go away by fiddling with the magnitude of feedback. They've tried, and failed. So how do they get the public to ignore the fact that they don't have a model, whereas the real climate scientists have subjected their own ideas to the scientific "sanity checking" of mathematical modeling. They spread the meme that they don't need to come up with a physically realistic model because their model is the "null hypothesis" (which sounds stupid if you known what a null hypothesis actually is, but most people don't, they think--like you--that it is just means some kind of default model). And they also spread the meme that these models can be "curve fitted" to say anything you want. Of course, that's not true--if it were, they would have their own, competing mathematical model that did not predict global warming--but the public are mostly unfamiliar with working with physical models, and imagine that it is something like fitting to an arbitrary mathematical formula. Of course, anybody who has actually worked with physical models knows this to be quite false. Very often, it is simply impossible to get such models to do what you wish, because physical models are constrained by physical principles--for example, if you want a model in which increased CO2 causes warming, but warming doesn't increase CO2, you run up against basic physical principles like conservation of energy.
The graphs are the same period of time, and the same vertical scale, simply unlabeled as to the absolute value. You cannot discern from either of them which one is supposed to be "natural" as per your AGW hypothesis, and which one is supposed to be CO2 driven.
Surely, if you actually think this through, you will realize how ridiculous this is. If your living room is uncomfortably wrong, do you imagine that you can determine, solely by looking at the thermometer, whether it is too warm because it is a hot day--or because you left the heat on? No, you gather other information; you step outside to check the weather, you check the thermostat, you check whether you left the oven on.
And it’s a miss! The stratosphere hasn’t been cooling in over a decade:
Now this is another very clever deception. Given any trend with random statistical noise riding on top of it, it is always possible to find a period of time short enough so that the noise obscures the trend. Of course, it is possible to examine the statistical properties of the noise, and determine what period of time you have to examine to be able to reliably detect a trend in the presence of the noise. This has been done for temperature measurements, and the required interval turns out to be a couple of decades or so. So we are deluged with claims that "the earth has not warmed" (or, in this case "the stratosphere has not cooled") over the last decade. Dishonest as hell, but very clever, because the public does not have a sophisticated understanding of statistical determination of trends.
Doesn't causality require that CO2 change first *then* temperature?
In a word, no. It is more the rule than the exception in physical processes for causality to be able to run in either direction, depending upon boundary conditions.
And they're not even sure of the *sign* of this feedback, much less the magnitude. That's a hole you can drive 100 years of warming right through.
The reason that they are not sure of the sign is that the evidence indicates that net cloud feedback is close to zero. When a value is close to zero, it is hard to determine whether it is a little bit negative or a little bit positive. And no, nobody has managed to come up with a model that creates the modern warming through cloud feedback and that is at all consistent with the known climate records. There are plenty of published climate models. Feel free to try.
No, it doesn't at all. The problem with the modeled cloud feedbacks (where they don't even know what *direction* it goes in), is that it essentially leads to error bars an order of magnitude larger than the effect we're trying to observe.
This is false. While models vary somewhat in their predictions due to differences in cloud feedbacks, every model that is consistent with the known climate record predicts statistically significant warming in response to the massive increase in CO2 due to burning of fossil fuels. Oh, by the way, "eyeballing" error bars is not a valid way to judge statistical significance--you must use the appropriate mathematical test.
Can you tell from these two graphs, which one is the "natural" warming and which one is the anthropogenic CO2 warming?
You can't tell much of anything from a graph in which the axes are unlabeled. Warming arising from "natural" sources can be distinguished because "natural" does not mean magic. Natural mechanisms of warming have effects other than just an increase in average temperature. For example, warming due to an increase in solar output predicts that the upper atmosphere will warm more than the lower (in fact, the upper atmosphere has cooled). It also predicts that the warming trend will be greater in the day than at night (it's the other way around). Etc., etc.
Listen to yourself carefully here - you're asserting that the dynamics of CO2 both lead and lag temperature changes.
Correct. That's what the physics predicts. Like many physical processes, it can run in either direction, depending upon circumstances (i.e. what initiates the warming). You are repeating one of those classic climate "skeptic" talking points that only sounds plausible to people who are ignorant of science. Real scientists just roll their eyes when they hear nonsense like this, because basically it boils down to, "Climate science must be wrong because over history, temperature and CO2 have behaved in exactly the way that the model predicts." To be fair, however, many of the standard climate "skeptic" talking points were carefully designed by lobbyists (as has been so thoroughly documented by historian Naomi Oreskes in "Merchants of Doubt") who specialize in creating specious arguments specifically designed to fool people with a superficial knowledge of science. You are hardly the first person to be taken in by this one, as foolish as it sounds to actual scientists.
Medievalist fantasy was basically invented by the British. Tolkien, who pretty much invented the genre, was British. So was C.S Lewis. If you're really pedantic, perhaps you might prefer to trace it back to Spenser or Malory. But they were British, too. I think it's kind of charming that a video of a modern fantasy by an American author retains this nod to the origins of the genre.
""The Null Hypothesis is the hypothesis that there is no relationship between two variables. Establishing that there is a relationship between two variables is the first step in establishing whether there is a causal connection between two variables. ""
Note the word *causal*. Go ahead, I'll wait for you to read it again:)
The part you missed is "first step." *sigh* I've explained this to you about 3 times already, but maybe the 4th time is the charm. Probably not, though. People seem to get very attached to crank notions, and the idea of a causal null hypothesis is certainly one of those.
The first step in establishing whether there is a causal connection between two variables is excluding the null hypothesis of no correlation. After all, if two variables are not correlated, it makes no sense to construct a causal hypothesis of how they are correlated. Once the null hypothesis has been excluded, it is then considered legitimate to construct causal hypotheses (all of which are considered on an equal basis, and must satisfy the usual criteria for a valid hypothesis (do I really need to repeat these?)
"In spite of this undeniable progress, the amplitude and even the sign of cloud feedbacks was noted in the TAR as highly uncertain"
Ah, another standard crank talking point: "If there is something that is not perfectly understood, then nothing is understood." If you think that cloud feedbacks can possibly account for global warming, then produce a model, and show that it can provide a reasonably good fit to the known climate record, and can also explain the modern warming. After all, many climate science models are publicly available. So plug in whatever cloud feedback you please, and show that the model can account for the difficulty. While do you imagine that nobody has done this by now? After all climate science "skeptics" love to clutch at the straw of cloud feedbacks. Actually, real climate scientists have tested models with different cloud feedbacks. As explained in the IPCC section that you quote-mined (but I suspect, did not actually read or comprehend) different cloud feedbacks have been modeled. But while uncertainty about cloud feedbacks creates some uncertainty about the magnitude of future warming, it doesn't make it away.
But think about this: clearly, global climate has been warming. So if this is due to some "natural" mechanism involving cloud feedbacks, that implies that cloud feedbacks have been changing. So what about clouds is changing? It doesn't seem too likely that the physics of water vapor is changing, does it?
A hypothesis that predicts *everything* predicts *nothing*.
So according to you, if the interaction between temperature and atmospheric CO2 displays the well understood dynamics of a positive feedback (i.e. if you increase one, the other will follow, and you can get it to go either way depending upon where you start), that means that it explains everything? Or are you just in a snit that something that you thought was a problem with climate science is actually a prediction? In fact, it would be a problem for climate science if it didn't work that way.
There are well established statistical methods of extracting long-term trend from short-term noise such as fluctuations in water vapor. As these fluctuations ride "on top of" the long-term trend, it is plausible that the rising trend could lead to more extreme weather events, although how much that is happening at the present time remains a challenging statistical question.
There have already been numerous tests of modern climate theory, which could have potentially have falsified the theory. For some of the tested predictions of climate science, see here. Indeed, every time a volcano erupts, it is another test of climate theory.
"The null hypothesis typically corresponds to a general or default position. For example, the null hypothesis might be that there is no relationship between two measured phenomena or that a potential treatment has no effect."
"The Null Hypothesis is the hypothesis that there is no relationship between two variables. Establishing that there is a relationship between two variables is the first step in establishing whether there is a causal connection between two variables. "
:
I see that you have found other sources that confirm what I told you: the null hypothesis is the default hypothesis of no correlation ("no relationship" is just another way of saying "no correlation") between two variables (i.e. one variable does not change in any consistent manner when the other one does). I note that although you have clearly been looking, you have found no example of a null hypothesis that mentions causality. You won't, because "null" (which means "zero") hypotheses are exclusive to statistical correlation, and correlation is incapable of addressing questions of causality. So your special pleading that "natural variation as a cause of climate warming" should be regarded as a null hypothesis, and therefore exempted from the normal criteria applied to competing scientific hypotheses (e.g. to be based on a physically realistic mechanism, to make falsifiable predictions, and to be consistent with the known data) is a non-starter.
So, are you asserting that you've considered every possible cause for every temperature fluctuation?
Yes, climate scientists have considered every physically realistic mechanism that could potentially account for the rise in temperatures, including the ones that were likely responsible for past episodes of (what you call "natural") climate change. If you think you have a new one, then provide the hypothesis, and show evidence that it is both consistent with the known climate record and acting today.
No, it simply isn't. Put another way, if we had any sort of ability to do accurate modeling at that level, we'd be able to create a model for the stock market
Sorry, but this kind of clutching at rhetorical straws is silly. There is no relationship whatsoever between the physical mechanisms responsible for climate change and the economic and psychological mechanisms that determine movements of the stock market. Here's one rather obvious difference: if somebody comes up with a model that accurately predicts movements in stock prices, then people will change their investments based upon those predictions, which will alter the movements of the stock market in such a way that those predictions are no longer valid. Fortunately, we don't have to worry that the physics of climate will alter in response to what we know about it.
So you're now asserting climate science predicts a time travel effect, where future CO2 can effect present temperatures?
It predicts that increases in CO2 can either lead or follow warming, depending upon whether the warming is initiated by release of CO2 or by some other factor, such as a change in solar output. This is very elementary, and you would already know it if you'd bothered to learn anything about the actual science instead of deriving everything that you think you know from crank web sites.
Neither of my series shows statistically significant correlation, yet *both* of them change. So the null hypothesis of natural climate change is clearly still the null hypothesis (we know climate naturally *changes*, and doesn't have to remain static compared to CO2 levels). In fact, the observed historical record shows a time lagged correlation, with CO2 *following* temperature. If we were going to propose a causality based on the majority of the data, it would go *from* temperature *to* CO2.
I'm not sure that it's worth the trouble of trying to teach you anything about basic, since your previous comments have suggested that you actively resist learning anything that might challenge your fixed opinion regarding climate science (e.g. you are highly familiar with climate "skeptic" talking points, yet you somehow have managed to learn almost nothing about the actual science). But I'll give it a shot.
A few key points to remember if you don't want to sound foolish when talking to actual scientists: 1. Statistics addresses questions of correlation, not causality. 2. The "null hypothesis" applies solely to statistical significance testing, and thus relates only to correlation, never to causality. 3. Hence, any statement with the word "cause" in it can never be a null hypothesis.
Now let's consider as an example one of the series that you came up with before.
Now it is important for you to remember that we are discussing statistical measurement. So you can't regard this as a mathematical series; you must regard them as measurements of statistical variables that have some "true" value (which you can estimate, but which you cannot in general know with perfect precision) plus some amount of random error or variation. So you do not expect the measured value to be the same every time you measure it.
So if you ask if a variable is "constant," or shows "no change," you aren't really asking it is the same every time you measure it, because for a statistical variable the answer to that is automatically "No". Rather, you are asking a question about correlation. Does the measured value of a variable depend upon the measured value of some other variable? For example, you may ask whether it shows no change over time, which in statistical parlance is equivalent to the statement that the measured value has no correlation with time. Or you may ask whether it shows no change from place to place, in which case you are asking whether its measured value shows no correlation with some measure of location. Or you may ask whether it shows no change when some other variable changes, in which case you are asking whether there is no correlation between one measured variable (e.g. temperature) and another (e.g. CO2). All of the above are examples of null hypotheses. What makes them null hypotheses is that in all cases, the hypothesis is that one variable does not change with the other--or more technically, that the correlation does not differ from zero more than would be expected from the random variation in both parameters.
Now your Series clearly shows no significant correlation; in fact the correlation is precisely zero. That doesn't mean that there is no correlation because the "null" or zero hypothesis cannot be proved, only disproved. Indeed, there is an obvious problem, because parameter (b) shows no variation at all. This could mean that it has not been measured accurately enough to detect a correlation. Perhaps a more accurate measurement would give values like 1.01, 1.02, 1.03, 1.04, 1.05, 1.06, 1.07, 1.08, 1.09, 1.1 -- which would new highly correlated. So the most we can say about the null hypothesis is that we cannot reject it based on the available data. Once you reject the null hypothesis, you may then consider the reasons why the variables in question (e.g. temperature and time) are correlated. At this point, all hypotheses, whether you choose to call them "natural"
Actually, most modern 3D technology does not require you to keep your head perfectly level. Older 3D glasses that used linear polarization showed crosstalk if you tipped your head but this is not the case with the modern technology. "Active" (shutter) glasses (the somewhat clunky ones) work perfectly well with moderate tips of the head, although your brain gets confused if you tip your head completely sideways (because the parallax is not in the direction that your brain expects from the position of your eyes). Most modern passive 3D systems use circular polarization, which is similarly insensitive to head angle.
So asking the question, "What kind of reform are you looking for? " is putting words in your mouth? That is ridiculous hyperbole. And I can't help noticing that you did not even answer the question. The notion that a having having limitations on the use of trademark--which under current law only applies in a very limited commercial context where some sort of plausible argument can be made that consumers can be misled--constitutes "ownership of a word" is also ridiculous hyperbole.
Do you need any more ammunition for patent and trademark reform??
Actually, yes. Companies are allowed to say whatever they want in a license agreement, whether or not it has any actual legal force. What kind of reform are you looking for? A "license agreement police" that reads every license agreement in the world and levies fines for overly broad license agreements? Do you really think the benefits would justify the cost of all of that extra bureaucracy?
After all, the only thing that actually matters is what the courts will enforce. I'll start worrying if any court, anywhere, enforces such an agreement or trademark against somebody using "book" in a generic manner or similar terms like "phonebook". But they won't. And Facebook wouldn't bring such a suit anyway, because they know they'd lose, and that would undermine their ability to bring the trademark suits that they really care about--the ones against social networking websites that are trying to ride on their coattails by calling themselves "visagebook" or something.
And the difference between a useful falsifiable prediction, and a silly one, is that a useful one (or set) will contain all necessary and *sufficient* components to imply that the proposition is true.
Once again, this is a logical impossibility. Since there is no way to enumerate the predictions of every possible model (indeed, the set is almost certainly infinite), there is no possible prediction that is sufficient to show that a model is true. This is the case for any theory about the real world.
In science, a critic doesn't need to replace an existing conceit with one of their own, they only need to show that the proposed conceit is false.
This statement is the hallmark of the scientific crank. You will never hear a successful scientist say this, or a historian of science. Who says this? Evolution deniers. HIV/AIDS deniers. Global warming deniers. It is a one of those deceptive notions that is obviously ridiculous to those who have done actual science, yet can sound plausible to the amateur. After all, it is true in logic, it is true in mathematics, shouldn't it be true in science? And indeed, it would be true in an ideal world without noise, without statistical variation, in which perfectly accurate measurement was possible, one in which it was possible to construct mathematical models at the subatomic level that perfectly reproduced nature and approximations were unnecessary. But of course, we live in the real world in which none of this is true. What a scientist will tell you is that all models are wrong. All models are imperfect representations of reality. That is why we call them models. But as was articulated perhaps most clearly by Isaac Asimov some are less wrong than others. The advance of science has been through a progression of successively less wrong models. It is very easy to nitpick somebody else's model, and pretty much worthless. All models are wrong. The question is whether it is wrong in a way that matters. To show that it is, you need to offer a better one--one that makes better predictions. Predictions are also important because there are some subtle statistical fallacies that it is possible to fall into (particularly for amateurs, but even experienced scientists can sometimes be tripped up) when you examine data after the fact (you'll see a lot of that on WUWT). This is why prediction is the gold standard of science. If you think that the current model is wrong, and wrong in a way that matters, you have to show that when you fix whatever is wrong with it, you are able to make better predictions. When one side of a debate has a clear theory that has been subjected to the discipline of mathematical modeling, with a long track record of making clear, falsifiable predictions, and the other side has nothing but model nitpicking, the former side will have far more credibility with working scientists--which is, of course, why every major elite scientific society in the world, and something like 97% of scientists with any kind of published track record in a relevant field, accept climate theory and its predictions regarding CO2 and climate change.
What is it? Zero change or zero correlation? You seem to believe that zero change and zero correlation are the same...let me demonstrate with some series again.
Zero correlation means that on average, when one variable changes, the other does not. Neither of your series shows statistically significant correlation. Zero correlation with time means zero average change over time. Again, I suggest that you consult a textbook on elementary statistics.
You're talking about perhaps *necessary* requirements
That's what a "falsifiable prediction" means.
You've given us no reason to believe that your predictions are indicative that your hypothesis is correct.
Real science never claims to prove that a theory is correct. Indeed, it is a logical impossibility to prove any theory correct. It is always possible that there could be some other theory that could be equally consistent with the data. In the case of climate science, critics have had a long time to come up with competing theories. The physics upon which climate science is based is published and the equations are well known. Code for many climate models has been made publicly available. Yet no critic has been able to come up with a climate model that is based on valid physics and is equally consistent with the existing climate data and that does not predict a worrisome degree of warming as a consequence of CO2 emissions. This is about as strong as evidence ever gets in science.
No, you've got it wrong again. "No correlation" (as opposed to "no causation", which is requires correlation, but then it also needs a bit more), doesn't mean *zero* average change in one variable when the other one changes. That's daft.
Wrong. Pick up a statistics textbook. The null model of correlation is that there is zero average change in one variable when the other one changes. Showing that the change in measured values is significantly greater than would be expected from chance variation constitutes rejection of the null model. Of course, correlation does not imply causality. Statistics is inherently unable to determine causality.
Again, you're playing a trick on yourself - you cannot simply prove your hypothesis by picking a trivial null to disprove, asserting that once it is gone nothing replaces it
The null model is the zero model--zero change or zero correlation. I didn't pick it; that's how it's defined. You are trying to call a tail a leg again.
hat's the kind of argument you get from creationists who insist that BAM, once you discard the null hypothesis of evolution, which breaks the 2nd law since it goes from disorder to order, that the only reasonable explanation is an intelligent designer.
Sorry, but this is scientifically illiterate nonsense. No biologist will tell you that natural selection is a null model. Remember, a null model is purely statistical, and therefore cannot be causal. One could of course create statistical null models that would be relevant to the theory of evolution. One such model might be "the genetic code of [insert species] has not changed over time." But the theory of natural selection is a causal model, not a statistical model, and hence can never be a null model.
You've managed to write a lot, but you've still failed the most basic test of science - the existence of the necessary and sufficient falsifiable hypothesis.
I can't help being amused when amateurs try to tell scientists how to do science--and get it ludicrously wrong. In this case, "necessary and sufficient" is not even wrong--it's impossible based on elementary logic.
But scientists do talk about what they believe, and they use the word in the common sense (as in "I believe it is about to rain"). They aren't going to stop doing so, and I think that it is not entirely honest to try to pretend that scientists do not form opinions about theories in advance of compelling evidence. Most scientists do, including many very good ones--they are just flexible and ready to change what they believe in the face of new evidence.
I also don't believe (there it is again) that we should be more strict in the usage of "hypothesis" and "theory." This just caters to the notion, held by some amateurs and often exploited by those who try to play semantic games to undermine scientific knowledge, that we have some sort of formal taxonomy of scientific certainty. In actual usage, scientists may sometimes use "hypothesis" to emphasize that an idea is particularly tentative. But the same person may in the next breath refer to the same idea as a "theory." Trying to get formalistic about it leads into all sorts of problems. What level of evidence do you need to upgrade a generalization from "hypothesis" to "theory"? One experiment? Two? Confirmation by two different labs? Three? Trying to get rigid about it is a nightmare, and you'll never get everybody to go along with any kind of strict criteria.
My view is simple. The facts are the data (or reports of data). All scientific generalizations, interpretations, and explanations are theory. Some theories have a lot of evidence to support them, some have very little, and some are known to be frankly wrong. We may occasionally note that we think that a theory is particularly tentative by calling it a "hypothesis," but we don't have a "theory nomenclature rating board". Every scientist makes his or her independent judgements regarding level of evidence
As for "law," the modern usage doesn't really relate to level of evidence at all--it's become pretty much synonymous with "rule of thumb," meaning a formula or algorithm that is simple, useful, and convenient, whether or not it is exactly correct. We aren't going to stop referring to Newton's equations of motion as "Newton's Laws" (even though we know them to be not quite accurate). And we're never going to start referring to Einstein's equations of motion as "Einstein's Laws" even though they have been verified to incredible precision.
Relativity only works when we load the Universe with materials that we guess exist (Dark Matter and Dark Energy). It's a huge amount of speculation.
Sorry, but this is completely wrong. Einstein's Special Theory of Relativity does not refer to any of them (it is the Special Theory that has supplanted Newton's Laws of Motion). Even Einstein's General Theory of Relativity (which is more accurate than Newton's Law of Gravitational attraction, but is clearly not a complete theory of gravity) predates these and does not require them.
We can make relativity work with items here on Earth much better, but not when we leave the atmosphere.
Also utter nonsense. Special relativity has been tested in multiple ways both in the absence of atmosphere here on earth and with respect to phenomena in outer space. No exceptions have been found anywhere.
Are many of Einstein's predictions seemingly correct? Sure, but to tout them as law when we know that it does not work is no different than someone pointing at a black book and saying "it's all in there" (I think you get the reference).
Who's touting them as "Law?" As I noted before, the way the word "Law" is used by actual working scientists is pretty much equivalent to "rule of thumb" -- a simple formula that may or may not be entirely accurate, but that is easy to apply and a good enough approximation to the truth to be broadly applicable. Einstein's Special Theory has been verified to an amazing degree of precision, but no matter how many decimals of precision its predictions are verified to, I can guarantee that scientists will never start calling it a "Law" -- it may in fact be absolutely true, but at lacks the simplicity and convenience of Newton's (incorrect) Laws of Motion.
Science can not answer many things, and lots of people claim it's the answer to everything.
As a scientist, this statement makes no sense to me. I don't know of any scientist who would claim that science is the answer to everything. On the other hand, denying scientific knowledge has historically been the answer to nothing.
You're doing it backwards. Don't tell me about the ten-thousand white swans you saw - tell me about what *black* swan would falsify your hypothesis. Climate models make numerous testable predictions which have *failed* to be observed...yet you ignore those as irrelevant?
All of the numerous predictions made from climate theory would have falsified the theory if the observations had come out other than predicted: If the nights were not warming more than the days, if the stratosphere were warming rather than cooling, etc., etc. So far, the theory has held up to every one of the numerous challenges
No, it doesn't. Null means "no relation". We have null hypotheses about all *sorts* of things that obviously change (say, the relationship between average gas prices and stock market averages) - you are truly torturing the definition if you're blithely asserting that it means "no change".
I'm glad to see that you now seem to have belatedly gained some comprehension of what I tried to explain to you a couple of posts back: ' "Null hypothesis" has a very specific meaning within statistics. "Null" means "zero" and it denotes that the null hypothesis is always the hypothesis of zero difference or zero correlation. ' Surely it is not at this point necessary to belabor the issue by repeating the whole definition every time I mention it? After all, "no correlation" merely means zero average change in one variable when the other one changes. Please feel free to substitute my previous more lengthy explanation if you find my shorthand too confusing.
The null hypothesis is that average global CO2 levels have no causal relationship to average global temperatures.
Sorry, here you start to screw up again. A null hypothesis cannot be causal, because the concept of a null hypothesis is exclusive to statistics and statistical correlation, and statistical correlation is logically incapable of establishing causality. A causal relationship implies correlation, but the reverse is not true.
Perhaps this will make sense to you if I explain the logic a bit more: It does not make sense to construct causal hypotheses about how two variables are related unless there is evidence that they are, in fact, related. So it is only legitimate to construct a causal hypothesis after the null (zero) statistical hypothesis has been excluded. So a null hypothesis could be zero change in temperature over time (excluded). Or zero statistical relationship between atmospheric CO2 concentration and temperature (excluded).
Once the null (zero) hypothesis has been excluded, it is gone forever--there is not a new, causal null hypothesis. All causal hypotheses (whether you chose to call them "natural" or not) are subject to the same fundamental criteria for scientific validity. There is a fundamental sanity-checking discipline that real scientific models are subjected to: are the statistical tests carried out in a way that is not subject to bias, or cherry picking? Has the hypothesis been formulated as a mathematical model that generates predictions that can be tested by observation? Is the proposed model consistent with the results of "natural experiments" such as volcanic eruptions? Is it consistent with observations of atmospheric gasses and climate conditions on other planets? Etc., etc.
What specific physical mechanism of climate change that has operated in the past do you imagine to be responsible for any other period of historical warming? Have you cataloged all of those possible mechanisms? Have you excluded all of those possible mechanisms?
This is far too large at topic to summarize in a blog post. Suffice it to say that many different physical mechanisms have been considered both with respect to past and current climate change--orbital changes, changes in solar activity or radiance, release of greenhouse gasses from different sources, etc., etc. You
Slashdot Mirror
It is an irreversible agonist that binds to nicotinic acetylcholine receptors and first activates then blocks them. At high doses it will paralyze muscles. At these low doses it would more likely act by interfering with cognition. Because it is irreversible, it likely has a cumulative effect.
No, the Biblical literalists are generally referred to as "young-earth" creationists. But it is still considered creationism if you believe that species were created in something very close to their current forms, as opposed to evolving from a common ancestor.
And substitution of 65:35 HFCS in place of sucrose (assuming equal sweetening and equal consumption of sweetened foods) would result in essentially no increase in fructose consumption, because 65:35 HFCS is about 39% sweeter than sucrose.
You seem to be experiencing a failure of reading comprehension. Which part of "None of the projections of climate science include what I would characterize as "catastrophic" global warming." did you fail to understand?
We seem to have another failure of comprehension here. Nobody is suggesting that the process of checking your work by doing the math magically causes your theory to be true. Rather, it is the way in which responsible scientists catch errors in their own ideas. Coming up with excuses for not doing the math to check his ideas is the mark of a crank, not a scientist.
It has been previously explained to you why this is nonsense. Try reading it again: "Surely, if you actually think this through, you will realize how ridiculous this is. If your living room is uncomfortably wrong, do you imagine that you can determine, solely by looking at the thermometer, whether it is too warm because it is a hot day--or because you left the heat on? No, you gather other information; you step outside to check the weather, you check the thermostat, you check whether you left the oven on." When there is "natural" climate change, there is other evidence of a change in "natural" mechanisms--a change in solar radiance, for example. Again, this is the sort of error that you fall into if you don't bother to "check your work" by mathematically modeling the physical mechanisms (whether you choose to call them "natural" or "unnatural").
Considering how little you obviously know about science, your instruction to scientists regarding "how science works" carries little weight. In fact, science virtually always progresses by competing models (in the unlikely event that you actually wish to educate yourself, you might consider reading some Thomas Kuhn). Remember: "All models are wrong; that's why we call them models." So nitpicking somebody else's model does not make your model right. No model is perfect; you have to show that their model is wrong in a way that matters. How do you do that? You construct a better model, and you show that your improved model makes better predictions. And here is where the climate science "skeptics" have so miserably failed.
He's right. It's pretty hard to come up with a plausible biochemical scenario in which such a modest difference in fructose content makes a big difference. Which is why the difference between HFCS and sucrose is not likely to be a basis of differences in health between Canadians and Americans. On the other hand, if Canadians consume substantially less sugary drinks overall than people in the U.S., that could be an explanation.
But sucrose is 50% fructose.
It's still pretty hard to imagine dramatically greater adverse effects with only 15% more fructose than sucrose. That would be an extraordinarily steep dose-response curve. And the actual increase in fructose consumption would be less than that, because fructose is 73% sweeter than sucrose, so with higher fructose content, less is needed to achieve a given amount of sweetness.
Seems unlikely; the commonly used form of HFCS contains about half glucose and half fructose. So does cane sugar. I suppose it could be an overdose of sugar in all forms, or of fructose whether derived from cane sugar or HFCS. Do Canadians consume as much sugary drinks as Americans?
Fiddling with the cloud feedback also does not eliminate the prediction of serious global warming that is likely to cause financial and human costs that far exceed the cost of mitigation. "Catastrophic" is a vague term. None of the projections of climate science include what I would characterize as "catastrophic" global warming. On the other hand, the greater uncertainties are at the upper end, particularly with respect to consequences. For example, biologists are not predicting that global warming will result in the emergence of a crop disease that will wipe out a major component of the world's food supply supply, resulting in the starvation of 10% of the world's population, but I expect that you will find few willing to assure you that it's beyond the realm of possibility.
I see. So in your mind, there are no physical mechanisms responsible for "natural" climate change that could be studied and mathematically modeled. It has nothing to do with solar radiance, cloud feedback the earth's orbit, volcanic eruptions, or shifts in ocean currents. It's just....magic.
Remember, mathematical modeling is the major ways that scientists "check their work," to see if their ideas actually make sense, or if they are fooling themselves. So if somebody tells me that they believe that the modern warming is due to something other than CO2 ("cloud feedback" for example) and I ask them, "Have you checked your work mathematically? Is your theory consistent with known data?" and they answer, "No, I haven't bothered, because I'm really, really sure that I'm right, and besides, cloud feedback is natural, so I believe that I should be exempt from having to check my work," I'll conclude that the guy is not a real scientist, but just a crank, and give his views little credence.
Well, that's all very interesting, but I'd be more convinced if you'd checked your work, and shown mathematically that it was possible to develop a physical model in which CO2 is not a strong climate driver, without losing consistency with the known climate data.
So first we have a straw man--no scientist has claimed that modern temperature records are unprecedented, only that the natural causes that have produced high temperatures in the past are not present today. And then you fall back on your old "null hypothesis" mistake. Remember, the null hypothesis must be a zero statistical correlation hypothesis. So what are you supposing to have zero correlation? Temperature and time? Temperature and CO2?
When examined over the period of time that statistical analysis indicates is necessary to be able to reliably detect the trend expected from global warming, it is highly significant. If you are asking, "But isn't it possible that global warming stopped for some unknown reason 9 years ago, or last week, or 10 seconds ago?" I will replay, "Absolutely. But absence of evidence is not evidence of absence."
Once I get past the visceral repugnance I (like most scientists) feel for dishonesty, I can't help but admire how well-crafted some of these climate science talking points are. It is quite clear that they were crafted by people who themselves understood the science, and who carefully crafted a set of fallacious pseudoscientific arguments designed to deceive people with superficial understanding of science. Of course, they know that it is not possible to get global warming to go away by fiddling with the magnitude of feedback. They've tried, and failed. So how do they get the public to ignore the fact that they don't have a model, whereas the real climate scientists have subjected their own ideas to the scientific "sanity checking" of mathematical modeling. They spread the meme that they don't need to come up with a physically realistic model because their model is the "null hypothesis" (which sounds stupid if you known what a null hypothesis actually is, but most people don't, they think--like you--that it is just means some kind of default model). And they also spread the meme that these models can be "curve fitted" to say anything you want. Of course, that's not true--if it were, they would have their own, competing mathematical model that did not predict global warming--but the public are mostly unfamiliar with working with physical models, and imagine that it is something like fitting to an arbitrary mathematical formula. Of course, anybody who has actually worked with physical models knows this to be quite false. Very often, it is simply impossible to get such models to do what you wish, because physical models are constrained by physical principles--for example, if you want a model in which increased CO2 causes warming, but warming doesn't increase CO2, you run up against basic physical principles like conservation of energy.
Surely, if you actually think this through, you will realize how ridiculous this is. If your living room is uncomfortably wrong, do you imagine that you can determine, solely by looking at the thermometer, whether it is too warm because it is a hot day--or because you left the heat on? No, you gather other information; you step outside to check the weather, you check the thermostat, you check whether you left the oven on.
Now this is another very clever deception. Given any trend with random statistical noise riding on top of it, it is always possible to find a period of time short enough so that the noise obscures the trend. Of course, it is possible to examine the statistical properties of the noise, and determine what period of time you have to examine to be able to reliably detect a trend in the presence of the noise. This has been done for temperature measurements, and the required interval turns out to be a couple of decades or so. So we are deluged with claims that "the earth has not warmed" (or, in this case "the stratosphere has not cooled") over the last decade. Dishonest as hell, but very clever, because the public does not have a sophisticated understanding of statistical determination of trends.
In a word, no. It is more the rule than the exception in physical processes for causality to be able to run in either direction, depending upon boundary conditions.
The reason that they are not sure of the sign is that the evidence indicates that net cloud feedback is close to zero. When a value is close to zero, it is hard to determine whether it is a little bit negative or a little bit positive. And no, nobody has managed to come up with a model that creates the modern warming through cloud feedback and that is at all consistent with the known climate records. There are plenty of published climate models. Feel free to try.
This is false. While models vary somewhat in their predictions due to differences in cloud feedbacks, every model that is consistent with the known climate record predicts statistically significant warming in response to the massive increase in CO2 due to burning of fossil fuels. Oh, by the way, "eyeballing" error bars is not a valid way to judge statistical significance--you must use the appropriate mathematical test.
You can't tell much of anything from a graph in which the axes are unlabeled. Warming arising from "natural" sources can be distinguished because "natural" does not mean magic. Natural mechanisms of warming have effects other than just an increase in average temperature. For example, warming due to an increase in solar output predicts that the upper atmosphere will warm more than the lower (in fact, the upper atmosphere has cooled). It also predicts that the warming trend will be greater in the day than at night (it's the other way around). Etc., etc.
Correct. That's what the physics predicts. Like many physical processes, it can run in either direction, depending upon circumstances (i.e. what initiates the warming). You are repeating one of those classic climate "skeptic" talking points that only sounds plausible to people who are ignorant of science. Real scientists just roll their eyes when they hear nonsense like this, because basically it boils down to, "Climate science must be wrong because over history, temperature and CO2 have behaved in exactly the way that the model predicts." To be fair, however, many of the standard climate "skeptic" talking points were carefully designed by lobbyists (as has been so thoroughly documented by historian Naomi Oreskes in "Merchants of Doubt") who specialize in creating specious arguments specifically designed to fool people with a superficial knowledge of science. You are hardly the first person to be taken in by this one, as foolish as it sounds to actual scientists.
Medievalist fantasy was basically invented by the British. Tolkien, who pretty much invented the genre, was British. So was C.S Lewis. If you're really pedantic, perhaps you might prefer to trace it back to Spenser or Malory. But they were British, too. I think it's kind of charming that a video of a modern fantasy by an American author retains this nod to the origins of the genre.
The part you missed is "first step." *sigh* I've explained this to you about 3 times already, but maybe the 4th time is the charm. Probably not, though. People seem to get very attached to crank notions, and the idea of a causal null hypothesis is certainly one of those.
The first step in establishing whether there is a causal connection between two variables is excluding the null hypothesis of no correlation. After all, if two variables are not correlated, it makes no sense to construct a causal hypothesis of how they are correlated. Once the null hypothesis has been excluded, it is then considered legitimate to construct causal hypotheses (all of which are considered on an equal basis, and must satisfy the usual criteria for a valid hypothesis (do I really need to repeat these?)
Ah, another standard crank talking point: "If there is something that is not perfectly understood, then nothing is understood."
If you think that cloud feedbacks can possibly account for global warming, then produce a model, and show that it can provide a reasonably good fit to the known climate record, and can also explain the modern warming. After all, many climate science models are publicly available. So plug in whatever cloud feedback you please, and show that the model can account for the difficulty. While do you imagine that nobody has done this by now? After all climate science "skeptics" love to clutch at the straw of cloud feedbacks. Actually, real climate scientists have tested models with different cloud feedbacks. As explained in the IPCC section that you quote-mined (but I suspect, did not actually read or comprehend) different cloud feedbacks have been modeled. But while uncertainty about cloud feedbacks creates some uncertainty about the magnitude of future warming, it doesn't make it away.
But think about this: clearly, global climate has been warming. So if this is due to some "natural" mechanism involving cloud feedbacks, that implies that cloud feedbacks have been changing. So what about clouds is changing? It doesn't seem too likely that the physics of water vapor is changing, does it?
So according to you, if the interaction between temperature and atmospheric CO2 displays the well understood dynamics of a positive feedback (i.e. if you increase one, the other will follow, and you can get it to go either way depending upon where you start), that means that it explains everything? Or are you just in a snit that something that you thought was a problem with climate science is actually a prediction? In fact, it would be a problem for climate science if it didn't work that way.
There are well established statistical methods of extracting long-term trend from short-term noise such as fluctuations in water vapor. As these fluctuations ride "on top of" the long-term trend, it is plausible that the rising trend could lead to more extreme weather events, although how much that is happening at the present time remains a challenging statistical question.
There have already been numerous tests of modern climate theory, which could have potentially have falsified the theory. For some of the tested predictions of climate science, see here. Indeed, every time a volcano erupts, it is another test of climate theory.
:
I see that you have found other sources that confirm what I told you: the null hypothesis is the default hypothesis of no correlation ("no relationship" is just another way of saying "no correlation") between two variables (i.e. one variable does not change in any consistent manner when the other one does). I note that although you have clearly been looking, you have found no example of a null hypothesis that mentions causality. You won't, because "null" (which means "zero") hypotheses are exclusive to statistical correlation, and correlation is incapable of addressing questions of causality. So your special pleading that "natural variation as a cause of climate warming" should be regarded as a null hypothesis, and therefore exempted from the normal criteria applied to competing scientific hypotheses (e.g. to be based on a physically realistic mechanism, to make falsifiable predictions, and to be consistent with the known data) is a non-starter.
Yes, climate scientists have considered every physically realistic mechanism that could potentially account for the rise in temperatures, including the ones that were likely responsible for past episodes of (what you call "natural") climate change. If you think you have a new one, then provide the hypothesis, and show evidence that it is both consistent with the known climate record and acting today.
Sorry, but this kind of clutching at rhetorical straws is silly. There is no relationship whatsoever between the physical mechanisms responsible for climate change and the economic and psychological mechanisms that determine movements of the stock market. Here's one rather obvious difference: if somebody comes up with a model that accurately predicts movements in stock prices, then people will change their investments based upon those predictions, which will alter the movements of the stock market in such a way that those predictions are no longer valid. Fortunately, we don't have to worry that the physics of climate will alter in response to what we know about it.
It predicts that increases in CO2 can either lead or follow warming, depending upon whether the warming is initiated by release of CO2 or by some other factor, such as a change in solar output. This is very elementary, and you would already know it if you'd bothered to learn anything about the actual science instead of deriving everything that you think you know from crank web sites.
I'm not sure that it's worth the trouble of trying to teach you anything about basic, since your previous comments have suggested that you actively resist learning anything that might challenge your fixed opinion regarding climate science (e.g. you are highly familiar with climate "skeptic" talking points, yet you somehow have managed to learn almost nothing about the actual science). But I'll give it a shot.
A few key points to remember if you don't want to sound foolish when talking to actual scientists:
1. Statistics addresses questions of correlation, not causality.
2. The "null hypothesis" applies solely to statistical significance testing, and thus relates only to correlation, never to causality.
3. Hence, any statement with the word "cause" in it can never be a null hypothesis.
Now let's consider as an example one of the series that you came up with before.
a) 1, 2, 3, 4, 5, 6, 7, 8, 9, 10
b) 1, 1, 1, 1, 1, 1, 1, 1, 1, 1
Now it is important for you to remember that we are discussing statistical measurement. So you can't regard this as a mathematical series; you must regard them as measurements of statistical variables that have some "true" value (which you can estimate, but which you cannot in general know with perfect precision) plus some amount of random error or variation. So you do not expect the measured value to be the same every time you measure it.
So if you ask if a variable is "constant," or shows "no change," you aren't really asking it is the same every time you measure it, because for a statistical variable the answer to that is automatically "No". Rather, you are asking a question about correlation. Does the measured value of a variable depend upon the measured value of some other variable? For example, you may ask whether it shows no change over time, which in statistical parlance is equivalent to the statement that the measured value has no correlation with time. Or you may ask whether it shows no change from place to place, in which case you are asking whether its measured value shows no correlation with some measure of location. Or you may ask whether it shows no change when some other variable changes, in which case you are asking whether there is no correlation between one measured variable (e.g. temperature) and another (e.g. CO2). All of the above are examples of null hypotheses. What makes them null hypotheses is that in all cases, the hypothesis is that one variable does not change with the other--or more technically, that the correlation does not differ from zero more than would be expected from the random variation in both parameters.
Now your Series clearly shows no significant correlation; in fact the correlation is precisely zero. That doesn't mean that there is no correlation because the "null" or zero hypothesis cannot be proved, only disproved. Indeed, there is an obvious problem, because parameter (b) shows no variation at all. This could mean that it has not been measured accurately enough to detect a correlation. Perhaps a more accurate measurement would give values like 1.01, 1.02, 1.03, 1.04, 1.05, 1.06, 1.07, 1.08, 1.09, 1.1 -- which would new highly correlated. So the most we can say about the null hypothesis is that we cannot reject it based on the available data. Once you reject the null hypothesis, you may then consider the reasons why the variables in question (e.g. temperature and time) are correlated. At this point, all hypotheses, whether you choose to call them "natural"
Actually, most modern 3D technology does not require you to keep your head perfectly level. Older 3D glasses that used linear polarization showed crosstalk if you tipped your head but this is not the case with the modern technology. "Active" (shutter) glasses (the somewhat clunky ones) work perfectly well with moderate tips of the head, although your brain gets confused if you tip your head completely sideways (because the parallax is not in the direction that your brain expects from the position of your eyes). Most modern passive 3D systems use circular polarization, which is similarly insensitive to head angle.
So asking the question, "What kind of reform are you looking for? " is putting words in your mouth? That is ridiculous hyperbole. And I can't help noticing that you did not even answer the question. The notion that a having having limitations on the use of trademark--which under current law only applies in a very limited commercial context where some sort of plausible argument can be made that consumers can be misled--constitutes "ownership of a word" is also ridiculous hyperbole.
Actually, yes. Companies are allowed to say whatever they want in a license agreement, whether or not it has any actual legal force. What kind of reform are you looking for? A "license agreement police" that reads every license agreement in the world and levies fines for overly broad license agreements? Do you really think the benefits would justify the cost of all of that extra bureaucracy?
After all, the only thing that actually matters is what the courts will enforce. I'll start worrying if any court, anywhere, enforces such an agreement or trademark against somebody using "book" in a generic manner or similar terms like "phonebook". But they won't. And Facebook wouldn't bring such a suit anyway, because they know they'd lose, and that would undermine their ability to bring the trademark suits that they really care about--the ones against social networking websites that are trying to ride on their coattails by calling themselves "visagebook" or something.
Once again, this is a logical impossibility. Since there is no way to enumerate the predictions of every possible model (indeed, the set is almost certainly infinite), there is no possible prediction that is sufficient to show that a model is true. This is the case for any theory about the real world.
This statement is the hallmark of the scientific crank. You will never hear a successful scientist say this, or a historian of science. Who says this? Evolution deniers. HIV/AIDS deniers. Global warming deniers. It is a one of those deceptive notions that is obviously ridiculous to those who have done actual science, yet can sound plausible to the amateur. After all, it is true in logic, it is true in mathematics, shouldn't it be true in science? And indeed, it would be true in an ideal world without noise, without statistical variation, in which perfectly accurate measurement was possible, one in which it was possible to construct mathematical models at the subatomic level that perfectly reproduced nature and approximations were unnecessary. But of course, we live in the real world in which none of this is true. What a scientist will tell you is that all models are wrong. All models are imperfect representations of reality. That is why we call them models. But as was articulated perhaps most clearly by Isaac Asimov some are less wrong than others. The advance of science has been through a progression of successively less wrong models. It is very easy to nitpick somebody else's model, and pretty much worthless. All models are wrong. The question is whether it is wrong in a way that matters. To show that it is, you need to offer a better one--one that makes better predictions. Predictions are also important because there are some subtle statistical fallacies that it is possible to fall into (particularly for amateurs, but even experienced scientists can sometimes be tripped up) when you examine data after the fact (you'll see a lot of that on WUWT). This is why prediction is the gold standard of science. If you think that the current model is wrong, and wrong in a way that matters, you have to show that when you fix whatever is wrong with it, you are able to make better predictions. When one side of a debate has a clear theory that has been subjected to the discipline of mathematical modeling, with a long track record of making clear, falsifiable predictions, and the other side has nothing but model nitpicking, the former side will have far more credibility with working scientists--which is, of course, why every major elite scientific society in the world, and something like 97% of scientists with any kind of published track record in a relevant field, accept climate theory and its predictions regarding CO2 and climate change.
Zero correlation means that on average, when one variable changes, the other does not. Neither of your series shows statistically significant correlation. Zero correlation with time means zero average change over time. Again, I suggest that you consult a textbook on elementary statistics.
The flight video could have fooled me. But I just couldn't believe it was real when they showed those little tiny batteries.
That's what a "falsifiable prediction" means.
Real science never claims to prove that a theory is correct. Indeed, it is a logical impossibility to prove any theory correct. It is always possible that there could be some other theory that could be equally consistent with the data. In the case of climate science, critics have had a long time to come up with competing theories. The physics upon which climate science is based is published and the equations are well known. Code for many climate models has been made publicly available. Yet no critic has been able to come up with a climate model that is based on valid physics and is equally consistent with the existing climate data and that does not predict a worrisome degree of warming as a consequence of CO2 emissions. This is about as strong as evidence ever gets in science.
Wrong. Pick up a statistics textbook. The null model of correlation is that there is zero average change in one variable when the other one changes. Showing that the change in measured values is significantly greater than would be expected from chance variation constitutes rejection of the null model. Of course, correlation does not imply causality. Statistics is inherently unable to determine causality.
The null model is the zero model--zero change or zero correlation. I didn't pick it; that's how it's defined. You are trying to call a tail a leg again.
Sorry, but this is scientifically illiterate nonsense. No biologist will tell you that natural selection is a null model. Remember, a null model is purely statistical, and therefore cannot be causal. One could of course create statistical null models that would be relevant to the theory of evolution. One such model might be "the genetic code of [insert species] has not changed over time." But the theory of natural selection is a causal model, not a statistical model, and hence can never be a null model.
I can't help being amused when amateurs try to tell scientists how to do science--and get it ludicrously wrong. In this case, "necessary and sufficient" is not even wrong--it's impossible based on elementary logic.
But scientists do talk about what they believe, and they use the word in the common sense (as in "I believe it is about to rain"). They aren't going to stop doing so, and I think that it is not entirely honest to try to pretend that scientists do not form opinions about theories in advance of compelling evidence. Most scientists do, including many very good ones--they are just flexible and ready to change what they believe in the face of new evidence.
I also don't believe (there it is again) that we should be more strict in the usage of "hypothesis" and "theory." This just caters to the notion, held by some amateurs and often exploited by those who try to play semantic games to undermine scientific knowledge, that we have some sort of formal taxonomy of scientific certainty. In actual usage, scientists may sometimes use "hypothesis" to emphasize that an idea is particularly tentative. But the same person may in the next breath refer to the same idea as a "theory." Trying to get formalistic about it leads into all sorts of problems. What level of evidence do you need to upgrade a generalization from "hypothesis" to "theory"? One experiment? Two? Confirmation by two different labs? Three? Trying to get rigid about it is a nightmare, and you'll never get everybody to go along with any kind of strict criteria.
My view is simple. The facts are the data (or reports of data). All scientific generalizations, interpretations, and explanations are theory. Some theories have a lot of evidence to support them, some have very little, and some are known to be frankly wrong. We may occasionally note that we think that a theory is particularly tentative by calling it a "hypothesis," but we don't have a "theory nomenclature rating board". Every scientist makes his or her independent judgements regarding level of evidence
As for "law," the modern usage doesn't really relate to level of evidence at all--it's become pretty much synonymous with "rule of thumb," meaning a formula or algorithm that is simple, useful, and convenient, whether or not it is exactly correct. We aren't going to stop referring to Newton's equations of motion as "Newton's Laws" (even though we know them to be not quite accurate). And we're never going to start referring to Einstein's equations of motion as "Einstein's Laws" even though they have been verified to incredible precision.
Sorry, but this is completely wrong. Einstein's Special Theory of Relativity does not refer to any of them (it is the Special Theory that has supplanted Newton's Laws of Motion). Even Einstein's General Theory of Relativity (which is more accurate than Newton's Law of Gravitational attraction, but is clearly not a complete theory of gravity) predates these and does not require them.
Also utter nonsense. Special relativity has been tested in multiple ways both in the absence of atmosphere here on earth and with respect to phenomena in outer space. No exceptions have been found anywhere.
Who's touting them as "Law?" As I noted before, the way the word "Law" is used by actual working scientists is pretty much equivalent to "rule of thumb" -- a simple formula that may or may not be entirely accurate, but that is easy to apply and a good enough approximation to the truth to be broadly applicable. Einstein's Special Theory has been verified to an amazing degree of precision, but no matter how many decimals of precision its predictions are verified to, I can guarantee that scientists will never start calling it a "Law" -- it may in fact be absolutely true, but at lacks the simplicity and convenience of Newton's (incorrect) Laws of Motion.
As a scientist, this statement makes no sense to me. I don't know of any scientist who would claim that science is the answer to everything. On the other hand, denying scientific knowledge has historically been the answer to nothing.
All of the numerous predictions made from climate theory would have falsified the theory if the observations had come out other than predicted: If the nights were not warming more than the days, if the stratosphere were warming rather than cooling, etc., etc. So far, the theory has held up to every one of the numerous challenges
I'm glad to see that you now seem to have belatedly gained some comprehension of what I tried to explain to you a couple of posts back:
' "Null hypothesis" has a very specific meaning within statistics. "Null" means "zero" and it denotes that the null hypothesis is always the hypothesis of zero difference or zero correlation. '
Surely it is not at this point necessary to belabor the issue by repeating the whole definition every time I mention it? After all, "no correlation" merely means zero average change in one variable when the other one changes. Please feel free to substitute my previous more lengthy explanation if you find my shorthand too confusing.
Sorry, here you start to screw up again. A null hypothesis cannot be causal, because the concept of a null hypothesis is exclusive to statistics and statistical correlation, and statistical correlation is logically incapable of establishing causality. A causal relationship implies correlation, but the reverse is not true.
Perhaps this will make sense to you if I explain the logic a bit more: It does not make sense to construct causal hypotheses about how two variables are related unless there is evidence that they are, in fact, related. So it is only legitimate to construct a causal hypothesis after the null (zero) statistical hypothesis has been excluded. So a null hypothesis could be zero change in temperature over time (excluded). Or zero statistical relationship between atmospheric CO2 concentration and temperature (excluded).
Once the null (zero) hypothesis has been excluded, it is gone forever--there is not a new, causal null hypothesis. All causal hypotheses (whether you chose to call them "natural" or not) are subject to the same fundamental criteria for scientific validity. There is a fundamental sanity-checking discipline that real scientific models are subjected to: are the statistical tests carried out in a way that is not subject to bias, or cherry picking? Has the hypothesis been formulated as a mathematical model that generates predictions that can be tested by observation? Is the proposed model consistent with the results of "natural experiments" such as volcanic eruptions? Is it consistent with observations of atmospheric gasses and climate conditions on other planets? Etc., etc.
This is far too large at topic to summarize in a blog post. Suffice it to say that many different physical mechanisms have been considered both with respect to past and current climate change--orbital changes, changes in solar activity or radiance, release of greenhouse gasses from different sources, etc., etc. You