Another good hint about deceiving people. Unless you want to go purely for gut feeling, don't accuse your discussion partner of dishonesty before at least the second paragraph.
Based upon your remarks, it is clear that you are not a scientist and have never studied statistics, so I imagine that you are deceived rather than deceiving.
Dear God man, do you actually believe this ?
Do I believe that I know the definition of the standard deviation? Yes, I do.
How random is the weather... well let's compare it with an actual more-or-less random thing everyone knows... so how much does the sky look like static on television.
A sufficiently detailed analysis could model static on a television, or the role of the dice, in a deterministic fashion. Almost everything modeled statistically as random processes (with the possible exception of quantum phenomena) is deterministic at some level. But many deterministic phenomena are also chaotic, and can be usefully modeled over longer time scales as random processes. Weather is one good example. Undoubtedly, given enough information, one could model the physical processes that dictate weather "noise," such as the precise timing of volcanic eruptions. But lacking that sort of information, one can usefully model weather statistically over the longer term.
As indicated above, the climate doesn't actually obey the law of large numbers - but we act like it does anyway.
It is not climate (deterministic long term changes in things like temperature controlled by the earth's energy balance), but weather (short term fluctuations due to chaotic processes) and measurement that is modeled statistically
*sigh* this is not true -at all- we do not simulate virtual earths to detect a warming trend.
Sorry, but you are mistaken. Every run of a statistical model of any physical process, whether it is weather noise or the spin or a roulette wheel can be regarded as a simulation out of a population of "virtual earths."
Keep in mind that these physical models do not model the complex social and economic factors that dictate the time course of atmospheric CO2 increase. As a result, they do not predict the trajectory of global temperature change—they predict the temperature change given a specific trajectory of CO2 increase. For this reason, the term "projection" is commonly used instead of "prediction" to describe the model output. So to evaluate the predictions of a climate model, it is necessary to plug in the actual CO2 values. When this is done, Hansen's 1988 modelstill looks quite good. Of course, as historically important as Hansen's pioneering work was in demonstrating that rising CO2 implies a problematic rise in global temperatures (a prediction that has been dramatically confirmed since then), there has been considerable scientific progress in quarter century plus since then, and he is no longer the only scientist doing this kind of work. There are multiple competing models from independent research groups, although they agree qualitatively in predicting that rising CO2 will result in climate change sufficient to cause multiple severe problems. For a full discussion of the modern state of the art in climate models, see the IPCC reports.
If scientists cannot agree on what the temperature really is and what it is doing, how is it possible to believe any prediction they make?
Duh. Scientists never unanimously agree about anything. But if you were going to pick somebody to set against the 97% of climate scientists and the National Academy of Sciences (and just about every elite scientific society in the world) who agree that the planet is warming, Spencer is a pretty weak reed. After all, this was the supposed expert on satellite measurements who insisted that the satellite measurements proved that the planet was not warming--until it was shown that he had failed to correct for orbital decay
Danish Astronomer Roemer was the first to assert that light did indeed have a finite velocity, even though the prevailing majority opinion (politically correct) at the time was that light travelled instantaneously from place to place. It took over 50 years before the scientific community as a whole finally admitted that Roemer was correct in his observations.
The speed of light? That's really the best you can come up with? A difference between "infinite" and so close to infinite that it made no difference for any practical purposes of the time? I'd say that falls into the category of a "tweak," just as Einstein's theories tweaked Newton's theories of motion and gravitation.
Roemer's work, which relied upon some difficult and rather arcane astronomical observations was never even published. Nevertheless, he did interest a few other scientists in developing more convincing ways of measuring the speed of light, which when published rapidly persuaded other physicists.
You don't think that this sort of remark... might be part of the reason global warming has such a bad rep.
I think that it is certainly true that early statisticians did not anticipate the extent to which their methods might be misused by dishonest people to deceive the public. If they had, they probably would have made a better choice of jargon than the term "statistical significance," because in common parlance "significance" is almost synonymous with "importance," whereas it's usage in statistics comes closer to "reliability." Certainly, no honest scientist would ever make a statement like "there is no statistically significant global warming," because it misleadingly suggests that absence of evidence (for global warming over a particular time scale) is evidence of absence. An honest scientist, if trying to argue that there was no warming, would state confidence limits on the trend. Why wasn't that done here? Because if you state confidence limits on the temperature trend over that time period, it turns out to be equally consistent with no warming and with warming even greater than climate theory predicts.
Then you test a hypothesis against that data. This does not result in "a warming trend" or "a cooling trend" it results in 2 numbers : chance that the temperature has risen -> p, chance that the temperature has not risen -> !p
Wrong! But unless you have actually studied statistics, it is easy to get this sort of mistaken notion, particularly when there are well-funded parties working actively to promote this sort of misunderstanding because it serves their own financial or political purposes.
A proper statistical trend analysis will result in a best estimate of the trend (which may be warming or cooling) given the available data, as well as confidence limits on that trend--a measure of how much the estimated trend would be expected to vary if that observation could be repeated (i.e. if you had a population of earths with a similar climate trend but with different weather, each of which could be identically sampled over the same time period using the same methodology). The measurements do not yield a probability that the temperature has risen and a probability that the temperature has not risen--they yield an estimate of how likely repeat measurements are to differ from the current measurement by a particular amount.
If 19 times out of 20, repeat measurements would be expected to yield a trend greater than zero, then as a kind of shorthand, statisticians say that the trend is "significantly" greater than zero (this is equivalent to saying that the 95% confidence limits on the trend do not include zero). But if that is not the case (let's say if repeat measurements would be expected to yield a trend greater than zero only 18 times out of 20), this is not evidence that there is no warming--it merely indicates that the data is not adequate to resolve the question.
By contrast, the canonical example of an exact science, physics, only considers a measurement reasonable when it passes a significance of six sigma (which is 99.9999998027% certain). That is *NOT* enough to declare something the truth within physics, the only thing that is enough for that is a mathematically consistent theory that passes repeatable experiments (and even then it usually takes 10 years or more).
Some types of studies in physics require multiple comparisons. Look at it this way: a probability of 1 in 20 of a conclusion being in error due to chance is pretty good if you are only measuring one thing. If you are doing a thousand measurements, however, that means 50 false positives. If even one false positive is enough to lead to a false conclusion on a matter of import then you need to set a more stringent criterion for statistical significance--like, in this case, 6 standard deviations. However, there is a downside to setting such a stringent criterion. When you minimize
Why did you cherry-pick one particular prediction?
It is axiomatic that "all models are wrong" -- that is, after all, why we call them models. The question is whether they are close enough to predict the things that matter. We know, for example, that Newton's model of motion is wrong--it makes slightly inaccurate predictions about addition of velocities, for example. But it is still accurate enough to calculate the trajectory of an artillery shell. In the case of global warming, the central question--the one that matters--is whether the trajectory of temperature change can be predicted based upon atmospheric concentrations of CO2. And Hansen's model predicted this (it predicted a lot of other details, about the warming as well, but this is the key point).
If you have another model that has made more accurate predictions, trot it out.
Clearly, we can do something about it; you are asking whether we will. Certainly, we have a piss-poor record so far. But assuming that we won't have the will, and thus that there is no point in making the attempt, is certain to be a self-fulfilling prophecy. If it is to be done, it will require a country with the courage, scientific credibility, and global economic clout to assume a leadership role , and set an example for the rest of the world. There are a limited number of candidates.
This is the part where the logic fails for me. It's a form of self-hatred to assume that everything that comes of Man's impact on the environment is necessarily unnatural and therefore bad.
This is an idiotic caricature of the actual concerns. It is more along the lines of "sea level rise resulting in the inundation of some of the most populated and valuable areas of the world is bad." It would be bad if it were due to a natural source, too. Fortunately, since it is a result of human actions, there is actually something we can do about it
But only if we are smart enough to understand what the actual concerns are, instead of indulging our prejudices with stupid caricatures.
The Earth was FLAT The Earth was CENTER OF THE UNIVERSE The Earth was CREATED BY $DEITY
Back in the day before science existed, you mean (and it is doubtful that it was ever the consensus of educated people the the earth is flat; even the ancient Greeks knew that was false).
On the other hand, scientific consensus has historically held up quite well, particularly with respect to major conclusions.
It DOES, however, tell us that the current climate change is (1) within the normal variation of climate, and (2) shoots some large holes in the "anthropocentric" part of the theory.
Strawman! Climate science does not claim the that the current climate change is not within the range of "normal variation of climate," (in fact, it states the opposite of that) but rather that the scientific evidence shows that the natural causes that have produced large variations in the past are not present today.
Especially as the predicted changes from the computer models have not occurred.
This is complete and utter crap. This kind of arrogance is why people are pushing back against you. You've created a theory that, rather conveniently can't be disproven.
Actually, while science is always being tweaked and corrected, I can't think of any scientific consensus this broad that turned out to be wrong in its major conclusions. Can you name one? (no fair going back to medieval times before the modern scientific method was developed).
It is very revealing that so-called "skeptics" of global warming reject the results of studies carried out by multiple different laboratories, using a wide variety of different analytical methods and many different types of data collected from around the globe, but uncritically accept as fact conclusions based upon 3rd hand accounts of agricultural practices in one small region of Europe. Summary and citations of the actual science can be found here
It is by the way, absolutely false that there has been "NO" temperature increase in the past 10 years. In fact, analysis of the data shows a clear upward trend over the past 10 years. The question is whether the increase satisfies the technical criterion of "statistical significance" -- which means showing that there is less than a 5% probability that an apparent increase of that magnitude could occur by random statistical variations. This is a particularly stupid argument, because statistical analysis of climate models (as well as weather trends) indicates that 10 years is too short an interval to reliably detect the predicted global warming trend even if it is real. (Although if you correct for known natural sources of climate "noise," it turns out that it is significant after all. So while we cannot prove that global warming did not end 10 years -- or 10 seconds -- ago, this is not evidence that it has stopped.
Ah, the myth of the debunked hockey stick. The actual facts are these. A paper by McIntyre and McKitrick identified a flaw in a statistical method used in the original pioneering "hockey stick" study, which when analyzing data could produce an artifactual upturn or downturn (with equal probability) when given random data. Although McIntyre and McKitrick never were able to get this artifact to produce an upturn with magnitude close to the reported "hockey stick," global warming "skeptics" (a term that ironically means people who are completely credulous to any argument against global warming, however weak) declared the hockey stick "debunked." Meanwhile, the real scientists did what real scientists do: they developed improved statistical methods that were not susceptible to the artifact, and showed that the hockey stick remained. They analyzed multiple different types of data, and showed that the hockey stick remained. A peer review by the nations pre-eminent independent elite scientific body, the National Research Council of the National Academies of Science, ultimately concluded that the major conclusions of the "hockey stick" papers were correct.
The Wikipedia article is fairly accurate and includes references to substantiate the above.
They were both right. During Newton's time we couldn't measure the speed of light, and there was no reason to presume it had some arbitrary finite value. Newton's theories completely described the measurable universe of the 1600s.
Science presumes that the universe has objective reality. So if two theories make different predictions, they cannot both be right (they can both be wrong, but even then, one can be "more wrong" than the other). If you want to say that Newton did as well as anybody could, given the knowledge and technology of the time, I'll agree. But that does not make him right.
If I ask you what 2+2 is and you say 4, I can't say your answer was wrong after I go back and carefully remeasure the original numbers and get 2.00001
Don't confuse science with math. You can't "remeasure" the value of the integer 2, because its value is a matter of definition, and 2 + 2 = 4 follows from the definition of addition.
Throughout history, ideas have warred it out through the process of open discussion and debate. Right now, this issue is totally Balkanized and neither side is talking to the other. Opening it up to discussion might allow us to get farther than trying to pick on side or the other.
Right. Have open debate in a high school class whether heat is molecular motion or phlogiston. Or whether Einstein or Newton was right. Whether disease is due to germs or evil humours. Or whether the planets revolve around the earth or the sun. Because science is so simple and easy to learn that there's lots of time to spend re-debating questions that were decided (at least as far as science is conceded) many decades ago.
Either way, there will be a concentration at which the inactivator will be ineffective as a drug. This "threshold" value may certainly be lower than is typical for a reversible inhibitor, but it is not infinitely lower.
I don't see that it makes any difference whether the binding or inactivation reaction is rate-limiting with respect to the rate of inactivation. Each binding event has a constant probability of leading to irreversible inactivation, and at any concentration, the fraction of inactivated receptors will increase continuously with exposure time. So the only case in a particular concentration of inhibitor will be ineffective regardless of duration of exposure is when it is so low that the rate of inactivation is small relative to the rate of receptor turnover.
Mizoribine and methotrexate are two that come to mind. You are certainly right about chronic exposure in the bees, and I too would be wary of it. I am just trying to point out that while the binding kinetics of an inhibitor are important, they are probably the least significant factor determining the efficacy of a drug. Take methotrexate, for example. Equilibrium dissociation constants for the human dhfr have been measured in the pico molar range
pM doesn't sound so low to me. I looked up the dissociation rate of methotrexate, and the numbers I found gave a dissociation t1/2 of a minute or so. It's hard to see how that could be at all rate-limiting for a drug with a blood t1/2 on the scale of hours. Bungarotoxin has a dissociation time constant on the scale of hours, but its Kd is in the fM range.
Yet, a typical clinical dose is ~20 mg/m2, which is far more than is needed to completely inhibit all of the dhfr in your body at any given moment, and it must be administered daily. The bioavailability and metabolic stability turn out to be the factors that matter most.
I'll bet a much lower dose would work if you gave it by continuous infusion. I don't think bioavailability is likely to be much of an issue with a lipophylic compound like imidacloprid. It looks like it will get in if you so much as spill it on your skin.
Generally if you are a couple of orders of magnitude below the dissociation constant of a drug for its target, it will have no physiological effect, because this puts the effect of the drug within the range of random variation that biological systems must necessarily be able to tolerate. But an irreversible inhibitor does not have a dissociation constant, and the magnitude of its effect is limited only by kinetic considerations (forward rate constant, turnover of the target, duration of exposure). Such drugs can safely be used clinically, where the exposure parameters have been worked out, and in some circumstances can have substantial advantages (e.g. the use of phenoxybenzamine in treatment of pheochromocytoma). But when they are chronically present (as in this case, where it was apparently a contaminant of corn syrup that the bees were fed with), the net effect is difficult to predict, and could be large. I don't know of any clinically used reversible inhibitors that bind so tightly that dissociation is rate-limiting on duration of action; what examples did you have in mind? There are certainly some toxins for which this is true, such as alpha-bungarotoxin.
With an irreversible inhibitor, the amount of irreversibly inhibited receptor will accumulate with duration or frequency of exposure and can greatly outlast the presence of the agent in the environment. There is no threshold dose below which receptor inactivation does not occur. Recovery will occur after exposure ceases, but will be slow and limited by the rate of production of new receptors. Although there is no equilibrium short of complete inhibition of all receptors, a type of steady state may occur at very low levels of exposure in which the rate of receptor inactivation is balanced the rate of production of new receptors. The latter is very slow, as the half-time of receptors is typically on the order of many hours or even days.
No processing them won't make much difference; processed or unprocessed, sugar is the same once it gets into the bloodstream. There can be some benefit to consuming them with other stuff that can slow absorption a bit so that blood levels of the sugar don't spike so high. But some fruit contain even higher levels of fructose than HFCS, and there is reason for concern that this may not be healthful
Especially with a drug that binds irreversibly, LD50 could be misleading, because the effect will depend on the duration of exposures and may be cumulative with multiple exposures over a period of a few days
Slashdot Mirror
Based upon your remarks, it is clear that you are not a scientist and have never studied statistics, so I imagine that you are deceived rather than deceiving.
Do I believe that I know the definition of the standard deviation? Yes, I do.
A sufficiently detailed analysis could model static on a television, or the role of the dice, in a deterministic fashion. Almost everything modeled statistically as random processes (with the possible exception of quantum phenomena) is deterministic at some level. But many deterministic phenomena are also chaotic, and can be usefully modeled over longer time scales as random processes. Weather is one good example. Undoubtedly, given enough information, one could model the physical processes that dictate weather "noise," such as the precise timing of volcanic eruptions. But lacking that sort of information, one can usefully model weather statistically over the longer term.
It is not climate (deterministic long term changes in things like temperature controlled by the earth's energy balance), but weather (short term fluctuations due to chaotic processes) and measurement that is modeled statistically
Sorry, but you are mistaken. Every run of a statistical model of any physical process, whether it is weather noise or the spin or a roulette wheel can be regarded as a simulation out of a population of "virtual earths."
Keep in mind that these physical models do not model the complex social and economic factors that dictate the time course of atmospheric CO2 increase. As a result, they do not predict the trajectory of global temperature change—they predict the temperature change given a specific trajectory of CO2 increase. For this reason, the term "projection" is commonly used instead of "prediction" to describe the model output. So to evaluate the predictions of a climate model, it is necessary to plug in the actual CO2 values. When this is done, Hansen's 1988 model still looks quite good. Of course, as historically important as Hansen's pioneering work was in demonstrating that rising CO2 implies a problematic rise in global temperatures (a prediction that has been dramatically confirmed since then), there has been considerable scientific progress in quarter century plus since then, and he is no longer the only scientist doing this kind of work. There are multiple competing models from independent research groups, although they agree qualitatively in predicting that rising CO2 will result in climate change sufficient to cause multiple severe problems. For a full discussion of the modern state of the art in climate models, see the IPCC reports.
Duh. Scientists never unanimously agree about anything. But if you were going to pick somebody to set against the 97% of climate scientists and the National Academy of Sciences (and just about every elite scientific society in the world) who agree that the planet is warming, Spencer is a pretty weak reed. After all, this was the supposed expert on satellite measurements who insisted that the satellite measurements proved that the planet was not warming--until it was shown that he had failed to correct for orbital decay
The speed of light? That's really the best you can come up with? A difference between "infinite" and so close to infinite that it made no difference for any practical purposes of the time? I'd say that falls into the category of a "tweak," just as Einstein's theories tweaked Newton's theories of motion and gravitation.
Roemer's work, which relied upon some difficult and rather arcane astronomical observations was never even published. Nevertheless, he did interest a few other scientists in developing more convincing ways of measuring the speed of light, which when published rapidly persuaded other physicists.
I think that it is certainly true that early statisticians did not anticipate the extent to which their methods might be misused by dishonest people to deceive the public. If they had, they probably would have made a better choice of jargon than the term "statistical significance," because in common parlance "significance" is almost synonymous with "importance," whereas it's usage in statistics comes closer to "reliability." Certainly, no honest scientist would ever make a statement like "there is no statistically significant global warming," because it misleadingly suggests that absence of evidence (for global warming over a particular time scale) is evidence of absence. An honest scientist, if trying to argue that there was no warming, would state confidence limits on the trend. Why wasn't that done here? Because if you state confidence limits on the temperature trend over that time period, it turns out to be equally consistent with no warming and with warming even greater than climate theory predicts.
Wrong! But unless you have actually studied statistics, it is easy to get this sort of mistaken notion, particularly when there are well-funded parties working actively to promote this sort of misunderstanding because it serves their own financial or political purposes.
A proper statistical trend analysis will result in a best estimate of the trend (which may be warming or cooling) given the available data, as well as confidence limits on that trend--a measure of how much the estimated trend would be expected to vary if that observation could be repeated (i.e. if you had a population of earths with a similar climate trend but with different weather, each of which could be identically sampled over the same time period using the same methodology). The measurements do not yield a probability that the temperature has risen and a probability that the temperature has not risen--they yield an estimate of how likely repeat measurements are to differ from the current measurement by a particular amount.
If 19 times out of 20, repeat measurements would be expected to yield a trend greater than zero, then as a kind of shorthand, statisticians say that the trend is "significantly" greater than zero (this is equivalent to saying that the 95% confidence limits on the trend do not include zero). But if that is not the case (let's say if repeat measurements would be expected to yield a trend greater than zero only 18 times out of 20), this is not evidence that there is no warming--it merely indicates that the data is not adequate to resolve the question.
Some types of studies in physics require multiple comparisons. Look at it this way: a probability of 1 in 20 of a conclusion being in error due to chance is pretty good if you are only measuring one thing. If you are doing a thousand measurements, however, that means 50 false positives. If even one false positive is enough to lead to a false conclusion on a matter of import then you need to set a more stringent criterion for statistical significance--like, in this case, 6 standard deviations. However, there is a downside to setting such a stringent criterion. When you minimize
It is axiomatic that "all models are wrong" -- that is, after all, why we call them models. The question is whether they are close enough to predict the things that matter. We know, for example, that Newton's model of motion is wrong--it makes slightly inaccurate predictions about addition of velocities, for example. But it is still accurate enough to calculate the trajectory of an artillery shell. In the case of global warming, the central question--the one that matters--is whether the trajectory of temperature change can be predicted based upon atmospheric concentrations of CO2. And Hansen's model predicted this (it predicted a lot of other details, about the warming as well, but this is the key point).
If you have another model that has made more accurate predictions, trot it out.
Clearly, we can do something about it; you are asking whether we will. Certainly, we have a piss-poor record so far. But assuming that we won't have the will, and thus that there is no point in making the attempt, is certain to be a self-fulfilling prophecy. If it is to be done, it will require a country with the courage, scientific credibility, and global economic clout to assume a leadership role , and set an example for the rest of the world. There are a limited number of candidates.
Please cite the part of the IPCC report that predicts that global warming will be "catastrophic"
This is an idiotic caricature of the actual concerns. It is more along the lines of "sea level rise resulting in the inundation of some of the most populated and valuable areas of the world is bad." It would be bad if it were due to a natural source, too. Fortunately, since it is a result of human actions, there is actually something we can do about it
But only if we are smart enough to understand what the actual concerns are, instead of indulging our prejudices with stupid caricatures.
Back in the day before science existed, you mean (and it is doubtful that it was ever the consensus of educated people the the earth is flat; even the ancient Greeks knew that was false).
On the other hand, scientific consensus has historically held up quite well, particularly with respect to major conclusions.
Strawman! Climate science does not claim the that the current climate change is not within the range of "normal variation of climate," (in fact, it states the opposite of that) but rather that the scientific evidence shows that the natural causes that have produced large variations in the past are not present today.
Oh, really?
Completely false. See here for a list of some of the confirmed falsifiable predictions of climate theory. And that includes the big one: predicting global warming before it was evident in the temperature record.
Citation needed. Please provide IPCC report references for the consensus climate science predictions that supposedly have not come true
Actually, while science is always being tweaked and corrected, I can't think of any scientific consensus this broad that turned out to be wrong in its major conclusions. Can you name one? (no fair going back to medieval times before the modern scientific method was developed).
It is very revealing that so-called "skeptics" of global warming reject the results of studies carried out by multiple different laboratories, using a wide variety of different analytical methods and many different types of data collected from around the globe, but uncritically accept as fact conclusions based upon 3rd hand accounts of agricultural practices in one small region of Europe. Summary and citations of the actual science can be found here
It is by the way, absolutely false that there has been "NO" temperature increase in the past 10 years. In fact, analysis of the data shows a clear upward trend over the past 10 years. The question is whether the increase satisfies the technical criterion of "statistical significance" -- which means showing that there is less than a 5% probability that an apparent increase of that magnitude could occur by random statistical variations. This is a particularly stupid argument, because statistical analysis of climate models (as well as weather trends) indicates that 10 years is too short an interval to reliably detect the predicted global warming trend even if it is real. (Although if you correct for known natural sources of climate "noise," it turns out that it is significant after all. So while we cannot prove that global warming did not end 10 years -- or 10 seconds -- ago, this is not evidence that it has stopped.
Ah, the myth of the debunked hockey stick.
The actual facts are these. A paper by McIntyre and McKitrick identified a flaw in a statistical method used in the original pioneering "hockey stick" study, which when analyzing data could produce an artifactual upturn or downturn (with equal probability) when given random data. Although McIntyre and McKitrick never were able to get this artifact to produce an upturn with magnitude close to the reported "hockey stick," global warming "skeptics" (a term that ironically means people who are completely credulous to any argument against global warming, however weak) declared the hockey stick "debunked." Meanwhile, the real scientists did what real scientists do: they developed improved statistical methods that were not susceptible to the artifact, and showed that the hockey stick remained. They analyzed multiple different types of data, and showed that the hockey stick remained. A peer review by the nations pre-eminent independent elite scientific body, the National Research Council of the National Academies of Science, ultimately concluded that the major conclusions of the "hockey stick" papers were correct.
The Wikipedia article is fairly accurate and includes references to substantiate the above.
Science presumes that the universe has objective reality. So if two theories make different predictions, they cannot both be right (they can both be wrong, but even then, one can be "more wrong" than the other). If you want to say that Newton did as well as anybody could, given the knowledge and technology of the time, I'll agree. But that does not make him right.
Don't confuse science with math. You can't "remeasure" the value of the integer 2, because its value is a matter of definition, and 2 + 2 = 4 follows from the definition of addition.
Right. Have open debate in a high school class whether heat is molecular motion or phlogiston. Or whether Einstein or Newton was right. Whether disease is due to germs or evil humours. Or whether the planets revolve around the earth or the sun. Because science is so simple and easy to learn that there's lots of time to spend re-debating questions that were decided (at least as far as science is conceded) many decades ago.
I don't see that it makes any difference whether the binding or inactivation reaction is rate-limiting with respect to the rate of inactivation. Each binding event has a constant probability of leading to irreversible inactivation, and at any concentration, the fraction of inactivated receptors will increase continuously with exposure time. So the only case in a particular concentration of inhibitor will be ineffective regardless of duration of exposure is when it is so low that the rate of inactivation is small relative to the rate of receptor turnover.
pM doesn't sound so low to me. I looked up the dissociation rate of methotrexate, and the numbers I found gave a dissociation t1/2 of a minute or so. It's hard to see how that could be at all rate-limiting for a drug with a blood t1/2 on the scale of hours. Bungarotoxin has a dissociation time constant on the scale of hours, but its Kd is in the fM range.
I'll bet a much lower dose would work if you gave it by continuous infusion. I don't think bioavailability is likely to be much of an issue with a lipophylic compound like imidacloprid. It looks like it will get in if you so much as spill it on your skin.
Generally if you are a couple of orders of magnitude below the dissociation constant of a drug for its target, it will have no physiological effect, because this puts the effect of the drug within the range of random variation that biological systems must necessarily be able to tolerate. But an irreversible inhibitor does not have a dissociation constant, and the magnitude of its effect is limited only by kinetic considerations (forward rate constant, turnover of the target, duration of exposure). Such drugs can safely be used clinically, where the exposure parameters have been worked out, and in some circumstances can have substantial advantages (e.g. the use of phenoxybenzamine in treatment of pheochromocytoma). But when they are chronically present (as in this case, where it was apparently a contaminant of corn syrup that the bees were fed with), the net effect is difficult to predict, and could be large. I don't know of any clinically used reversible inhibitors that bind so tightly that dissociation is rate-limiting on duration of action; what examples did you have in mind? There are certainly some toxins for which this is true, such as alpha-bungarotoxin.
With an irreversible inhibitor, the amount of irreversibly inhibited receptor will accumulate with duration or frequency of exposure and can greatly outlast the presence of the agent in the environment. There is no threshold dose below which receptor inactivation does not occur. Recovery will occur after exposure ceases, but will be slow and limited by the rate of production of new receptors. Although there is no equilibrium short of complete inhibition of all receptors, a type of steady state may occur at very low levels of exposure in which the rate of receptor inactivation is balanced the rate of production of new receptors. The latter is very slow, as the half-time of receptors is typically on the order of many hours or even days.
Actually, it sounds like farmers may just need to stop feeding their bees with corn syrup
This is a neurotoxin. Corn does not have neurons, so all corn will be equally (and probably highly) resistant
Wrong. Even ordinary corn will not be harmed by this pesticide. It's a neurotoxin, and corn does not have nerve cells
No processing them won't make much difference; processed or unprocessed, sugar is the same once it gets into the bloodstream. There can be some benefit to consuming them with other stuff that can slow absorption a bit so that blood levels of the sugar don't spike so high. But some fruit contain even higher levels of fructose than HFCS, and there is reason for concern that this may not be healthful
Especially with a drug that binds irreversibly, LD50 could be misleading, because the effect will depend on the duration of exposures and may be cumulative with multiple exposures over a period of a few days