On Tue, May 7, 2013 at 1:36 PM, Jed Rothwell <jedrothw...@gmail.com> wrote:
> > > Null results in some fields far exceed positive results. Beaudette pointed > to the early experiments cloning mammals. He said it took about 1000 > attempts for one success. I have pointed to the number of collisions > required to detect a few examples of the top quark. I believe the tests ran > for about a year and there were billions of collisions per second. That is > a very small success ratio but no one claims the top quark does not exist > for that reason. > > Reproducibility doesn't necessarily mean you get the same result every time. That's theoretically impossible in quantum mechanics. Reproducibility means that if you do the same experiment, the results will be the same on a statistical basis. For example in Rutherford's famous experiment, not every alpha was reflected back from the gold foil. But anyone could do the experiment as prescribed, and after enough time, would get the same *distribution* of scattering angles. Examples of (early) transistors or cloning are often cited as experiments that only have a statistical reproducibility. But cold fusion does not even have this. If the transistor or cloning recipes are followed, the success rates are the same within experimental error. But with cold fusion, they aren't in the same ballpark. If you made a transistor that worked, anyone could make it work, but if you get a LENR cathode that works, it only works in one lab, with one experimenter. Of course, even statistical reproducibility is not needed if a single observation is sufficiently indisputable, like the Wrights' flight in 1908, or the explosion of a fission bomb, or even levitation of a superconductor. As long as the effect is reliable enough so that it can be widely demonstrated, or so that anyone can follow a prescription and with suitable patience see the effect. But cold fusion has neither a reliable indisputable demonstration (at any statistical level), nor a more subtle, but statistically reproducible effect. And if it were real, there *should* be indisputable demonstrations. Energy density of GJ/g should be as obvious as a light bulb turning on.There are apocryphal stories of heat after death and things melting, but they can't be taken on tour to convince the world, no matter how many cells you use. Otherwise, the DOE panel would have been presented with a suitable demonstration. And there is no inter-lab reproducibility of any kind. That's why in 2008, McKubre wrote: "… we do not yet have quantitative reproducibility in any case of which I am aware.", and " in essentially every instance, written instructions alone have been insufficient to allow us to reproduce the experiments of others." Miles has said the same thing. To most scientists, this means there is no reproducibility in the field. In 1994, McKubre claimed to have all the criteria to get a positive result every time the criteria were met, but a few years later the Toyota IMRA lab in Japan reported negative results in 27 of 27 electrolysis cells. Evidently, they could not follow McKubre's recipe. That's not surprising since in 1998, McKubre wrote: "With hindsight, we may now conclude that the presumption of repeatable excess heat production was premature…" , given that only 20% of his cells worked. That's consistent with this quotation of an executive director at the Office of Naval Research, who had funded experiments by Miles and others (from a NewScientist article in 2003): "For close to two years, we tried to create one definitive experiment that produced a result in one lab that you could reproduce in another,” Saalfeld says. “We never could. What China Lake did, NRL couldn't reproduce. What NRL did, San Diego couldn't reproduce. We took very great care to do everything right. We tried and tried, but it never worked." " Lately, you've touted a 2012 conference report from NRL, as irrefutable, but it claims the energy of a drop of gasoline, and admits a dismal 5% reproducibility. Only in cold fusion are steadily worsening (unrefereed) results cited as the latest irrefutable evidence. The reality is that there is not a single experiment in the field that a qualified scientist can perform with expected results (other than null results), even on a statistical basis. There is not a single nuclear reaction that people in the field can agree is occurring. There is not a single credible example where the energy from cold fusion can power the experiment itself, let alone the world. > It is ironic that some people in high-energy physics have demanded a > higher success rate from cold fusion, and they have condemned it because > they claim it is based on statistical proof of existence. That is not the > case. Their own research is often based on statistics. > > Their research is always based on statistics. There is no received wisdom about what level of reproducibility makes a phenomenon credible. Scientists use experience and their wit and make judgements. The situation in high energy experiments is very different from cold fusion experiments. Their statistical results would be the same every time they do the experiment, or they'd reject the results. But cold fusion results are no better than random. And the judgement of the smartest and most experienced scientists is that the cold fusion evidence falls far short of being credible. That's just the way it is, and in spite of a huge preference for it to be real, as was obvious in 1989. > Some skeptics claim that a low success rate or an effect that is difficult > to replicate should thereby be considered pathological science. As I said, > this would expand the definition of pathological science. Langmuir never > said anything like this. > The definition is simply science of things that aren't so, and Langmuir described it as an area of research that simply will not "go away"—long after it was given up on as 'false' by the majority of scientists in the field, and that certainly fits cold fusion, although of course believers will immediately label those who have given up on it as not being in the field and therefore tautologically convincing themselves it's not pathological. Thus self-delusion permeates their entire thinking process. Langmuir may have coined the term, but it has taken its own life, as English words do, and acquired a meaning by usage. The characteristics he listed were never intended to be written in stone, and to me, the best indication is not the difficulty in seeing an effect, but the failure of the effect to become more manifest, of the field to make progress, and of the research activity, as measured by refereed papers to stagnate or gradually approach zero (before anything has been learned). Cold fusion fits.