User talk:Abd/cf

Reproduciblity

The following texts have been changed:[1]

• Irreproducible Difficult to reproduce.
• They were unable to replicate their process publicly after the press releases.
• The effects are not reproducible [difficult to reproduce] under independent verification, or even consistently reproducible at all. Even among cold fusion researchers, reports of substantial excess heat are rare, as shown in the 2010 review. Where excess heat is found, the level of heat produced can vary greatly.
• The great majority of written-up experiments fail [failed] even to obtain excess heat.

Arguments for "irreproducible"

• Pons and Fleischmann reported excess heat and substantial neutrons. The neutron report was artifact, and, though very low levels of neutrons have been reported, the neutron artifact was irreproducible.

• When others attempted to run the Pons and Fleischmann experiment, in early 1989, no anomalous effects were seen, other than some isolated reports of effects that were retracted.

• Even where other investigators find excess heat, the levels of heat are not consistent and vary from experiment to experiment, even if conditions are apparently the same.

Arguments against "irreproducible"

• The basic discovery of Pons and Fleischmann and what they first found was unexplained heat. Later, looking for evidence to support their conclusion of "nuclear" -- from the heat being beyond what they could explain with chemistry -- they tried using a neutron detector and thought they found neutrons (at low levels, not commensurate with the heat). They were not expert in neutron detection, and that was an error, and was quickly retracted, and is irrelevant to the longer history of cold fusion, which moved on, based on the main finding of anomalous heat, as reported in their original paper (the neutrons were acknowledged, then, not to be associated with the "main reaction," which they ascribed to an "unknown nuclear reaction."

• The finding of excess heat was at first irreproducible. However, the early replication attempts did not follow the Pons and Fleischmann protocol, largely, and there were many problems. The chief problem was haste. The U.S. Department of Energy apparently funded many of these attempts and wanted quick answers. The reports that the Department of Energy depended on, for its finding that there was no conclusive evidence for the effect, had not had adequate time to generate the effect, using the Pons and Fleischmann method. It took months. When Melvin Miles, who was one of the contracted replicators, mentioned in the report has having negative results, started to see positive results after his original negative ones, and before the final ERAB panel report had been prepared, his report was ignored.

• If there is a single replication of a finding, it cannot be said that it is "irreproducible." It's been reproduced. One may still challenge it, one might possibly note that it is not "widely reproduced," if that is true, but it is no longer, obviously, "irreproducible."

• "Not consistently reproducible" would also be correct, though misleading. Research groups that persisted found the effect. Sometimes it took more than a year.

• The argument has largely become moot. The *effect* (anomalous heat in palladium deuteride) has been confirmed by other means, specifically correlation of the heat with helium.

unable to replicate publicly

This one isn't even wrong. Nobody does this kind of work "publicly," other than a few isolated demonstrations that have been run, such as by Schwarz at MIT in early 2012. This is scientifically irrelevant.

Were Pons and Fleischmann unable to replicate their experiment after the press conference? Not exactly.

After the press conference their experiments continued, with positive results. However, at a certain point, they ran out of the original material. Nobody suspected at that point that the palladium batch was crucial, and the lab tech used up the last palladium from their original batch without keeping any. They obtained more, and it didn't work. They were, shall we say, consterned! However, they were able, eventually, to get experiments working again, and much successful work was done, especially in France, that was published. That's "public" as far as what is normally done.

It looks like someone might have heard of the incident and mangled it into an overstatement.

independent verification

We are talking only about independent reproduction, not reproduction under the control of the original researcher. Note that there is a conflict between "not reproducible" and "not consistently reproducible." If we say that something is "not consistently reproducible," we mean that it is reproduced, but that such doesn't happen with every attempt. Or it happens with results differing in magnitude.

The original text is cobbed together from various ideas and reports, not coherently written. Suggested as a replacement is something accurate. The effect was not consistently reproducible, that could be said, but this is incompatible with irreproducible. Further, plenty of effects are not consistently reproducible yet are considered real.

Consistent reproducibility, also called "reliability," is very important for a commercial application but not so much for establishing the reality of an effect, which can still be managed through correlation, for example. It's done all the time.

even among cold fusion researchers

The full phrase was Even among cold fusion researchers, reports of substantial excess heat are rare, as shown in the 2010 review.

See [2]. This is based on Figure 4, pdf page 7, which has been misinterpreted. That figure was based on data compiled to show how often large amounts of power were found. It was not designed to show how often significant heat is found, the lowest category for the electrolytic method is roughly 0 - 1.25 W. There is a problem with the chart in that it talks about "energy" but the chart shows power, in watts, not energy (which would be in joules or KWh or something like that). Storms grants that most cells don't show excess heat, but it is entirely unclear what the chart means. Papers in peer-reviewed journals reporting excess heat may have outstripped those showing none, yearly, starting about 1991, and accumulated until they exceed the number of reports showing no heat. (However, for this, Abd has relied upon the Britz classification of papers, and this might be misleading.) Given, however, that Pons and Fleischmann were using calorimetry with a precision of about 1 mW, and other methods might be 50 mW or 250 mW, many reports showing excess heat would be in the first bar of the histogram. Storms' purpose was to show the scatter of reports of high heat, that it was inconsistent with a hypothesis of "measurement error."

This is thus an attempt to reinterpret a report to derive conclusions opposite to what the report actually shows. Even if true, it would be irrelevant to the issue of reproducibility. Here the preamble to the 1989 ERAB Panel report may be useful:

"Ordinarily, new scientific discoveries are claimed to be consistent and reproducible; as a result, if the experiments are not complicated, the discovery can usually be confirmed or disproved in a few months. The claims of cold fusion, however, are unusual in that even the strongest proponents of cold fusion assert that the experiments, for unknown reasons, are not consistent and reproducible at the present time. However, even a single short but valid cold fusion period would be revolutionary."

So, it took longer than a few months. As Storms wrote in his 2010 review, "evidence accumulated." Only by neglecting the evidence that was found after the initial flurry of negative replications can the position of "irreproducible" be maintained.

The great majority of experiments

This may not be true, unless severely qualified. The Storms paper doesn't establish this for reasons stated above.