I think there is a big difference between the "infinite energy crowd" and the experiments going on in "cold fusion". One can argue whether there is "zero stage energy" or not, but one can surely agree, that if it is, its energy density (erg per cubic centimeters) is infinitesimal and thus one needs to harvest that "infinite energy source" from am infinite volume of space, not practical. However, fusion, we all know exists and produces vast amount of energy (exactly the mass difference between the fusing nuclei multiplied by c^2), how do we know, we see its (energy) extraction in the sun (what is called the Hydrogen/Carbon cycle), and of course we have seen it on earth in hydrogen bombs, We have even spent something of the order of $50 B on various schemes involving attempting to replicate the sun fusion process with high power lasers in the Tokamak. No question that fusing hydrogen/deuterium into helium is a source of energy. Can it be extracted without bringing the fusing nuclei to a million degrees and hundred of thousands of atmospheres pressure is the only question. Without going into too many details, 10^5 atmospheres of H2, does not require, necessarily, to have such "physical pressure" present. Other means can get you there.
Oh, as for Pons and Fleishman, they have demonstrated a strange effect (excess heat evolution, which in one case was so intense, the electrode bath boiled and the experiment essentially "self destructed"), similar strange effects have been (though, irreproducibly) reproduced at seven other labs that I know of. I know you'll laugh at "reproduced irreproducibly", but that is the nature of the beast, I spent a lot of time myself to reproduce some type of experiments, just to find out that they "die" on me and have to start all over again. These are intermitent effects, and it will take a long time to determine what are the optimal parameters (out of about 80 plus such parameters, count the permutations) that will yield the optimal conditions, and then, you'll find that the process itself changes some of those parameters making reproducibility a little difficult, but, not impossible, once one understands the role of each of those parameters.