As a physicist, it's interesting to me how people cite the 2nd law of thermodynamic as proof these machines don't work. In fact, the 2nd law is not a law at all but only a statement of what is overwhelmingly likely to be the case.
You can, for example, calculate the non-zero probability that a pot of water at room temperature will inextricably start boiling, as the air around it gets colder.
The chances of that happening are something like 1 out of 10^(10^200) or some other absurdly small number (depends on a lot of factors). Still, if it happened, the universe probably wouldn't dwell on the apparent 'violation' of the 2nd "law."
The machine is bogus though.
