(Apologies if this is overly rant-y. I’ve been suffering though a discussion from a creationist that boils down to “information” can’t increase, therefore evolution is impossible. I’m therefore a little sensitive to the topic of information and thermodynamics)
Demonic device converts information to energy
The laws of physics say that you can’t get energy for nothing — worse still, you will always get out of a system less energy than you put in. But a nanoscale experiment inspired by a nineteenth-century paradox that seemed to break those laws now shows that you can generate energy from information.
Of course, the key word is seemed. This is my main peeve here. The all-to-common insinuation that some law of nature has been violated.
Of course, later on (not paragraph 19, though), they admit
The experiment does not actually violate the second law of thermodynamics, because in the system as a whole, energy must be consumed by the equipment — and the experimenters — to monitor the bead and switch the voltage as needed. But it does show that information can be used as a medium to transfer energy, says Sano. The bead is driven as a mini-rotor, with a information-to-energy conversion efficiency of 28%.
I’m not sure how they get to “information is a medium to transfer energy,” and from that to “information is converted to energy.”
There’s a somewhat better article, via Dr. SkySkull
[T]here is energy in information. To store a bit of information a system like a computer memory needs to be put into a defined state, either a ’1′ or a ’0′.
That seems more reasonable — storing information (and all information needs to be stored) requires energy, so there is energy in information storage. But what is being presented as “information” is just the state of a system; one could just as easily say there is energy in e.g. an electron’s orientation in a magnetic field (or the location of a polystyrene bead, as in this case), and skip the discussion about information.
It sounds like a version of the Brownian ratchet, where a paddle would spin in only one direction from random collisions, because the ratchet would impede motion in the other direction. It fails because the ratchet, too, would be subject to collisions, and fail to work if everything were at the same temperature. Here, the mechanical ratchet has been replaced by an electric field, which is not in thermal equilibrium. You expend energy determining when to change the field.
I’m glad I didn’t run into any stories that called it “pure energy.” Oh, crud.
Edit (11/22) Sean does a summary, which makes a lot more sense.
The connection is not that “information carries energy”; if I tell you some information about gas particles in a box, that doesn’t change their total energy. But it does help you extract that energy.