I do follow the fusion research in some moderate detail because I do think that it's the most likely future advance that could make a "qualitative abrupt change" in the methods that we exploit to obtain energy. If you look at the fusion category of TRF blog posts, you will find some texts debunking Andrea Rossi's cold fusion but many more "real experiments" such as NIF with its lasers, the Z-machine, and – obviously – ITER, a French-international version of the tokamak concept.
Yesterday, University of Washington released a press release on their dynomak paradigm.The picture above shows that it is a nicely compact prototype. I haven't been able to see what the successes have been but they say that it could be cheaper to be built than coal power plants etc.
See also: Alaska Native, Phys.ORG
The design originated from term paper in the classrooms of University of Washington which is pretty cool. In April, Elsevier's journal Fusion Engineering and Design published their
The dynomak: An advanced spheromak reactor concept with imposed-dynamo current drive and next-generation nuclear power technologiesby 10 UW authors including Derek Sutherland and Thomas Jarboe.
The design is known as "spheromak" – it's a word used at least since the late 1970s – or "dynomak" – which is their new name for this particular University of Washington incarnation of the idea. The word "spheromak" has mostly meant that the magnetic field needed to confine the plasma is generated... by currents running through the plasma itself. The abstract of that paper tells you quite a lot what molten salt they propose for first-wall cooling and tons of other technical issues as well as some economic and efficiency estimates.
I wish them a lot of good luck. If electric energy were produced mostly by similar gadgets in the future, I think that it would get cheaper and people would finally start to switch to electric cars and other things that are economically inferior these days.