We've been cooking bread for at least ten thousand years before thermostatic control came along, so I can understand that not being part of the design requirements.

Your bank balance doesn't appreciate; the bank pays you an income at a particular interest rate on the total value of your deposits that they hold. Stock dividends are similar - its a payment, not paper appreciation - and dividends are also taxed as income. Paper appreciation is different in kind from interest and dividend income.

16 tons/yr for m(b)illions of years is in aggregate a lot more than a few hundred tons a year for the last few decades.

Contrary to popular belief, diplomatic missions do not enjoy full extraterritorial status and are not sovereign territory of the represented state.[5][6] Rather, the premises of diplomatic missions remain under the jurisdiction of the host state while being afforded special privileges (such as immunity from most local laws) by the Vienna Convention on Diplomatic Relations. http://en.wikipedia.org/wiki/Diplomatic_mission

I worked on a "tiny" particle physics experiment ... in about 4 months of running, we collected 150TB of data. My current experiments will collect PBs of data, and LHC is expected to collect EB of data over its lifetime. 10TB would be considered peanuts these days :-)

The US government didn't spend a penny on LHC? Really? A few seconds with your favorite search engine would have dispelled that myth.

In reality, the U.S. government has contributed over $600M to date in direct and in-kind contributions, with significant support for University and National Lab groups that participate in all aspects of construction, operations, and analysis.

The bigger problem is that Ni62 is the most tightly bound nucleus known, http://hyperphysics.phy-astr.gsu.edu/hbase/nucene/nucbin2.html#c1 or http://www.nndc.bnl.gov/chart/ Fusion or fission of Ni62 require an input of energy; they clearly aren't measuring spontaneous release of energy in a fusion event...

You certainly CAN use GPS to synchronize clocks at the level of a few ns. While the GPS timing signals themselves are not accurate enough to form a stable timebase at this level, there are multiple methods which use GPS to implement time transfer between ground stations at much better than the 10ns level, when use in conjunction with an external high precision oscillator. See, for instance, http://tf.nist.gov/time/commonviewgps.htm or http://www.nist.gov/pml/div688/grp40/tmas.cfm and the references therein.

Because you can't just steer these beams to arbitrary locations ... you need to build a beamline that points from CERN to Fermilab. There are a number of these types of experiments worldwide, but they are all point-to-point experiments.

Better to think of Newton's laws as an "approximation" to the laws of special relativity, rather than the other way around.