chrismb wrote:I've been told that it was Newton that proved elliptical orbits [whereas Kepler, as you say, 'demonstrated' the possibility] and is a relevant inclusion in this discussion because in Newton's day all such proofs were done by graphical means, after Greek geometry. He firstly proved elliptical orbits by his newly invented calculus, but had to then go back and repeat the proof geometrically before disseminating it into the public domain, which he did using his calculus as a guide.
The point being, you can't use your own sited ideas to prove themselves (as per BlackLight) but you have to frame them in the conventions of the day. This is not unreasonable and provides the contiguity of evidenced logic required for well-founded modern science.
Proof is a slippery word, and in science often relative.
Kepler's analysis of Tycho Brahe's observations was enough to make ellipses the preferred explanation over epicycles. Graphical or not, I can call this proof, though Kepler did not have a complete theory.
Newton went one step further, and showed that ellipses were the result of a much simpler theory, which also explained other phenomena - planetary gravity. This theory thus becomes massively more likely. It has a lower prior because F=-Grm1m2/r^3 is very simple, and it explains very precisely (though not perfectly) multiple observations which would otherwise require independent ad hoc explanations with a combined much lower prior.
And Newton with Leibnitz and others developed the framework to prove that ellipses were the result of this theory - and many other things.
Scientists are limited by the ideas that are current, no doubt, so that 'best' theories are not necessarily obvious. One clear example now would be a GUT-gravity unification where things look complex now but the correct theory, when understood and presented in the correct framework, may turn out to be very simple. Then again, it may not...
I don't alas think the BLP excess heat is likely to be the result of any such theoretical advance. The BLP 'theories' could of course be preferred to existing ones if they had stronger weight of interlinked confirmatory evidence - impossible - and would at least be interesting if they made significantly better predictions than existing theory (they do not). Coming up with new independent theories that will be taken seriously is difficult when the evidence for the existing theory is so strong, and much easier when there is no well developed existing theory (e.g. GUT-gravity).
Best wishes, Tom