The 19th century 'ether' research failed because there were two models built on speculation, both wrong. Maxwell's ether was initially a solid, but after a disaster with the elasticity of the ether in displacement current (see three posts back), he converted to gear cogs and idler wheels. Opposing him was Kelvin's ether, a jelly like substance which could sustain permanent vortex rings, hence Kelvin's vortex ring atomic model.
The problem you understand started with J.J. Thomson and Henry Becquerel, who discovered evidence for a subatomic particle (electron) and for the break up of the atom (radioactivity). Kelvin then simply used his powerful reputation to dismiss these as speculative nonsense, rather like modern work on the spacetime fabric is treated by most experts today.
Anyone speaking out in favour of science was ridiculed. Progress was denied. Then, just as science was getting into a serious rut, Einstein came along, an obscure patent clerk on the Continent, and showed the way forward very clearly: damn the mechanical models and mechanisms. All that matters, Einstein's philosophy showed, is mathematical models that can be tested and that are useful.
Einstein saved radioactivity and the electron, not merely by working on the mathematical theory of the photoelectric effect and relativity, but giving the world an excuse to ignore the ether. Maxwell's and Kelvin's ether models were not only contradictory in mechanical detail, but virtually untestable. If you have a vague, non-mathematical, idea of vortices or gear boxes in the fabric of space, you can keep adding new idler wheels or new properties to keep it from failing in tests, just as epicycles were added to the Earth-centred universe to keep it 'fixed'.
This is the great problem with the over-simplistic Popper-Kuhn conception of science. You have a theory and make a prediction. The test is done and the result is unexpected. Now, does that disprove the theory, or does it prove that you need to modify the theory to bring it into line with the latest experimental finding? This is where the big controversy occurs.
Kelvin's vortex atoms were mathematically shown - by Kelvin and Tait, great mathematical physicists of the day - to be permanent and indestructable. Rutherford therefore had problems with Kelvin over radioactivity. Rutherford also competed with J.J. Thomson who had a plum pudding (mixed, i.e., non-nuclear) theory of the atom. In the end, Rutherford won recognition, in part because his nuclear atom led to Bohr's early theory of quantum mechanics, later developed into the modern theory by de Broglie, Schrodinger, Heisenberg, Born, and others.
Sadly, Bohr, Heisenberg and Born opposed Einstein, de Broglie and Schrodinger over the interpretation of the mathematical result. Einstein's fame of 1919 (confirmation of general relativity in the solar eclipse) allowed him to become heretical in 1920, where he said in his inaugural lecture at Leyden University: 'According to the general theory of relativity, space without ether is unthinkable.'
So we are, all things told, indebted to Einstein. Even if special relativity is not the final word.
One severe problem with 'special relativity' is that it is presented as the last word, as a disproof of reality. For example, in this big bang universe we are at an age of 15 Gyr, and everything we see is younger. The farther the star or supernova we see, the further back in the past it is, because of the time taken for the light to get here. So if we see two simultaneous supernovae, we can work out absolutely which happened first by knowing their distances! This is the exact opposite of the popular writings on 'special relativity' which say that you cannot tell which star exploded first.
Another issue is absolute motion; the 2.734 K cosmic background radiation is 3 mK blue shifted in the direction of our absolute motion and 3 mK red-shifted in the other direction.
So we can tell our absolute motion from that (about 400 km/s, partly due to the attraction of the Milky Way towards a big galaxy nearby, but if for sake of argument we have been going 400 km/s since the big bang - which is an order of magnitude approximation - we are only 0.3% of the radius of the universe, in other words within 0.3% to the middle of the big bang).
This effect was called 'the new aether drift' (the title of a Scientific American article on the subject in the late 1970s). Critics responded by fiddling Copernicus' discovery. They claimed that Copernicus did not discover or work on the solar system, but instead had discovered that 'the earth is not in a special place in the universe'.
In other words, the anthropic argument - that everything is the way it is because if it wasn't, it would be different! This anthropic argument was usefully used by Fred Hoyle in the 1950s to predict that because we contain carbon, a nuclear reaction producing carbon in stars must be possible. The prediction was confirmed afterwards by nuclear physics! So this 'mum's argument' - the anthropic principle - has made useful scientific predictions.
However, it is a cheap explanation for everything to say that 'everything must be the way it is because otherwise we wouldn't see it the way it is'. What people want are scientific facts, not arm-waving waffle. One way of getting useful information out of the anthropic principle is to define it precisely as the set of circumstances needed to create humans: how different would the laws of nature need to be to prevent humans from existing (and observing nature) now?
By trying calculations using different values of fundamental constants, for example, it can be ascertained that small differences would prevent the synthesis of some elements and thus prevent humans from existing as they do now. However, is this really a useful way to approach science? Some prople try to use these results to 'prove God exists' by saying that the universe must have been deliberately designed, or we would not be here to observe it.
Others (Susskind and Kaku like the multiverse) turn this argument on its head and claim that there are an infinite (or very large) number of parallel universes (the 'multiverse', since the definition of universe is at odds with many universes), and say that the reason this universe is the right one for life is simply because it is the one with life in it. If you have only one universe which is precisely right for life, then that would suggest deliberate design, but if you have a great many universes which are all a bit different, then it could have arisen by accident.
So modern physics is being driven by religious ideas. In this way it is, indeed, crackpot.