À propos de l'absurdité qu'un interféromètre puisse détecter des ondes gravitationnelles : 

---

If a gravitational wave slightly stretches one arm and compresses the other, the two beams would no longer completely subtract each other. LIGO does not, however, measure the change in path-length because the gravitational wave compresses or expands the light’s wavelength too. Instead, what the device reveals are tiny shifts in the period of the two light beams. If the crests or troughs of the wave arrive out of synch, they produce an interference pattern, meaning that the light acts as a clock and not a ruler.

-- vs --

I've seen several questions asking wouldn't the gravitational wave stretch the light just like it stretches the length of the arm. If the light and arm are equally stretched than no change would be observed. It's quite true that a change in the metric does stretch light, but remember that the lasers are continually shining new light into the arms. The length of the arms is 4 km so the light makes the round trip in about 27 microseconds. However the highest frequency of the detected wave was 250 hz, making the shortest period 4 milliseconds. So the length of the arms is changing more than a factor of a hundred times more slowly than the light is measuring that length. The end result is that the frequency of the light is not affected to any significant extent by the gravitational wave, and that's why it can detect the change in the length of the arms.

-> Cohérent avec l'interprétation n°2 de https://physics.stackexchange.com/questions/153657/ligo-flawed-by-the-identical-expansion-of-laser-wavelength-and-arms-in-presence/235822#235822

---

Jetter un œil au §27.6 du cours de mécanique de Thorne pour deux dérivations rigoureuses.