I’ll throw this out. If I recall correctly, one random pop science article claimed the amount of thrust that was measured WAS in the ballpark of what was predicted.
For what that is worth.
I mean, predicted according to what? They don’t even know how it works, how would you know what theory to use to determine how much thrust would be generated?
As far as I understand, this is a purely experimental observation, which makes it even more exciting because it’s something completely new!
Pretty sure they were saying that before relatively, probably before gunpowder, too.
Maybe, but the scientific discoveries in recent times (say, in the last 100 years) haven’t really been of the revolutionary sort. Mostly evolutionary. Even Relativity, as you say, didn’t really do much to advance the current state of technology. Sure, it explains a whole lot, and enables us to calculate things better, but do we really care what a black hole looks like?
Even the last revolution, the computer age (arguably), evolved from transistors, which had been around for decades at that point. The age of Galileo deducing the laws of physics by dropping apocryphal cannonballs are long past. Science these days are designed top down, with experiences to confirm or refute theories that are based on other theories, and which are confirmed or refuted by the presence or absence of particles literally smaller than an atom.
This though, this is really exciting. An unexplained physical phenomena! I doubt we’ve had anything like this in the last century.
Unexplained physical phenomena? We’ve had thousands of those in the past century. Some of them eventually turned out to be experimental error, some turned out to be unexpected but mundane effects, some of them eventually led to new science, and some of them are still unexplained.
I dunno, how many of those unexplained phenomena literally shake the foundations of science?
I mean, telling me that “the law of conservation of momentum is now wrong!” is pretty major in my book. Even though Newton’s laws are inaccurate due to relativity, we still teach F=ma because in most cases, F=ma is good enough. But in this case, it’s not a clarification, it’s not a refinement, it’s literally “nope”. This is literally like “nope, the atom is not like a plum pudding” when centuries of experimental evidence and all forms of theory we know state it to be the case.
Sure, in the end, maybe it’s experimental error, but if it’s not… I’m happy to be alive. 
Yes.
I firmly agree that Einstein’s Relativity (Special and General) wasn’t revolutionary; in fact, I’d argue that it’s the capstone of Classical Mechanics, the project begun by Newton to put physics on a sound footing by creating deterministic mathematical descriptions of all physical properties. General Relativity completes this to the greatest extent it can be completed by bringing gravity, electromagnetism, and kinematics all into the same tent, which rounds out the macroscopic phenomena humans can directly observe.
Except…
It’s incomplete. It gives theories which fail to explain things like, say, the photoelectric effect, which can be explained once you bring what’s now called Old Quantum Theory into your universe, which is an ad hoc correction to Classical Mechanics which explains why electrons don’t spiral into atomic nuclei emitting hard radiation but isn’t really a coherent worldview.
Old Quantum Theory was developed into Quantum Mechanics, which uses a surprising amount of the same mathematical machinery that was developed for Classical Mechanics (the Lagrangian and Hamiltonian formalisms, the finest jewels of Nineteenth Century mathematical physics, are quite useful in QM) but really is a discontinuous jump in terms of the underlying philosophy. Suddenly, you can’t put your hand on what your equations are describing. It’s an open question as to precisely what parts of your equations are mere formalisms and what parts correspond to physical reality.
The jump to Quantum Mechanics was revolutionary. Electromagnetism, the study of which first inspired Special Relativity, also inspired the death of Classical Mechanics. It is impossible to account for photons and electrons with a fully deterministic theory; you must introduce true randomness into your universe; if you take the path integral formalism seriously, the basic laws of physics are only ultimately followed, as opposed to being absolutely true at all times in all places at all scales; the center cannot hold, mere anarchy is loosed upon the world, three quarks for Muster Mark, and all that Modernist stuff. The math actually isn’t that difficult, it’s just that it describes a world which is bizarre by any macroscopic sensibility and it doesn’t look like it’s ever going to un-bizarre itself as we learn more about it. Recent thread about the subject in which I made quite a good post.
So we still haven’t agreed on a consistent worldview for Quantum Mechanics, as you’ll see in the thread I linked to. However, I will argue that you’re somewhat incorrect: There was a revolution in physics in the past century or so, and it was called Quantum Mechanics. Whatever we come up with for that worldview, it can’t be a simple evolution of Classical Mechanics.
Presumably when the peer reviewed paper comes out we will see a number of other institutions attempting to duplicate the effect. Also, putting one on a cube-sat and launching it to test in orbit is pretty cheap, hopefully soon.
Ok. I missed QM; you’re right. But as a non-scientist, I’ll get excited about QM when they make my ansible. 
Although, I suppose entanglement counts…
Eh. If I were alive when QM was developed, I’d be just as excited, I guess.
It isn’t like they’re flipping a switch and the thing hovers in midair; they’re claiming a level of thrust so tiny that they have to make sure it isn’t someone breathing on the device. When it comes to unconventional physics, microscopic results are microscopic evidence.
satellite testing just exchanges one set of experimental uncertainties for another. If the effect is real (it most likely is just experimental error, especially given the reported “thrust” values are barely above the system noise), better to nail it down on earth first.
Again, if these thrust numbers are errors, wouldn’t they be expected to statistically fall in both directions roughly equally? If the “error” is always in one direction, that sounds like thrust to me.
The error in question may not be random. There may be some unknown bias to the error in one direction or another.
I don’t actually see why we need equations or even a theory. We just need data that is too strong a signal to be chance. Specifically, since this is allegedly a “thruster”, we launch a microsat with one and watch it. If it actually changes velocity without fuel, boom, then we can work out theories as to why.
Isn’t that the same for an ion drive?
Well their best device so far has thrust that - if it exists - is below the noise. And it sucks kilowatts to get there. This isn’t a cubesat. Worse, LEO is a rotten place to test it. You are not able to avoid the effects of the Earth’s magnetic field (shielded in the tests on the ground) nor are you in a vacuum. Cube sats re-enter after a week or so. Thermal control is dire to maintain, and the possible thrust the device creates so small that measuring it will be seriously difficult even if present.
Given the current tests have not ruled out thermal issues, it seems very premature to be looking to test in orbit.
More like a year-ish, though it depends heavily on the orbit and the ballistic coefficient of the cubesat. Still, I suspect you’re right that the air drag will dominate any expected thrust level. It’s hard to model, too, since the atmosphere expands and contracts, and the satellite doesn’t necessarily keep a consistent orientation, and so on.
One could in principle launch it further out. It doesn’t need comms or anything. But solar wind and radiation pressure are still complicating factors.
Ion thrusters produce thrust that’s several orders of magnitude higher than the Embdrive. Just going from a quick glance at wiki, the thrust measured for the Emdrive is 100 microNewtons. In comparison the ion thrusters used on the Dawn spacecraft produce 90 milliNewtons, nearly 1000 times more thrust.
For another (silly) comparison, 100 micronewtons is approximately the force of gravity between two 1000 kg spherical cows standing 80 cm apart.
And how does that compare with the force of a breath?
It’s the difference between pursing you lips and blowing (ion drive) and sighing somewhere in the same room (EmDrive).