There is an answer that is--in some sense--even more trivial than nic's argument: the whole point of relativity is that the result of the experiment (does it float or sink?) is not dependent on the observer's frame of reference.
I know, I know: your first introduction to relativity is always about how things do depend on the frame, but it is always about how things you thought were invariant (because there are under Galilean transforms, really aren't)
- How long are the cavities?
- How long does the beam take to get from the hall entrance to the target?
- Is the speeding spaceship ever full closed into the dock (i.e. does the rear door close before or after the front door opens)?
These kinds of questions ask about only part of a 4-vector (the space component and the time component respectively) between two events. Those question are not Lorentz invariant, so the answers depend on the frame you choose.
But consider questions about
- Does the electron beam have enough energy to produce charmonium on a hydrogen target?
- What was the mass[*] of the neutral particle that decayed to produce the two charged particles we see in the detector?
- Does the spaceship hit the front door?
- Does that stuff sink or float?
These kind of questions must have the same answer in every frame.
Please, take away the idea that physics does not depend on the choice of frame. This is very important.
[*] The invariant mass, guys. This is why particle physicists don't talk about "relativistic mass": because the real mass is the Lorentz invariant mass.