Why don't the laws of physics hold in non-inertial frames?

[Dale]

Of course, you do not need to do that but you can as well conveniently stay in a fixed inertial frame,

This is my point. The choice of reference frame is dictated entirely by the physicist's convenience, not the measurement apparatus.

 

[vanhees71]

Again, to interpret the pointer readings of your measurement devices within your arbitrarily chosen reference frame (the "lab frame") you need to know, which quantities these pointer readings represent. The most simple example in relativity is an ideal clock. It shows its proper time, which is different to a clock that is at rest wrt. to the lab frame. An ideal clock at rest relative to the lab frame shows the lab frame's coordinate time (by convention!), while a moving clock shows a different time. Of course, you can map these times to each other IF you know the relative motion of the clock wrt. your lab frame!

 

[Dale]

Yes. None of that is controversial.

 

[loislane]

This is my point. The choice of reference frame is dictated entirely by the physicist's convenience, not the measurement apparatus.

Maybe what you mean is that defining an inertial frame by measurement suffers from the same problem as the one-way speed of light, that cannot be measured without assuming a previous fixed synchronization between two distant clocks, so assigning just one or two inertial frames to two objects spatially separated can then only be a convention(physicists convention), not something derived from measurement devices?