Well, saying that it's "compressing as it falls" doesn't quite explain the observation either. What you see is the original slinky with a sudden catastrophic "sound wave" at the top propagating across it. It doesn't look like the whole slinky is contracting as it falls.
It's important to understand that as you hold it, each part of the slinky exists in a condition of equilibrium: since it is not accelerating upward or downward, the force of gravity and the force of the spring above pulling the slinky up must exactly cancel. If you slice the Slinky up into different "chunks" as a function of height, each chunk (call it dm) has the relationship (up-force) - (down-force) = g dm. It only connects to the parts immediately above and below it, and only moves when those parts move.
So actually, as long as that bottom part remains stretched just the way it is, the bottom edge cannot go up or down; the tension up top is the same as the tension on the bottom.
With a little more reflection, the stretching changes must propagate at the speed of sound in the slinky. That's just what sound is -- changes of pressure (in this case, tension) propagating through the solid. If you could send this sort of information faster, then sound would propagate faster commensurately. The speed of sound is fundamentally thus seen to be a connection, for the spring force, between "wiggle this side over here" and "that side wiggles."
Now, there are other forces at play in the slinky in theory which can propagate much faster: I imagine that a sound wave which spirals through the plastic (using it as a waveguide) for example, might actually travel to the bottom much faster. The top pushes on air which has a faster speed of sound, and emits light which goes even faster. But, those forces are not the crucial force which holds the slinky up and therefore needs to change.
It's important to understand that as you hold it, each part of the slinky exists in a condition of equilibrium: since it is not accelerating upward or downward, the force of gravity and the force of the spring above pulling the slinky up must exactly cancel. If you slice the Slinky up into different "chunks" as a function of height, each chunk (call it dm) has the relationship (up-force) - (down-force) = g dm. It only connects to the parts immediately above and below it, and only moves when those parts move.
So actually, as long as that bottom part remains stretched just the way it is, the bottom edge cannot go up or down; the tension up top is the same as the tension on the bottom.
With a little more reflection, the stretching changes must propagate at the speed of sound in the slinky. That's just what sound is -- changes of pressure (in this case, tension) propagating through the solid. If you could send this sort of information faster, then sound would propagate faster commensurately. The speed of sound is fundamentally thus seen to be a connection, for the spring force, between "wiggle this side over here" and "that side wiggles."
Now, there are other forces at play in the slinky in theory which can propagate much faster: I imagine that a sound wave which spirals through the plastic (using it as a waveguide) for example, might actually travel to the bottom much faster. The top pushes on air which has a faster speed of sound, and emits light which goes even faster. But, those forces are not the crucial force which holds the slinky up and therefore needs to change.