In response to #2, consider that every material is fundamentally "springy"[1], and many engineering materials deflect a human-noticable amount when enough force to move them is applied. Thus, you can model every connection between an actuator and an object as a spring. When the actuator starts accelerating, it applies a force through that spring, which causes the spring to extend, and the force actually applied to the object is provided only by the extension of the spring. It's only once the spring is applying the same force as the actuator is that the two objects are moving at the same speed. However, at that point, the actuator and object are moving at different speeds, so the spring is still extending. As a result, you end up with an oscillation in the velocity of the object, which is almost never desirable. For a start, if one of the parts is metal, this causes fatigue, which will cause the part to fail much sooner. Secondly, you generally want the object being moved to follow a precise path, and that oscillation will show up as ringing[2]

[1] Yes, this is a vast oversimplification, but the model I'll build using it is reasonably accurate.

[2] Most 3d printing enthusiasts are familiar with this issue; e.g. https://www.simplify3d.com/resources/print-quality-troublesh... . However, most of the advice you see amounts to "make everything stiffer", which helps, but the real solution is to be less jerky.