And ramping up weight near the bottom is indeed a pseudo bottom out. Lets say full travel is 4mm. at 3.8mm it takes 1 ton of force to go from 3.8mm to 4mm. Will you ever get to 4mm? No. Have you technically bottomed out? No...So on switches where the weight required is ramped up to keep you from bottoming out, while technically you have not bottomed out, all you've done created a situation where the force to bottom out > than the what you're pushing while typing.
But, surely, you understand the difference between a large discontinuous jump in the force and a region of the F(x) diagram where the derivative is simply sufficient to cushion the stroke?
With that said, if what you're trying to say is that one does not
consciously react to the bump, and then decrease the applied force to avoid bottoming out, I would have to agree.
However, the same thing could be said for audible feedback. You do not instantaneously change your force output in response to the click. But that is not the point.
One of the advantages of a tactile bump, as well as a click, is that it teaches you when the switch actuates. That is why it is important that the tactile event (or click) is not decoupled from the actuation.
Over time, this teaches you to apply the correct amount of force over the desired distance (=do the correct amount of work, the integral of force with distance). It becomes muscle memory, so to speak.
However, if the drop-off after the tactile event is very large and no subsequent ramping up of the return force happens, then it can be difficult not to bottom out. And then we're back to the idea of cushioning ...