hi katz, thanks for sharing your patch, I was wondering how to do something similar a few days ago. As an exercise in curiosity, I changed the way it works with an exponential sigmoid curve. I think it works well and is closer to what I had in mind, it still has some bugs that make it not 100% functional.
With certain values of the parameter that enters through the right inlet (the one that controls the slope of the curve) the departure and arrival values of the line-e do not reach their destination completely, and in the console there are two division by zero errors that do not I managed to detect to correct them.
Maybe someone can help?