The apparent cubic spline used to join data points on http://embed.ly/stats looks a little ridiculous, especially when the popularity of something takes off from 0 and your spline shows it first dipping to some negative number.
Maybe if you try using a quadratic spline? Mathematically it is almost just as trivial, so your performance should be unaffected, and in my experience it gets rid those embarrassing artifacts left by using cubic splines to join points representing exponential data.
Comments
The apparent cubic spline used to join data points on http://embed.ly/stats looks a little ridiculous, especially when the popularity of something takes off from 0 and your spline shows it first dipping to some negative number.
Maybe if you try using a quadratic spline? Mathematically it is almost just as trivial, so your performance should be unaffected, and in my experience it gets rid those embarrassing artifacts left by using cubic splines to join points representing exponential data.
Whoops, you're right! I don't think I've ever noticed that before -- just changed it to a monotone cubic and it looks much better. Thanks!
http://en.wikipedia.org/wiki/Monotone_cubic_interpolation
Yeah that looks great.