Lots of real-world spatial data is actually spatial point process data. For example, the coordinates of lightning strikes. Kriging is not a useful tool in this context--there is nothing to interpolate. Typically, a point process model is fit such as the inhomogenous Poisson point process, with the rate surface being the object of interest. We could implement geoms/stats for calculating and visualizing the rate surface in a way similar to the kriging geoms/stats.
Lots of real-world spatial data is actually spatial point process data. For example, the coordinates of lightning strikes. Kriging is not a useful tool in this context--there is nothing to interpolate. Typically, a point process model is fit such as the inhomogenous Poisson point process, with the rate surface being the object of interest. We could implement geoms/stats for calculating and visualizing the rate surface in a way similar to the kriging geoms/stats.