There are two perspectives we often get on a ribbon disk for a knot K in the 3-sphere: A ribbon disk is defined as either an immersed disk in S^3 with only ribbon self-intersections, or an embedded disk in B^4 with no maxima with respect to the radial Morse function. These definitions are often called "equivalent," but more technically, the difference between them is like the difference between a knot and a knot diagram; one is an embedding, and one is a lower-dimensional projection of an embedding. The "ribbon number" r(K) for a ribbon knot K is the smallest number of ribbon intersections among any projection of any ribbon disk bounded by K, and r(K) can then be seen as a sort of crossing number for ribbon disks. The Alexander polynomial of K can be used with rather surprising accuracy to give lower bounds for r(K), and we discuss our work in understanding and cataloguing ribbon numbers for low-crossing ribbon knots. Parts of this talk are joint with Stefan Friedl, Jeffrey Meier, Filip Misev, and lots of undergrads.