The Jordan canonical form of a matrix $A$ assigns a partition $p_\lambda$ of $n_\lambda$ to each eigenvalue $\lambda$ of $A$, where $n_\lambda$ is the multiplicity of $\lambda$ in the characteristic polynomial of $A$.
Most undergradute-level textbooks on linear algebra record $p_\lambda$ using a "dot diagram", which is like a Young diagram but transposed and using dots instead of boxes. Does anyone know why this notation is preferred or where it came from? Is there any good reason to use it (other than tradition)? Young diagrams have the advantage that you can easily label the boxes, which comes up when explaining how to compute a Jordan basis.
I actually only have one printed example: Linear Algebra by Friedberg, Insel, and Spence. But I've heard that, for example, dot diagrams were also used in linear algebra courses in Sofia Univeristy in the 80s, and I think they are widespread.