Effective resistance matrix and nonedgesonly=True
For undirected graphs, there is a method effective_resistance_matrix() which returns the matrix whose (i,j) entry is the effective resistance between vertices i and j. The current implementation has an optional boolean parameter nonedgesonly=True, whose default behavior has the effect that if (i,j) are adjacent then the corresponding matrix entry is 0.
I was wondering, what was the reason behind the choice of default value for this nonedgesonly parameter? I haven't come across cases myself where this feels natural, and there are some nice theorems regarding the effective resistance matrix with nonedgesonly=False.
This interesting question regards the implementation choices made by the pakage's author(s), is indeed a mathematical question but could nevertheless be adressed to these author(s) (probably via sage-support or sage-devel) with better chances of getting a good answer.
HTH,