Now, one of the packages I maintain is graphviz which suffices at what it does: use Graphviz to visualise graphs using static images. Despite its various problems, I keep using Graphviz because – unlike most of the flashier graph visualisation programs that I’ve found – it doesn’t require a fancy GUI just to convert a pre-existing graph into an image (admittedly, others may have library versions, but most seem to be written in Java and Python, which do not seem as useful in terms of writing Haskell bindings). However, one thing that Graphviz cannot do is let you dynamically visualise graphs, which is especially useful for extremely large graphs (e.g. call graphs).
So, how crazy am I?