Some sort of simulated annealing? #64
Description
I have found that the force layout tends to frequently get stuck in local optima when processing my graphs (namely, heap graphs when characterising software behavior). I have found that it can useful to emulate a simulated annealing process by manually modulating the local and global force-directed layout parameters over time. However, it would be nice to have some automated support for this as it can take many minutes to get the graph into a shape that reveals its structure.
It would be great to have either a proper simulated annealing process to the force layout or even just have a way of modulating the existing local/global parameters over time to help the force-directed layout find a satisfactory optimum more quickly.