A node I'm developing has reached a point where I start to have doubts about its complexity. I was wondering from your experience, what have you established as personal rules concerning nodes. For example... when do you decide that the compute method has reached a level of complexity where it would be wiser to divide it up into two nodes. Or is it more goal based? Such as if your compute method has become very complex, but if it all serves a single end purpose..then there is no problem about complexity.
So basically how specific should a nodes function be? I'm trying to keep nodes simple and focused on a specific task...but then I don't want to have a large set of nodes. So how do you find middle ground between tons of nodes...and few very complex nodes computing tons of different tasks.