I am new to pd and have tried to search for an answer, but I think may question is more about the paradigm shift to graphical programming than it is a specific technology question ... so apologies if I'm unclear cause I'm still getting my head around the concepts.

I'm going to talk in terms of OO as I have used it in java, and hopefully this will explain what I am trying to achieve in pd.

I would like to create and modify a large number of objects at runtime. Each of these objects needs to maintain references to a set of other objects - and these links need to be maintained at run time as objects associate or disassociate with each other. Additionally, these objects need their own internal set of functionality - either inherited from a parent abstraction, or overridden and/or specialized as needed. Everything needs to be typed and parametrised so that I can write algorithms to make choices on what type to instantiate on the fly.

Each object in itself needs to hold a large amount of data, and be able to perform operations on itself and others based on it's internal state. I could go on, but I think I've waffled on long enough.

Should I be translating this onto the graphical programming model, and do I just need to nut away at it more until the penny drops?

Or should I be looking to externals to model these objects and algorithms?

Thanks in advance!

Chris