The bliss engine - how do I share what I figured out. A novel approach to machine learning based on memory and motivation.

So, over the course of the last 8 years or so, I wandered to a form of machine learning that is a bit novel, and I'm trying to figure out how best to share it with the world. I do intend to open source it, but I'm not sure how to approach the ML community properly.

Essentially it is an extension on term-vectorspace search (similar to how you compare two documents using a cosine similarity), combined with a few very simple realizations about motivation... except written in C. (with a good start in C#, Erlang, and Javascript).

My initial idea was to write a book that leads the reader, even if they are short chapters, through the thought processes and code, and directions I think it can go.

I've also considered making a public demo website of the code running a simulated African savannah, but I just don't have the resources/time to integrate everything.

Thoughts on how to share an idea if you are an outsider?