Peer Groups And the JXTA Platform

Introduction

Over the last few weeks there were a number of questions about PeerGroups, why it is the way it is, and were it may be going. I'll try to classify the various issues and give some insight about the way the code is put together and what we could and could not do with it. This is not a spec nor a manifesto, rather it is brain-dump, food for though (though for food ?). I tried to summarize here all that's been on my mind about JXTA, as an experimental platform, more than as a set a protocols, and how my biases translated into code.

Most of the following topics are quite related to each other, I'll mostly stay clear of analysing these relationships so that the lot is somewhat readable. I trust the reader will easily figure-out how each choice influences the options available in other area.

I chose to split the rest of this document in one page per topic, so that we can keep using these pages to track comments, new ideas, and in general the evolution of the design.

Here they are now. All comments and (more) heated debate most welcome !



Jean-Christophe Hugly