BDPD — The Why, the What, and the How
A narrative tour of the Be Different Play Differential project
Preface
This is a narrative tour of BDPD — Be Different Play Differential, an open-source laboratory for studying how shared resources fail (and sometimes survive) when the agents using them are heterogeneous, imperfectly informed, and only intermittently good-willed.
It is written for the curious newcomer. No code. No equations beyond what a sentence can carry. One reference per idea at most. The aim is to give a sensation of BDPD — what it asks, what it contains, and how the pieces fit together — so that the four research papers, six blog posts, two pitch decks, and ten-lecture mini-course that make up the project become easier to navigate afterwards.
The tour is in three acts.
The Why asks why BDPD exists. It begins with Hardin’s tragedy of the commons (Hardin 1968) and the Seneca cliff (Bardi 2017), two old questions about how cooperation breaks, and explains what was still missing in the literature when the project started.
The What describes the laboratory itself: the multi-agent Arena with its three swappable physics engines, the heuristic and LLM agents, the Forest of Humbaba card game, the governance toolkit (pacts, cheap talk, sanction ladders), and the four research papers that surround it. A single mind-map at the top of the chapter holds the geography together.
The How is about the method. Two substrates run in parallel: a simulation that sweeps thousands of cells, and a tabletop game whose cards can be drawn in front of a class. They are kept deliberately in tension so that a finding has to survive both before it counts.
Every chapter is short. Every figure that follows was generated by the project itself, either by the simulator or by hand for the card game. Where a deeper treatment exists, a pointer to the relevant paper or blog post is given inline.
The full project — code, paper sources, card-game artwork, and the documentation site — is available on GitLab. Code is AGPL-3.0; papers, course material, and game rules are CC BY 4.0; AI-generated card art is CC0.
Transparency and authorship. This document was authored by Roberto Brunelli with the support of frontier large language models; model outputs were source-grounded by prompt, validated by extensive triaging with different models and finally integrated under author supervision.