Open-source Sustainability
A multi-year initiative investigating the long-term health and sustainability of open-source communities through agent-based modeling, active inference, and cybernetic approaches.
Open-source Sustainability is an ongoing initiative within Orthogonal Research and Education Laboratory (OREL), conducted in coordination with JOPRO’s Society Ethics Technology (SET) working group. The initiative investigates how open-source communities can be made more sustainable in terms of contributor involvement, governance, and long-term success. Its work combines agent-based modeling (ABM), active inference, reinforcement learning, and cybernetic personality theory to study contributor behavior, governance dynamics, and the institutional health of collaborative software projects.
The initiative spans multiple years (2022–present) and has hosted a rotating cast of contributors, with much of its work supported through Google Summer of Code (GSoC) and the OpenWorm Foundation / INCF.
Workstreams
- LLAMOSC — LLM-powered agent-based model for open-source community sustainability (GSoC ‘24 and beyond)
- Sustainability Auditing Tool (SAT) — an auditing framework for evaluating community health
- Active Inference & Cybernetic Approaches — modeling contributor behavior through active inference
- Reinforcement Learning Techniques — ML approaches to community dynamics
Contributors
2022 GSoC cohort
Sustainability Auditing Tool team
2023 GSoC — R. V. Rajagopalan 2024 GSoC — Sarrah Bastawala, Shubham Soni 2025 GSoC — Vidhi Rohira 2026 GSoC — Kavya Zala
Repository
github.com/Orthogonal-Research-Lab/Open-source-Sustainability
Projects
LLAMOSC: LLM Agent-Based Modeling for Open Source Communities
A framework that pairs Large Language Models with agent-based modeling to simulate and sustain open-source communities.
SustainHub: Adaptive Agent-Based Model for Open-Source Community Sustainability
An adaptive agent-based model for studying and supporting the sustainability of open-source communities.
Exploring Open Source Sustainability via Agent-based Modeling
Agent-based modeling approaches to understanding the long-term sustainability of open-source communities.
Agent-based Collective Cognition for Open Source Communities
Agent-based models of collective cognition applied to open-source community dynamics.
Ethical Regulators for Open Source Communities
Applying the ethical regulator theorem to the governance of open-source communities.
Reinforcement Learning and Collective Cognition for Open Source Communities
Reinforcement learning approaches to collective cognition in open-source communities.
Sustainability Auditing Tool (SAT)
A web-based auditing platform for evaluating the long-term sustainability and community health of open-source projects.