Five years ago, we open-sourced a storage project. Like many teams, we believed at the time that if the code was good enough, a community would gradually form around it. Over the following five years, however, we came to realize that the growth of an open-source project is not only about code evolution. It is also about how a community is built, how it is perceived, and how it can remain sustainable over time.

During these five years, we continuously observed and reflected on many signals commonly used to evaluate open-source projects: GitHub stars, issue activity, contributor numbers, user adoption, license choices, donations, and commercial support models. All of these signals are valuable, but they are also easy to oversimplify. For open-source infrastructure projects, community health does not always mean higher visibility or more activity. The real condition of a project often lies in more complex relationships: whether user adoption continues to grow, whether feedback is high quality, whether contributions are sustainable, whether the license builds trust, and whether the business model strengthens the community rather than weakening it.

In this talk, I will share five years of practical experience from an open-source storage project, together with public data from sources such as OSS Insight, to discuss how open-source storage projects have changed in terms of community building and sustainability. The talk will focus on several questions: Which metrics should open-source projects really care about? Should infrastructure projects pay more attention to contributor growth or user adoption? How do different license choices and forms of commercial involvement affect community behavior and external perception? And in the reality of “open source + commercial operations,” how can a project build long-term trust?

This is not a ranking of projects, nor a simple judgment of which model is better. It is a community retrospective based on real-world practice. I hope this talk can provide other open-source projects, especially infrastructure projects, with a more practical framework for observing and understanding community health.

Speakers:

Rui Su: Open-source advocate and co-founder of JuiceFS, a cloud-native distributed file system

Rui Su is the co-founder of Juicedata and creator of JuiceFS, an open-source distributed file system built on object storage. He has spent the past seven years building storage infrastructure for AI, analytics, and cloud-native environments, and leading JuiceFS from a proprietary product into a globally adopted Apache 2.0 project.

Rui speaks regularly on topics including AI/ML storage, Kubernetes-native data infrastructure, POSIX access over object storage, and open-source project strategy. He has presented at KubeCon 2023 Shanghai, SNIA SDC 2025, Community Over Code 2025, and KubeCon 2025 Atlanta.

Selected talks:

SNIA SDC 2025 — Why s3fs Fails in AI/ML and How to Achieve Scalable POSIX Access Anyway https://www.youtube.com/watch?v=deQKBSAvt_M Community Over Code 2025 — Why We Chose Openness: The Story Behind JuiceFS Switching to Apache 2.0 https://www.youtube.com/watch?v=DtyJHVz4CJE KubeCon 2023 Shanghai — How to Accelerate Model Training and Eliminate the I/O Bottleneck for the Cloud (in Chinese) https://www.youtube.com/watch?v=pK9pUxRpBSM