Shay Bannon, the CEOs of Elastic and Grafana, explicitly referred to AWS and Amazon Elasticsearch Service as the reason for the license change, while the other CEO, Raj Dutt, described the move as balancing value creation and revenue creation while maintaining an open-source philosophy.
Elastic had previously filed a lawsuit against Amazon for trademark violation, and Amazon quickly responded to the license change by forking Elasticsearch. Grafana, on the other hand, recently entered a partnership with Amazon — a first for Amazon. Nevertheless, Dutt also referred to the need to protect Grafana.
The people working on the CHAOSS project, standing for Community Health Analytics Open Source Software, under the auspices of the Linux Foundation want to change that. In 2013, a company called Bitergia was founded to provide services around the software, as the interest in understanding how open source communities work was growing. In 2017, the CHAOSS project was adopted by the Linux Foundation, and Link joined as a co-founder.
A virtuous cycle of community health metrics evolution
The process of defining and managing those metrics is a virtuous cycle of evolution. There are Common Metrics, metrics for Diversity and Inclusion, Evolution, Risk, and Value. Some metrics are raw numbers, counting things such as number of commits, number of issues, or number of people. Relying on metrics alone can be deceiving — context is always needed to interpret metrics.
Measuring community-related value
The CHAOSS project also includes aggregate, synthesized metrics that pull different threads together and take them to the next level. A good part of those metrics apply to open source projects just the same as they would apply to an internal software development team. It all begins with asking the right questions, which should help ensure the right metric is used to answer those questions.
Metrics and side effects
Besides metric gamification, there are more things to be careful about. For example, when we talk about diversity, equity and inclusion, we start dealing with personally identifiable information that can be sensitive.
Toward more fine-grained open-source economics?
In 2020, ZDNet wrote about an analysis based on Github commits for various commercial open source projects. What that analysis showed was that vendors backing the development of those projects were the ones contributing the majority of code.
Commercial open source vendors have a legitimate interest to be concerned about capturing the value they generate. Third parties like Amazon may also have a legitimate argument in claiming they contribute in ways that go beyond code.
License engineering is one-way commercial open source vendors have to deal with the situation. However, it seems like a rather blunt instrument, with considerable side effects. Link noted that in terms of seeing how license changes affect the community, community metrics may help.