Free software users and authors tend to be highly aware of the risks of large-scale collection of personal data, and often consider software telemetry to be incompatible with user privacy. But it is possible to collect anonymous data about how software is used in a way that respects its users and protects them from harm – and by doing so, the software's authors can make better-informed decisions about how to improve it. In this presentation, I will make a case for why free software projects should want to collect data on how the software is used, and how this can be done in a privacy-respecting fashion. I will describe the anonymous metrics system in Endless OS, a desktop Linux distribution: our goals for the system, its design and implementation, the types of usage data we collect, and what we have learned from that data, as well as some shortcomings of both the system itself and the data we have. I hope to encourage other free software projects, on the desktop or otherwise, to consider adopting similar techniques to better understand how their software is used.