The /metrics endpoint is there for the operator, so you can monitor/measure your pod’s performance and adjust accordingly, or file bug reports for me to fix broken shit™ 🤣 – It is never pushed anywhere, only pulled. And I only ever pull two key pieces out dao and mao of known single and multi-user pods, but I ASK beforehand 🤣 – For example I’m not pulling DAU and MAO stats from @kaniyama_t@yarn.takuma-csirt.com ’s Twitter2 pod because I never got a response to whether I could or not 😅

@prologic@twtxt.net I will play devil’s advocate. The telemetry being proposed for the go tool chain is opt-out. I think opt-out is better that being “based on honour”. It is also meant to “see, and fix broken shit” 😜😂

Also, I bet most running Yarn do not even know about it. 😬😂

@bender@twtxt.net You make an excellent point. I might just put it behind an --enable-metrics flag 🤔

Although performance metrics that measure CPU, Memory and such aren’t really the same thing as “telemetry”