I think you're well aware that people aren't upset at the distinction between training on Copilot data versus training on private repo data (at rest). People are upset because GH is using an opt-out model. Your response is disingenuous not to address this, and the "hope this helps" comes across as condescending (not sure if that was your intention.)