TL;DR — usage reports now include one boolean: whether the run was started from your phone. It exists to answer whether a Codeep iOS app is worth building with behaviour rather than a poll.
Changed
- Usage reports now include whether the run was started from a phone. One
boolean, alongside the model, token counts and estimated cost already sent.
It answers one question — is a Codeep iOS app worth building — and it answers it with behaviour rather than opinion. Asking would answer the wrong thing: "would you like an iOS app?" costs nothing to say yes to, so a poll of it comes back positive, looks like evidence, and is worth nothing. 3.1.0 already shipped the cheap version of that app — starting a task from your phone over Telegram — so what is worth knowing is whether anyone actually does it.
Said out loud rather than left in the code. Finding out from a diff that something new is being recorded is a reasonable thing to be annoyed by.
Nothing about your prompt, the answer or the files is sent — this is a yes/no about where a run came from, and it goes only if you have linked Codeep to an account and left usage reporting on. /telemetry turns the whole thing off, as before, and CODEEP_NO_TELEMETRY or DO_NOT_TRACK in the environment force it off regardless of the setting.
Fixed
takeRunFromPhone()moved out of the notification branch. It was read
inside if (noticeCredentials), which depends on having Telegram configured and on the run being long enough to notify — so any run that sent no notification left the flag unread. Consumed once per run now, before anything branches on it, which is what its own comment had always claimed.
Docs
docs/ios-decision.md— the iOS app is paused, with the reasoning written
down: why a poll was rejected, what is being measured instead, and the three things that would take it off pause. There is no iOS code, so the pause costs nothing; what it buys is not re-arguing this every few weeks.