Christian Couder <christian.couder@xxxxxxxxx> writes: > For debugging and statistical purposes, it can be useful for Git > servers to know the OS the client are using. > > So let's add a new 'os-version' capability to the v2 protocol, in the > same way as the existing 'agent' capability that lets clients and > servers exchange the Git version they are running. > > This sends the same info as `git bugreport` is already sending, which > uses uname(2). It should be the same as what `uname -srvm` returns, > except that it is sanitized in the same way as the Git version sent by > the 'agent' capability is sanitized (by replacing character having an > ascii code less than 32 or more than 127 with '.'). > > CI tests are currently failing on Windows as it looks like uname(1) > and uname(2) don't report the same thing: > > -os-version=MINGW64_NT-10.0-20348.3.4.10-87d57229.x86_64.2024-02-14.20:17.UTC.x86_64 > +os-version=Windows.10.0.20348 > > (See: https://github.com/chriscool/git/actions/runs/9581822699) I think we already heard from enough people to cover the spectrum of opinions. I'd have to say that needs to be carefully kept to the minimum what we send in the 'user-agent' like manner. The "git bugreport" is an opt-in "these should help you in helping me" feature, designed to allow further redacting by the user before sending it out, and should not be compared with "on by default for everybody" telemetry data. I personally like the idea to add to user-agent, instead of adding a new capability. What is the true motivation behind this? Is this thing meant to gather statistics from potentially non-paying general public from hosting providers, or is this primarily for $CORP IT folks to make sure that nobody is being too stale?
