[07:54] <duflu> alan_g: Hey I only just discovered DisplayReport, which I had partially duplicated. Would anyone complain if I changed the default to be for every vsync instead of summarised per second?
[07:58] <RAOF> duflu: I'm +1 on that.
[07:58] <duflu> I'm guessing not. No one other than us would be using it
[08:00] <alan_g> duflu: I doubt anyone else uses it. An alternative is to add multiple reporting options.
[08:06] <alan_g> RAOF: do you want me to split this into two MPs? https://code.launchpad.net/~alan-griffiths/miral/hide-streams-from-Window-Management-API/+merge/304626
[08:09] <RAOF> alan_g: Commented on MP.
[08:12] <alan_g> RAOF: thanks. I'll split it to separate the discussion. (Not just here, but we really need to get some clarity around the distinction between "WM", "Shell" and "Compositor")
[08:14] <alan_g> I'm just going with "adding later doesn't break ABI"
[08:17] <alan_g> duflu: do you remember when/how we could remove this hack? https://code.launchpad.net/~vanvugt/mir/fix-1391976/+merge/262678?
[08:18] <duflu> alan_g: bzr blame will tell you where it got removed
[08:18] <duflu> If you specify a source file and look at the relevant lines
[08:18] <duflu> Which means I don't remember
[08:19] <duflu> It might have been Alberto's switch to protobuf-lite
[08:19] <alan_g> duflu: yes. I don't remember either.
[08:19] <alan_g> I don't think it was that - the problem comes back with protobuf3 - so I was wondering if we patched protobuf. I'll have to dig it over.
[08:21]  * alan_g thinks alf_  went looking at the protobuf code. But he's "away".
[08:22] <alan_g> And "blame" isn't so good at when lines were deleted.
[08:22] <duflu> alan_g: Yes r2798
[08:22] <duflu> bzr blame is broken but I found it anyway
[08:23] <duflu> bzr log <specific source file> sometimes works and did in this case
[08:24] <alan_g> thanks
[08:26]  * duflu is still in the school of "if you need it to keep working into the future write it yourself and don't rely on 3rd party code"
[08:27] <duflu> Because in the worst case you have no blocker to fixing any issues
[08:28] <alan_g> So, write everything from scratch?
[08:28] <duflu> Well not "everything" :)
[08:29] <duflu> But protocol code I would have