I'm not the author of this CL. Re-assigning (though I think a test here is not practical or useful).

Leonard is OOTO until mid Feb. We need another owner. 

[mac triage]  Issue 667274  isn't a regression, so the statement in the description is false.

I also agree that tests are not practical.

Hmm, it seems like a test might not be practical, but I'm not sure it's clearly the case. We want to make sure that the browser window controller is not a touch bar provider - it's possible that you need a touch bar Mac to test for this, but given that I can call up the touch bar simulator via Xcode (i.e. no rebooting or anything) I suspect that all of the touch bar stack is in place, just dormant waiting for some piece of a framework to call it.

(I don't know if tapted@ has a different reason for thinking this test is impractical.)

So the first thing to try is writing an app that checks to see if a window controller responds to a touch bar method, and running that on a non-touch bar 10.12 Mac. I'm guessing that it will respond to the method. Then add the code to opt out of the two protocols listed in the cl, and confirm that the window controller no longer responds to the method. Assuming that works, the test would be a browser test that fails if the window controller responds to the touch bar method.

Archiving old bugs that have only received trivial updates for some time.

If you feel this issue should still be addressed, feel free to reopen it or to file a new issue. Thanks!