[Affected Platforms]: - All Windows - All Mac - All Linux [Prerequisites]: - Have the "Firefox Notes" experiment enabled from testpilot.firefox.com. - Have the "Firefox Notes" sidebar opened. [Steps to reproduce]: 1. Open the browser with the profile from prerequisites. 2. Click the address bar. 3. Press the "Alt+Shift+W" keys and observe the behavior. [Expected result]: - The sidebar is closed. [Actual result]: - Nothing happens. [Regression window]: Last good revision: c200124f2eecc1dae1bae361085b64985e7ccd2f First bad revision: 57e0c8a166d2b378eed7c0008643ade6b184145f Pushlog: https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=c200124f2eecc1dae1bae361085b64985e7ccd2f&tochange=57e0c8a166d2b378eed7c0008643ade6b184145f [Notes]: - The behavior is reproducible if you start with the sidebar closed. - The issue is also reproducible with the TxP version of "Firefox Notes". - The issue is no longer reproducible if the e10s is disabled. - This behavior is not encountered if the "Ctrl+B" and "Ctrl+H" shortcuts are used instead of the "Alt+Shift+W" one. - Attached a screen recording of the issue:

Masayuki, the regression window seems to point to Bug 1333459, do you have an idea of what's happening?

Hmm, I have no idea for now. However, the shortcut key combination is too bad. Alt + Shift is standard accesskey modifiers of web contents on Windows and Linux. When this combination is detected on focused chrome document, the keyboard events are sent to all remote children: https://searchfox.org/mozilla-central/rev/7fb999d1d39418fd331284fab909df076b967ac6/dom/events/EventStateManager.cpp#1281,1288,1302 I'm not so familiar with this case, though. And comment 0 said this is reproducible even on macOS...

If Alt + Shift is a standard modifier already used for web contents, then I guess that one would assume that we shouldn't use that for UI? Is this a correct interpretation? Marius, would you agree to that?

Redirecting the question to Ryan, as he should be able to answer this.

(In reply to Andreas Farre [:farre] from comment #3) > If Alt + Shift is a standard modifier already used for web contents, then I > guess that one would assume that we shouldn't use that for UI? Is this a > correct interpretation? > > Marius, would you agree to that? You didn't ask me in the comment, but yes, I think so. Conflicting a shortcut key and a web content accesskey is really bad. User may get different result when they uses it.