You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Right now, an experiment could drive users to uninstall Firefox entirely -- surely a decisive vote of non-confidence -- and that would not register in our outcomes (beyond measures of user engagement and retention). This ticket tracks counting users that uninstall Firefox. I think this should be a baseline/default outcome, just like, e.g., search volume. (Note that uninstalls are a Windows-only concept; there is no uninstall on other platforms. And not all Windows users installed in the first place, so not all can uninstall.)
Right now, an experiment could drive users to uninstall Firefox entirely -- surely a decisive vote of non-confidence -- and that would not register in our outcomes (beyond measures of user engagement and retention). This ticket tracks counting users that uninstall Firefox. I think this should be a baseline/default outcome, just like, e.g., search volume. (Note that uninstalls are a Windows-only concept; there is no uninstall on other platforms. And not all Windows users installed in the first place, so not all can uninstall.)
Technically, this is a little awkward but not impossible. Uninstall events send a special uninstall ping. That ping contains the (legacy) client ID of the profile that last ran that Firefox installation. I manually counted such uninstallations for a native notification experiment: see https://docs.google.com/document/d/1zuFv8LX0o-5lTjxQzrEfDHPJZVKNQfsDrjOJagGWa2k/edit# and my notebook at https://colab.research.google.com/drive/1-kXArtZMU0wrsieez60iqXZom19XPOOG.
The text was updated successfully, but these errors were encountered: