> You're looking at virtual memory and not resident memory. That does not change anything in the conclusion. 685M kwin_x11 417M plasmashell

> You're looking at virtual memory and not resident memory. That does not change anything in the conclusion. 685M kwin_x11 417M plasmashell 161M krunner 74820k kded5 58612k kdeconnectd 53368k polkit-kde-authentication-agent-1 52460k ksmserver 50364k kaccess 49488k plasma-browser-integration-host 48568k kwalletd 44268k kglobalaccel5 44044k org_kde_powerdevil 44020k klauncher 40384k kactivitymanagerd 26068k kdeinit5 24432k xembedsniproxy 21508k kscreen_backend_launcher 18964k file.so [kdeinit5] file local:/run/user/1000/klauncherwpWDlK.1.slave-socket 11672k kio_http_cache_cleaner 3984k kdesud 3652k xsettingsd 96k start_kdeinit 144M kate 96464k konsole ---- 73828k xfdesktop 59788k xfwm4 44508k xfce4-screensaver 38552k xfce4-appfinder 36260k xfce4-panel 29392k xfce4-session 29216k wrapper-2.0 /usr/lib64/xfce4/panel/plugins/libactions.so 27156k wrapper-2.0 /usr/lib64/xfce4/panel/plugins/libsystray.so 27064k Thunar 25284k xfsettingsd 25284k xfsettingsd 25140k tumblerd 17656k xfce4-notifyd 15248k xfce4-power-manager 5880k xfconfd 54492k mousepad 51244k xfce4-terminal The findings from the article are non-reproducible, too. When I run the desktop environments with a text editor and terminal each, `free -h` reports a clear difference in the "used" column: KDE 5.19.4: 1.2Gi Xfce 4.14.5: 521Mi