I run Debian 9 with PulseAudio on a computer with two sound cards. The screenshot below shows the Volume Control application when no audio is played back. When audio is played back, a drop down is displayed where I can select the output device. Is there a way to select the output device *before* the audio stream is started? To me this seems more natural. Edit: Here is my typical use case. I want to play a song from my computer on the living room stereo (sound card B) instead of through the desktop speakers (sound card A). Since I cannot select the output in advance I have to start the song and have it played back on the desktop speakers, then select the living room speakers once the option shows up in the playback tab in the Volume Control window. Then I need to play the song from the beginning now that I have selected the desired output. This is bad UX.

I use the [quake drop-down terminal](http://guake-project.org/) a lot. Recently, it started to switch between tabs when I use my mouse wheel anywhere in a terminal. Usually, I use the mouse wheel to scroll within the terminal, and _not_ to switch terminal tabs. Now the behaviour ist, that it does both: It passes a scroll event to the currently opened terminal tab, and switches to the next tab, and so on, until I stop scrolling. What might have caused this, and where can it be switched off; where is the corresponding configuration switch? I am not sure if this is a guake issue at all or some GTK3-"magic" which was silently imposed upon me with some upgrade. In the settings of guake (Right click -> "Preferences"), I don't see that I have mouse scroll assigned to any action, and especially under "Keyboard shortcuts", "Go to previos tab" and "Go to next tab" are assigned to `+ and +`, respectively. I am using guake version 3.9.1.dev0 (according to guake --version) from [Arch Linux](https://archlinux.org/packages/community/any/guake/)) . I am not using gnome, but I am using xfwm4 as window manager, without a desktop manager, and lightdm as session manager. I do start up everything I want to have ready at Xorg startup via ~/.xinitrc. I have already reported an issue upstream: [Here](https://github.com/Guake/guake/issues/2142) . Regards!

Many command line utilities accept long ("GNU style", according to ) options such as

--version

. To my surprise, truncated versions are often interpreted as the full option. For example,

from GNU Coreutils gives

user@computer ~ $ df --version
df (GNU coreutils) 8.32
Packaged by Gentoo (8.32-r1 (p0))
Copyright (C) 2020 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later .
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.

Written by Torbjorn Granlund, David MacKenzie, and Paul Eggert.

The same output is obtained by replacing

--version

with any of the truncated variants

--v

,

--ve

,

--ver

,

--vers

,

--versi

and

--versio

. However, when I add additional letters to the option, it suddenly realises that the option is invalid:

user@computer ~ $ df --versionn
df: unrecognized option '--versionn'
Try 'df --help' for more information.

Is this behaviour intentional? What is the reasoning is behind it? My understanding is that most utilities rely on

or

(or similar), so I have looked into their manpages for clues, but not found any explanation. E. S. Raymond. *The Art of UNIX Programming*. Pearson Education, Inc, 2004

Debian and Ubuntu provide .deb package files that can be installed directly. what about Fedora? Can all tar-balls be one day .deb files for ease of installation ... many people lose themselves installing after gunzipping. There is a lot of code which commoners don't get an idea of when doing ./make etc.

In Debian 10, but since older versions (always using Gnome), the scrollbar usually follows the mouse (the scrollbar knob is always below the cursor), but sometimes it scrolls the page contents much slower (the knob lags behind the cursor - e.g. the cursor have "walked" 500 pixels, but the knob only "walked" 50 pixels). Not a Gnome image, but you got the idea of what I'm calling "scrollbar knob": Most of the time I get the first behavior. Rarely, and unpredictably, I get the second. Actually I'm more used to see it happening on vertical scrollbars, but I guess it may happen equally on horizontal. And it usually happens when the content is much larger than the window, so that the knob is actually much smaller than on the image. After five years using Debian+Gnome, I still can't figure what causes the difference. Can anybody explain it? I didn't find anything on Google (maybe I don't know the right words to search?).

When I press Super on Gnome shell, I get the overview of the current workspace with all open windows. However I can't select any window using the arrow keys unless I press one time on the DOWN key or the TAB key. Is this a UI / UX bug ? I tried on Fedora 31 and Ubuntu 18.04 ; same issue.

I was wondering what or how we can do to develop a better Linux kernel or variant (if necessary). What problems does Linux have and how could they be solved or if Linux needs some feature? **Justification:** Due to the system of usefulness of response, the great community that is here and the interest that can have in this response both to a passionate newcomer and to an expert. I think it is a question that could strongly favor the community, its development and the future of Linux development and variants.

I have Mi Notebook Pro (15.6 inches), it has very big, usable touchpad (apparently, manufactured by Elantech). But it's palm detection works very poorly. I started to bang my head with it in hope to tweak my touchpad and found that there is actually _two_. Here is the list: $ sudo libinput list-devices | grep -A18 Touchapd Device: ETD2303:00 04F3:3083 Touchpad Kernel: /dev/input/event5 Group: 6 Seat: seat0, default Size: 125x78mm Capabilities: pointer gesture Tap-to-click: disabled Tap-and-drag: enabled Tap drag lock: disabled Left-handed: disabled Nat.scrolling: disabled Middle emulation: disabled Calibration: n/a Scroll methods: *two-finger edge Click methods: *button-areas clickfinger Disable-w-typing: enabled Accel profiles: none Rotation: n/a -- Device: ETPS/2 Elantech Touchpad Kernel: /dev/input/event8 Group: 10 Seat: seat0, default Size: 125x81mm Capabilities: pointer gesture Tap-to-click: disabled Tap-and-drag: enabled Tap drag lock: disabled Left-handed: disabled Nat.scrolling: disabled Middle emulation: disabled Calibration: n/a Scroll methods: *two-finger edge Click methods: *button-areas clickfinger Disable-w-typing: enabled Accel profiles: none Rotation: n/a The touchpad which actually produces input events is ETD2303:00 04F3:3083 Touchpad (/dev/input/event5), I found it out using libinput debug-events. It raises a bunch of questions. The general one is why there is two touchpads? The touchpad in my laptop have fingerprint scanner (I'm not using it), could be that the second one is a fingerprint scanner? The second major question is as follows. It seems like there is no proper driver for my touchpad in the kernel because of strange name of listed touchpad (is it PCI ven:dev IDs?). How to check that? And last, but not least question is how to tune palm detection so it will work more conveniently? Is it possible in the first place? I've Budgie 10.5 (Gnome stack 3.28.1), xorg 1.20.3, xorg-driver-input-libinput 0.28.1, libinput 1.12.3, and linux 4.19.8. # UPD I accidentally looked at kernel logs and found very interesting thing: it is _fuuuuuull_ of this event: i2c_hid i2c-ETD2303:00: i2c_hid_get_input: incomplete report (14/65535) When I did it there was almost _billion_ of this event! I decided to reboot and then measure frequency again: $ uptime 01:13:31 up 13 min, 1 user, load average: 0,14, 0,61, 0,67 $ journalctl -k | grep i2c-ETD2303:00 | wc -l 24219 As you can see, ~24 thousands of same event in 13 min is insane. So, here is another questions: what's it to hid-i2c here? Isn't that this touchpad driver is elantech? I found this question , there is advice in the answer to disable "HID over I2c" feature. I've no option in firmware setup to do that. Can it be done programmatically? Is that the touchpad "phisically" wired over I2c?

I am looking for a way to create download-and-install style programs for Linux. To clarity what I mean by "download & install", I mean that the installation process proceeds like this: 1. User downloads graphical installer (NOT a deb, rpm, etc. file... I'm talking about an actual ELF executable) from website 2. User runs graphical installer (preferably by double-clicking it) as a regular user (not as root) - installer puts program & its dependencies (included with the installer) in a nice place in the home directory and adds things to the desktop environment menus to make running easier 3. User can now run the program from the desktop environment menus 4. Uninstaller is installed alongside the program, which can be run at any time This style of download & install is very similar to the method the majority of traditional personal computer users (in contrast to mobile PC users which are accustomed to app-stores) are used to and I wish to distribute any programs I make using such a method.Plus I just really like this method of software distribution. It just feels right. I predominantly use C++ and FLTK for Linux programs. I am interested in all of the different options available for replicating this experience, so feel free to mention different methods of "getting there" Assume that a program has already been created and that it needs to be "retro-fitted" into this installation style

I recently installed Ubuntu-Mate on an old Acer laptop. It runs greats and apps like VLC (as an example) run great. There are fewer crashes and buggy behaviour. On VoidLinux/musl I have to constantly use killall -9 vlc to force it to shut it down. On Ubuntu-Mate and Linux Mint, it's more stable. In other non-Ubuntu OSes, I had to configure the X.org driver for Radeon graphics to eliminate all of the screen tearing during video playback. However, Ubuntu-Mate takes care of (most of) the tearing. (Although, Linux Mint does not automatically take care of the screen tearing during video playback.) So the question is: Are applications heavily modified and patched to work with the more common Ubuntu-based OSes? Is it because Ubuntu devs are willing to use proprietary or closed-source drivers? Or is it because the software is just compiled differently (e.g. different linked libraries and compilation options)? (Disclaimer: I'm not a programmer, but I've installed different OSes over the uses for my own personal use and curiosity.)

I have multiple browsers on my system (Firefox, Google Chrome, and Chromium). I normally use one of them at a time based on my needs, but when other applications want to open a browser they always open their/system default browser. Is there an application or script that is able to detect if a browser is already running and use it as current default browser? **EDIT** I don't want a browser to detect its own instance! I want a browser requester program/script to detect any already open browser. - For example assume I have Firefox, Google Chrome and Chromium, and I click a link in a PDF file, and Chromium is already open. I want the link to be opened in Chromium. - At another time, Firefox is open. Then I want the link to be opened in Firefox. Actually, I want already open browser have more priority than the system default one.