Linux sound isn’t so bad if you’re willing to work at it. Right.
Actually Linux is notorious as being a difficult platform for audio professionals to work with/tolerate. Even many of the people who do podcasts and videos about Linux for a living have admitted that it falls short when it comes to the sound system. Too many who don’t wind up using MacOS for audio production revert to Windows.
A couple of things I’ve found along the way that some people may find valuable.
1. Before making any change in the defaults first determine how (a) to back up those settings; and (b) how to restart things after resetting.
2. Don’t spend a lot of money on expensive audio hardware like microphones or mixers until you’ve had some time to experience the limitations of sound on Linux. While I’ve got a fairly good pair of headphones (an old Sony MDR-XD200 whose ear pads are fast wearing out), I’ve recently settled on just using the USB mic built into my Logitech QuickCam Communicate STX because none of the many other mics I’ve tried work any better.
3. USB microphones do seem to work better than traditional phono connector types. Static and crackling due to poor electrical grounding of phono connections is so common in most PCs that I suspect bad design or manufacturing processes by computer (and computer component) makers are to blame.
4. If you’re using a USB mic, be prepared to deal with having to adjust some applications (like Audacity) to accomodate changes in the USB device ID after disabling the device or restarting the system.
5. There are variations in how sound has been implemented among the various distributions of Linux, and even between major versions within each distribution, so be careful when following advice given on mailing lists, forums and even official documentation.
6. The shipping defaults in some distributions, particularly Fedora, can be awful. For example, I’ve found that out-of-the-box mic volume is always set too high, resulting in overmodulation and background hissing.
7. PulseAudio is an inextricable part of every leading distribution. It cannot be removed. It is also difficult or impossible to bypass. Pulse’s shortcomings have in fact been the major impediment to Linux becoming a professional audio platform. If every blind defender of Pulse were forced to give up part of their salary towards fixing it I believe some balance of justice in the universe could be achieved.
8. By all means try the Jack Audio Connection Kit, but be prepared for things to go wrong. In my case integration of Jack with PulseAudio worked just fine, except that all my volume levels, output and input, fell to a whisper. After looking for a solution over several hours I finally backed out of the integration because life is just too damn short.
9. There are far too few real-world configuration examples and diagnostic procedures offered by most Linux sound projects. If developers are going to continue to force this kind of horrendous complexity on us, they have an obligation to provide us with the tools to deal with it.
10. Many audio applications, like Audacity, wind up forcing users to install pretty much every Pulse, ALSA and Jack related package in the repository. Keep that kind of insanity in mind when looking to cull what you might reasonably expect to be “excess” software from the system.
But I’m not holding my breath.
 The Fedora 18 Musician’s Guide is the most complete single source of information about sound on that distribution. But it is in fact already dated and things in later versions of Fedora do not work exactly as described there.