To be fair, the Audacity UX designer made a massive video about the next UX redesign and how he tried to get rid of "modes" and the "Audacity says no" problem:
https://www.youtube.com/watch?v=QYM3TWf_G38
So this problem should get better in the future. Good UX (doesn't necessarily have to have a flashy UI, but just a good UX) in free software is often lacking or an afterthought.
You're making application for yourself and somewhere down pipeline you decide that it could benefit others, so you make it open-source.
People growl at you "It's ugly UX but nice features" when it was originally designed for your own tastes. The latter, people growl at you for "not having X feature, but nice UX".
Your own personal design isn't one-fits-all and designing mocks takes effort. Mental strain and stress; pleasing folks is hard. You now continue developing and redesign the foundations.
A theming engine you think. This becomes top-priority as integration of such becomes a PITA when trying to couple it with future features later.
That itself becomes a black hole in how & schematics. So now you're forever doomed in creating something you never desired for the people who will probably never use it. This causes your project to fail but at least you have multiple revisions of the theming engine. Or you strike it lucky and gain a volunteer.
Pre-emptive anti-snark: yes, the old version will still exist... if you can dig up the right github commit and still make it compile in 2030.
Now he got rid of the modes by adding handles and border actions - so 1) wasted some space that could be used for information 2) required more precision from the users because now to do the action you must target a tiny handle/border area 3) same, but for other actions as now you have to avoid those extra areas to do other tasks.
While this might be fine for casual users as it's more visible, the proper way out is, of course,... MODES and better ones! Let the default be some more casual mode with your handles, but then let users who want more ergonomics use a keybind to allow moving the audio segment by pressing anywhere in that segment, not just in the tiny handle at the top. And then you could also add all those handles to visually indicate that now segments are movable or turn your pointer into a holding hand etc.
Same thing in the example - instead of creating a whole new separate app with a button you could have a "1-button magicbrake" mode in handbrake
What is there to see? You add a bar that takes space. That space can be taken up by something useful. Just like you have apps that hide app title bar and app menus so you can have more space for your precious content. This is especially useful for high-info-density apps like these audio/video/photo authoring ones. Note how tiny those handles are in the video, why do you think that is?
> tradeoff is an incredible degree of customisation
You don't have that tradeoff, neither of the 2 solutions are anywhere close to "incredible customization", so you can pick either without it.
> In terms of precision, they're working on accessibility issues
Working towards what magic solution?
> but I'm not sure how this change is any special than any other UI.
why does it have to be special? Just a bog standard degradation common to any UI (re)design, nothing special about it.
> the modes were horrid
Of course they were. Just like they were horrid in that MS Paint app the dev worked on before. But you can make any UI primitive horrid, even buttons, that's no reason to remove them, but to improve them!
The space is used for information. The fact clips in Audacity finally have names and you can see those names is a fantastic improvement. The space taken up by the clip title is the handle.
"Audio 1 #1" "Audio 1 #1" "Audio 1 #1"
"Audio 1 #1" "Audio 1 #1"
names replacing the audio wave height!
But sure, if you need names permanently right there and are ok to lose space to show them, and if the handle is inconveniently small to only fit the text, then yes, you wouldn't lose space in that case. You'd only have other issues.
But that coupling likely has other design implications, e.g., you're unlikely to get an option to only show names on hover instead of having a bar, or to show names as an overlay (in many cases the names aren't that long to need to take the height of the the whole segment)