I would rather have the same "random" order than what Spotify does; random never plays some songs, and overplays other. I get true random has the possibility of that, but when I've heard the same song for the 4th time in an hour, on a 20-hour playlist, something is wrong.

I have this issue with youtube music only for some reason. For me "Shuffle" seems to be starting random, then after a handful of songs entering a loop of 6 or 7 songs. I never figured out why or looked it up, I just refresh the page. I know it must be only me somehow, but don't know what's happening.

My car's player (reading from a flash drive full of MP3s) doesn't have "shuffle", it has "random". The difference is subtle, but it means that each time it advances to a new track, it selects that at random. The result is that I never hear all the tracks in an album/playlist without hearing a bunch of repeats. Infuriating!

But when I play from my phone or PC, I use PlexAmp (backed by my Plex media server), which is the opposite extreme. Not only does it know what randomly shuffling is, but the server has analyzed the sonic profile of every track, and knows about similar artists and similar albums. So you can play an "artist radio" featuring that artist and other related stuff. And you can choose an auto DJ, which inserts between tracks the closest sonic match to the song it just played, or a brief detour of about 3 tracks that are spaced along the path from the track that just played to the one that's about to play. It's fantastic for discovering stuff you haven't heard in a long time within a very large library.

Pretty much every USB music player has the same faults:

1. shuffle play does not work

2. crashes if there are more than N songs

3. does not find songs in subdirectories

All of these are trivial to fix. I do not understand the programmers who make the firmware.

Amperfy for the Iphone randomizes the shuffle.

I just looked at the code and it's apparently using the Fisher-Yates shuffle with a call to arc4random each pass. It's literally eight lines of Swift!

I've never suggested this before, but if the Android app store has truly been as bad as you claim for as long as you claim, you should seriously consider buying an Iphone to solve your problem.

Is what you call the "shuffle button" a "shuffle playlist" button or a "shuffle mode" button? There is a subtle difference between those two and for what I understand you want the second

> the main reason is that LLMs are generators, they like to generate code where code removal or reduction would solve the problem

Don't over-index on this choice of words. LLMs can just as easily generate a diff that removes or refactors code, they don't have any intrinsic "preference" to always add more code, unless it's been trained into them.

A lot of the early training had concentrated on "wow factor" one-shot style demoable use cases, where the models indeed ended up with a tendency to assume they were in a "produce something new for me right now" situation and often or always started churning out additional code, even without all relevant information. As training regimes improve, and as customer demand evolves, this is changing, and the latest big models place a lot more emphasis on maintenance tasks - still not enough, but it's at least in the mix. As they mature, I'd expect the models' "instinct" to gradually align with what everyone's idea of a senior software engineer is, taking time to figure out if they can simplify and refactor rather than always adding more.

It's a new field, it will take time to get it right, but I don't doubt it will eventually be right.

> Because they are trained on mediocre code that is the vast majority of what's available on the internet

Again, this is outdated. It's true that the earliest training took this "spray and pray" approach, but data quality filtering is now a huge focus and the labs have made a lot of effort (and progress) in figuring out what good code is and getting the models to follow that. They've massively improved, in my experience, and it'll only get better. Again - give it time, this is new.

And Winamp.

I mean they mentioned they had the code on codeberg initially, so it must have been in a git repo before. Maybe they just did the ol' `rm -rf .git && git init .`

The point is that the author wrote the music player app because they wanted specific features and for their own pleasure/coping.