I had to use the Arch Linux live boot usb today and I was pleasantly surprised to hear a short beepy music play when it got booted. It felt quite welcoming!

In my particular case, it served quite useful too as my laptop currently does not have an internal display, and nothing shows up on the external HDMI monitor until after the OS boots up, so it was quite reassuring as well.

I have now configured my GRUB to play a small sound as well (since I have multiple bootloaders installed).

How To

1. You need to edit the config file /etc/default/grub.
2. If you simply un-comment the line starting with GRUB_INIT_TUNE, it will enable the default tune which is a single and very short "beep" sound.
3. Or, you can edit that line to have it how you want. Some suggestions: https://jdnash.com/api-makers/grub_init_tune-necropolis/
4. Then update the grub config (requires root): grub-mkconfig -o /boot/grub/grub.cfg

And you're done.

Relevant Arch Wiki: https://wiki.archlinux.org/title/GRUB/Tips_and_tricks#Play_a_tune

I'm currently using Navidrome to self-host my music collection, while using DSub to listen on-the-go.

This works very well for most genres, except for the bulk of my music which is classical music. This presents its own host of problems pertaining to cataloguing and using metadata, since there are often multiple recordings of the same musical composition, recorded by multiple conductors with different orchestras and/or soloists. There may also be different instrumental arrangements of the same musical piece. Merely sorting by "Artist" is therefore quite unsatisfactory in this scenario.

Some streaming services have come up with quite satisfying solutions in my experience (notably Apple Music Classical and Idagio), but I am not sure how to go about listening to my own self-hosted classical music.

Is anybody here on Tildes familiar with this organizational problem? I would be very eager to hear how you have tackled this. Is there any self-hosted software more suitable to cataloguing musical collections with extensive metadata?

I'm working on an application that allows a user to view playlists belonging to a particular radio show and stream/download/favourite the tracks in them. It has 4 core entities: User, Show, Playlist and Track.

• Each show has multiple playlists (one-to-many)
• Each playlist has multiple tracks (one-to-many)

To be able to reference a playlist belonging to a particular show. I gave those playlists the same uuid as the show they belong to. A few questions though.

1. Is this the right/best way to associate data?
2. As a track could potentially belong to multiple playlists, I can't take the same approach as I do for (show/playlist) How would be best to handle this? Ideally I would like to have a single "Track" table containing all tracks for all playlists.

For any experienced database designers out there, how would you structure this data? What would you consider in designing the schema and why? If I did go with 4 tables only, presumably there would be performance implications given the potential amount of data in any one of those tables, particularly tracks. If that is the case, how best to structure this kind of thing with performance in mind? Thanks in advance for any help :)

For reference, in case it's of importance, I'm using sqlite3.