Raspberry Pi Zero 2W running Moode Audio. It's perfect. I have an old iPad running a kiosk-like browser to control but the norm is via web UI remotely.
Hey @hebocon I've been using jellyfin for both music and movies for close to a year...but keep hearing of navidrome (for music) as well. So, can i assume that you don't really use jellyfin for music, and instead use navidrome for that? How does that work for you? Any issues of interference, etc? I guess if i wanted jellyfin only for movies/tv, i could point it only at the folder where such media is saved, and then separately manage music via navidrome...but wondered what the benefits would be? Genuinely curious if you woiuld kindly share more info about your setup. Thanks! :-)
I set up JF for music first, then added Navidrome. I then added Moode for my separate stereo speakers and Home Assistant too. Never had an issue with everything having access at the same time.
Navidrome has worked well for me for the last couple years. My collection (~80 GB) is pre-organized FLAC but Navidrome will transcode to MP3 if needed. I use Substreamer on Android to connect to it (Airsonic API/protocol) or the WebUI at home or work.
Just the right balance of simplicity and features for me.
if not for the work requirement, at 80gb you could likely do what I do: use syncthing to make there be a full copy of the files on your phone. I've got a media terminal, my laptop, and my phone each keeping each other up to date. it's never broken or been frustrating. it works offline perfectly.
> Navidrome does not support browsing by folders, but simulates it based on the tags with a structure like: /AlbumArtist/Album/01-Song.ext
I don't think I have seen this tag simulation when I tried it around 2 years ago. But in any case, is this good enough? And does it recognize artists and songs from MusicBrainz like Jelly does flowlessly?
I don't know what you mean by "recognizing songs from musicbrainz", I've never used jellyfin to be able to compare. In my case, all files are tagged outside Navidrome.
All my players allow browsing by either Album Artist, Artist or Album. My folder layout follows this principle, so I'm happy with that.
But I can imagine layouts which don't adhere to this (classical music comes to mind), in which case I can see how not being able to get to the folders can be annoying.
Recognizing from MB is this: you have title1.mp3 and you can use MB fingerprint to detect song/album. Or song simply has subset of tags and MB fills up the rest, along with the links to MB details of artist/album where you can get tiny little details.
Discogs is a bit better in that, but its proprietary, so no.
I am more proficient with folders and have the number of tools to do so. Any GUI that streaming servers present is very unusable for me, I use it only for major happy case. Anything more advanced, I drop to the file system.
For Musicbrainz I use beets to import/organise the songs into my library directory structure.
For stuff not on mb I use fb2k as it has a fairly decent tagger and move it to the local external drive, which is synced to the server.
At home I usually just use fb2k to play to my sound system via an interface, and on the go I use play:Sub connected to the navidrome instance transcoded to 160k OPUS (initially over tailscale but now via portforward/cloudflare and soon cloudflare tunnel)
I use FB2k with picard and MB. If its not in the MB, I add it myself. With bookmarklests and picard this is very fast process.
Beets is too much work. I don't always have shell around nor I want to remote for this. This thing I use works on whatever machine I am currently.
I use jelly for convenience to connect to my media server when I am not at home. At home, I always use foobar2k which simply rocks for precise search and randomly generated lists (I even use SQL for this, via plugin). Its playing capabilities are far from any jelly like streaming server. Jelly is very bad at non-typical case, you can't even share a link to the current playlist and if left alone, after a day or two I have to reload jelly home page and go from there again, as anything that was left in the browser for a couple of days stops working until I reload from home.
I used to add stuff to MB but it's a bit involved as a process and honestly I'm too lazy a lot of the time. Picard is fine as a GUI tagger but I like foo.
If I can't pick what to listen to I quite like radiooooo, Radio Paradise, Radio Meuh, etc.
The lecture is the explanation.
The tutorial/lab is a guided experience (Microsoft's fake Contoso company comes to mind).
A how-to guide is what you need when you want to search/chatgpt a solution.
Reference material: thesaurus, encyclopedia, owner's manual.
Video games have built-in tutorials and yet walkthrough guides are still necessary for tricky puzzles. You check the key bindings as reference.
Your videogame example shows the problem. Tutorials are effectively just a different form of walkthrough. You are getting walked through the basics. So that's my point, you can't have different words on your website that is supposed to help people gain understanding and then start off by confusing everyone with words that are all so close in meaning.
The purpose is always the same: To learn. It's how you learn and what your learn for that are different. A tutorial is for getting exposed to something. A guide is for a very specific purpose in something you're already familiar with. An explanation is for a deeper understanding and references are for not relying on memory. They all refer to different things that ultimately falls under the same concept (and intersect with each other). If you're building a system, you should have all of them otherwise, the documentation is lacking.
My experience trying to implement this for internal use in a tech-adjacent industry is that "how-to" guides dominate because of a company culture of expediency over quality.
With time and care, those can be turned into tutorials and explanations. I use Obsidian to embed reference PDFs.
The end result is too hyperlinked for most to understand and the all-in-one outdated mega page in OneNote gets passed around instead much to my disappointment.
I don't know what that would look like for software but in another talk he was bemused by the narrow-mindedness of programmers who only saw the value of A Pattern Language as a mechanical tool rather than as a way to create a more beautiful world.
I take it one step further: I'm not only invisible but am being actively targeted with prejudice by those who can see me. You very quickly develop a knack for predicting what "cages" will do at any given moment.
As a pilot, there is a sense of "falling behind the aircraft" when your situational awareness diminishes --- getting ahead of the plane by 30 seconds, then 5 minutes, and then 30 minutes translates to non-flying tasks/jobs too. It's a great mindset for "incident management" in any field.
reply