Ask HN: Is there a cloud music library that doesn't keep deleting my music?
In just an hour, iTunes Match (a feature of the paid Apple Music subscription) tried to match the songs and deleted almost all of them and just left 3 songs in my playlist. I tried it a couple of times and this was still the result.
Apparently some songs aren't licensed for my location/country and/or aren't available in the catalog of Apple music for my country and therefore they were all deleted except the ones that Apple already has in their catalog.
That got me thinking - Is there a reliable paid service where I could just upload MY music I legally purchased and they wouldn't be deleted just because of some stupid location based licensing or some other reason?
Spotify is no better as my songs in their library have disappeared a couple of times as well.
Thank you in advance.
126 comments
[ 3.0 ms ] story [ 189 ms ] threadPossibly the final straw was when I wanted to purchase an album that existed in iTunes, but only in the iTunes "store" of another country.
Since then, I've just went back to maintaining my own library. I purchase lossless music from Bandcamp and Ototoy. I also rip CDs. I had been importing my files into VLC on iOS but recently switched to Sailfish and just copy my files onto an SD card. For streaming I setup my own instance of Koel.
Does it suck to maintain almost 200GB of audio files? Yes. Have I lost a single track since doing so? No.
In what way does is suck? I have about 1TB of audio files on local disks, which I manage with beets + mpd. No troubles at all.
I don't know what ongoing maintenance there would be after the initial import.
Backups? Automate them. Integrity checks? Use a better file system.
Right now, for me, its an external SSD drive and I have a Blaze backup that, in theory, when it works, when it hasn't decided to not backup for a week, preserves it all up in the cloud for me, while my Time Machine manages my main system drive locally.
I dread the day I'll have to restore all of this when the inevitable disaster strikes, but I SHOULD be protected (knock on wood).
It fits on one drive, so you can do all that with a few minutes of effort once every two or three years. Or just when you change drives, even less often.
> paying for, installing, maintaining the technology black hole that you fall down into trying to mitigate all of the above from happening in the first place.
Is that more than having a backup? You'd better be doing that anyway.
You could go fancy and set up Z-RAID but if you're really inclined to do that you'd probably already have it set up for your main files and then dumping the music on top is near zero effort.
There is a non-zero amount of "care and feeding" that goes into it.
For example:
Moving them around and re-indexing. I recently copied my library onto an SD card for a new phone, which had a new media player, which then had to re-index them all. As I add new songs to the main dataset, I'll have to sync them over to the phone.
I also manage a Koel instance so I can stream the files from wherever I want. But I like the ability to have them offline as well. And Koel is great but, if I ever decide to move to another utility I'll probably lose playlists or have to convert them over.
Ripping CDs, then moving files over to the main dataset takes time, not a lot but it is a thing. Also I have no shortage of CDs with incorrect metadata in the CD databases or don't exist in the CD databases. Or often there are multiple matches and I have to find the right one.
Scanning the CD album art which I've done for some rarer albums in which I could not find artwork for online (or what I did find was blurry and over compressed)
Managing metadata: For example, I've some CDs that were ripped as "TALKING HEADS", others as "Talking Heads". It's a little more painful with Japanese character sets e.g. "Tatsuro Yamashita" vs "山下 達郎".
If I purchase albums from providers like BandCamp or Ototoy, I then need to make sure the metadata matches what I've decided to go with.
I use ZFS with redundancy, and take regular snapshots but I also make occasional backups to external drives but I could also use something like S3 as well.
This is really a hobby at this point, not a convenience. With a streaming provider you fork over a few bucks a month and listen to whatever they have in their library.
Having a song silently removed from your "collection" after you discovered it on a streaming service and because it is streaming you only marked but don't remember the name is the most infuriating sucking thing ever.
Ever had a song you heard a long time ago and you never knew it's name then one day you find it? It's a gem, and then Google takes it away from you and you don't know the name and can't ever listen to it again. It's lost again...
https://www.bhphotovideo.com/c/product/248440-REG/Sony_CDPCX...
on Ebay, connected it to my home theater through a TOSLINK cable, and I've been loading it up with DTS encoded 5.1 discs. Also I got into the minidisc hobby
https://www.minidisc.wiki/
Like cryptocurrency, cloud music always ends in tears.
Why should that matter? As long as you're not making songs uploaded by one user available for another user to listen to without them also uploading the same song, it's fair usage. You're allowed to store the bytes wherever you want, it's the (re)distribution that is illegal.
YOU can store your bytes wherever YOU want. This would be someone else storing your bytes for you. You could get around it by using Dropbox or something but if you want a good streaming music system/app/experience, then you are locking into "knowing" that it is music, not just a user's files are a user's files.
For audio this is covered under the format shifting permission provided by the AHRA.
[0] https://funkwhale.audio/
Luckily, unlike many open source projects, there are a couple of (niche) managed service providers for it:
- https://cloud68.co/instances/funkwhale
- https://weingaertner-it.de
If you're willing to pursue the self-hosted root, there are loads of options beyond just Plex:
https://jellyfin.org/ https://funkwhale.audio/
I'd like to add Roon and Roon ARC to the list. Some people seem to like it a lot, but they do not support FreeBSD for their server, meaning it's quite tricky to install to my NAS...
https://roonlabs.com/
These are no longer popular, so not many services are doing this anymore. It's such a wasteland that Wikipedia has a page for closed services [1]
A few small services like medialeap.com still exist, but it is anyone's guess how long they will stick around or how good they are. But I guess you could try that.
[1] https://en.wikipedia.org/wiki/Comparison_of_online_music_loc...
4 years of Spotify pivots is missing from their[0] Wikipedia page (2006-2010). I signed up in 2007 and stopped using once I got annoyed with them removing songs from my playlist. They then pivoted to music streaming subscriptions, with the recording industry becoming a "partner".
[0] https://en.wikipedia.org/wiki/Spotify#History
https://github.com/navidrome/navidrome
I'm using DSub on my android phone to get a nice mobile player, which also sync's tracks locally to the phone if I wander into an area with low or no signal. Navidrome being compatible with the Subsonic API means there are several players on all platforms that can hook into it.
I drive through areas with no service for a few hours per week.
The other alternative is self hosting, plex and plexamp are popular.
https://support.google.com/youtubemusic/thread/52908732/faq-...
It's purely a loss for them financially and is a giant thorn architecturally for YouTube's backends since it's very different rights- and lifecycle- wise than their other content.
They pretty much only brought it over from GPM because they were worried about the user outrage if they didn't, so as fewer people care about curating music files they will eventually decide pulling the plug is worth it.
I'd uploaded all of my offline library to Google Play Music, back when it was relatively easier to do so, and before it became YouTube Music.
Unfortunately the service seemed to do some sort of acoustic fingerprinting to make uploads more efficient, and I ended up with the wrong language versions of several of my film soundtrack songs! Loads of them got randomly converted from Tamil -> Telugu, Tamil -> Hindi, Hindi -> Telugu, Hindi -> Tamil, etc.
My tags for Hindi songs were always in English. Because of the way Hindi songs are credited, searching by tags like artist was always annoying, but that's not a problem specific to google music. I did have times where Japanese track names got translated to English in the old Google Play Music, but I haven't experienced that with Youtube Music.
Now I either buy music via Bandcamp and download the FLAC or replace the ones I lost via other methods and store them locally in Plex and play with Plexamp + Tailscale. That should avoid them getting googled again from any service.
Pretty irritating, tbh. I haven't bought a license to listen to a song, I've bought a license to use destructible chunks of plastic or bits...
I can also recommend Youtube Music for that.
[1]: https://github.com/akrylysov/bsimp
https://github.com/navidrome/navidrome
https://github.com/sentriz/gonic
these have good app support so this could be nice
Thanks!
On a phone's tiny port, storage will stick out very far, snag constantly, and something is probably going to break.
So external storage is just not viable if you're going to be plugging it in for more than a few minutes at a time. And buying more internal storage won't be reasonable until someone puts out a phone where it's "only" twice as expensive as high-end microsd cards. (Right now iPhone storage costs 8x as much, Pixel storage costs 8x as much and requires buying the top model for larger sizes, and Galaxy storage costs 6x as much and requires buying the top model for larger sizes.
For subscription services, I get it. I don't love it, I think the contracts they've all negotiated are stupid, but fine. I get that a bunch of lawyers leads to me not being able to play almost any of my music while I fly over the Pacific.
But the combination of locked-down-OS (something I'm actually for) and legaleses / constant access negotiations means the effort I would have to go through to legally pay for music and actually hold onto it for good and then play it from my phone is too hard.
If you want to save a few dollars, you can get it close to $5/month if you go with cheaper S3 compatible storage from Backblaze[3] or Cloudflare[4] + the cheapest $2.5 instance from Vultr[5].
[1]: https://www.digitalocean.com/pricing/droplets
[2]: https://www.digitalocean.com/pricing/spaces-object-storage
[3]: https://help.backblaze.com/hc/en-us/articles/360037814594-B2...
[4]: https://developers.cloudflare.com/r2/platform/pricing/
[5]: https://www.vultr.com/pricing
[1] https://swinsian.com/
https://www.navidrome.org/
If you're already hosting Plex, what's one more Docker container running Subsonic? :)
It's best to just store it all locally (and make it read-only).
[0] https://www.reddit.com/r/Piracy/wiki/megathread/music/
It's cute that you care, but money is a lot better spent on merchandise from the band's store than on records if you're trying to financially support the artist.
“Your job is not to sell music. Your job is to sell merchandise, and your music is just advertising for your merchandise.”
If I were to rationalize it, I would say “stealing music” is essentially “stealing advertising”.
(By the way, I don’t think you should be downvoted so much. I believe in good-faith discussion, even if I disagree)
I'll buy a vinyl from the merch booth, maybe. But I have no reason to purchase digital content. And I certainly don't feel bad about entertainment piracy.
I play music. I don't care if people copy my music and share it. Every musician I personally know feels the same.
If you want a cloud service that keeps your CDs and doesn't fiddle with them it'll have to be an encrypted S3 bucket.
My perspective on this was formed when I sold a chemistry program I wrote on a TI-85 to another student in 1997. It felt ridiculous to charge a lot for something I still had after selling it so I charged $1. Inadvertently beating Apple to the $1 mobile app market by 8 years, but I digress...
Access to a physical item isn't the only reason we need something like copy right.
We also need it to successfully amortize the creation cost of things across many buyers, else you'd have to sell copy for the full creation cost else it'd become a first mover disadvantage.