| > In what way does is suck? 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. |