“New Distraction” by fanclubwallet

Super catchy pop-rock tune. Beautiful production, great hook, really strong vocals. 📺

“Waterfight” by Bruiser and Bicycle

I love the way this song moves in and out of the beat. Really interesting melody and instrumentation. Good vocal harmonies. And the little synth line that appears here and there sits beautifully on top of the otherwise acoustic feel. 📺

“Lasting Impression” by Ava Luna

This song has such a cool rhythm. Boy / girl vocals always win with me. And the bass on this achieves an enviable tone. 📺

“Honest Man” by Freak Slug

Some songs have a chewy feel that makes me just want to eat it. The hook on this track is infectious. 📺

“Animal Kingdom Chaotic” by Jesca Hoop

Not technically new to me — this is from a forthcoming re-imagining of Hoop’s 2017 album Memories Are Now. But after 7 years, this song has incredible staying power and remains one of my all time favorites. The vocal melody here is off the hook, interesting lyrics and some really nice guitar work. Also check out Simon Says from this album. 📺

“Adalynn” by Ava June

I may be slightly biased on this one, but my daughter released her third single in September. She’s not on Bandcamp (yet!) but she’s on the major streaming platforms. I’m incredibly proud of the work she’s done on these and cannot wait to hear what she does next. Be sure to check out the other two, Eliza and Part of Your Life.

I use Spotify and YouTube for discovering and listening to new music, whether new releases or just new to me. That probably amounts to 30–50% of my listening time. The rest of the time, I listen to my collection of digital music. I have been collecting MP3 files for decades. Much of it was transferred from my large CD collection in the late 90s and early 00s. I bought a lot of music from MP3.com and eMusic back in the day. And from Amazon and iTunes. What I couldn’t find in digital format, I’d buy from CD Baby or elsewhere, rip the CD to MP3 and never touch it again.

When our child was old enough to start discovering music, I realized I was in a music rut and hadn’t really paid attention to new music in years. I discovered new music again along with my daughter as she started getting into music. My collection needed some new blood, and I wanted to at least be familiar with what is popular.

I used Google Play Music at first and just started listening to new releases across various genres and saving what I liked. Turns out I liked a decent slice of the new music coming out. Later on, because what had become YouTube Music kept dropping the ball on important midnight releases (my daughter, Taylor Swift), we switched to Spotify.

Here is my process for finding and collecting digital music:

• Listen to new release and recommended playlists each week. Listen to genre specific new release playlists. I often don’t have a lot of time for this, as it requires me to pay attention to the music rather than just have it playing in the background. I just do it when I can and feel like it.
• Add songs I like to my Liked playlist. New songs tend to fall in one of three categories: 1. I’m not interested. 2. It’s interesting. 3. I like it. Songs that fall into buckets 2 and 3 go into the Liked playlist.
• Listen to the Liked playlist. Remove anything I’ve added that I’m not interested in after a few listens. Maybe half of the “interesting” songs end up dropping off over time. Sometimes even a song I really liked at first listen sours on me after a few more listens.
• Every three months, purchase the songs that have been added to the Liked playlist since the last quarter and add to my collection. I do this seasonally, breaking up the year into Spring (Mar-May), Summer (Jun-Aug), Fall (Sep-Nov) and Winter (Dec-Feb).
• After purchasing and downloading, I tag the files for consistency, apply a standard set of genres, cover art, and add it to a new quarterly playlist, so I can always go back to that time.
• The music is then transferred to my Navidome instance, which runs on the same server that hosts this site. Navidrome is an open source self-hosted streaming service, and I love it.

I’ve been doing this methodically for 5 years now, adding around 100–150 songs each season.

You’re probably wondering why I go through this trouble and added expense instead of just making playlists on Spotify. There are several reasons:

• A preference to own my music and have full control over it, rather than renting it from a service. Streaming services are generally considered bad for artists. They can also raise their prices, change features or shut down at any time.
• The ability to organize things as I like and decide what does and does not make it in.
• Not feeling locked into a streaming service. This process works on any streaming service without having to migrate or rebuild my collection.
• The joy of tinkering, customizing Navidrome, grooming and otherwise maintaining the collection. Navidrome keeps a database that I can access to do lots of interesting things, and direct access to the files enables me to write scripts to do just about anything I want.
• The inclusion of music that isn’t available on streaming, including that of my own creation, friends’ creation, and local artists.

Sadly, purchasing digital music has become more difficult as streaming services have taken over. I buy mostly from Bandcamp, Amazon and Mp3Million. Aside from the subset of music available through Bandcamp, Amazon is the only legitimate source for digital downloads. If Amazon were to follow their peers and stop offering digital sales in favor of forcing listeners into their streaming service, there would be nowhere to buy most music digitally that legitimately compensates artists (or more specifically, their labels), and that’s a scary thought. I also hate putting money in Bezos’s pocket, so I’ll buy older music from Mp3Million. If a song is not available anywhere digitally, I record it from YouTube or Spotify. Those instances are limited to a few songs per quarter, though. If I like most of an album that isn’t available on Bandcamp, I may also purchase it on vinyl or CD and convert to MP3 from there.

Here are some stats from my collection, as of this writing:

• 1,527 artists
• 2,808 albums
• 12,747 tracks
• 6,465 “liked” tracks kept in regular rotation
• 4,700 archived tracks that are not in Navidrome at all

Tracks (total / liked) by decade:

Navidrome supports the ability to “like” tracks, and I use that in a Smart Playlist that I call “Like FM” to dynamically serve up 150 random liked songs that I haven’t listened to in 180 days or more. This is what I usually listen to — it’s like my own personal radio station available from anywhere and always serving up stuff I haven’t heard in a while. And since my music tastes are fairly broad, it always makes for an eclectic ride. Here’s the definition for Like FM:

{
  "all": [
    {
      "is": {
        "loved": true
      }
    },
    {
      "notInTheLast": {
        "lastPlayed": 180
      }
    }
  ],
  "sort": "random",
  "limit": 150
}

{
  "all": [
    {
      "is": {
        "loved": true
      }
    },
    {
      "notInTheLast": {
        "lastPlayed": 180
      }
    }
  ],
  "sort": "random",
  "limit": 150
}

Navidrome provides a nice web interface for listening, similar to streaming services. It also implements the Subsonic API, which is supported by a wide variety of clients and players. On the desktop, I like Supersonic. For Android, I like Simfonium.

Tagging for genres is a never-ending process. MP3 tags support multiple values, so I can tag tracks with multiple genres and use Navidrome or Simfonium to filter on them. My current strategy is to pick one or more primary genres that best describe the track, then add additional descriptive genres. Here are the primary genres, with track counts indicated in parentheses:

• Rock (6,038)
• Pop (2,128)
• Soft Rock (764)
• Electronic — not Pop or Dance (638)
• Jazz (577)
• Funk (318)
• Reggae — includes Dub, Dancehall, Ska, etc. (298)
• Folk (272)
• Hard Rock (205)
• Hip-Hop (187)
• New Age (175)
• World (154)
• Punk (153)
• Classical (137)
• Bluegrass (86)
• Dance (81)
• Fusion (67)
• Metal (32)
• Blues (33)
• Country (29)
• Ambient (14)

Then I add any of these modifier genres that apply, though most tracks haven’t been augmented with these yet:

• Mainstream (1,492)
• Instrumental (1,254)
• Jam — improvisational (799)
• Vinyl — ripped from a record (491)
• Soundtrack (193)
• Alternative — artsy/avant-garde/quirky (107)
• Minimalist (80)
• Childrens (41)
• Oldies (62)
• Vocal — solo vocal or highly vocal focused (29)
• Holiday (12)
• Chill (7)
• Cover (6)
• Psychedelic (5)

These have changed over time and are still changing. I’d love to get even more descriptive with genre tags like Drums, Guitar and Bass to denote interesting solos or parts, or tags to better describe the mood of a song like Happy or Melancholy. I use Puddletag for managing tags in my files, and I’ve also written a custom Python web app that I call Multigenre for doing large scale genre tag management. I’ll post about that in more detail sometime.

I’ll leave this here for now. There will definitely be more posts on this topic in the future, and I may promote this post to a page that I can better maintain going forward.

If you’re a fellow digital music collector, hit me up in the comments or email me. I’d love to talk shop and exchange notes.