How essential are statistics and databases for enjoying music?

[hiccup]

It's probably me.

And, it's a philosophical wandering.
So it's perfect for Beyond...

I have honestly been wondering about this for a long time:

Why do so many users put so much value in things like play-count?
Why are that many concerned about guarding and maintaining it till the end of times?
If you are managing a radio station, I might understand.

But if you are enjoying music, and enjoy exploring the incredible amount of music available these days, why worry about statistics?
Your brain is (supposed to be) X times more powerful than most current algorithms.
And it is X∞ times more personal.

Why unload your brain capacities to some database?
Isn't this about music, memory, emotion, spur of the moment?

[frankz]

For me, the stats are critical, and the way MB allowed me to utilize the stats is why I'm such a big fan.

My main listening playlist has 22000 songs. These are what I listen to on shuffle continuously through the day and night. For me, weighted variety is key to enjoyment.

My listening has always been based on date last played and date added.I've recently begun the task of rating my music and integrating that into my autoplaylist decisions. I like to focus most on new music, so newly added songs are generally available to be played for 5 months with no wait period after a play (unless they're rated below 3 stars). After that, general music (3 star or unrated) won't get shuffled again for 6 months. Songs rated lower than that go away for longer. Songs rated higher come back more quickly. 5 star songs do not go away at all. This way I get a good variety weighted to things I know I enjoy but with plenty of surprises to keep me interested.

.

[hiccup]

Thnx, that's interesting input.
I'm gonna digest it.

The musical comparison that comes to mind for me here is Bach.
Some find it distant and mathematical, others revel in it's mathematical beauty. (guilty)

Systems like yours may contain some mathematical beauty that I have not yet trusted to serve me ;-)

[psychoadept]

I don't worry too much about holding onto statistics til the end of time, but I do like having a way to be sure I cycle through most of my 10K+ tracks over time.

I'm VERY protective of my custom tags, though, many of which have been entered manually and losing them would be a huge setback.

For example, I like having playlists that I can let other people pick from to suit their tastes, such as when they're riding in the car with me. I'm still struggling with genre, but I've had pretty good luck creating playlists based on Billboard charts and decade. The fine details are more to satisfy my perfectionism and curiosity than for any practical use.

Then I have a whole collection of (mostly) industrial music that I use when I want to dance.  I use play counts/dates, a custom tag, and a playlist mixer to turn MusicBee into a reasonably competent personal DJ.

One of the few "special collections" I have that's actually complete is the Bond themes, which have their own playlist and custom album cover view. There are others in various states of incompleteness.

There are several other custom tags, some purely informational and some to help me keep track of my workflow and the status of certain files.

[frankz]

Your brain is (supposed to be) X times more powerful than most current algorithms.
And it is X∞ times more personal.

Why unload your brain capacities to some database?

Systems like yours may contain some mathematical beauty that I have not yet trusted to serve me ;-)

Thanks for bringing this up.  The part from your first post struck me as being entirely opposite of the way I think about my collection and listening habits.

Here's the way I think about it.  I listen to 3-5 new albums (albums I've never listened to) per day along with 3 Spotify playlists (New Music Friday, Release Radar, Discover Weekly) per week.  From those, I take the songs that speak to me in some way into my listening library.  At that point, my brain has already done the work of deciding what I want to listen to.  Then I rate them to determine how much I want to listen to them (how often).  That's more brain. At that point, I offload the grunt work - managing when things get played based on the criteria I've established - to the algorithm. For me, that keeps it exciting.  It's like listening to the radio, but the station only plays what I like, and it plays things I like a lot more than it plays things I only like a little.

I can't decide whether curating this library is my hobby or listening to this library is my hobby.  Probably a little bit of both.

[Freddy Barker]

My listening has always been based on date last played and date added.I've recently begun the task of rating my music and integrating that into my autoplaylist decisions. I like to focus most on new music, so newly added songs are generally available to be played for 5 months with no wait period after a play (unless they're rated below 3 stars). After that, general music (3 star or unrated) won't get shuffled again for 6 months. Songs rated lower than that go away for longer. Songs rated higher come back more quickly. 5 star songs do not go away at all. This way I get a good variety weighted to things I know I enjoy but with plenty of surprises to keep me interested.

Pretty much spot-on to the way I organise my music library, "Last Played Day And Time" and Ratings are key to how my listening is curated purely via auto-playlists and playlist-mixers with the results very much like a pseudo radio station.

Freddy