As I sorted the other day though my materials folder, for the first time in a long time, with real curiosity and fascination, a division that I was thinking about for a while finally clicked, a general but very handy rule of thumb that helps you classify the jungle.
I have this folder called "motifs"; --kinda vague and badly use of the word, but the term 'speaks' to me and that's what matters--. It's the place where I put conceptual ideas for songs, stuff that is in the abstract form rather than actual recording snippets or written word. For example, let's say, I listen to a song with mandolin and synthesizers in a movie and I think "hey, I'd like to compose something with that vibe". That's just an idea for when I'm a mandolin+synthesizer mood, all I need is a reminder. Or matters that I want to expand as lyrics.
So here's my aha moment: the composing stage of music creation can happen in plenty of ways, but in my case all those ways fall into two main big fat categories. Either there's pull (the melody that comes in the shower and tells you "write me! write me!"), or there's push (the day that you wake up and say to yourself "hey, I feel like composing a new song today").
The motifs, then, would be push - you can review them when you feel you're in "the mood", like reading through a menu of nice foods you've selected for yourself-. And most of the rest would go in the pull category (the riff pulled itself out of you, and when you play it again under the right conditions it will keep on developing, like in a game). Of course, speaking of "pure pull" would paint too nice a picture; in most cases, usually, there is always some "push" moments involved in finishing anything; those moments where you tell yourself "I just have to finish this now, I know it's the time", and you get down to it. in other cases, once the pull is exhausted, all you can do is wait, let it cool down, wait for the next pull "wave" to happen. Some songs are quick to "manifest" and others can travel with you for years.
I don't know yet how this division will translate into actual processes, but it's a great step towards the clarity I'm always looking for: getting to understand better what I do, getting to see it in a different way.