December 4ths

12/4/2008            

It will be interesting to see how music will be reshaped by economic and/or social upheaval. During the Depression the movie industry saved the music industry to some degree. Regardless of a bailout of one medium for another I think we will have different uses of music, and musicians will be responding to those needs (whatever they are). 

For most people music is encoded to emotions, not extraneous data that the brain is picking up. (This is similar to the theory that video frame rates cause more anxiety.) That said, I think the application of cognitive science to music marketing can affect our musical preferences to some degree. Whenever you say to yourself, “I like this but I don’t know why”, it has something to do with how the brain has processed it. I’m not so sure that an awareness of fidelity affects the musical product in any profound way, but a study of psychoacoustics and music preferences might be revealing.

[12/4/2025: There have been many times when I’ve generated music with AI and have liked it even with its flaws–garbled lyrics, bad audio quality, etc. If it’s pop and upbeat it will immediately lift my mood, even when I know I should be rejecting it. It is definitely a different use of music. What’s remarkable is how much more I listen to music now, having to wade through all the iterations of a song. In a studio, you’d be playing the same mix over and over and then make a final mix, but now you play multiple mixes and choose a final one–or several]. 

“Discuss musical preferences” 

12/4/2015

It’s interesting that the “labels” for being “represented” by someone in art or the music industry are becoming one and the same. I was thinking that music might be more interesting in a “gallery” metaphor, but the boundaries between the domains are more blurred now. Gallerists are becoming Producers and Labels.

In music, working with synths makes it much easier to compose a piece of music because it shortens the time between inspiration and straight to the resultant production, without hiring orchestras for rehearsals and performances. Similarly in digital art, the artist can conceptualize and produce work within hours, and use all the new tools to market it. But this is the case in many industries now, where all the middle operations are being squeezed out by the Internet.
                        
In any event, representation in the future will be much more independent, perhaps using AI, the blockchain, and mobile payments: You will represent yourself through your own algorithms. Algorithms as of now are in the black box of the “siren servers”, but it would be somewhat of a new Zeitgeist to create your own algorithms, separate from the ones pushed at us from social media. (The Internet works better as a “pull” technology).        

[12/4/2024: Lots of people want their own LLMs like they’d want their own website or server. I want my own way to generate music based on my own rules and algorithms from my Composition Strategies].    

[12/4/2025: Mostly a good prediction. On YouTube for example they should use a prompt format where the user can describe the experience they want, which is more “pull” and not “push”].

12/4/2022

Creativity is timeless, so there is no old or new. When I look at Roman art I still see great creativity. That said, to become “great” in the eyes of the world means you must create those things that are considered to be contemporary. In my view, it’s having the correct intentions with what you’re doing, which comes with time. Ultimately, creativity should have wisdom embedded in it.
            
[12/4/2024: I’ve always considered myself contemporary, even to this day. This is why I like seeing what I can do with new technologies like AI. I don’t think I’m too old to be checking out what the young artists are using to make art. But my boundary is whether it is done with the integrity of an artist’s overall praxis. If you’re young and starting with AI, in 20 years you have to be able to look at it as a period, and you move on to other things that interest you. That’s where wisdom is in the equation].     

***

The hardest thing to teach is how to read music--not so much knowing what the notes are on the staff lines and spaces, but how to read rhythms, and has to be done interactively, not just by watching a video or thinking about it intellectually. It has to be done through actually “rapping” it. 

The interesting thing about language is that its coherence lies in the rhythms of the words and whether certain syllables are on upbeats or downbeats. If you put upbeat words on downbeats and vice-versa it starts to sound like a foreign language. The first stages of learning to read rhythms are actually quite embarrassing but there’s no way around rapping them. #riff 

12/4/2024

What seems to be an effective process for using AI to generate music is to start with two or three lines and let it generate something. Then listen to that and come up with more lines, then regenerate it and edit the result accordingly. Sometimes I like to make another acoustic version with different music. two versions, one generated by AI and one generated by you. In any event, it’s kind of a strange feeling of being carried away by the fun of using an AI tool to make something within minutes and play around with the results. But you know you really need to get back to the serious work: 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

12/4/2025

January in December continues--frigid. The snow that fell days ago is still in its original forms on surfaces, with dollops of snow on tops of fence posts and other objects like tiny Buddhas.