March 28ths

3/28/2014
 
The sonic landscape of a city has its own ecology of “organisms”: the screeching sounds of metal against metal, background hum from power lines, buzzing neon, passing elevated trains, pneumatic machinery—the natural sounds of a city landscape that becomes a quotidian experience, yet unobtrusive and perhaps has a calming effect. (This was mixed to not interfere with the top layer of attention). 

3/28/2015

In the old radio days one had favorite DJs pre-packaging potential influences, now musicians have to wade through it all. In visual art there’s the load of Art Forums to pursue new interesting work.

[3/28/2026: A good way to understand a new technology is through the lens of another. Broadcast radio would be a way to consume content peripherally in the background like a radio playing without our having to interact with it. This is essentially McLuhan's theory. Scrolling is the bane of a medium being too cool. But you don't want it too hot either. That would be totalitarian control].

3/28/2023

One of the biggest cliches is “3-chord rock”. People always quote it when they talk about something being very simple and easy to digest, and is used as a polemic against prog rock or jazz, which has four or more chords. It's too simplistic because even if you're using three chords maybe it's not harmonic at all. It's rhythmic. The complexity comes from the rhythm or the melody, and there could be something rigorous and difficult about it as well. If you try to play it, it's only three chords of course, but there are other things that are complex.
 

3/28/2024

Drop 31 songs. Interesting: It has nothing to do with AI, yet is more innovative. AI is supposed to look like the future (as it does in dystopian films) but lacks real innovation. I doubt the musical AI churn will ever sound innovative, and thus really is not the future.

[3/28/2025: What a difference even 6 months makes. I’ve already spun out at least 50 AI tracks. Is it innovative? No. Is it interesting and fun? Yes.]

3/28/2025

The routine for almost the past year is to go through the diaries and find riffs for new pieces of music. This morning I used a title idea from 3/28/2004, “The First Thing I’d Do”. This process will never be able to be done by AI. You can ask it to generate a rhythm, perhaps based on words as I'm doing now, but it will never be the same. But at the same time, you can compare them and decide whatever works better as a compelling piece of music. These A-B comparisons are interesting to do. Perhaps both will be equally as good. But like a Chicago deep dish pizza, you can't replicate it at home, the corollary being AI won’t have je ne sais quoi. Humans always hold the trade secrets or imprimatur for consciousness.

 A doowop version: