Back around the turn of the century my wifi card died when I was reinstalling my system and I had no money for a new wifi card, no internet at home so I ended up with a very a basic console only Arch install, only audio software I had installed was SoX. I started out using SoX and Bash to make music, explored Lame's ability to encode anything as an mp3 and eventually discovered what TFA talks about. I never made anything I would call good, it just is not a method all that compatible with my interests but it has stuck with me all these years and has left me feeling that much of computer music has stagnated (in method, not output) and we have a great deal of room to explore yet.

Stagnated is not quite the right word, I think what computer music has been doing in the last couple decades is establish its primary instruments and techniques, the various audio DSLs, which is a fairly important thing musically speaking, it builds the culture and repertoire. Computer music is strongly rooted in how the musician interacts with the code, it is the strings of their guitar and I think we have barely touched on exploring that relationship yet. What is the prepared piano of computer music? how do I stick a matchbook between the strings of the code or weave a piece of yarn through it?

I hope more go back to exploring these very basic and simple ways of generating sound with computers and start experimenting with it, there is more out there than just ugens.