Can confirm, I recently gave up on learning anything new re: data visualization and have just been using matplotlib instead. Training data for it has been pili

Can confirm, I recently gave up on learning anything new re: data visualization and have just been using matplotlib instead. Training data for it has been piling up since 2008. The AI's are so good at it that you hardly ever have to look at the code, just ask for changes to the graph and iterate. Honestly it's been kind of fun, but I do feel like the door is closing on certain categories of new thing. Local maxima are getting stickier, because even a marginal competence is enough to keep you there--since the AI will amplify that competence in well-trained domains by so much. Emacs lisp is another one. I'd kind of like to build a map of these.