was a quick build so no guarantees. but it's pulling our repo live via github api. rest is next.js + react with canvas 2d for the graph and d3-force for physics. zustand manages state between graph and sidebar, framer motion for animations. fuse.js for search. background is some custom webgl shader. hmu if u need any specific context.
grab your popcorn boys,
academia is about to POP.
honestly i’m so excited to see what’s going to happen to the already rusty and super dysfunctional institution of academia, now that AI adaptation is through the roof.
Beyond crystallization, an underrated function of a knowledge artifact is that it triggers new cognitive cascades when activated ...
Here an interesting article from Cornelius which goes further 👇
trace viewing prototype
video shows mock data to test more of the UI (its ugly)
but already works on @arscontexta
trace start injects temporary hooks into the target project, waits for a session to start, and traces all tool calls and messages etc
trace end removes the temporary hook but keeps the trace
(can resolve files and git later)
@ecoMLdev@arscontexta repo growth tree demo for the visualisation
it shows @arscontexta plugin starting to build a new knowledge base on this.
it already did some surface level research on epistemology etc and now high level input on Hofstadter books (GEB, Strange Loop, S&E)