Conversation

I was thinking about an ontology for scenes. First attempt was a false start that didn’t hold together — a hierarchy from a handful of top-level scenes to zillions of low level subcultural ones. Second attempt didn’t work either — large superscenes subsiding sub scenes...
4
25
But I think I have it now. The right structure is a temporal DAG defined by exit-to relationships. When you exit a scene what scenes can you go to next?
4
14
Replying to
Even if individuals never return to an old scene, it's hard for me to see why the overall graph would be acyclical. Aren't some Republicans ex-Democrats, and some Democrats ex-Republicans? There's something here, though. Paths of least resistance? Maybe an electrical analogy?
1
2
Replying to
Quote Tweet
Replying to @zem42
Well if there’s a net direction locally (A to be is more likely than B to A) for every pair with significant flow, you’ll get at least a local order. I suspect if you impose a minimum net flow threshold to filter random noise you’ll get a proper DAG.
1