4 ms·
Unwanted emergent behaviour in complex systems: Ant Death Spirals
- JoeAltmaier 16y agoNot a spiral! Simply a cycle.
- Qz 16y agoIf you add the dimension of the lifetime of the ant, then it becomes a spiral, hence death spiral.
- JoeAltmaier 16y agoI don't get it - where is the spiral? The ants are circling.
- Qz 16y agoYou're only thinking two dimensionally.
- pavel_lishin 16y agoNow you're thinking with Portals.
- JoeAltmaier 16y agoI guess the TShirt guy is too.
- sprout 16y agoCan't put time on a T-Shirt. Clearly he's thinking 3-dimensionally.
- silentOpen 16y agoTake the origin as the end of life. The angular position of the ant is maintained as the angular position of the ant-life. The distance to the origin is a measure of the ant's remaining life. As the ant walks around the origin in a cycle, the ant's remaining life decreases. When the ant-life reaches the origin, the ant dies wherever it was in the cycle.
- JoeAltmaier 16y agoI get it - its a spiral if you draw it that way. Then anything is a spiral - an ant marching thru the forest, me biking across Iowa, we're both approaching the 'origin'. I was hoping something about the ant motion would work into the spiral-ness.
- j_baker 16y agoHmmm... interesting. I suppose this could be thought of as a real-life example of livelock. Or maybe a deadlock with busy waiting.
- Jach 16y agoInteresting behavior. I like the one commenter's explanation that it emerges from ants simply being selected to follow ants in front of them.