Live Site: https://indeterminatemultitudes.glitch.me
Code: https://glitch.com/edit/#!/indeterminatemultitudes
Drawing on concepts from both our Understanding Networks and Biomaterials classes, Sam and I are creating a web application that generates a durational simulation of the pattern formation behavior of physarum (slime mold) network transport systems. The simulation is affected by nodes produced from geoIP location data intercepted from traffic over the network, which is monitored by a raspberry pi running wireshark (a free and open-source packet analyzer).
Physarum, a kind of slime mold, is known to exhibit behaviors that indicate the organism has a form of cellular intelligence. While researching slime mold simulations, I came across Sage Jensen's work with reference to a 2010 paper by Jeff Jones, “Characteristics of pattern formation and evolution in approximations of Physarum transport networks.” This paper provides a theoretical breakdown of the algorithm, which models both continuum and agent-based simulation systems, and draws parallels between computation, information processing, network infrastructure and physarum growth. Our project seeks to visualize this resonance between computational network infrastructure and its biological analog - while being critical of the naturalization of the artificial in bio-computation. In addition to the nodes being added as they come in from wireshark, the corresponding location data (City/State/Country) will show up on screen, giving observers a topological sense of how the autonomous systems that make up the internet are their own geopolitical territory.
Effect of sensor distance on pattern scale and granularity src
“pattern formation is intrinsic to growth and behavior. Inspired by one such organism, the true slime mold Physarum polycephalum, we present examples of complex emergent pattern formation and evolution formed by a population of simple particle-like agents. Using simple local behaviors based on chemotaxis, the mobile agent population spontaneously forms complex and dynamic transport networks. By adjusting simple model parameters, maps of characteristic patterning are obtained. Certain areas of the parameter mapping yield particularly complex long term behaviors, including the circular contraction of network lacunae and bifurcation of network paths to maintain network connectivity” -“Characteristics of pattern formation and evolution in approximations of Physarum transport networks”, Jeff Jones
We divided our to-dos between configuring the raspberry pi and reconfiguring the simulation algorithm to suit our needs. After encountering trouble running the MaxMind GeoIP plugin on WireShark on the pi, we decided to manually make calls to the geoIP API to serve back a response containing the city, state and country information. But, after setting all that up, we rebooted the system and the plug-in began to work. So, with this data, we needed to find a way to continuously capture relevant IP fields from WireShark and pipe it to a text file. The tool we decided to use was TShark.
In terms of the simulation, we needed to find a way to generate representations of multiple 'nodes' (collections of agents/particles), that would be added over time as the script pulled from WireShark. Jones’ model employs an agent-based layer (the data map) and a continuum-based layer (the trail map). The source code we were drawing on only depicted one group of agents, which were initiated in a function called regenerate(). Playing around with the parameters of the simulation produces a great variety of behaviors and patterns. The paper groups the patterns that can be produced into these classes: reticular, labyrinthine, island, and hybrid or incomplete.