Social Network Paintings

Code Societies, Day 9 with Melanie Hoff & Dan Taeyoung

written by stud1nt

Day 9 of Code Societies begins with an introduction to Melanie Hoff and Dan Taeyoung’s collaborative work as Small Data Squad, “an internet forensics agency” that “research[es] our *patron’s* past and current internet selves and online data trails”. In 2016, Melanie and Dan set up shop SDS at an all too appropriate event, Internet Yama-ichi, an internet black market of sorts, to provide their internet reincarnation autobiographies and other consensual stalking services.

Melanie’s research “investigates social conventions, political rhetoric, and the ways in which representation actively shapes our interactions and consciousness” which accounts for how fantastically she curated our Code Societies program! Dan is Adjunct Assistant Professor in Architecture at Columbia GSAAP and co-founder of the guild for social good Prime Produce.

Today’s class is “an exploration of the most common representations of social networks and the ways the database as form has influenced these representations”. In other words, how do we understand and visualize our connections as a society of networked individuals? Are nodes and edges sufficient to describe the relationships we have with each other, systems, and culture as a whole? (Hint: probably not.) So, how can we reuse and reimagine our experience within and of society?

Melanie and Dan pose a question: what is the self-portrait of a society? How can we represent a network of networks? With regards to social networks like Facebook and Twitter, we discuss the difference between directed and undirected graphs. In directed graphs, all the edges are directed from one vertex to another whereas in undirected graphs, edges have no orientation.

We are introduced to the concept of the sociogram, graphical depictions of social networks, developed by Jacob L. Moreno in order to analyze decision making in groups. In the 1934 book Who Shall Survive, Hudson uses sociograms to explain a series of runaways from the New York Training School for Girls in Hudson. He posits that the social structure and relationship between the girls rather than their individual personalities compelled their choices.

Moreno’s sociogram

If data structures are attempts at modeling our lived realities, how can they do so in a way that is inclusive and creates space rather than delineating a binary? For instance, let’s consider the data structures used to capture ‘gender’ or ‘race’ as data points. If the data structure is binary or created without input from anyone in a position of marginality, the data is predisposed to capture inaccurately, if not unfairly. It reflects a way of being that does not speak to the multiplicity of identities.

Our first prompt of the day was to map collectively SFPC ideas and practices to and from each other. Students’ submissions ranged from “code / care” to “laughs / flirts”. We python to download these responses from Google Sheets, turn it into a network graph with Graph Commons, and create poetry by traversing / wandering around on the graph.

Prompt: Make a spreadsheet and populate it with ways that your body interacts with technology.

Next, we applied the same method to mapping the ways in which our body interfaces with technology to generate more poetry.

Prompt : Invent a taxonomy of edge relationships using charcoal / graphite

Melanie and Dan distribute large sheets of paper and markers so we can individually map how we arrived at the School for Poetic Computation. After, we form groups of 3–4 and share our connections to the space. Some of us found it was easiest to tell the story of our journey to class through our relationships with people and some took a much more literal approach in accounting for waking up late and delays in transit in their models. Next, we are asked to create a taxonomy of edge relationships on index cards with charcoal or graphite (inherently diffuse substances).

This proved to be much more challenging to represent: what if the unit of modeling a relationship is not a node and line, but the line itself?

We share our drawings with each other on the wall and consider if it is possible and useful to model them computationally. If we tried to represent networks as precisely as possible, would we just be simulating reality? We come to a stopping point with the idea that all models are wrong but some are certainly helpful.

The key both you and I are trying to find is the solution to the critical problems (wo)mankind will face in the coming decades. Each of us is searching with sincerity and devotion. What is profoundly different, however, is our basic strategy. You stand in the light, trying to move the lightpost closer to the place where the key might be. (fable from Groping in the Dark: The First Decade of Global Modeling)