Media Precedents: From Flow Charts to Systems Diagrams
A genealogical primer on representing process and movement through diagrams
For lab {L03}, the focus is less on technical workflows than on using tools and techniques we’ve already covered (or are closely related to those we’ve covered) for more complex compositions. The key media precedent we’ll be breaking down and building back up is the systems diagram. This brief primer on diagrams of system and process should give you a sense of how you might deploy the techniques shown below in your own work using relatively simple workflows between QGIS and Illustrator from the previous labs (including this week). The real work is in composition, and thinking carefully, critically, and creatively about how you classify the elements of a system, how you draw a boundary around it, and how you connect those elements within and beyond the system.
First let’s return to the diagrammatic work of systems ecologist H.T. Odum (and his collaborators) we explored in lecture: the systems diagram. Systems diagrams are a subgenre of the more widely-recognized “flow” or “process chart.” How we understand flow and process — and the physical analogues and metaphors we use to model them — is crucial for how we understand what constitutes a “system.” This apparent in the stylistic changes of Odum’s diagrams over time: he moved from a more mimetic style of diagram (in which flows of energy appear sinuous and fluid, like a river and its tributaries)…
…to a more technical energy systems “language” mirroring contemporaneous computing circuit diagrams, in which different symbols represent different functions and entities in the system.
The key point here is that as a genre of spatial media, the resulting systems diagram and its operational “grammar” allowed for a generalized application of the diagram beyond ecology as long as the object of representation could be portrayed as a system. By that logic, entire countries could be represented as systems of energetic inputs, outputs, and socio-metabolic processes…
… and it was not long until these diagrammatic representations were rescaled to the world system and planetary socio-metabolism. In the case of Jay Forrester’s work at MIT on a series of “world models,” the diagrams also represented the logical circuitry of computer programs built to simulate different outcomes based on various resource allocation, energy production and consumption, rates of population growth, and other key sociometabolic variables.
While energy systems diagrams date from the early post-war period (during Odum’s) involvement with U.S. Atomic Energy Commission and its nuclear weapons testing program), they borrow heavily from at least four earlier lineages: i) geographic diagrams of movement (especially of people, exemplified by Gemelli’s 1704 Aztec migration map and the works of Charles Joseph Minard)…
ii) control engineering for thermodynamic systems (exemplified by Irish engineer Matthew Riall Sankey’s diagrams first published in 1898)…
iii) early-20th century industrial and supply-chain standardization (exemplified by Lillian and Frank Gilbreth’s “process charts”)…
…and iv) the design of analog electronic computing circuits during the build-up to World War II (exemplified by mathematician John von Neumann’s development of electronic computing apparatus for the U.S. military), which were closely associated with early “cybernetic” efforts to represent the brain as a computer and model neuronal pathways as circuits (exemplified by Warren McCulloch and Walter Pitts’ “logical calculus” of neuronal activity).
This abbreviated genealogy suggests a close affinity between the diagram styles (and they way they represent systems, process, and flow) and dominant theoretical systems, political economic concerns, and the science and technology of the time (a small part of a broader “technical” regime, in the more expansive sense we discussed in lecture.) We can see these legacies at work in contemporary attempts to diagram how systems work across increasingly indistinguishable social and natural spaces, like Kate Crawford and Vladan Joler’s Anatomy of an AI project.
This “anatomy” takes the form of a systems diagram of intricate detail and idiosyncratic uses of diagrammatic convention from across the examples above. For instance, they develop their own symbolic grammar to represent elements of the labor process from a Marxian perspective (labor power, products/commodities, the means of production, and waste).
They deploy different drawing styles (realistic wireframes, maps, cross-sections, a periodic table, geometric diagrams) to capture the heterogeneity of the parts of the system while using a variety of linework styles and widths to represent different kinds of connections between those parts.
Finally, they organize it compositionally such that reading from left to right follows an expanded “lifecycle” of the particular artifact they’re interested in mapping, the Amazon Echo Dot, from mineral extraction and logistical distribution to internet infrastructure through to abandonment and disposal.