- This discussion of platforms draws on an ongoing discussion with Benedict Singleton, Nick Srnicek, and Alex Williams, on their potential as technical models and model technologies. Singleton, “ On Craft and Being Crafty ” (PhD dissertation, Northumbria University, 2014); Singleton, “ The Long Con, ” in When Site Lost the Plot, ed. Robin Mackay (Urbanomic: Falmouth, 2015); Srnicek, “ Eyes of the State: Machine Thinking and the Production of Economic Knowledge ” (presented at the Incredible Machines conference, Vancouver, March 7–8, 2014); and Srnicek and Williams, Inventing the Future: Folk Politics and the Struggle for Postcapitalism (London: Verso, forthcoming).
- For a nontechnical overview, see James Gleick, The Information: A History, A Theory, A Flood (New York: Pantheon Books, 2011), especially for a history of longitudinal demarcation, network synchronization, and time zones.
- The history of notational systems deserves more than a few dissertations. In order to engineer his difference engine, Charles Babbage first had to invent a now long-forgotten notational system for how parts would be manufactured to make sure that one worker ’ s gesture would be compatible with that of another. Architects know Louis Kahn ’ s notational system to analyze and plan the movement of autonomous agents through pathways, used but once for his unrealized project to redesign automobile traffic flows in Philadelphia. His beautiful diagram of how cars might turn left and right hangs on the wall of the Museum of Modern Art in New York may prove to be a particularly breathtaking prototype for future shipping manifests.
- See Rem Koolhaas, Delirious New York: A Retroactive Manifesto for Manhattan (Oxford: Oxford University Press, 1978).
- “ Generative entrenchment ” comes from William Wimsatt ’ s philosophy of engineering. See his Re-Engineering Philosophy for Limited Beings: Piecewise Approximations to Reality (Cambridge, MA: Harvard University Press, 2007). Thanks to Benedict Singleton and Reza Negarastani for the original reference.
- See Srnicek, “ Eyes of the State. ”
- For a Marxist perspective on this asymmetry, see Matteo Pasquinelli, “ Google ’ s PageRank Algorithm: A Diagram of Cognitive Capitalism and the Rentier of the Common Intellect, ” in Deep Search, ed. Konrad Becker and Felix Stalder (London: Transaction Publishers, 2009).
- This example comes from conversations with Bruce Randolph Tizes.
- The stack model referred to is the network protocol stack, but this is not the only form of a software/hardware stack. Memory stacks based on FIFO (first in – first out) are taught in basic data structures courses. Application programming stacks are central to how web applications are engineered. Any complex software system is built from layers of code designed to perform different low-level or high-level functions, from the operating system kernel, to system utilities, to core and application services, to applications themselves and the graphical user interfaces that translate human-cultural interest into machine-executable instruction and back again. In systems designed for a densely networked computing environment where groups of machines may be serving applications, accessing shared stores of data, and parsing data for distribution over the web to provide specific solutions, the programming architecture becomes more layered, more generic, and more modular. Programming stacks are sometimes an aligned combination of an operating system: a server, a database system, a programming and scripting language for the authoring of unique applications. For example, the LAMP stack (Linux/Apache/MySQL/PHP, or Python) is (was) a widely used, open source set of programming languages and related tools and libraries for building general-purpose web applications. (This generic structure allows programmers of code on one layer, the Python scripts, to rely on the code constructed at the other levels by other programmers, knowing that the entire system will work as planned. Available open source code written for one project may be repurposed for a particular project, and as we move up the stack, that code requires more specific programming for specific applications.) For all such programming stacks, layers divide the computational labor of input, processing, storage, and machine control in regular and predictable ways, and so the performance of one is dependent on the other.
Notes
383