Data is changing the world around us. Or, is it? In Data & Society, we will explore how digital data is produced, communicated, disassembled, reassembled, sold, and acted upon in the early twenty-first century. Our objective is to develop a critical yet informed account of why data matters in contemporary societies and how it might affect our collective futures. The class requires students to read, participate, discuss and engage with current debates about ‘data’ in general terms and digital data in particular.
Grading
Data audit (Due week 3): 20%
Data platforms (Due week 5) 20%
Algorithmic surveillance/audit exercise (Due week 7): 20%
Final project: 40%
Week 1: Where does data come from?
Session 1: A pre-history of data
In this session, we’ll try to understand what makes ‘data’ so unique. Is data the same as ‘facts’? What does ‘data’ represent? Where does data come from? How is it generated? How is it reproduced? Is it stable? Is it unstable? When? How? Why?
The relevant reading for this first session is the introduction of JoAnne Yates’ Structuring the Information Age (Johns Hopkins University Press).
Session 2: The constraints of digital data
In this session, we will look at the specific ways digital data is generated. Our discussion will be mostly autoethnographic and will require each student to compile examples of how data represents him/her in different domains of life (credit, grades, health records, etc). We will use these as opportunities to think about how data collection practices, whether automated or not, reproduce existing categories and, with them, moralized conceptualizations of society.
The relevant reading for this session is Fourcade, Marion, and Kieran Healy. “Seeing like a market.” Socio-Economic Review 15, no. 1 (2016): 9-29. And, Lupton, Deborah. “How do data come to matter? Living and becoming with personal data.” Big Data & Society 5, no. 2 (2018): 2053951718786314.
Week 2: Why big data is so ‘powerful’
Session 1: From statistics to prediction
Contemporary discussions about data often refer to ‘big data’ or ‘data science’. What do these concepts mean? In this session, we will explore the recent shift from older discourses of statistics onto newer vocabularies about ‘data’ and ‘prediction’. Why is prediction so powerful? And why is it so fraught?
Readings for this week will include Janet Vertesi’s now classic account in Time Magazine: http://time.com/83200/privacy-internet-big-data-opt-out/ and Jäger, Kai. “Not a new gold standard: Even big data cannot predict the future.” Critical Review 28, no. 3-4 (2016): 335-355.
Session 2: The end of the end of theory
A famous claim about the widespread availability of data in contemporary societies is that it heralds the end of ‘traditional’ theory. How valid is this claim? In this session, we will think of theory’s place in big data.
The relevant reading for this session is Anderson, Chris. “The end of theory: The data deluge makes the scientific method obsolete.” Wired magazine 16, no. 7 (2008): 16-07.
Week 3: Inequality, through data
Discussions for this week will be based on Automating Inequality, by Virginia Eubanks.
Session 1: In the name of efficiency
What compels organizations to adopt data-intensive practices? What happens when they do so? In this session, we will think about how the imperatives of scarcity, efficiency, and objectivity interact within organizations to automate their data-processing practices.
Session 2: The road to hell is paved with predictions
Are designers evil? In this session, we will look at whether systems designers explicitly code inequality into their systems.
Week 4: Biased data
Session 1: Why designers matter
For this session, we will discuss how design of systems affects outcomes. We will start by discussing the origins of computing through Marie Hicks’ Programmed Inequality (MIT Press). We will then think of how organizational make-ups translate into data blind spots and the choice of data categories that may encode biased, unfair, or unequal ideas about the world. We will discuss the case study represented by this Guardian article: https://www.theguardian.com/technology/2019/mar/28/big-tech-ai-ethics-boards-prejudice
Session 2: Racist algorithms
In this session, we will look at how the definition of a ‘training set’ in Machine Learning can have consequences on an algorithm’s predictions. In particular, we will see how biases in data can generate biased estimations. We will discuss several case studies to be uploaded to TritonEd, including Noble, Safiya Umoja. Algorithms of oppression: How search engines reinforce racism. NYU Press, 2018.
Week 5: What’s the value of data?
Session 1: Addictive clickbait
Here, we will consider the way digital platforms entice users to remain within constrained, valuable, monetized spaces. Specifically, we will think of how algorithms create cages, traps, and calculated connectivities in order to generate financial value. Readings will include: Gillespie, Tarleton. “The relevance of algorithms.” Media technologies: Essays on communication, materiality, and society 167 (2014) and Seaver, Nick. “Captivating algorithms: Recommender systems as traps.” Journal of Material Culture (2018): 1359183518820366.
Session 2: Platform capitalism
How is data an object of capitalist economies? Here, we will address the broader connections between digital data industries and new (but oldish) configurations of capitalism. The relevant reading will be Langley, Paul, and Andrew Leyshon. “Platform capitalism: The intermediation and capitalization of digital economic circulation.” Finance and society 2, no. 1 (2016).
Week 6: Project workshops – how to audit algorithms/data (Case study: Facebook)
Week 7: Data infrastructures
Session 1: Railways, underwater cables, and hacked routers
This session is about the material substrates of data. How does data travel? Where does it reside? And how does this reflect the path-dependencies of history and the possibilities of situated action? For this week, we will discuss chapters from Dourish, Paul. The stuff of bits: An essay on the materialities of information. MIT Press, 2017 and Starosielski, Nicole. The undersea network. Duke University Press, 2015.
Session 2: Controlling revolutions
As a way of exploring the connections between data and technological infrastructures, we will discuss the case of Huawei and its relation to the rollout of 5G technologies.
Week 8: Data privacy
Session 1: All those traces
Who knows digital data? And what happens with this? After a brief reflection on the social and political history of privacy, this first session will look at how data ‘privacy’ is constructed, negotiated, and presented. Readings will include fragments of Kitchin, Rob. The data revolution: Big data, open data, data infrastructures and their consequences. Sage, 2014. And the paper by Amoore, Louise. “Biometric borders: Governing mobilities in the war on terror.” Political geography 25, no. 3 (2006): 336-351.
Session 2: Data as geo-politics
In this session, we will consider the global regulation of data. We will discuss Amoore, Louise. “Cloud geographies: Computing, data, sovereignty.” Progress in Human Geography 42, no. 1 (2018): 4-24 and the rift between US and European approaches to data regulation.
Week 9: Alternatives?
For this week, we will consider alternatives to data as we know it. Can we produce participative modes of data generation? Can we rethink property rights a world of collective data ownership? Can we design platforms that produce, rather than extract? For this week, we will read extracts from Lanier, Jaron. Ten arguments for deleting your social media accounts right now. Henry Holt, 2018. And Dobusch, Leonhard, and Jakob Kapeller. “Open strategy between crowd and community: lessons from wikimedia and creative commons.” In Academy of Management Proceedings, vol. 2013, no. 1, p. 15831. Briarcliff Manor, NY 10510: Academy of Management, 2013.
Week 10: Presentations/closing discussion