Over the last couple of weeks, I have been conducting a scoping review of data in the home. It has been interesting for many reasons – mainly because there is just so much data circulating! But also, because the connections between devices, people and infrastructures are so complex and shaped by a host of factors including culture, education, religion, economics and geography. The review brings together some disparate fields – from critical data studies, computational science, platform studies, communication, education, as well as parental mediation and screen time. At times, they are odd bedfellows but empirical studies in each of these fields sheds some light on data flows and frictions in the home. In this blog post, I reflect on the different types of data materialisations in the home and how their interconnections complicate the dominant understandings of datafication processes.
To organize the studies reviewed in the scoping review, I use an adapted version of Dourish and Mazmanian’s (2011) framework for materializing information. I have written about this in an earlier blog post, but in the review I use it to explore three ways that data are materialized in the home – infrastructures, devices and goods, and representations. What I’m interested in is how data is ‘apprehended’ by householders or how they experience it in their everyday life. For example, when the green light is flashing on the modem, the family knows that the internet is connected and data is flowing into and out of the home. This is a materialisation of data. In many respects, the modem is like a bottle neck through which all data going into and out of the home must pass.
Data is also materialised through devices like laptops and mobile phones as well as goods like smart fridges and security systems. Another growing source of data generation in the home are datafied toys. While all these objects exist in the world – i.e. they can be seen and touched – without data their function is greatly changed or compromised if data is not flowing. The media objects that exist in the home use data in different ways and this, in turn, shapes the more direct representations of data that householders encounter in the form of dashboards, visualisations and graphs. There are obvious differences between data materialisations that take the form of objects (i.e. infrastructures, devices and goods) and those that are representations (i.e. dashboards, graphs and visualisations). Representations are, in a sense, an interpretation of data that has been generated, collected and processed.
Each materialisation effects the other. The quality and performance of the broadband network will influence the devices and goods that a family can use in the home; and what devices and goods a family has will influence the kinds of representations of data they encounter. Data circulates through these different materialisations, however, it is the representations that are decisive. Through representations all other materialisations are registered and interpreted. For example, the efficiency of the broadband network will show up in how devices and goods perform, which are represented through visualisations and dials on a dashboard or bill.
Another interesting point, is that the interpersonal, familial relations that exist in each home complicate understandings of what constitutes ‘personal’ data. For example, a connected toy or baby monitor is often coupled with an app on a parent’s phone. Data flows from the device and is processed elsewhere, but the representations of this appear on the app. Data that is collected and processed through the device or good and accompanying app yield information not on just one individual, but rather on the dyad (parent/baby). Research into parent control apps has shown that many of the most downloaded apps violate privacy laws, sharing personally identifiable information of child and parent with an array of third parties. This is what Goulden (2018) calls ‘interpersonal’ data – data that is ‘generated by and observable to the members of a group or setting’ (p.1582). But how is this data processed? In the case of the baby monitor, who is profiled? Do smart devices complicate notions of the end user, or provide information on more than one person? The processing of personal data is based on the premise that it will lead to reliable predictions about an individual’s behaviours. Materialisation of data in the home complicates this equation and highlights the need for research into how interpersonal data is processed and used in the digital economy.
Dourish and Mazmanian’s (2011) framework adopts a layered approach to understanding data materialisations – resonant of Bratton’s (2015) ‘stack’ – where each builds vertically on the other from infrastructures to devices and goods to representations. What is missing however is how data flows between these layers and the factors that influence these flows. In the case of the family home, the generation of interpersonal data and the relational dimensions of data are (understandably) not a focus. Furthermore, materialisations higher up in the ‘stack’, such as representations in the form of apps and dashboards, actually influence materialisations beneath, like devices and goods, meaning the drivers for adoption and adaptation of technologies are complex and interconnected. Further research is required to understand how data flows across discrete settings like the home and how it mediates the relationships of those that live there.
Thanks to Julian Sefton-Green for talking through some of these ideas and introducing me to new ones – I’m grateful!
References:
Bratton, B. (2015). The Stack: On Software and Sovereignty. Cambridge, MA: The MIT Press.
Dourish, P., & Mazmanian, M. (2011). Media as material: Information representations as material foundations for organizational practice.Paper presented at the Third International Symposium on Process Organization Studies, Corfu, Greece.
Goulden, M., Tolmie, P., Mortier, R., Lodge, T., Pietilainen, A.-K., & Teixeira, R. (2018). Living with interpersonal data: Observability and accountability in the age of pervasive ICT. New Media & Society, 20(4), 1580-1599. doi:10.1177/1461444817700154
new ways of understanding digital data 