Carbon Footprint of Streaming: Illusion of Objectivity

Extract

[...] These values include manufacturing, unlike international models. But this development is hampered by a recurring problem: the lack of data, due to companies' refusal to disclose the characteristics of their infrastructures. This point directly refers to the concept of "data voids" studied in class: the absence of reliable data is not neutral, it allows certain actors to remain invisible or to limit criticism." In conclusion, the carbon footprint indicator of streaming is a telling example of the impact of quantification on the formation of public opinion. [...]

[...] The infographic of the [...]

[...] The second effect is the choice of the allocation formula, by Go, by minute or by subscriber. As Roussilhe points out, this attachment to Go, the most commonly used administrative option, allows for a 'proper' and aggregated accounting. This problem fits perfectly into the logic of Desrosières, according to which an indicator is never neutral but carries the interests of those who construct it. The symmetrical nature of the attribution by Go symbolically returns the environmental impact to the user - since, of course, it is assumed to consume a volume - while most of the effective impact is actually attributable to fixed equipment such as television or boxes, whose consumption hardly varies with the volume of data transfer. [...]

[...] The core of the problem is therefore the following: Does the quantification of the carbon footprint of streaming create an illusion of objectivity? The article by Roussilhe shows that the numbers never fall from the sky: they are produced by a set of actors, conventions, and technical or political constraints. This deep inscription in social choices echoes what sociology of quantification studied in the course reminds us: numbers are not innocent, they 'shape the world' (Desrosières) by making certain dimensions visible and others invisible. [...]

[...] We learn that in Europe, the average carbon footprint of an hour of streaming would be 56 gCO?, but this number falls to 10 g CO? in France due to the energy mix. Roussilhe emphasizes that these values do not take into account the manufacture of equipment, which is often the most environmentally costly phase. We see here a first classic effect of quantification: what the indicator does not measure disappears from the debate, as explained by sociological approaches to the construction of data. [...]