Networked infrastructure and the black box

Greetings from Montreal, soon Medellín… some thoughts on opening the “black box”.

In the recent article Small technologies, big change (PHG 2011) that Nicolas drew your attention to, I was trying to bring geographical and STS approaches to networked infrastructure into interaction for the specific purpose of moving away from the “black box” metaphor of networked infrastructure.

In the geography literature, a variety of metaphors about infrastructure (e.g. exoskeleton, cyborg urbanism…) place networked infrastructure as stabilized and isolated from users. It supports their daily lives, but they do not interact with it or influence it. Similarly, in the large technical systems literature, LTS are taken as black boxed, and purposefully so. They are built for durability and immunity from users.

However, what I found in my research was that many utility managers actually seek to engage a variety of user groups in the management of the infrastructural network through the introduction of relatively simple technologies into homes and businesses. These technologies, which I dub mediating technologies, can have a significant impact of the network. For example, they can reduce the strain on network capacity and on the environment. The technologies that I examined included a variety of home and business retrofits to improve water efficiency (reduce consumption) as well as different types of software to assist homeowners and large industrial consumers to detect leakage beyond the property line, encouraging them to fix problems themselves.

So, not only was the LTS (in my case, water and sewer infrastructure) malleable rather than rigid, managers actively sought to open up the black box and integrate users into its management, rather than striving for invisibility. All this could be accomplished, not through a gargantuan unearthing and remodeling of the system, but through the addition of relative simple technologies to its peripheral nodes.

Here, STS theory on the interaction of users becomes very important because it tells us that users can interact with technology in a variety of unintended ways, producing results that were not the intention of the developers of that technology. Thus, with the purposeful integration of users into the management of LTS, they become both more malleable and less predictable.

Such a shift, from stabilized black box to malleable and interactive, has the potential to generate a variety of progressive benefits. In Montreal, where I live, for example, there is no water metering and thus very little user information about their relationship to the system. The black box is retained, as are high levels of consumption and leakage. In other communities in Canada, utility managers found that by increasing user information, through metering, and giving users the tools to manage their consumption (e.g. low flow devices), a variety of positive effects resulted. These included reduced consumption (and sewage outflows) and delayed infrastructure expansion.

In Medellín, where I’m heading, users are integrated into the system in a variety of ways. Beyond, the standard mediating technologies that I discuss in the article, users are integrated directly into the construction of the infrastructure. In order to create jobs in Medellín’s low-income barrios, in 1998, the local utility EPM began contracting to the barrio councils (the JACs) to build needed infrastructure. EPM guides the JACs through the process and the JACs hire all local labor. Both residents and EPM staff find that by employing local people to build the infrastructure, it is built to a much higher standard than when the utility contracted the work to private construction firms. The process also develops the capacity in the JAC and the community to monitor the new systems and alert EPM of any problems. Thus, the black box can be opened up in a variety of ways with a variety of interesting consequences.

1 thought on “Networked infrastructure and the black box

  1. Temporality might be the answer. I was thinking about how, if we could track, for example, workarounds over time that we might get a new view of "when" and under what circumstances lead to stabilization, or, as the years pass, what sorts of support are easiest to achieve ….Sociology is not so good with temporality….


Comments are closed.