October 7, 2011
This week I attended “Living in a de-material world: The design and maintenance of sustainable social networks,” a talk by Tom Finholt of the University of Michigan School of Information. Refreshingly, Finholt employs the broader, old-school definition of social networks which considers people’s relationships both online and offline. For Finholt, a sustainable social network might exhibit itself as academics walking to campus meetings instead of driving.
According to Finholt, when cradle-to-cradle energy consumption of ICTs is taken into consideration they are not always as green as their analog alternatives. When the embodied energy– the energy involved in mining of source materials, manufacture, and transport- is counted as part of the overall energy footprint of an MP3 player, it casts a bigger energy shadow than ye olde vinyl record.
The organizing principle of Finholt’s talk- dematerialization– is bandied about by economists who seek ways to replace more material-intensive consumption with less material-intensive consumption without tanking the economy. For example, extracting and producing a new copper wire for a communication network is more energy and material intensive than extracting and producing the material for a new fiber-optic cable though it can do the same job (and then some). But, according to Finholt, evaluating the environmental impact of information and communication technologies is still a tricky business. While other important domains such as durable goods and transportation have well accepted and well-understood ways to model energy consumption in their domains, there are many obstacles to creating valid models to spell out ICT’s impact. Just-in-time procurement of parts- a common practice in ICT manufacturing- means that two seemingly identical devices can have identical-looking components which in fact come from different factories with different energy footprints. The way that communication is routed over the internet makes it hard to know whether a message passed through server farms that were energy hogs or energy misers. Was a message sent through a server farm cooled by a 1970s-era cooling system running full tilt? And we can imagine that companies would balk at an academic’s attempt to determine whether Facebook runs a greener set of servers than Google or Flickr.
While the energy footprint models for ICTs need work, some things are strikingly clear. Finholt and his NSF funded collaborators hope to supplant academic conferences and meetings with remote teleconferencing because they are confident that the carbon savings derived would be significant. But wide adoption of remote technologies over in-person conferences is nothing less than a sea change in the way that academics carry out their professional lives. The fact that the talk I attended was in Seattle while Finholt is based in Michigan just underscores how uphill this mass behavior change will be. (Even researchers working on remote collaboration in a race to save the planet still jet around.) The solution that Finholt’s team has been experimenting with gets scholars to gather in smaller clusters closer to home.
Here is a video of a talk that Finholt gave recently at Microsoft: Living in a de-material world: The design and maintenance of sustainable social networks.Infrastructure, Manufacturing, Systems
© 2020 deeptech