In computer science, we often guide and evaluate work by metrics such as efficiency (of execution, of task performance, of maintenance, etc). But such metrics do not make sense for many types of computational media work. Fox Harrell’s talk at the Media Systems gathering, “Matching Methods: Guiding and Evaluating Interdisciplinary Projects,” suggests that, rather than there being one answer to evaluating computational media research, part of the work is in identifying values and goals, which can then point to the methods that might be appropriate.
For example, he discusses Mimesis, a game exploring identity representation and prejudice. One goal of such a project could be helping conceptual change happen for players. This kind of goal is a value in both computer-supported cooperative work/learning and activist contexts, which suggests that methods from these contexts might be appropriate for guiding and evaluating the work.
Of course, rather than this sort of innovative search for the right match, we often see mismatches running from the amusing to the disturbing. Harrell’s talk opens with some humorous dramatizations of this potential. He also presents two dichotomies: “cultural critique versus evaluation” and “creating culture versus serving needs.” These are both useful starting points for consideration, and yet also break down in revealing ways. For example, is building a new tool for representing dynamic and transformative identities (as opposed the normative tools in standard social media) a case of creating culture or serving needs?
Another of Harrell’s examples is a powerful discussion of the Living Liberia Fabric, an interactive narrative peace memorial initiated in affiliation with the Liberian Truth and Reconciliation Commission and Michael Best at Georgia Tech. This project was created together with Harrell’s students, building on his GRIOT system. It began with HCI/media approaches, these ranged from scenario-based design and semi-structured interviews to iterative prototyping and stakeholder analyses. At the same time, the students were utterly daunted by the material, memorializing a war that only ended in 2003. They spoke with people such as a Truth and Reconciliation commission member, a peace museum expert, and a man whose son was abducted into being a child soldier. They finally found entry through what Harrell calls “cultural computing” — explicitly grounding computational work in diverse cultural perspectives (including those currently less privileged in computer science and engineering practices).
What are some of the values, aesthetic models, mourning models, and memorialization models inherent within Liberian culture?
The answers to these questions provided a starting point for the design of the interface and experience. The eventual design — which presents itself as an interactive digital cloth, telling stories of post-conflict Liberia from multiple identity perspectives — points to a goal: engineering for subjective experiences. This then points to the appropriateness of using methods from both engineering and the arts to guide and evaluate the project.
For more on Harrell’s work, look for his brand new book Phantasmal Media — a short description is in an MIT News article. Also, as with Anne Balsamo’s talk posted last week, which also delved into issues of mourning and memorialization, PDF slides are available on the main Media Systems page for this talk. Feel free to discuss here in the comments or on Twitter with hashtag #MediaSystems. Also, watch for Bill Gaver’s eye-opening talk here next week!
This material is based upon a project supported by the National Science Foundation (under Grant Number 1152217), the National Endowment for the Humanities: Exploring the human endeavor, the National Endowment for the Arts, Microsoft Studios, and Microsoft Research.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation, the National Endowment for the Humanities, the National Endowment for the Arts, Microsoft Studios, or Microsoft Research.
