This presentation is based on the paper that I submitted for the alt.chi proceedings and that has been available online since early January. I am not going to rehearse its arguments in detail, since the paper is rather discursively heavy and it is not possible for me to go into detail given the time that I have. Rather, what I would like to do is to focus on two points: first, how the assumed relationship between HCI and industry sets up a frame that necessarily excludes certain types of problem spaces that might be politically important to consider, and second, how this frame might exclude different ways of approaching problems themselves. To do this, I'm going to (click) first revisit my paper through the lens of a related historical discussion, namely that involving the "conduit" metaphor that has been prevalent in information sciences and HCI. This is heavily based on a paper written by Ronald E. Day entitled "The 'Conduit Metaphor' and The Nature and Politics of Information Studies". And then I am (click) going to again return to the work of the industrial designer Victor Papanek and talk about how one of his projects involving a building of a radio from local components can be instructive for a non-commercially-based version of HCI.

What I'm returning to is the history of the development of cybernetics and information science, namely the work of Norbert Wiener, Claude Shannon, and Warren Weaver, among others. (go through image) The conduit metaphor set up a variety of equivalences between what were previously disparate concepts. First, there is the assumption that you can perfectly transmit information from a source to a receiver, if only you were to remove the noise. Thus, (click). Secondly, there is an equivalence between information and communication. The assumption is that as long as you have perfectly transmitted the information, you have also perfectly communicated the message. (click) The last equivalence I want to point out is the one of translation; namely, that messages in a variety of media, modes, and senses can be converted into a certain code (binary code, perhaps) and that this translation is perfect. Thus, binary code becomes a kind of master-code that can "perfectly" represent all other types of media (click). This metaphor, and its embedded equivalences, became the guiding metaphor for the information sciences. In the last twenty years, of course, we have witnessed a number of attempts to counter this and open up information science to other forms of representation and understanding of communication. What I want to stress is that the metaphor itself is not ideologically neutral; it makes assumptions about what types of communication are privileged over others. And when it gets embedded into funding and professional structures it further constrains the types of projects and research that are likely to occur. As Day writes (click)

What is important to understand here is his focus on the mechanisms of "research funding", or the ways that the metaphor gets embedded within a whole series of processes that work to constrain what is possible. Additionally, the conduit metaphor as described removes the political from the equation: data, information, and communication become equations to solve, rather than something that must be continually negotiated. He wants us to understand what is lost here, and how certain strands of poetic or artistic practice do not make these sorts of compromises.

He goes on to write: "That art and poetics remain problematic areas for vocabulary (e.g., in the assignment of metadata to artworks or poetry, outside of an historical or archival framework), points to conflicts in how vocabulary is understood and how it functions in a “scientific” paradigm."

Day further connects this explicitly to politics in the sense of the limitations this metaphor constructed for how the information sciences related to society. He writes of the need to open up the metaphor, to not use it as a bracketing off of politics, if information science is going to meaningfully respond to the complexities of language in society. (click)

What I am arguing by analogy is that the commercial relationships that are at the heart of HCI are similarly limiting. What I mean by this is that by framing within the confines of a traditional capitalist market-based view of the world, we are necessarily limiting the types of projects we can consider as well as what types of research topics are consequently valorized. In my paper I analyzed a concrete example of this in the Microsoft Research "Being Human 2020" booklet, chosen both because of its relevance for HCI, in terms of its topics and use in HCI courses, as well as for the participation of many known in the field. In it, we could see how with certain examples the problem space was limited by the assumption of creating a marketable technology. Taking to heart what we have learned from looking at the conduit metaphor, we can recognize the historical contingency of certain configurations of practice and can begin to question what was assumed to have been true for all time---or at least for all recent memory. This "framing" of the research in this way tends to further the push of new and ever-more-"powerful" technology, leading to job advertisements such as this (click):

Go through next slides, talking about the language used.

Again, "customers".

Why is a financial services company advertising here during this financial "crisis"?

More importantly, however, is the real physical impact of the need to always increase demand for computational and digital materials. This requires not only the discard and recycling of old equipment, but also the search for, and conquest of, new, better performing materials. First, with regards to computer recycling, I want to show you this image (click):

This image encapsulates one consequence of the linking of computing and the commercial world. The photograph is from the photographer and filmmaker Edward Burtynsky who documents how humans are having an impact on the natural landscape. His film, "Manufactured Landscapes" is an especially haunting account of how, among other things, computer components are polluting the environment around the world. Here we see computer circuit boards---which are extremely toxic---being recycled (often by the women and children) in China. In a way, we are all responsible for this, and the intertwining of HCI practice and the commercial market is especially involved, given the relentless push to use and develop new technologies, rather than utilizing what already exists. On the other hand, newer, faster, more powerful machines require this (click):

The relationship of coltan to tantalum capacitors, digital economies. The "blood diamond of the digital age".

Complexity of the coltan trade (small "artisanal miners" versus large producers, local warlords, multiple steps that the ore goes through before it gets out of the country, local understanding of the demand and wealth of the material). There is much, much more to talk about regarding this point, and I think HCI has sadly been ignorant of the political economy of the material processes that form the basis of the equipment we use, and the equipment that becomes necessary as we develop more and more resource-hungry and pervasive applications. We should be educating ourselves about this issue, as well as working with others, especially in anthropology, that are looking at what Appadurai has called the "social life of things". But, now I turn to Victor Papanek and cow dung to consider one possible alternative to a relentless push for the new.

As I went into detail in my paper, Victor Papanek was an industrial designer, first active in the 1960s, who was especially concerned with the ways in which design was implicated with the market. He detested the idea of "shroud design", meaning the design of new coverings and cases in order to push new products onto the market, and was interested in how local materials, labor, and handiwork could be involved in the making and design of products.

Given a brief to design a radio for the global south, how might we do it? If we were doing it within a commercial frame, we might say, well, let's design a cheap hand-crank radio built in a factory in China and subsidize it with advertising (either on the outside of the device or on air during broadcast). Papanek, however, wanted to develop a design that was not dependent on external manufacturing nor would contribute to profits outside of the locality. Thus, he and his graduate students designed this radio: it is made of a used tin can (a local one, not one transferred from the West), wax, paper, or cow dung for power, an earphone, a hand-woven copper antenna, and a nail for a ground wire. Total cost (in 1972 terms): 9 cents.

Important here to understand is how, following Papanek's lead, we could similarly reconsider HCI projects using reconfigurations of existing technologies and materials---a methodology that indeed would not be conducive to production of more and more technological devices, but that could potentially portend a different configuration of people and machines that have a different relationship to the commercial than the ones that are prominent now.

To finish: Firstly, I have briefly tried to show how the relationship between HCI and industry is a framing metaphor that necessarily excludes particular types of projects from being valorized. I suggested two material consequences of this with regards to computer recycling and coltan manufacture. Second, by making reference to the work of Papanek, I have tried to suggest how technological devices can be built simply and cheaply using local and recycled materials, something that is difficult to envision within a commercial frame that is predicated on the continual cycle of purchase/use/discard/purchase. Now, my talk should not be seen as denying nor diminishing the work of HCI researchers that are similar to what I have presented here. However, I believe that we need to foreground how particular framings and relationships between HCI and industry necessarily lead to particular configurations of ethics of problem choice. I hope that this paper and presentation open up discussion about key questions that have either have not been raised or have been on the periphery, namely: (click) You may disagree strongly with how I have characterized the relationship between HCI and industry, but I hope nevertheless that my provocation opens up this necessary space of discussion. Thank you.