Is it possible to design for appropriation and unintended uses? If so, how? If not, why not?

In this post, my primary goal is to present the notes from that discussion. At the same time, my notes are incomplete. It was hard to capture everything and I found myself mixing up my own interpretations of what was said and what the person saying might have said. Therefore, I have decided to also go ahead an add some additional commentary. If you participated in that conversation and think that: a) I have missed something completely, b) misunderstood a point that you or someone else made, or c) have any reactions or thoughts since then, please comment below! (And, of course, to everyone who didn’t participate, feel free to weigh in however you see fit).

My motivation for raising these questions is based on my interest in investigating the design of socio-technical systems while somehow coming to an understanding of groups’ everyday practice in relation to media production.

Framing the discussion
I framed the conversation by raising three recent examples/incidents that have been on my mind:

Recently at the Designing for User eXperiences conference (DUX’05), a designer in the audience asked a question that boiled down to this: “How can I design for user experience if once I put my products out into the world, they pretty much ‘die.’ They are no longer mine and they are being used in ways that I never intended” (not an exact quote except for the ‘die’ part). My thought was that once a product is finally out in the world being used in ways never intended it can finally be described as being “alive.” Either way, there is a point that “user-centered design” practices often try to design systems for specific purposes for specific people (or types of people) and not for unanticipated uses by unanticipated users. In what ways does designing for appropriation fly in the face of user-centered design? In what ways could it fall nicely in line with user-centered design?

If anyone has seen a teenager’s MySpace page, you may notice that it has a particular “aesthetic” that comes partially as the result of people customizing pages by cutting and pasting HTML and CSS code into various profile-editing forms. It’s not clear to me if this was ever intended by the MySpace designers (my thought is that it was not, but that’s not from personal conversation with them), but it certainly is an important part of current social practices on myspace.

In Changing Minds: Computers, Learning, and Literacy (2000), Andy DiSessa (in UCB’s education department) argues that appropriation is a critical aspect to systems that are designed to promote learning and literacy. He also points out numerous examples of things that his research team saw in the classroom with kids and teachers using his software that he never anticipated and were critical parts of the groups’ learning experience with the software (see Chapter 3 if you’re really interested in this topic in his book…). One argument he makes is that a system has to be open in various ways and this is critical to learning:

“A computational medium that does not let all its users create to suit their own instincts will miss probably the most powerful principle of appropriation and arguably one of the most powerful principles of learning. Owning and knowing are fraternal twins. Owning leads to knowing and vice versa. (p55).”

Before the discussion, two other people in the group provided some resources that would help frame our discussion. One person pointed out an interesting workshop from CHI 2005 (ACM’s Conference for Human Factors in Computing) called “Designing Technology for Community Appropriation” (organizers’ papers here, participants’ papers here). Another said that the topic reminded him of Paul Dourish’s new paper, to be presented at CHI 2006, called “Implications for Design”, which argues for a new way of considering the relationship between ethnography and design practice:

…Ackerman critiques the intuition that people adopt and adapt technologies because the technologies are poorly designed, and that better designed technologies would obviate the need for such adaptation and appropriation….

…Certainly, though, what [ethnography] does is to refigure “users” not as passive recipients of predefined technologies but as actors who collectively create the circumstances, contexts, and consequences of technology use. HCI research has, of course, long had an interest in aspects of the ways in which people might configure, adapt, and customize technologies. This ethnographic view, though, focuses not simply on how people explicitly transform or program interactive technologies, but how those technologies take on specific social meanings through their embedding within systems of practice. As a focus of HCI research attention, “design,” in this sense, goes beyond giving form to technologies to encompass appropriation – the active process of incorporation and co-evolution of technologies, practices, and settings.

Recap of the discussion
The conversation began with a discussion of what the word “appropriation” actually meant. There was some debate as to whether or not we were talking about making something “one’s own” or making something one’s own for purposes not intended or anticipated by the thing’s designers. Now that I look back at how DiSessa and Dourish use the word, it seems to have to do more with making something one’s own. And to some extend this aligns with definitions I have found in The American Heritage Dictionary: “To take possession of or make use of exclusively for oneself, often without permission” and in Webster’s: “1. to take exclusive possession of…3) to take or make use of without authority or right.”

(Aside: Note that both of these definition seems to also include the notion of “misappropriate,” a word, I pointed out in the discussion I had seen in court rulings which seemed to imply steal. Looking at both definitions, I think it’s interesting how the “mis-” seems redundant. But if you go with just the notion of “making something one’s own” then the “mis-” is an essential distinction. Also, I wonder at their emphasis on appropriation as an act of one individual as opposed to a group. Also, I am starting to wonder about the relationship between appropriate as an adjective and appropriate as we are using it here. A tricky question could be: “Is it appropriate to appropriate?”)

But, people in the group did point out a common use of the word did have a lot to do with the idea of using something in a way never intended, and this would be one way of making something one’s own. A canonical example brought up had to do with the Roland 808, a product released in the 1980s that was intended to allow musicians to record drum rhythms (we might have talked about guitar tracks in our discussion, but I am having trouble finding any articles that go beyond drums here…). But people did not like the sound and other products seemed to do a much better job. It was not until techno’s rise in the mid-late 1980s that a new community found a whole new use for the product, which ended up making Roland a whole lot of money as they turned the company towards developing the 808 as a tool primarily for the unanticipated uses for which these new communities had appropriated the product.

Later in the conversation, someone made the suggestion that appropriation may have to do as much with unanticipated “users” as it did “uses.” Perhaps making this distinction would help us tease out the right definition of the word appropriation. I raised another example of “appropriation,” when fans of TV shows appropriated the content of those shows in the creation of works of fan fiction (see Henry Jenkins’ Textual Poachers for one academically-oriented account). These fans were the intended audience of the shows’ producers. But, the uses to which these fans were putting these shows were not a) anticipated and b) welcomed. Now that I think about it, one could also argue that these fans were appropriating VCRs in ways that may not have been intended: as video editors rather than just recording or playback devices.

A member of the group brought up a tension between products that that were designed to be potentially used for a wide variety of things, but perhaps they were “mediocre” for those uses, and products that were designed really well for one thing. I later called this the problem of the swiss-army knife. Does designing for appropriation mean that you can’t try to design something really well for one purpose? Do you have to try to design swiss army knives?

Along these lines, someone talked about notions of “resistance”: when a product is really good at what you are designing it for, it has not resistance for that use and a lot of resistance for other uses. I wonder if this also implies that a product that is not really designed well for one purpose partially resists all purposes? Or if perhaps the uses for which a product can be appropriated are the ones that are the least resisted?

This discussion links to one we had later in the meeting about the relationship between the word “appropriation” and the word “affordances” (Don Norman, who introduced the term to me through his book “The Design of Everyday Things” has a good article summarizing recent (?) thoughts on the use of the word here). Is “affordance” a better word, or a more useful word, for what we’ve been talking about? Is designing for appropriation about maximizing the affordances? We talked about the affordances of paper and the unanticipated uses of paper that the material affords (crumpling up and throwing at people).

Someone mentioned the intended uses of a hammer in relation to a subtle warning about the use of hammers that I had never considered. While a hammer is great for hitting in nails (and removing nails) and probably hitting other things, there is also a warning sign on many hammers that says, “Do not hit hard objects.” Why? Because you can chip the hammer.

The discussion turned at some point to learning. Someone brought up the fact that discovering how to use tools, what tools to use when, was something that was a part of an apprenticeship. People watch, learn, and participate with teachers, mentors, and others, when they figure out how to use tools. What happens when you separate consideration of the artifacts from the processes of coming to learn how to use them? Does that lead to appropriation? Or is appropriation a part of the apprenticeship process?

A member of the group raised another provocative question: is appropriation the result of a misinterpretation or misconception of product? Or is it the result of an intentional “misuse.” We found examples of both (CD tray as full coffee cup holder, parking meter as empty coffee cup holder were brought up at various times which raises the interesting question of caffeine and misuse).

Someone put forward the idea that there is a relationship between art and appropriation, that art is an example of “extreme appropriation” because artists often ask the question: What can I do to specifically be interesting, to specifically use something for some purpose in order to explore the a variety of questions related to that thing, such as its place in culture.

Is “designing for appropriation” a contradiction in terms? This was another fun mind-bender for the group. Once you put in your head the idea that appropriation may be possible and that you should consider it, are you designing for “appropriation” or for some intended use which would no longer make it appropriation, even if you can’t predict what that intended use might be. Is designing for “openness” or “flexibility” the same thing? So, as companies are now releasing APIs for all of their products in the hope the creative people will combine services to create new products, is this a way of designing for appropriation?

(As another aside, this is a practice that someone decided should be called a “mash-up” the use of a term that is in itself an appropriation of a term to describe a genre of music having to do with DJs combining music, often from wildly different artist-genres. My favorite is a Wings-Eminem mash-up called “A Slim McShady” from “Go Home Productions.” When I first heard the use of the word “mash up” to refer to combining APIs, I cringed, though perhaps the continual cycle of appropriation here makes the use of that term, unfortunately, appropriate. Uh oh. Making myself dizzy.)

We discussed whether or not there was a physical counterpart to APIs? One possibility that was posited was that of standardized parts. Standardized parts could be considered the physical equivalent of hackability. Something being hackable (software) is fun and challenging, perhaps encouraging appropriation.

We ended the meeting by turning more towards the question of methods. Does “Participatory Design,” a method for designing products in which people who may be the intended users of those products and designers work together in the design process. Perhaps the notion of “participation” should be extended. Do people need to have a better understanding of what it means to design or of what can be designed given the constraints. Would examples of appropriation come out earlier?

This led to a rather challenging question for all of us in the room: do the traditional methods of designing and evaluating design ensure that designers could never observe the possibilities for appropriation? Usability assessments, laboratory experiments, task analysis, and heuristic evaluations frame the issues to be evaluated so carefully, that how would anyone be able to use these methods to even consider the possibilities of appropriation? I followed this up with linking this issue with participatory design methods: despite the goal of creating a good team from participants and designers, if it is not possible to truly eliminate the power differences between participants and designers, would even participatory design help design for appropriation?

Another method posited was that of the “technology probe” something that is put in the world (or given to individuals) that is equipped to record data about its own use and then can be analyzed as to how it’s being used.

One person suggested that even designers brainstorming on this issue could help lead to characterizing a product along the following dimensions: a) what does our product do according to the uses that we have prescribed and intended? b)what could someone do with our product given my features? c) what could people not do with our product?

Finally, right near the end of the discussion, someone said that the conversation was leading towards three different levels of being able to design: a) a technical level, b) a process level, and then c) a higher level, perhaps a philosophical level. At that last level, designing for appropriation might involve a philosophical position. The example raised had to do with the photo-sharing site Flickr and its decision to remove any drawings from being public or searchable (discussion here). Flickr here is either taking a philosophical or economic stance on this issue that is inherently about member appropriation of the site. What is the relationship to “big” things like “democracy” and “economics” and designing for appropriation.

The conversation ended appropriately with someone saying something to the effect of: “Wow, we raised many more questions and didn’t resolve any.” That seemed like a fitting ending to a great discussion.

To anyone who has seen a teenager’s home page on MySpace, neither the above anecdote or the insights drawn from it may be all that surprising. I recently read an article in BusinessWeek Online entitled MySpace: Design Anarchy that Works that makes the following argument: MySpace’s success can be attributed partially, if not totally, to the founder’s choice to provide people nearly unlimited customizability in the look and feel of their pages:

Regardless of its aesthetic consequences, this customizability is one of the site’s most attractive features, and the do-it-yourself sensibility of the site resonates with the audience’s desire for self-expression.

I had a chance to see this upclose while doing participant observation at a local youth program. During their free time, kids frequently logged in to MySpace and showed us their pages and some of their friends’ pages. The pages were cluttered with pictures. Embedded audio and video clips pounded my senses. And pages that required lots of horizontal scrolling were common.

One of the most interesting points that Jesse James Garrett makes in the BusinessWeek article is that the rather poor design of the default pages on MySpace encourages this practice:

If the default presentation and the common areas of MySpace had cleaner, more professional designs, users might hesitate to customize their spaces, feeling intimidated by having their amateur design work side-by-side with the professional-looking defaults. Instead, the unpolished style invites users to try things out, telling them they don’t have to be professional designers to participate.

Garrett admits that it’s a “mystery” if these decisions were intentional. But regardless, he’s on the verge of claiming that the breakdown of one part of what is thought of as a component of good interaction design, specifically the visual design aspects, is what leads to people’s enjoyable experience with the site. Of course, good visual design is only one aspect of good interaction design, so I am curious to understand how well designed the tools that enable this “design anarchy” are.

Furthermore, I am still not sure how much it matters. Academic researchers who have explored the importance of good usability for children and kids have argued that it does matter. But, I have to think that very few kids and teenagers care very much that their friends’ MySpace sites are designed poorly according to current standards of HCI. There are a lot of nice research questions to investigate here. From the perspective of kids and teenagers using MySpace, how much does usability matter? Despite the wide adoption of MySpace, what, if anything, don’t they like about the site? What, if anything, do they find frustrating in the process of design?

(thanks to Ypulse for the pointer to the BusinesWeek article).