Interactions

I have just been trying to figure out how to explain some of our design ideas inside the various ACM/CHI-ish communities in the upcoming paper-writing season. 

I have found that if I keep my head down and focus on the work of shaping ideas and things, enough comes along career-wise to keep body and soul together, but I’m not really all that good at getting my ideas published. (My second-most-cited paper, “The Three Paradigms of HCI” [1], was never properly published because, for structural reasons, it was routinely evaluated by people who disliked it. Warning: If you write a paper called “The Three Paradigms of X,” you can expect it to be reviewed by three people, each an esteemed proponent of one of the paradigms, and each equally likely to dislike the premise of the paper, since no claim is made about the primacy of any of them.) 

But the fact that I’ve made a career despite problems like this does not particularly help my students. In general, I love my students and would like to tell them Useful Things that lead to their Growth, Well-Being, and Success. 

So, in the course of positioning work, I turned to a slowly unfolding debate to figure out how to present some of the ideas that our group has been working with and to explain to myself why one of my most important papers, reflecting on what important topics design should address in the next 10 years, was recently rejected. This note reflects a kind of a selfish approach, less concerned with what people actually have been saying and more concerned with reading how papers will be or are being read. In this post, I call that their shadow, but I might also call it connotation rather than denotation, or, coming from a different philosophical position, perlocutionary meaning.

I went back and re-read some of the Bardzells’ work on critical design (“What’s 'Critical' about Critical Design?” [2] “Analyzing Critical Designs” [3]) and then Pierce et al.’s response [4] in the 2015 CHI proceedings. And then I asked myself what purpose human-computer interaction research fulfills in the world—and who gets to decide. A key idea to me is to answer through research and thought, “what is important to design?”, “how do you know that you have done it?” and, increasingly, “how does/can design shape the world?” Many of these are ideas that I share with Steve Harrison. 

Critical design: reflexivity and beyond

The notion of critical design is very welcome and very familiar to me—my first degree was in English and American Literature and Language, and I love the idea of provoking design ideation through critical inquiry. The Bardzells develop the ideas of critical theory for HCI design in “What is Critical…”, a CHI paper that advocates critical theory as an important lens for both designer and consumer: reflection, perspective shifting, and theory-as-speculation as methods of understanding what is around us and taking design action, and supporting these with dialogical methodology. I or my students can take that paper, use it to propose a design, and say “This is what we were thinking. What do you think?” 

This is in line with Schon’s reflective practitioner, of course. More broadly, this has a strong family resemblance to the design tensions framework [5] that I proposed in 2007, which was also a method of getting people to think more deeply about the meaning of their design choices and what constitutes value in a design situation. The design tensions framework was concerned with how easy it is in HCI to overlook design success if that success is due to an accretion of small factors rather than a single big idea. But it is only one piece of the puzzle, and “What is Critical…” seems to me to provide another piece. 

Canonical—and non-canonical—examples

Following on “What is Critical…”, a graduate student, Gabriele Ferri, working with the Bardzells, wrote a DIS 2014 paper that could be interpreted as playing a nice speculative game, taking critical design to the next level. On one hand, this paper addresses absolutely crucial questions for design researchers and for teachers: What constitutes a contribution in the field and how do we recognize one? One component consists of examples, particularly teaching examples, of critical design. This caught my attention, in part because in the “Three Paradigms” paper, we criticize HCI for a lack of key teaching examples, and note that teaching examples were one of the chief elements that Kuhn noted in identifying something as a paradigm. The absence of teaching examples in HCI makes teaching HCI seem like trying to land on Jupiter; we can point to an accretion of stuff that moves around the sun in an orderly way, but, hey, it’s all gas! Does a person really want to go to a planet where there is nothing to land on or grasp? So, I respect the fact that the paper took on the question of examples. Furthermore, I appreciate that part of the paper is saying, “Let’s openly talk about what is in and what is out.” It accepts that people’s careers rise and fall on judgments about their work (that, at least, is graspable!). This is very important for discourse in the field. 

On the other hand, there was something troubling about the Ferri et al. 2014 DIS paper. As I read it, I could hear unpleasant echoes of family holiday meals long ago at which my various relations would defend the scholarly canon of knowledge, scoff at “the so-called social sciences”—“They’re not sciences!” “Physics is the only really science!” “Math is the queen of disciplines”—and otherwise revel in their refinement and insight. 

OK, the paper does not, strictly speaking, propose a canon. In fact, the canon that the paper does not propose (that is, a non-canon) is very intimidating. If you, newbie, claim to be doing critical design, are you going to get smacked down because your design isn’t in the non-canon? What if your design does not rise to the level that a person would call, for instance, transgression (transgression being part of the non-canon)? If you say that your design is in the non-canon, then does it sound less like research? 

The notion of canon that is not proposed by Ferri et al. is scary enough so that reading the paper, I do not know what to tell my students. Anxiety is not always a bad thing—maybe my students and I should be anxious, maybe we all should be a lot more anxious about the contributions of our field, maybe this is precisely what I helped call for in a different paper, “Making Epistemological Trouble” [6]—but, indeed, the shadow cast by this paper is quite anxiety provoking! 

Trying to illuminate HCI using shadows from design 

So, with these musings, I came to the Pierce et al. paper from this past CHI that, I’m told, is widely interpreted as a response. The authors propose a bunch of problems, some deep, some shallow, with the approach proposed and explored by the Bardzells’ and their colleagues. Like me, Pierce et al. also appeared to think that the prospect of a canon, even the shadow of a canon, is scary. They point out that critical design is not the only way for design and HCI to interact. Pierce et al. are certainly correct that only a trickle of ideas about design has made it into HCI thinking. True, but somehow unfair; I have to point out that only a trickle of ideas about experimentation and modeling have made it into HCI, yet few people associated with the Third Paradigm trouble themselves to obtain anything more than a stereotypic, fifth grade view of the Second Paradigm. The authors then go on to propose some possible futures. There are some nice ideas about openness. As with the earlier papers, there is a denotative content that is interesting and the sense of active minds at play. 

But, like the Ferri and Bardzell paper it criticizes, the Pierce et al. paper also has shadows, and to my mind they are darker ones. It really does not take on the problem that the Bardzells and company address about how HCI should be open to design. The shadow of the Pierce et al. approach is that, while it promises freedom (“open up HCI to all kinds of ideas from design”), it threatens domination by the few, the arbiters of design. Is design in HCI nothing more than translation from design to HCI? If it is only translation, that does not sound like freedom to me. And then the injunction to “focus on tactics, not ontology” makes a major power move that seems to threaten to shut down discourse. Although the authors disclaim the desire to control speech about design themselves, they write that “the issue of the designer’s intention needs to be handled carefully.” What does that mean? Who gets to say it? 

Pierce et al. calls for metadata about the designer’s intention to be taken into account in discussing ideas. 

Really? What justifies that position?

Addressing the designers’ intention cannot find its justification in arguments about how design works, because when a designer creates an artifact it goes out into the world with whatever juju marketing and circumstance give it. Michael Graves was recently memorialized in Metropolis by someone who had worked on his low-end consumer product line for Target. His colleague praised Graves’ focus on usability. That’s nice. Nonetheless, I am perfectly free to tell you that I was relieved when my Graves-designed hand-mixer finally died. The balance was wrong and my kitchen walls were frequently spattered. The only reason I know anything about the designer’s thought is because Graves’ achievements lay towards the art side of design, and artists may sometimes be offered a little blurb and an occasional token of respect or, in this case, a tribute. I concede that Graves’ purpose was not to promote egg-based wall decoration in my home but this is irrelevant to my experience as a user or even a client of the design work. 

So this reverence for the designer’s intent is not, I think, a claim about design. Is it a claim about design research? Well, on the one hand, in any kind of research, we make reference to prior work, but, in HCI more than most of the other fields of my concerns and expertise, we are also highly selective. For goodness sake, we write ten-page papers! Why should the designer’s intention, which is, importantly, also the researcher’s intention, be so prioritized? The direct thought that Pierce et al. present is about respecting authorial voice. That is an important discussion to have, a discussion that I would recast as concerned with developing schools of thought as we move forward in design research in HCI. But let’s go back to the shadow and the question of what I tell my students. 

I see something much less discussable and more draconian in the shadow of the Pierce et al. paper than in the Bardzells’ papers. I am sure that Pierce and company themselves would be horrified were I to tell my students, “Do not reconsider the meaning of (for example) the Drift Table because Gaver hath spoken and he hath named it Ludic!” That is a very dark shadow indeed, because unanswerable.

Years ago, Steve Harrison, Maribeth Back, and I wrote a paper called “It’s Just a Method!” [7]. The idea was that design methods are not important in-and-of themselves, but only because of what they enable the designer to perceive about the situation. In parallel, I would like my students to be able to use theory to inspire themselves and to describe their projects to others. But it’s hard to use much of these theories this way. We put our little 10-page barque to sea and theory threatens to fall on our heads as though we floated beneath a calving iceberg. 

Meanwhile, none of this gets the field closer to the issues that so trouble me about the status of technology in shaping our views of ourselves as subjects. Students, back to shaping our little research boat! 

Endnotes

1. Harrison, S., Tatar, D., & Sengers, P. (2007, April). The three paradigms of HCI. In Alt. Chi. Session at the SIGCHI Conference on Human Factors in Computing Systems. San Jose, California, USA (pp. 1-18).

2. Bardzell, J. and Bardzell, S. (2013). What is "critical" about critical design? In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI '13). ACM, New York, NY, USA, 3297-3306.

3. Ferri, G., Bardzell, J., Bardzell, S., & Louraine, S. (2014, June). Analyzing critical designs: categories, distinctions, and canons of exemplars. In Proceedings of the 2014 conference on Designing interactive systems. (pp. 355-364). ACM.

4. Pierce, J., Sengers, P., Hirsch, T., Jenkins, T., Gaver, W., & DiSalvo, C. (2015, April). Expanding and Refining Design and Criticality in HCI. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems (pp. 2083-2092). ACM.

5. Tatar, D. (2007). The design tensions framework. Human–Computer Interaction, 22(4), 413-451.

6. Harrison, S., Sengers, P., & Tatar, D. (2011). Making epistemological trouble: Third-paradigm HCI as successor science. Interacting with Computers, 23(5), 385-392.

7. Harrison, S., Back, M., & Tatar, D. (2006, June). It's Just a Method!: a pedagogical experiment in interdisciplinary design. In Proceedings of the 6th conference on Designing Interactive systems (pp. 261-270). ACM.

Deborah Tatar