Hand waving and the "real work" of design

| | TrackBacks (0)

I've been studying interaction design practice at San Francisco consultancies for my dissertation, and I'm beginning to publish. In the interests of making my work available to people who don't go to academic conferences and who don't read academic papers, I'm going to try to translate what I publish into more general-audience-friendly blog posts. I realize that it usually goes the other way for academic work, with blog posts serving as a way to work out arguments pre-publication, but that's not really how I work. So instead of moving from more casual writing to more formal, I'm actually going to move from formal to casual.

As a first offering, here's the blog-ification of my new interactions article, for those of you who do not subscribe. If you do subscribe, I encourage you to read it!

This article emerges from a conversation I had while observing a six-week website redesign project. I wasn't spending full days at the office, since it was a smaller project and the team members had other clients to worry about. So I felt like I'd lost touch with what my study participants were up to. I especially was a little unclear about what one of the project leads was doing on the project. When I asked him, he said:

"Oh, I'm not doing any real work on the project any more. I'm just showing up at client meetings and hand waving."

Which sets the stage for this article. At the time, I took what he told me for granted. But when I typed up my fieldnotes, it struck me: Why is handwaving not "real work?" And what is "real work," in interaction design, anyway?

If you do interpretive research, that's the sort of definitional question that should grab your attention. When people say that something isn't "really art", or that someone isn't "really white," they're defining a category by what it excludes. They're naming outsiders and insiders. In this case, my participant was defining "real work" by what it's not: hand waving.

"Hand waving" is a pretty apt description of what interaction designers do with clients. Typically, those meetings are intended to review the course of the project and decide what to do next. These decisions are generally based on representations of the planned future system, such as wireframes, visual comps, flows, site maps, paper sketches, etc. The problem is that there's an inevitable gap between representations of what is to be built and the experience of the finished product. This is most obvious when you think about the difference between static wireframes and interactive systems. But the problem lies deeper. If you see the goal of interaction design as supporting human goals and activities through various interactions between humans and machines, then even a clickable prototype will not match the experience of the live system. Imagine Amazon without the recommendations, or World of Warcraft without the thousands of players per server.
Hand waving - ie, a combination of verbal explanations and evocative body movements - is a way to bridge that gap between intermediate representations and the experience of the built system. Hand waving supplements visual representations in order to make up for what they lack - interactivity, movement, emotionality, etc. Matthias Arvola calls this "making sketches behave." It involves more than just hands, of course - hand waving is a full-body activity, including tone of voice.

Yet, as this story suggests, this kind of "hand waving" for clients is both treated as a routine part of their jobs, but is often dismissed as not "real work."

While they see client meetings are important, many designers don't see them as integral to the craft and discipline of interaction design. For them, the "real work" of interaction design is the work of creation - of collaboratively envisioning future products and services, then creating artifacts that represent them, such as wireframes, videos, and site architectures. "Real work" takes place in generative, free-flowing team meetings and in focused, solitary "heads down" work on computers as they move from paper sketches and post-its to InDesign and Keynote. So, despite the unavoidable necessity of communicating with clients, designers don't seem to talk much about hand waving as a part of interaction design as a profession.

However, with more than 100 hours of project work analyzed through video and audio, I'd say that successful hand waving looks more and more like an important and hard-won accomplishment. Instead of dismissing it or taking it for granted, educators, students, and certainly working designers should understand it as a critical part of successful interaction design. Successful hand waving requires a range of practical skills -- from role-playing client reactions, to choreographing movement through complex visual presentations, to staging convincing demonstrations of expertise and effort.

So, what do all those practical skills require? More and more, I've been thinking of hand waving as a kind of embodied performance.

Embodied performance

So, obviously, what I'm talking about as "hand waving" happens in the context of a presentation. There's a performer (the designer) and an audience (clients and other designers). There's a stage (the conference room or, more specifically, a projection space or whiteboard). That's the performance part.

I say "embodied" in order to remember that hand waving involves not just hands, but also full bodies, architectural spaces, and tools. It requires conference rooms, whiteboards, sharpies, wall projections, presentation documents, conference telephones, et cetera. Not to mention Post-its! Lots of post-its.

Lilly Irani has a great point about all of these resources: as infrastructures for design, these resources aren't just tactically useful for getting tasks done, they're also symbolic of a certain kind of design work. Having and using post-it notes in front of clients is important as a kind of professional performance, in which access to Post-it notes is seen as necessary to present a firm as innovative and competent. In her study of design work in Delhi and Bangalore, she saw designers going to some trouble to get their supplies of Post-its and sharpies. That's a good example of the staged nature of something even as seemingly simple as putting a post-it note on a whiteboard.

Here, I'm drawing on the work of sociologist Erving Goffman. Goffman was interested in the production of "the interaction order" between people. In his classic The Presentation of Self in Everyday Life, Goffman proposes a theatrical model of action, in which people consciously (and unconsciously) produce "performances" that influence how others see them. Goffman's work on performance suggests two concepts that help in understanding hand waving.

The performed self First, identity is the result, not the cause, of the performance. That is, social performance doesn't result from an objectively existing stable identity. Rather, the interaction between performer and audience produces that seemingly stable, well-bounded identity.

Stages and staging Second, in Goffman's world, performers move between what members of a team or group consider a "backstage" and a "frontstage." Team members create a backstage by controlling access by outsiders, and often engage in conduct "backstage" that contradicts the impressions they are trying to create on the "frontstage." For example, Goffman describes a restaurant in which the wait staff behave quite differently in the kitchen and in the dining room. The location and attributes of a frontstage and a backstage can't be taken for granted; part of the performance lies in enacting those different regions. As the settings for performances, the appearance and configuration of those stages may be carefully managed.

Hand waving in practice
I've drawn three dimensions of hand waving as embodied performance that seem particularly significant from my work with designers.

Backstage work
What I'm discussing here is a frontstage, client-facing activity. Clients or visitors rarely see the backstage work that supports the frontstage of client meetings.

In part, that's because a considerable amount of work goes into creating divisions between the stages. Through digital tools such as project blogs, research wikis, and even daily video tours of the project space, clients may feel as if they are keeping on top of design work. But, as one might imagine, client access to the physical and digital spaces of interaction design consultancies is carefully managed: their visits are scheduled, and they are given access only to specific project spaces on company servers. When clients do make in-person visits, information about other projects is physically hidden - doors to offices are closed, and foamcore boards are pulled up to conceal post-it covered walls. In the case of the designers that Lilly Irani is studying, their clients certainly don't know the lengths they go to in order to get supplies and keep abreast of current Silicon Valley trends. The goal is to make the performance look effortless.

Part of hand waving, then, is the hard work of building a frontstage on which to perform it.
On the other hand, imagined clients make frequent guest appearances in backstage design work. Studies of gesture and explanation in design more usually discuss how designers work through the implications of their sketches by role playing future users, or even components of an interactive system (One of the most useful ones is Matthias Arvola's study of interaction design students.) It's a really common event: one designer will say something like, "So then I click on this , and you go to the next page

But clients are also role-played in design meetings. You can imagine, after the sheet of paper is pulled out, how another designer in the meeting might say something like, "Oh, that's never gonna fly with [insert client name here]."

Designers imagine how clients will respond to various ideas, speculate on client team relationships, and strategize ways to handle difficult news. They're trying to figure out how to be maximally persuasive, but it's not necessarily Machiavellian or deceptive. It's just the result of a collaborative relationship in which the people with the money and IP (ie, the clients) are different from the experts in design. Obviously, the teams of experts are going to have their own desired outcomes. But since clients have the final say, there's a lot of more-or-less overt persuasion towards the "right" decision. Depending on the project and client, that decision-making process may be more or less casual and collaborative. But contract work creates an economic relationship in which designers cannot afford to pretend that clients don't exist.

In practice, this means that the presence of hand waving is not just limited to frontstage activities. It permeates "real work": anticipated, rehearsed, and then used to influence internal design team decision-making.

Performing deliverables

Deliverables underpin these performances, as in a walk-through. As the name "walk-through" suggests, a third dimension of hand waving is choreography - the movement of participants around and through a set of documents.

Take a set of wireframes originally created in InDesign. As part of a walk-through, they could be printed out and mounted on a wall. In that case, the presentation consists of a designer physically moving the group across the wall, and directing her attention to specific areas with her hands. The wireframes could also be imported into Keynote as sequential slides, then projected onto that very same wall. In that case, the audience is seated, and watch as the mouse cursor moves them sequentially through the frames. Alternatively, they could all reside on a single, large digital document. The designer then digitally jumps around the document, zooming in and out of each region as s/he walks through an activity or use case. Contrast this to a situation in which the clients and the designers are reviewing the wireframes over a distance. At that point, the designers have no control over the wireframes, and potentially may find that their deliverables are being reviewed on the small screen of an iPhone, with the client skipping around an 80-page presentation at will. (That happened in one project I observed.)

To recap: there's always an experiential gap between the deliverables and the future product or service. Bridging that gap requires choices about staging. There is no one best choice - but sensitive designers are aware of the consequences.

Staging the project

Providing persuasive evidence of expertise and effort is part of what makes client meetings so compelling and convincing. After all, as one designer said, "Part of my job is to make people feel better about spending half a million dollars on a process that isn't predictable."

Hand waving, for him and others, involves a physical performance of confidence in his recommendations, from tone of voice to body language. People who don't act "professionally" in front of clients - who roll their eyes dismissively, for example - are going to get reprimanded. Complaints about "unprofessional" behavior, like the idea of "real work," is a good way to explore how designers define a field.

Staging is also important in providing evidence of hard work. As a backdrop for a presentation of a set of personas, for example, one group of designers used a 10' x 7' whiteboard entirely covered with clusters of handwritten post-it notes. They did not present details of their process. Instead, at key points in the meeting, they pointed at the whiteboard while justifying their conclusions. The hundreds of hand-written notes, carefully arranged into labeled groups, was a visually overwhelming demonstration that otherwise would not have been as visible to the clients.

Choreographing movement through deliverables are what we usually talk about in terms of design - the designers are bridging the gap between what is here (ie, wireframes) and what is absent (the people, the system, and the resources and goals both bring to bear on the interaction between them.) However, hand waving also supports project-level stories about why, when, and how certain decisions were made - which in turn helps make sense of those deliverables.

What embodied client performances do

So, in my definition, hand waving can involve physical pointing and gesturing at paper print-outs, navigating through a digital presenting and pointing with a mouse, or even verbally talking a remote client through the navigation of their own copy of the presentation. The goal of this, however, is always the same: to ensure that all parties agree not just on what they are looking at, but what it means. One could say that they are trying, literally and metaphorically, to keep all parties on the same page in terms of how a visual representation relates to a future interactive system.

Again, the issue for the designers, the clients, the developers, et cetera - is that the future system doesn't exist. This gap between deliverable and final system is one of the reasons why interaction design resembles architecture more than it resembles graphic design or fashion. The logo or typeface that clients approve in graphic design is the very same logo or typeface that they will later use. A one-off garment for a fashion show will largely resemble that same garment in mass production. What you see, in graphic and fashion design, is much closer to what you finally get. (David Fleming's study of graphic designers and their clients is a good academic study of the relative stability of graphic design deliverables in relationship to final outcomes, even if the project itself is in flux. It matches what I remember of being a print graphic designer in the 1990s.)

In interaction design, however, just as in Goffman's notion of the self, there is no stable, objective system out there to be performed in deliverables. Nor do the deliverables themselves exactly match the final system. The goal of the client meetings is to use conversations based on representations to "stabilize the object" - to accumulate enough agreed-upon specifications for the final system in order to make it buildable. (Sometimes, this involves actually building the system. Often, though, interaction design consultancies will pass the specifications to another vendor to built it). In this case, sketches and other representations of digital structures are "material anchors" for the more transitory and ephemeral actions of talking and gesturing. We tend to treat these documents as the authoritative representations of the system to be built. But if that were true in practice, if the wireframes and specs and scenarios and site maps actually were adequate to the purpose of building an app or a website, meetings wouldn't run so long... In fact, there'd be no need for meetings at all. But meetings do run long, and documentation does need to be explained.

In combination, the staging, the deliverable, the gesture, and the verbal explanation allows a group of people with different backgrounds and interests to come to some agreement. Otherwise, we'd have a situation much like the famous blind men and the elephant -- except that there is no actual elephant against which to compare their different explanations! Designers are, in a very real sense, hand waving imagined elephants -- new products and services -- into existence. When projects fail - when the client hates the final presentation, when the developers can't built the website on budget, when the prototype doesn't work - "miscommunication" is often blamed. "We didn't properly align our expectations," people might say. Or: "The developers didn't understand the implications of the wireframes."

One of the provocative implications of taking on Goffman's notion of performed selves is that in hand waving products into existence, designers are also performing themselves as interaction designers into a stable identity as well. What's being stabilized by all that talking and pointing is not just the system to be built, but also the relationships and identities of the people making it.

If that's so, then more skillful hand waving benefits not just individual projects but interaction design as a profession.

Hand waving is the kind of activity that has been called "articulation work." Articulation work is an incredibly useful concept - it's the kind of meta-work that keeps cooperative work on track when the spread of tasks between various groups threatens to derail it. Articulation work, as is often pointed out, tends to disappear from "rationalized models" of work process. It doesn't seem like "real work." Hence, the fate of hand waving.

Role-playing clients, staging effort and choreographing narrative movement may not necessarily feel creative or innovative, but it is key to decision-making in commercial interaction design.

Why does this matter?

I suspect most of these observations won't come as any surprise to working designers. What I'm hoping, however, is that reframing hand waving as "embodied performance" helps people recontextualize that effort and take it more seriously as a worthwhile art and craft.

After all, part of the work of design is paying attention to how we work. I believe that taking hand waving more seriously as a legitimate part of design as a service can help us (and here I count myself as a designer) think more specifically and concretely about what kinds of values, interests, and priorities we are building into systems and services through what we choose to present, and how we present it.

Managing relationships with decision-makers such as clients lies at the heart of how designers makes change in their organizations and in the larger world. It does so through assisting in stabilizing future systems - what they do, how they do it, what they mean - throughout the design process.

As a fulcrum in the design process, it can provide some leverage in changing processes and outcomes. For example, in trying to introduce greater consideration for environmental sustainability, or ethical and just employment conditions. Which is why we can't ignore how deliverables are delivered - staged, choreographed, and carefully performed.
Sure, hand waving feels ephemeral and possibly even fake. But if it helps successfully stabilize the features and functionality of a service or product - or the continuing relationships of designers, developers, and clients which product those services and products - then it has significant consequences. It is real work.

0 TrackBacks

Listed below are links to blogs that reference this entry: Hand waving and the "real work" of design.

TrackBack URL for this entry: http://www.confectious.net/mt/mt-tb.cgi/963


  • /thinking
  • projects

About this Entry

This page contains a single entry by Liz published on July 19, 2011 2:16 PM.

links for 2011-05-06 was the previous entry in this blog.

links for 2011-07-25 is the next entry in this blog.

Find recent content on the main index or look in the archives to find all content.

Creative Commons License
This weblog is licensed under a Creative Commons License.