USC Annenberg Online Journalism ReviewUSC

Thinking About Interaction Design for Online News Delivery


Online journalism needs better design for active readers rather than passive consumers. The author's research indicates that Web content can be made far more meaningful and useful through better use of interactivity, or "productive interaction."

The recent proliferation of blogs and other content management systems (CMS) has eliminated the Webmaster as a mediator between the writer and reader, turning the Web into a read and write medium, as Tim Berners Lee -- the inventor of the Web -- originally imagined it. This development is creating a new and more direct channel of communication between writers and readers and has given more power to the writer. Unfortunately, the typical Web page is still constrained by limited interaction and a lack of affordances -- designer-supplied capabilities and tools -- that empower the user.

In many ways, online publishing design has stalled. Content producers seem happy with the raw benefits of the medium -- timely access, self-publishing, links, and search -- while barely scratching the surface of interactive publishing design opportunities. As a system of communication, online interactive documents are close to the days of spinning zoetrope filmstrips (a precursor to cinema), where the online form and technique lack the power and quality of mature mediums like cinema and print. Writers, designers, software developers, publishers and readers must experiment and push the medium beyond its seminal beginnings, just as film moved far beyond the zoetrope.

This article and its associated design explorations compose a system of design research that's engaging a dialog about the future capabilities and possibilities of the interactive medium. We need to ask a series of questions:

  • In what directions do we want interaction to go, and what are the ideologies behind these visions?
  • What are some principles and techniques that can help get it there?
  • What are some working and speculative prototypes of these new directions?
  • What are the implications for designers?

I favor writing and designing in directed research that aims to get at these basic questions. It's a way of pushing and pulling design and designers forward through the process of making, then analysis, and more making.

In the end, these questions boil down to a single one: how can designers make interactive systems more meaningful, rich and user-enabling? This is an acute question because interaction design can do much better. Given this, what are some robust design directions that can lead us to this future?

Improving online communications can take many approaches. In particular the focus on usability and experience design has greatly influenced interaction design. But I believe these techniques fail to address some of the basic benefits of interaction. They mostly position the user as a passive consumer following a scripted, "easy to use" path in a medium that offers something different.

By contrast, I propose a system of design I call "productive interaction," which views interaction as a medium that enables the user as producer of his or her own outcomes and meanings. Productive interaction aligns the design of non-linear content, context and affordance in an open, collaborative fashion, enabling the direct manipulation of the work's material. Users can create custom, personally significant meaning spaces of their own.

This alternative approach switches the audience from consumer to producer. Instead of designing prepackaged experiences, the designer focuses on a range of user goals and builds systems that facilitate the personal production of meaning(s). It flips the focus from consumerist manipulation through experience, to a facilitated, productive output via active dialog between the work and the user.

This model does not however, exclude or reject usability, story, experience, and other design approaches. But to improve the enabling characteristics of interactive media, the author/designer should put greater emphasis on the particular affordances that encourage active meaning making, rather than passive consumption of experiences and predetermined association of hyperlinks.

What is productive interaction?

Productive interaction is a recasting of the author/designer's position in relation to the audience. Instead of laying out a linear narrative in an enveloping experience, the productive interaction designer frames an exploration of a meaning space, making sure the audience has the affordances to create their own "take."

How is productive interaction different? Certainly all media rely on a dialog between the work and the audience. People view a painting and spend minutes, hours, even years absorbing, rethinking, feeling, and interpreting the work in their minds. Yet this psychological interaction is separate and detached from the painting, which stands fixed on the wall. Books, magazines, and other print media have a more tangible interaction, where readers actively turn pages, perhaps choosing their own path through the work. Still, the book retains its original linearity (a strength of the medium), and the reader is left thinking, imagining or making separate notes to create their own take on the material.

Even much of today's interactive media fails to reach beyond the interactive quality of print media, and pales in comparison to print's usability, information density, and compelling materiality. For example, the benefits of reading the news at The New York Times on the Web have mostly to do with features around the edges of linear content -- timely updates, links, search, and anywhere availability. Understanding the "the news" is a superior process with the print edition of The New York Times. This is because of the poor quality and few benefits the Website has over the powerful affordances of print that enable scanning large dense pages, spreading multiple sections out, folding it up to focus, using the meaning implied by typography and headline formats, passing a section to a friend, and so much more.

In contrast to traditional media, productive interaction's strength is facilitating and provoking the dialog. It enables juxtaposition, and supports the remixing of the actual content.

Productive interaction gives the reader a pair of scissors and permission to cut up the book. It's a system of direct manipulation, where the user becomes a co-designer in the creation of a custom content stream suited to their immediate desires, purposes and intents. And because the interaction is mediated by a computer rather than a static bookbinding, many affordances can be built to help the user create more meaning.

In broad terms, this paper is a call for designers to live up to the visions of seminal interactive thinkers such as Vannevar Bush, Ted Nelson, and Alan Kay. How can we move beyond the stagnation of the Web's context-disruptive hyperlink and video games' engaging but superficial 3D navigation systems?

Our goal should be to create systems that live up to the quality of printed newspapers and books. But the mission is not to displace previous mediums; rather, it's to find a new niche alongside existing media. The interactive designer has to find the strengths of the new medium and exploit them to their fullest, creating a high quality means of communication with a unique voice.

Ideally, designers will create interactive works that enable the user to develop an ad-hoc exploration with techniques such as dynamically created juxtapositions, faceted and multiple views of the material, active serendipity tools, and smart content that reconfigures itself in meaningful and useful ways. The user exploits these affordances to construct a personal outcome with unique significance because she created it herself.

Futurist visions of interface like those in the film "Minority Report" give us some clues about the possibilities. We can imagine systems that have very high-resolution displays, tangible interfaces, embedded presence, and flexible, smart interactions. But what will actually work, and how will designers function in this new space?

Design research

I'm working on a series of design prototypes to explore the issues raised in this article. This approach helps ground the thinking and analyses, and provides new insights and design directions. In fact, designing for me is another method of thinking and analyzing, using the process and intuition of design alongside the real-world constraints of making something actually work. So to help ground the discussion and analysis raised here, I'll describe several relevant interactive works. First, I'll discuss two early-stage (as of June 2004) screen-based prototypes -- Topic Explorer and Text Facets -- which are part of my current research. Second, I'll describe a Web work by design group Future Farmers called They Rule.

Each of these projects demonstrates some of the principles and techniques of productive interaction discussed in the following section.

Topic Explorer

This project explores ways to provide a deep exploration of a topic. One part of the prototype uses the ideas of serendipity, excerpting, and smart content to provide the user with an exploration affordance that strives for the kind of serendipitous grazing that often happens in a library. A "serendipity-stream" of short phrases from the main texts cascades down one side of the screen. The user can click on any phrase resulting in the main text immediately scrolling to the section of text the phrase comes from. The serendipity phrases are generated in several ways: randomly, from author tagged pull-quotes, or from headings in the text. This same technique is applied to images, where image fragments cascade down the screen as a method for serendipitously finding pictures.


Keyword search (right) with two texts (Kerry & Bush) scrolled to match

In another feature, the user clicks on keywords, and two main texts will simultaneously scroll to the next match of that word or phrase. Using the idea of juxtaposition, the user can compare how two different texts discuss various keywords. For example, if the user clicks on "tax cuts," two texts describing the positions of presidential candidates Kerry and Bush will scroll and show the phrase highlighted in context. Subsequent clicks on the key phrases scroll the texts to the next occurrence (or recycle back to the first occurrence). Preset key phrases are supplied by the author, but the user can type their own in and search the texts in same manner.

Text Facets

This simple prototype explores the techniques of content faceting, simultaneity, juxtaposition, and center/periphery to enable the user to quickly work with different sets of related information. The user is initially presented with four narrow columns of small text. If they click on any column, it expands by getting wider and growing in font size. Any column can be expanded or collapsed, and the text facets can be moved around as a group. With this system, the user can easily foreground or background a text facet, focusing on just one, or juxtaposing several selected texts.


Text facets with second facet expanded

They Rule

Josh On of design firm Future Farmers developed They Rule a system for understanding and researching corporate and institutional governance in terms of how they are ruled, particularly in terms of the common, yet unreported connections the organizations have between their shared board members. They Rule uses techniques such as remixing, filtering, smart content and audience contribution to present flexible views of the corporate board members and their relationships.


They Rule

Users select one or more companies or institutions and they appear on the screen as a boardroom table. Each company icon has options, the most important of which is the ability to display its board members. Once a member is displayed, the user can expand the member to see what other companies they are on the boards of. In this way, the user creates a map of companies/institutions and board members that graphically shows their inter-relationships. The created maps can then be saved and shared with other site visitors, creating a growing library of map presets with different themes (e.g. "cable and RBOCs", "IBM -- Microsoft -- Intel", "Halliburton and The Media", "I see no conspiracy here [shows very few connections between oil and military contractors]", etc.)

Principles and techniques

Productive interaction requires a different approach to design, and a different view of the audience. To help frame these differences, we can look at the development of productive interaction systems through four major vectors:

  • Content: Information, narrative elements, meanings and sensations as communicated in text, image, video, sound, tactile and other modes.
  • Context: The integrated presentation of content in form, decoration, attitude, organization, selection, values, and experiences.
  • Affordance: The handles that enable the audience to work with and manipulate the content and context.
  • Audience: The users as integral elements of the total system, who operate it through the affordances, and who create the final expressions.

I've created an extensive list of these principles and techniques including ideas such as smart content, real-time juxtapostion, serendipity, faceting, filtering, dynamic excerpting, social design and user as content.  While too long for this article, they are available in a longer paper that's posted on my Web site for this topic --

Implications for designers

Productive interaction changes the role of the designer, shifting the balance from remote, one-way communicator who is represented by fixed linear presentations, towards present, two-way impresario and facilitator, who is represented by a context that initiates collaboration with the user. This has several implications in the way a designer approaches the creation of a work.

Software and data structures

Because dynamic affordance is an integral part of this medium, interactive design uses structures of data and software that require a clear understanding of the benefits and lexicon of these computational techniques. This does not mean designers should be engineers. But whether tagging text copy with metadata, or describing the process (algorithm) for displaying pull-quotes from that metadata, interactive designers should be facile in using the computer as mediator of their expressions. They must also be capable of communicating with software designers when a work calls for additional expertise.

Authoring systems

Because of the complexity of the software and data structures, it's important that designers use and create authoring systems to implement particular versions of their interactive approaches. Implementing these complex interactions from scratch on each new project is impractical, and authoring systems allow the author to amortize their thinking and work over a series of projects.


The interaction designer weaves a complex tapestry of content, context, affordance and audience. The only way to do this successfully is to be deeply engaged with each of these vectors and the relationships and mechanisms that bring them together. This means creating, editing, and reorganizing content. It means doing the same with context and affordance, and this engagement requires a full understanding of, and advocacy for the audience.

Collaboration and detachment

Productive interaction implies a greater level of responsibility to the user and a new detachment from the work. As a collaborator, the designer can't simply make a statement and walk away. He or she has to be present in the interactive work, providing the contexts and affordances for the user. This is not to say the designer shouldn't have a point of view. But a point of view has to be presented in a manner that assumes some give-and-take -- a collaboration.

In this sense, the designer must be more detached from their point of view, providing a level of respect and trust for the user, helping them come to their own conclusions. This detachment often extends to the point of allowing for user failure. Without the freedom to fail, the possible outcomes may all seem safe, pre-determined, and boring. This does not mean the designer dumps a box of junk on the floor for the user to sort through. Nor does it mean offering an infinite set of affordances and options, or an ugly, aesthetic-free presentation. But it does set up a tension and challenge for the designer?how can an interactive work provide freedom and still maintain an authorial integrity? How can high quality form support changeable content? How do aesthetics hold together when the audience can make changes?

First, it means designing everything from the ground up to be interacted with. So when the user remixes the expression, the content, context and affordance continue to work at all levels?form, meaning, function, and aesthetics --with some room for failure. Second, as a diligent facilitator, the designer can provide presets,like directed museum tours, with a high degree of coherence and aesthetic integrity, which serve as good starting points for the user's explorations.

Third, productive interaction requires a high level of authorial comfort with the unresolved outcomes, disorder, contradictions and occasional aesthetic breakdowns of interaction. It's a shift from the idea of fixed, finished design to multifaceted, evolving, user- mediated output. The designer has to embrace the user as a collaborator and member of the design team.

Breaking constraints through design research

Interactive designers have had it beaten into them that they must design for the constraints of the medium. Pixel dimensions, processor power and bandwidth dominate the design process to an overwhelming extent. This is completely valid for commercial product delivered today. But the medium and technology are too young to limit the creative range within these constraints.

Interaction designers should devote part of their practice to breaking the common constraints; designing for very large displays, moving away from the "mouse crouch," incorporating tangible interfaces, and experimenting with new delivery systems. I believe this design research can be done in an effective way by focusing on principles of interaction design and practicality rather than making artistic or "flashy" work. Similarly, the emphasis has to be on effective expression and communication rather than technology.

Design innovation can be made with practical intent and built on existing leading edge technology, showing the way for near-future applications. It can also be approached in a grounded form of design futurism, which opens the imagination for the medium's long-term future, and influences near term design through trickle down design ideas. The impacts that science fiction has on contemporary culture form an interesting example for those wishing to influence design culture.

In the end, a serious approach to a new medium requires a rethinking of expression and what is best communicated in that medium. It also requires a rethinking of the designer's role and approach to the medium. Interactive media takes designers in new directions, where they need to understand software and data structures, utilize custom authoring systems, engage and design the entire system, and adopt a different responsibility and relationship to their expression and audience. It challenges designers to imagine and produce work on unfamiliar and unknown platforms with attention to a set of design principles and an eye towards innovation.

Designing the future

Let's put it this way: The interactive medium is a lousy one. It can be trite, clumsy, self-conscious, and ugly. In other words, it's poorly designed. And here I'm talking about the whole field -- the Web, games, cell phones, media art, operating systems, all of it. As a young medium, it lacks a range of technique and effective design language. Is it entertaining, informative, useful and profitable? Often. But that doesn't make it a mature medium.

The downside of the early success of the games and the Web is complacency. It's easy to get stuck using common design clich?s and ruts and innovating around the margins. On the other hand, and often because of the perceived success, some designers ignore real applications and create obscure, "experimental" work that has no ongoing relevance. Instead, interaction designers should take up the creative challenge and move beyond the limitations of our own rudimentary spinning zoetropes, inventing more powerful principles, techniques and platforms for a more mature and newmode of communicating. We should find the unique and effective characteristics of interaction and develop design approaches around that.

Ultimately, progress calls for designing the future of the medium itself. Recently, the Library Foundation of Los Angeles honored author Susan Sontag with its Literary Award. At the event, Sontag said, "Reading offers you a different model of how to feel and think than is offered by the 'televisual' world. Reading is a producer of inwardness, personal discovery". Perhaps designers can envision a new interaction that offers a similar system of personal discovery, but one that's more outward and expressive, using the direct experience of working with content to create a productive discovery within a media rich environment.

Related Links
Alan Kay Will the future look like 'Minority Report'?
Ted Nelson
They Rule
Vannevar Bush

The zoetrope, a precursor to film


Serendipity stream