Sunday, March 23
Knowledge Compass: opening windows, punching holes in stovepipes, forming communities, connecting people to people
Jane Starnes, Sr. Information Specialist at Intel, described herself as a taxonomy specialist, and the heart of her presentation about implementing a “knowledge sharing” solution is a vocabulary of categories that everyone in her organization understands. The problem: 79,000 employees, many departments and organizations, no central document repository, and a real need to communicate across these divisions. Her challenge: to facilitate expertise sharing, open windows between organizations, and create reusable best practices. The solution: a repository containing categories of links, related resources and best practices, with experts who can answer questions for each topic.
Jane described her design process, how she got the organization involved, and how the project rolled out (she had usage numbers to meet). She explained how the screens work, and how experts are chosen. Key takeaways: Management needs to value and reward participation in the project, but collaboration in any company usually requires a big culture change.
Kids in the Mi(d)st – the process of building a design team that includes people whose ages range from 9 to 55
Nancy Kaplan and her team from the University of Baltimore are doing interface design for the International Children’s Digital Library. Nancy talked about their unique design process. They have gathered a team of children who are considered design partners, not just research subjects. When the team (adults and children) does contextual inquiry, for example, the children do the interviews, ask follow up questions, and take pictures. The adults take part in the research, but having children trained to participate in the research adds value and unexpected insights to the process.
Teaching the children how to do research was not easy: they have limited attention spans and making the value of observation clear to them was hard. Only when they had gone through the entire process of using their observations in an actual design exercises did they understand the value and get better at it. Another challenge was the group dynamics: the group was made up of children of different ages (8 to 11, approximately), and was a mix of boys and girls. It took a lot of team building to turn them into a design team.
The children were also active partners in the design of the user interfaces. Again, they had to be taught how to come up with ideas that go beyond the obvious. They became real designers once they mastered a few tools (i.e., Flash and Dreamweaver—teaching them Flash only took an hour!). Once they could start making their own designs, their design ideas became exciting and insightful. Once they began designing, they were taught various exercises in design prototyping.
A big advantage of having a kid design team is that they haven’t yet been socialized like adults. They have high expectations of technology, and that is an advantage. Kids say, “I want it to do X,” where X can be anything.
At the end of the presentation, Nancy said this has been the most fun research she has done in her life and there was a lot of nodding in the panel. They also called for other universities to start using a similar process. More information about their design process (and pictures!) can be found at their website (http://www.icdlbooks.org/adults/adult_info.html).
—Peter Van Dijck
Persuasion Architecture — Waiting For Your Cat to Bark
Bryan Eisenberg, John Quarto-von Tivadar
Bryan Eisenberg and John Quarto-von Tivadar focus their work on commercial websites, of which there are four types: e-commerce, content, lead-generation, and self-service. Each website tries to persuade its visitors to take action, the measure of which is the conversion rate. But the website must meet the visitor’s goals before the business’s goals can be met. Conversion rates for retail stores are around 40%, but for websites it is only around 2-3%. Obviously websites are not meeting their users’ goals as well as they could. The web industry has a 70% failure rate in web-related projects. In any other industry that would be completely unacceptable.
The Minerva Architecture Process (MAP) is something used at FutureNow to help improve the success rate (and conversion rates) of web projects. It prompts non-expert users to create persuasive systems by allowing non-technical business people to learn and manage the development process. There are six steps in the MAP process: uncovery, wireframe, storyboard, prototype, development, and optimization. Creating different personas is very important. Marketing personas help the sales process, while design personas help the development process. There are three elements that go into persona creation: demographics (the attributes of the persona, like age), psychographics (the buying decision process), and topographics (how demo- and psychographics mesh with similar selling processes).
Cross-Cultural Information Architecture: Lessons from Japan
Adam Greenfield discussed the differences, challenges, and some of his experiences of practicing information architecture in Tokyo, Japan. He broke the differences he sees into three types—cultural, social, and cognitive. The media environment is very image-dense in Japan. Motion on the page is greatly privileged. They like things to be animated and pop off the page. Every company wants its website to reflect the heart of the company and the dreams of the users. He told the story of the Farting Salaryman, an animated mascot that farts or burps when you mouse over it. It shows people that the company is human and has a heart.
These differences have implications for the architect.
- First, humility: you must admit you are embedded in your culture.
- The Japanese are a risk-adverse culture. CMS and dynamic sites are seen as risky. You have to work to defuse that.
- Expect to produce non-standard deliverables, such as “before” and “after” diagrams, heavily annotated schematics, and use cases.
- Deliverables we are used to don’t contain enough information and details. Proposals that are 50-60 pages that say little are more important than shorter reports that say everything. You need to show that you have worked hard on the client’s behalf.
- Expect to justify all of your decisions. It’s not enough to say something is a best practice. You must provide supporting documentation. Names are important (such as Jakob Nielsen, Lou Rosenfeld, and Peter Morville) and credentials count. You don’t.
- Dare to be suboptimal. This is a challenge. You may not be happy with the results of a project (since they may not match your ideals of a “good” site), but if the client is happy, then you have to accept that.
Panel: Making Connections with Techies: Five IAs and one IA-friendly programmer share their tips and tricks for making connections with technical types to build sites
Moderated by Chris Farnum and a panel discussion by Margaret Hanley, Kristen Truong, Dennis Schleicher, Jodi Bollaert, and Simon Wistow
Chris began by stating he wanted the panel to be a bit like “Politically Incorrect,” and described how the “useratti” and “techies” have a fair amount of misconceptions about each other. Margaret focused on the right time to design together and to design alone. It’s important to know who “owns” the documents, when to bring them in and collaborate, and how much detail is required to satisfy techies. Kristen explained what technologies IAs should understand, and what IAs should explain to techies. Her tip: IAs can help resolve disconnects between teams. Dennis posed a question, “who’s playing in what sandbox,” or, where does IA end and tech begin? He says we both look at the same data but ask different questions about it. His diagram, showing IA between Creative and Tech sparked much discussion amongst the panelists. Jodi discussed communication problems, especially among project managers. Her tips for managers and for IAs are all about getting involved in each other’s processes. Her plea, “Techies are people too!” got a big audience chuckle.
Then the techie, Simon, came up to the podium. “Techies love food,” he said, explaining how to improve relationships. His point: Techies and IAs work on the same problems, both dealing with interface design challenges. It’s extremely valuable to work together, a lesson he learned when he was moved away from the design team and productivity dropped severely. His self-deprecating comments, calling techies “aesthetically-challenged trolls with a tendency to patronize,” got the biggest laugh of the morning. But his point was well taken by the audience.
Employees’ Experience Levels and the Relation to Usability in a Web-based Information System
Mike Alexander presented the results of a usability study conducted to determine if the level of experience that a user has with the system affects the usability issues uncovered, the success rate of tasks, the use of work-arounds to complete tasks, and the level of perceived usability issues. The study was conduced by Mike along with Hsin-Liang Chen, both of the School of Information at The University of Texas at Austin.
The system being tested in the study was a web application used to configure settings and features used in monitoring industrial equipment. Nine experts and nine novices were each asked to perform ten tasks using the system. The success rate, time on tasks, and perceived level of success and difficulty was recorded and analyzed.
The results showed that expert users were not any more likely to successfully complete a task. However, experts were able to complete almost all tasks significantly faster than novices, and were less likely to use the Help feature. Users were not always consistent in their expressions of difficulty in relation to their ability to successfully complete the required task. In some cases, little difficulty was expressed though they were in fact not successful; in other cases, users could easily complete tasks but expressed a considerable level of difficulty.
Impact of Behavioral-Based User Research on Site Design
Richard Omanson of User Centric, Inc. presented different ways to gather behavior-based information that can be applied to a user-centric methodology.
The techniques were grouped into three categories. In each of these three categories, Richard mentioned several techniques and technologies that can be used to gather and analyze information. An explanation of the technique was provided, along with an example, and a summary of the benefits and limitations.
The categories and their associated techniques were:
- User assessment (for inferring goals and expectations from behavior)
- Card sorting
- Search term analysis
- Site traffic analysis
- Design assessment (for conducting usability tests on the design)
- Reverse card sorting
- Usability testing
- Site assessment (for inferring reasons for shortcomings from behavior)
- Page ratings
- Traffic path analysis
- Eye movement analysis
- Link click analysis
For many of these techniques, Richard demonstrated or showed screenshots of various shareware and proprietary technologies that may aid in the process.
The presentation concluded with a graph showing all of the techniques mentioned, plotted by Actual/Predicted Behavior vs. Reasons Explained/Not Explained. The slide can serve as an excellent starting point for determining the best techniques to use depending on the type of information that needs to be gathered.
Teaching Information Architecture
Chris Chandler, Matt Fetchko, Eric Reiss
Is information architecture a course or a curriculum? Can IA be taught at all? Three instructors from three different institutions set out to give us insight into their approaches, methodologies, and successes in teaching information architecture.
Chris Chandler, adjunct lecturer at UCLA, gave a comprehensive overview of the undergraduate “boot camp” he team-teaches with Lynn Boyden. Although students come to his class with only basic web development knowledge, coming out of his class they are good at “talking their way into a job.” Over the years, Chris has learned to tailor his course—at first, the duo assigned a lot of reading. A 300-page reader was daunting for the students, however, and now the reading is much lighter. At the end of the semester, students engage in a mock project where experts from the community come in and present a scenario. Students have one hour to ask as many questions as they can, another hour to figure out the problem, and another hour to present.
Matt Fetchko, adjunct assistant professor at the Interactive Telecommunications Program, a graduate program at New York University, presented his approach next. Matt gave a survey of information, starting with clay tablets and continuing into the present. Matt offered us tips and strategies for keeping students involved throughout classes. Making each lesson interactive and bringing real examples is key. As with other teachers in the group and in the audience, Matt advocated for group work in class, despite student protests.
Finally, Eric Reiss, an independent consultant who teaches courses primarily to business students, challenged us with the question “Can anyone teach IA at all?” He pointed out that some may say it can’t be done, but he believes it can, although it’s not about lectures, exams, or grades. Reiss’s approach is very different, since, instead of a captive audience over a series of weeks or months, he sees his business school students over a concentrated day or two. His goal is less to teach students IA, then to get them to think about IA. Thus, he likened his role as more of a “midwife than teacher.”
He pointed out that trying to encourage students to simply read a book on information architecture without instruction (with all respect to Peter Morville who was moderating the session), is like asking them to read a cookbook. Teach them how to think. “Set free the rules,” he urged. Rationality means knowing the rules; creativity means knowing when to break them.
The conversation that took place after the “official” presentation was just as active. Many of the attendees in the room currently teach information architecture, making for a detailed discussion with real problems and resolutions. Topics ranged from the skill level needed for IA students to the concern with working with real clients from the community in a class. Victor Lombardi, from the Asilomar Institute of Information Architecture, was identified as a point of contact. He is currently championing a research initiative for AIfIA with the goal of recommending an information architecture curriculum.
Selling the Deliverable: Presenting IA to the Client
Sitting in this session I really felt like a good friend was sharing some hard-earned wisdom he’d gained through years of presenting his work to clients. Seth gave very practical, usable advice on both the content of an IA meeting and the dynamics of the client/consultant relationship (although I believe his advice is just as applicable to internal client relationships). His guidance included tips on establishing the relationship and gaining credibility and trust, setting meeting goals and a realistic agenda, logistics of controlling the room, and even the positive side-effects of an unexpected coffee spill. Throughout, he emphasized the importance of remembering that the client wants you do well, that you were hired because they trust your experience and expertise, and that you should always strive to make your team look competent, and to make your clients look good.
“A Spirit of Simplicity”: What Information Architects Can Learn from the Arts and Crafts Movement
Michael Magoolaghan’s presentation on the similarities in the goals and ideals of the Arts and Crafts movement to that of the work of the IA was a refreshing look back to the past. IA’s are constantly looking forward to the future and new technologies, but Michael reminded us that there are important lessons to be learned from our recent history. The Arts and Crafts movement of England and specifically of the United States is one such period that offers lessons for the modern IA.
In the Arts and Crafts movement, buildings and objects were intended to be useful, beautiful, and affordable. This is not that different from the desire that our work be usable, desirable, and useful. He then went on to expand on the Arts and Crafts ideals—Simplicity, Honesty, Utility, Organicity, Craftsmanship, and Harmony—taking each and comparing them to desirable traits within the work of an IA.
Finally, he cautioned that while the Arts and Crafts movement upheld these ideals, it was often within the context of a socio-political movement. He concluded by asking some important questions: How can we take an honest approach to design? How can we utilize structural decoration? How can we treat a homepage like a hearth? How do we find and provide joy in our work? All good questions to ponder as we make our way in this fast paced modern world.
Hypertext Gardens, Architecture, and IA
In contrast to Michael Magoolaghan’s gentle and retrospective presentation, Mark Bernstein took charge of the space and his presentation much the way a sideshow barker commands that you attend his latest show. Mark was entertaining, provocative, and at more than one point downright antagonistic. He began by proclaiming that we do not want to recreate the failures of the Arts and Crafts movement. He challenged the label Information Architect and whether or not there was architecture involved.
He then proceeded to show the audience a flurry of images of important buildings to illustrate his points, making analogies from the architecture to the work of the IA. His presentation was fast-moving and contained a lot of quotes from the famous architects Walter Gropious and Louis Sullivan. Mark also spent some time reviewing the IA literature and taking the authors and the community to task for only talking about all that is wrong and bad out there.
When he was finished, many in the audience were stunned into silence. A few people asked some very good questions, but most folks left with a lot to think about.
User Experience and IA: Panel
Moderated by Jesse James Garrett with panel discussions by Terry Swack, Jess McMullin, Chris Fahey and Peter Morville
This panel had quite a large crowd, and Jesse James Garrett started things off by questioning the audience on their expectations about the panel. Is Little IA/Big IA really a runaway train?
Terry said she’s glad she can concentrate on what she does best, and that there are always people who can do the other things. Peter questions whether there is a meaningful difference between UX and IA, and states that the dispute is really a political issue. UX vs. IA is not the only model, and he rejects monolithic views.
Then there was some audience discussion about IA vs. UX beginnings, and why the IA term was assumed given its legacy. Peter said he just liked it.
Chris spoke about cross-pollination within the UX field as a whole, while Jess described his experience of adding UX into an already working engineering process.
After that there was much audience participation. The McDonald’s example of methodology was discussed. (You can still create exceptional results, but not by following the methodology.) And our work should be about the end results, not deliverables. Unfortunately time ran out on this lively debate while it was still in progress.
Karl Fast’s presentation on Metadata Harvesting gave a glimpse into a new way of understanding metadata. He opened with the basic definitions of metadata and buildt on those definitions with metaphors of metadata that exists everywhere (music notation, library catalogs, HTML headers, passports, product brochures). We live in a distributed, heterogeneous network of electronic/digital information and our information environment is constantly changing. Metadata is likewise changing to reflect this new environment. After his descriptive opening, he compared the Open Archives Initiative to search engines. He provided several other examples of supporting the bridge between people and information: Napster, Gnutella, Z39.50, and RSS. All these forms of technologies are not new; they are just different ways for providing an infrastructure for metadata across networks of information.
Karl did not offer new models but his analysis of metadata standards and models could offer us insight into future ways to connect people and information. He emphasized the need of Clifford Lynch’s recombinant and migratory metadata in order to support the dynamic nature of information. He offers a starting point for all of us to see how metadata is really all around us and how it could be useful as we produce more information.
Information Architecture versus Visual Design: The Movie or When Worlds Collide: Collaborative friction between information Architecture and Visual Design
After being saturated by two very full days of talks on every IA topic under the sun, it was a welcome relief— for my last session of the summit —to sit back and watch a well-produced video presentation starring some of the colleagues I had seen that weekend. Kim Ladin of Hot Studio presented a video she produced consisting of a series of interviews with eight or so design and IA professionals on the topic of visual designers and information architects. Despite a technical glitch in the very beginning which left Peter Merholz’s first segment silent (I know…Peter Merholz, silent…eerie), the series of interviews gave me new insight into how IAs are perceived by designers, and vice versa.
After shedding light on how IAs and designers differ in their expectations and approach, Kim provided suggestions on how we can all just get along. Many suggestions involved developing mutual respect, assuming competence on both parts, and getting everyone involved as early in the process as possible. Interestingly, Simon Wistow (sacrificial techie on the IAs vs. Techies panel earlier) pointed out that many of the issues between IAs and designers parallel the sore points between techies and IAs. The common lesson: we need to work together, collaborate, and appreciate and respect our differences.
Five Minute Madness
A favorite tradition from past summits has been five minute madness and this year was no different. Everyone is offered the opportunity to come up and say a few words—about the conference; about the profession; about themselves; about their other agendas (i.e., contribute to Boxes and Arrows, contribute to Digital Web, join AIFIA, volunteer for an initiative, attend DUX, etc.). It didn’t really matter. This year, the lineup to the mike started out slow, but once people started coming up there was no lack of commentary and thoughts. All aspects of IA were addressed—from thanking the summit presenters, to thanking the community at large for being together, particularly in these early days of the war, from an observation about the international attendees to shout-outs about the AIFIA organization initiatives. Boxes and Arrows was wished a happy birthday and some folks offered their services within academia to address the research needs of practitioners. The topics were as varied as the speakers and were a showcase of the diversity of this community.
Conference Chair Wrapup
Christina Wodtke, this year’s summit chair, wrapped up the weekend with a few words of thanks to Richard Hill and some food for thought about being an IA. She challenged us all to live! To be passionate about life and all it offers and to continue to learn and be curious. And with that, the summit was over for another year.
- Mike Lee’s photos from the Summit
- Views of Portland: Erin Malone
- Friday evening cocktails: Erin Malone
- Saturday Sessions: Erin Malone
- Sunday Sessions: Erin Malone
- Victor Lombardi’s Photos
- Thomas Vander Wal Photos 1 and Photos 2
- Christina Wodtke’s photos
- Lisa Chan’s photos
Other Summit Summaries & Thoughts:
- Summit Blog setup by Adam Greenfield and Joshua Ellis
- Lou Rosenfeld
- Victor Lombardi
- Thomas Vander Wal
- James Spahr
- Anne Galloway
- Mark Bernstein
- Lieke van der Meer
- Amy Lee
- Mike Lee
- Adam Greenfield
- Long Now Foundation
- Long Bets
- Rosetta Project
- Wayfinding and Navigation in digital spaces
- Model of Attraction
- Sociobiology of Information Architecture
- James Spahr’s poster (PDF)
- Representing Many Voices: Challenges in Government Site Design
Brian Arbogast de Hubert-Miller is the Information Architect for the School of Information Studies at The Florida State University. He is also a full-time member of the faculty and specializes in information architecture theory, experiential learning, and professional practice.
Dan Brown has been practicing information architecture and user experience design since 1994. Through his work, he has improved enterprise communications for Fortune 500 clients, including US Airways, Fannie Mae, First USA, British Telecom, Special Olympics, AOL, and the World Bank.
Madonnalisa Gonzales-Chan aka. Lisa Chan wrangles the volunteers for Boxes & Arrows. During the day she is Metadata Services Manager at Stanford Graduate School of Business.
Liz Danzico is a Product Manager of Search and Browse at BN.com and teaches Interface Design at the Fashion Institute of Technology. Liz has a B.A. in English from Penn State and an M.A. in Professional Writing from Carnegie Mellon. She is a copy editor at Boxes and Arrows.
Seth Gordon uses his understanding of user research and IA to improve user experiences and solve business problems. He has recently completed consulting projects for the Nielsen Norman Group and Razorfish. Visit him at www.gordy.com, where there isn’t a drop of content about user experience.
Brenda Janish an editor for Boxes and Arrows, is an information architect at iLeo in Chicago where she has helped design consumer-facing sites for clients including Kellogg’s, Procter & Gamble, McDonald’s, and the U.S. Army.
Jeff Lash is a Usability Specialist and Information Architect at MasterCard International and writes the IAnything Goes column for Digital Web Magazine. His personal web site jefflash.com proudly has no IA-related information.
Erin Malone is currently a Product Design Director at AOL (America Online). She has been a practicing interaction, interface and information designer since 1993. She is editor in chief of Boxes and Arrows.
Dorelle Rabinowitz has over 15 years experience working as an information architect, designer, producer, and a storyteller in new and old media. She’s been at SBI and Company (formerly Scient) for three years as a lead information architect.
James A. Spahr works as a designer and programmer for Designframe Incorporated. His design work has been showcased in Graphis Books and in ID Magazine. He teaches undergraduate Information Architecture and Graphic Design at Pratt Institute.
Peter Van Dijck is a Belgian information architect who specializes in metadata and international information architecture. He is the author of a book called “Information Architecture for visual designers” that will be published in September 2003. His site is http://petervandijck.net.