The definition of information architecture is still up for grabs—sometimes the discipline seems to have as many definitions as practitioners. According to Louis Rosenfeld and Peter Morville's Information Architecture for the World Wide Web (the bible of the profession) the information architect "clarifies the mission and vision for the site, balancing the needs of its sponsoring organization and the needs of its audiences; determines what content and functionality the site will contain; specifies how users will find information in the site by defining its organization, navigation, labeling, and searching systems; [and] maps out how the site will accommodate change and growth over time." Since IA is a relatively new and wide-open field, it attracts professionals with backgrounds in Web design, usability, library and information science, human/computer interaction, journalism, and other areas.

In recent years, ASIS&T (formerly ASIS) has been reinvigorated by reaching out to emerging disciplines such as information architecture, and the audience at this meeting was very different from the crowd one would see at past ASIS events. After some brief remarks by ASIS&T officials, Rosenfeld, who is with Argus Associates, set the tone for the conference. He noted evidence of the increasing growth and maturity of the field, such as advances in tools and metrics;a professional salary survey; the debut of the listserv SIGIA-L, which now has over 1,700 subscribers; and the growth of IA communities on the local level. He was also very upbeat about the state of the industry. Despite the slowing economy there were numerous postings for information architects at sites such as Monster.com and HotJobs.com. In Rosenfeld's view, the field is "countercyclical," with the amount of information growing exponentially and the number of information architects growing in a linear fashion.
 

It's the Content, Stupid
Jared Spool's keynote address was entitled "It's All About Content." Spool is a well-known usability consultant and the principal of User Interface Engineering. If you have never seen him, he's an extremely entertaining and thought-provoking speaker. Amidst a collection of Web site examples that were mostly amusing and occasionally shocking (users of the Disney.com site can quite easily go astray and have been known to appear in Anaheim carrying reservations for Orlando and vice versa), Spool pondered the paramount importance of content and the challenges it poses to information architects. He called for information architects to develop an in-depth understanding of the content users are looking for and showed examples of how they can be led toward their goal by following "scent." Scent, an idea originally developed at Xerox's Palo Alto Research Center (PARC), is a way of describing information-seeking cues and behavior that Spool's company is now exploring. He also detailed the dismal state ofWeb interactions, which have been frozen at a 42-percent success rate for several years now, and called for better tools to measure user success at Web sites, manage content, and plan information architecture. 
 

Nuts and Bolts
In keeping with the "practical" approach of the conference, many presentations were of a pragmatic nature, such as case studies or nuts-and-bolts discussions of development topics, including tools and metrics. The talks were short, with presenters limited to 20 minutes to allow time for questions, and the extremely engaged audience members took full advantage of their question-and-answer time. 

There were seven of these brief presentations during the course of the day. I won't go through all of them here, but will briefly describe two examples. First, Phil Oye of Sapient walked the audience through a complete redesign of the United.com Web site, which included interviews and exercises with real customers and a period of prototyping using paper and Post-it notes. He mentioned the benefits of a modular approach and urged the audience to begin collaborating with graphic designers as early as possible in the process. Next, Adam Polansky, a senior information architect at rareMEDIUM, demonstrated the close relationship between information architecture and content management by detailing how his company handles the planning process from content-requirement gathering to organizing information to interface design.
 

Vendors, Where Are You?
One striking difference about this conference was the lack of a vendor exhibit area. In fact, the cry, "Vendors where are you?" literally went up from the attendees. Although there are tools available for site development and content management, for the most part information architects are doing their jobs with homegrown or poorly adapted software tools, and a heartfelt cry for better tools went up more than once. No one seemed to feel that an all-encompassing IA-in-a-box application was warranted, but some areas that could be addressed better are site mapping, site diagramming, thesaurus management, and content management. A killer app would dynamically update site maps and navigation paths as changes were made in the architecture.

The meeting took a reflective turn at the end of the first day with a "reflections and projections" panel that tried to take a look at what had happened since last year's meeting and where the field was headed. Some issues raised included the controversial question of IA's relation to usability, whether there was really a need for separate professional credentials, and the need to agree on problems and methods. One panelist mentioned the upcoming challenge of portable devices and Internet appliances, which were brought up elsewhere at the conference several times. 

Usability expert Alison Head observed that IA is a large part of her work and decried the lack of practical hands-on solutions. She also said that this is a time when information architects have the ear of the Web community and that they'll have to learn to talk to people outside their own IA world (which is difficult, as this is a complex field that overlaps many others). She encouraged information architects to reach out to other disciplines. 

Peter Merholz noted that the concept of modularity in development was coming up more and more. He expressed some disappointment that there were no discussions of business at the conference, and challenged the community to do a better job of addressing business needs and explaining how IA is relevant to the business world. To a certain extent this is making a virtue of necessity, as he also observed that "money isn't flowing the way it used to." After thesefairly brief observations by the panelists, the floor was opened up to the audience members, and they had quite a lot to say. Among other things, they were concerned about how to sell IA, and the friction or resistance that IA is meeting from other disciplines. There was a certain amount of intellectual jousting, mixed with a lot of laughter.

There were four concurrent topical presentations after dinner. The best-attended seemed to be a session entitled "What Do You Do All Day?" led by Jesse James Garrett and Christina Wodtke. There was a very lively discussion about the IA's role in various companies, the gap between the published job description, and how the information architect actually spends his or her day. I didn't catch enough of the other presentations to do them justice, but there was certainly enough variety to provide something for everyone.

We were right back at it early the next morning, with a session on criteria and metrics for evaluating architectures. MartiHearst ofthe University of California­Berkeley detailed an attempt to develop an empirical and automated approach to evaluating Web site usability on a basic level. This might be used as the basis for a development tool analogous to the Microsoft Word Grammar Checker, which could aid the many people creating or managing Web sites who are not professional Web developers. Judy Cantor of Giant Step presented some work that she and her colleagues had done at AT&T on an automated approach to gathering data about what users were actually doing on a site. This data could be used for evaluating usability among other purposes. Right now, usability testing tends to be very labor-intensive and expensive, and both of these talks pointed toward ways to streamline and automate expensive parts of the development process in the future. 

Perhaps as a reaction to all the practicality going around, Peter Morville (a founder of Argus Associates and a "big thinker" on information architecture topics) apparently scrapped his intended speech and instead gave a pep talk to the troops, urging them to avoid relying on metrics as a magic bullet in a time of uncertainty or succumbing to the sins of simplification that might result in dumbing down the profession. Instead, he gave examples of successful sites, such as REI and Wine.com, that show an excellent understanding of what the user might be looking for, and then point him or her toward the needed information. In a sense, this brought the conference full circle to Spool's keynote address, since he was calling on information architects to develop this type of in-depth understanding of content and how to lead the user to it.

There were yet more case studies that morning, and after lunch, two more talks. Professional search engine consultant Avi Rappoport spoke about the consequences—both intended and otherwise—of adding local search to a site, and Amy Warner of Argus Associates discussed the linguistic foundation of labeling. These were both very good, substantive talks, and I was a bit disappointed on the speakers' behalf that they were left for the end of the meeting, as the audience was starting to thin out by that point.

This conference had a very high energy level—both the presenters and the audience cared passionately about their profession and where it was headed. The industry is growing up quickly, and there were some signs of growing pains. Due to the economic slowdown and the dot-com shakeout, some information architects were looking for work when in the past they had been pursued by employers. There was also some uncertainty about where the field is headed, especially whether information architects will become more specialized, how they will relate to other disciplines, and whether excessive attention to pragmatic concerns will lead to lower-quality work. All of this debate seemed thoughtful and healthy. I did have one minor quibble, which was that very few presentations were available as handouts. If the organizers do get them put up later at the conference Web site, that would certainly help.

Overall, this conference was a breath of fresh air. After attending many larger conferences and trade shows over the years, one becomes a bit jaded and begins to wonder glumly if the talk with the enticing title is really a thinly veiled sales pitch, or whether the exhibit area will hold anything new or different. At the ASIS&T Summit, however, the passion and intelligence of the participants were the main exhibit, and the talks were generally very worthwhile. And while ASIS&T conferences are sometimes accused of being too theoretical for some tastes, the practical focus of this meeting made it into an unusual mix of intellectual excitement and useful career information. Several attendees came away saying that this was one of the most productive conferences they had ever been to, and that they had come away with a language for discussing what they were trying to do with their companies.

Given the speed at which this field is evolving, it's too soon to predict what next year's meeting will cover. Based on current evidence, one thing seems certain: It won't be dull. 
 
 

Alice Klingener is a freelance writer and information consultant. Her e-mail address is grovelandgroup@yahoo.com.