I think therefore IA?
Andrew Dillon, for the ASIST Bulletin, December/January 2001*
'Information architecture' (IA from now on) has rapidly become the first ASIST hot button of the 21st century, seemingly launched upon us by the mid-year summit 2000 in Boston (reported in previous issues of the Bulletin) and fueled by a flood of discussion on the SIGIA listserv (details at: www.asis.org/Conferences/Summit2000/Information_Architecture/listserv .html). Now, in this, the first of a series of columns on IA that I will write for the Bulletin, it is my pleasure to cover the emergence and shaping of this topic in regular time.
Of course, like all 'new' topics, IA one has its own roots and longer-term adherents, and many have not been shy about finding references to the term in old books, or articulating their track-records and pedigrees as 'information architects' (IAs from now on). I mean, how new can IA be when some people claim to have been doing it for over 20 years? While definition of IA is at best fuzzy, ranging from the narrow view that it is a web-specific profession, to the dismissive view of it as just another name for library science, it is clear to me at least, that the IA label excites many people. Furthermore, IA appears to provide some degree of identity to people who struggled to see themselves as part of any distinct group within the collective information design and analysis professions.
For me, the most important aspect of the Summit 2000 meet was the enthusiasm people shared for the ideas. Having entered that meeting slightly dubious of the IA concept, I left convinced that this was one label that meant a lot to a lot of people. There were numerous testimonies from those who claimed to have found a home with this group, practitioners and even academics who were delighted to hear others' tales of identity crises and cross-disciplinary work practices. Interest was piqued and desire given form at the summit, and being there, it really felt like one was present at a historical event. So far, so good!
However, a couple of residual doubts remain for me. First, while I accept IA as a concept and even a meaningful description of a design process, I need more convincing that any one person could be described accurately as an Information Architect. Just as I value the process of user-centered design, I would view with some cynicism a person who told me their profession was 'user-centered designer'. So there are architectural properties in information, of that I am sure, but are there any information architects? Most people want to think so, and who could blame them? Information Architect is a very sexy title, bound to impress one's friends and family, and certainly a more lucrative professional title than web designer or librarian. But I find it hard to shake my sense that information architecture currently represents a collective process more accurately than it describes what any individual does. Maybe that doesn't matter. After all, what is it exactly that 'information scientists' do?
Second, I am wary of drawing boundaries between disciplines and skill sets that so clearly overlap despite the job titles or formal qualifications of the participants. In our attempts to define IA, both at the summit and the subsequent online discussions, I heard and read many definitions based on what IA was not, or what is was different from, rather than what it truly was. What inclusionary definitions have been provided seem so narrow to me as to become exclusionary. Can we really create a field whose sole purpose is, as some suggested, to ensure adequate navigation of websites? And even if we could, would it warrant the title 'architecture'? I don't think so. As people spend more and more time in information spaces it is clear to me that the architecture of such spaces will take on increased importance. As such, I long to see a more inclusive definition of IA emerge that incorporates a richer sense of the full user experience in information space. As I have said before, how many users go to a web site just to navigate through information space? Probably the same number of people who visit buildings just to try out the elevators!
Among the birthpains of any discipline are the problems of definition and scope, since people's search for something new precedes any firm knowledge of the precise form it must take. But I am uncomfortable with raising IA as some kind of all-conquering alternative to existing disciplines. For me IA is best seen as an umbrella term under which we will find many concerns shared with researchers who describe themselves as information scientists, interaction designers, usability engineers and so forth (there's those labels again!). We can add to this the methods and skills of journalists, educators, market researchers, system analysts and many more who have ideas and knowledge of what it takes to make information accessible, desirable and consumable by people.
Exploring and resolving these issues will be important for the long-term health of IA. An overly narrow focus might ease the burden of definition, but it will likely only trivialize the concept over time and reduce it to a label. Claiming to have launched a new field on such a definition will garner attention initially but will only disenfranchise many others whose work and approaches really are IA-relevant in a broader sense. To limit IA to just the web (and only current conceptions of the web at that) and even further, navigation and structure within such web environments, seems a doomed strategy, excluding from our ranks many people whose concerns with information design in the broader sense can only enrich the true meaning of information architecture. More than this, the simple web-centric view of IA blinds many to the existing knowledge we have gleaned about interaction and information design from studying these issues elsewhere.
So where does this leave us? Well, the enthusiasm from May has been sustained. The SIGIA group is official within ASIST, the listserv lives, job ads now include the term IA in their descriptions and Kent State has announced a new masters program in IA and Knowledge Management (see www.slis.kent.edu/programs/iakm.php3). We even have a special issue of JASIS on IA planned. By all accounts, this has been a pretty rapid development cycle. However, as we know only too well from design studies, the final product often looks very different from the early prototypes. We have established important requirements for IA, performed some perfunctory stakeholder, user and task analysis, and even seen initial designs floated. Now is the time for some testing of ideas, and as we know only too well, testing invariably leads to re-design and often the revisiting of original assumptions. Now it starts to get really interesting.
Comments and reactions to this column can be emailed to firstname.lastname@example.org
*This article originally appeared in the Bulletin of the American Society for Information Science and Technology, Vol. 27, No. 2, December/January 2001, a publication of the American Society for Information Science and Technology, 1320 Fenwick Lane, Suite 510, Silver Spring, MD 20910 USA. Phone 1 (301)495-0900 http://www.asis.org