IA of Potentiality: Toward a Grounded Theory of Information Architecture Philosophy, Theory and Research, The

Bulletin of the American Society for Information Science & Technology, Aug/Sep 2006 by de Hubert-Miller, Brian Arbogast, Burnett, Kathleen

Invitation to Open a Conversation

Although information architecture is, as Adam Greenfield has pointed out in Boxes and Arrows, whatever information architects say it is, we will not be seen as taking IA seriously until we can demonstrate we are willing to be responsible for what we say. I agree with Andrew Dillon's assessment in one of his 2001 IA columns in the Bulletin that "the biggest obstacle to IA becoming a distinct discipline remains its lack of unique methods and theories." Therefore, I propose we begin a formal conversation focused on methods and theories. Doing so does not deny or disparage the interesting informal everyday conversations that have been taking place on our listservs, at community meetings or during professional practice. They are vital sources of ideas. I feel so strongly about their importance that I chose to adapt grounded theory as my dissertation research method. My adaptation of grounded theory has allowed me to integrate these informal conversations with more formal precursor and contemporary discourses that surround IA. The formal conversation that I propose has, in this sense, already begun.

As a participant observer in the development of the emerging discipline of IA, I was motivated to instigate a formal conversation because of some practical problems I have encountered as a teacher, a researcher and the manager of a professional practice in IA. As I have traversed each of these roles, I have found it difficult to consistently describe and explain IA. Although my interests have always landed me in positions where I was required to wear multiple hats simultaneously (sculptor, designer, manager, leader, businessman, etc.), I found myself at loose ends more often with IA. I lacked a philosophical and theoretical base to ground my understanding regardless of the hat I happened to be wearing at a particular moment. The problem became so acute for me that in the spring of 20041 decided to stop teaching and practicing IA temporarily so that I could turn my full attention to addressing this problem through my research. I set myself the task of developing a framework that would provide me with the philosophy and theory that I lacked to ground my work.

It now seems that what I conceived of as a problem particular to my multi-role situation has become an acute problem for the community at-large. Practitioners are performing excellent work combining complex ideas, curricula are emerging to prepare new professionals and new venues for publication are being established. Each of these work products speaks to the seriousness and rigor with which the community is now prepared to approach IA, as well as to the extent of the community's aspirations for its future. The everyday, informal conversations on listservs and at IA-related conferences, especially the most recent IA Summit in Vancouver, are increasingly nuanced. Vexing issues that require the sort of examination one would expect to be provided by academics keep resurfacing. In 2002, Greenfield "with a groan and a resigned, Charlie-Brownish smile" lamented the seemingly unending reoccurrences of what he called

...the "What is IA" discussion. No matter what the stated agenda, it seems, some well-intentioned newcomer - and occasionally a knowing provocateur invariably raises the question of the field's provenance. Is it traditional library science retrofitted and renamed for a digital age? Is it a subset of interaction design? Isn't it just something that all good web designers do anyway, unconsciously? What's the difference between it and usability? How does it relate to "architecture architecture"? And what's an "XMOD," anyway?

I confess to being one of Adam's knowing provocateurs of the past. Now, however, I think the time is right to turn the conversation from one held with static texts (which can't talk back) to one with dynamic human beings (who certainly will).

Starting the Conversation

To begin this conversation, I offer the approach that I have developed over the past several years working on my dissertation. This is only one of many possible approaches one could use to compose a philosophical and theoretical basis for purposes of IA, but I believe it will provide a rich and inclusive opening. I will try to be clear about why I have chosen this particular approach as I describe the constellation of concepts that ground it.

The most immediately obvious way to approach philosophizing about IA is to first compose a philosophical argument about the nature of information, then extend the rationale of that argument to account for/describe the set of characteristics of information that, for purposes of IA, can be said to constitute its architecture. If IA is a discreet domain, it will choose to describe these characteristics and their relationships somewhat differently than other disciplines have.

There is no lack of philosophical argument about the nature of information. While many disciplines have been engaged in this argument over time, computer science (particularly the field of artificial intelligence) and library and information science are the modern disciplines that have sustained engagement with the nature of information most consistently over time. The arguments are too various and too complex in their specifics to recount in this context. Suffice it to say that while neither exhibits what I would call a disciplinary position on the nature of information, there is nonetheless virtually no cross over between the two. Every argument about the nature of information is built on three postulates, which although often either not stated at all or only obliquely, are essential to the argument's development. The first postulate regards state (what is information?); the second purpose (why is information?); and the third form (how is information?). To avoid what I have found to be the major stumbling block in understanding arguments about the nature of information, I offer my postulates as transparently as possible: