Joshua Porter picked up on my comments and blog post about social design and gulped at what he saw. When talking on IAs, Joshua writes:
“IA at its most basic is the wrong frame with which to approach Design”
“The problem is that IA models information, not relationships. Many of the artefacts that IAs create — site maps, navigation systems, taxonomies — are information models built on the assumption that a single way to organize things can suit all users … one IA to rule them all, so to speak.”
“One technologist and designer … even referred to this ever-growing set of definitions as the ‘IA land-grab’, referring to the tendency that all things Design are being redefined as IA.”
To be honest, I found Joshua’s comments quite amusing.
And if Joshua had’ve wrote most IA models information, not relationships, I would have agreed with him. I’ve done quite a number of work in the area of information classification and know first hand that most standard approaches to organising information (by some IAs) simply don’t work because not all people think about their information in the same way.
As knowledge workers, most of what we do is use and process information, use other people’s knowledge, and produce more knowledge and information, in the course of our day-to-day working lives. What I keep seeing, time and time again, are systems that try to help people to do this work, but are badly designed.
Most systems architects, designers, and business analysts, therefore, try to solve the problems people have with accessing knowledge and information by coming up with ways to make systems that are easy to use. Many designers (IAs, UX specialists, and the like) deliver taxonomies and other rigid expressions of information classification and just expect users gulp them down believing them to be inherently usable. A recent experience with a developer reinforced this for me — he was going to use the same systems template he had always used and expected people to just learn how to use the system.
For the most part, there are very few cases where “one IA to rule them all” will work (specialist fields of science immediately come to mind). Generally, I get indigestion when I think about delivering only one way to find information.
To deliver systems that support knowledge work, ones that are truly usable, all designers (IAs included) need to first understand:
- why do people do what they do
- how do they think about the knowledge and information in their heads
- what information do they need, and
- how do they share it with others
Most of this is User-centred Design 101, as articulated by Jesse James Garrett.
To answer these questions, I draw on theories and models from my psychology background. I find that understanding how people think, how associative memory and recall works, and the behaviour people exhibit when looking for and processing information, is vital to my design work. This helps me determine how navigation systems need to work for a particular group of people, for particular users, and for a particular exercise. My expertise in social psychology also gives me insight into the group dynamics and interactions, so that social behaviour, particularly the sharing of information, can be modelled and supported.
I know many IAs say that what we do is both “art and science” [1], but I don’t agree. It draws heavily on the science of cognitive, behavioural and social psychology (or at least, if it doesn’t, it should do so more of the time).
Lastly, and probably most importantly, most modern models of how people store information in their heads, and how they relate to information (as Joshua rightly suggests), is about relationships, and not about classifying information itself. This is why I believe that topic maps are so important in the information classification paradigm. Put simply, with topic maps, you can represent and articulate the relationships between pieces of information, rather than having to classify them. And these are pathways that have built in redundancy — just like the web of memories in people’s heads, navigation models should allow multiple ways to find information, rather than using just one.
Does this mean an ever widening definition of IA is or does? Are IAs doing the land-grab? Hardly. IAs need to evolve their discipline to ensure that as new ways of thinking emerge they incorporate them into the systems they designs. For me, being an IA means bringing psychology, knowledge management and almost 15 years of experience working and designing for the web. Ultimately, adapting an evolving discipline will mean better systems for knowledge workers — and isn’t that the point?
M
- - - -
[1]. Information Architecture. Wikipedia. Online at: http://en.wikipe….rmation_architecture
