Information Architecture
Information architecture (IA) is so ubiquitous it can be difficult to put one’s finger on it. Information architectures (IAs) are in the websites we use, the apps and software we download, the printed materials we encounter, and even the physical places we spend time in.
(The Information Architecture Institute)
My young children currently enjoy playing the game 20 questions. They used to guess wildly...
Are you a car?
No.
Are you an eagle?
No.
Are you the world?
No.
Are you that tree?
No.
They have little hope of getting to the right answer in only 20 attempts unless they learn to think in categories.
Are you alive?
Yes.
Are you an animal?
Yes.
Do you live in this country?
No.
Are you a type of bird?
Yes.
With this approach, even if the answer is no, they will be much closer to the solution. By thinking logically in categories they narrow it down to the right answer. They’re learning information architecture.
My wife teaches cooking and is particularly good at information architecture in the kitchen. When friends move house, they sometimes ask her to help by unpacking the kitchen things and organising them in the new drawers and cupboards. She knows what every gadget is for, when it will be needed and how to organise them along with pots and pans, knives and wooden spoons so that when someone is preparing food in the kitchen he can intuitively put his hand on the right tool right where he expects or needs it to be. My wife intuitively practises information architecture.
When I moved to Germany I easily got lost on Autobahns for two reasons. The first was the lack of roundabouts at exits. In Britain, after exiting a motorway, there’s usually a roundabout, which gives drivers the option to rejoin the motorway (in either direction) if they take the wrong exit. Within a few months in Germany, I’d taken a wrong exit and caused an accident doing a dodgy u-turn because there was no easy way to correct my mistake. The second reason for getting lost in Germany is the lack of compass directions on the road-signs. When one road crosses another, there will be two exits, one for each direction and the great thing in Britain is the addition of North, South, East or West so that if you don’t recognise the place-names in the direction you want to go you can still head off in the right direction. In Germany, it sometimes takes many miles before I realise I’ve gone wrong. I once had to use the sun to navigate! I’m not trying to say that Britain is better than Germany, I’m just saying that as an experienced road-user, I’m used to certain features in the system. This too is information architecture.
Definition of information architecture (IA)
Information architecture can be summarised as organising information and helping users understand that organisation. That is:
- categorising and grouping information into a coherent system (what my children are learning about life),
- labelling those categories, groups and structures in the end-users’ language so that they can easily and intuitively understand how to get to where they want to go (what is missing on German roads),
- positioning those structures and labels in a way that the user will understand, recognise and expect (what my wife is particularly good at).
usability.gov also provide a clear definition:
Information architecture in practice
IA begins before the beginning and ends after the end. Ideally, the work of the information architect begins before the information exists, or before it has been compiled, because part of his or her responsiblity is to create the blueprint for how the information is to be organised. If this doesn’t happen then there is a chance that categories will have been missed or information will have been arbitrarily assembled or data will have been poorly labelled. And the blueprint for the information should be future-proof - part of the plan will cover “how the site will accommodate change and growth over time.”
If you were teleported randomly into the middle of a large city you’d never been to before, your obvious first question would be, “Where am I?” The users’ obvious first question when they land on a web-site or see an application’s interface for the first time is, “Where am I?” Therefore, one of the most important jobs of the information architect is to make sure wherever a user lands in a cyberspace-city there are signposts and recognisable landmarks by which to get orientated as quickly and easily as possible.
Actually, the first job of the information architect is to create the blueprint of the city’s structure before it is even built. This will make creating navigation (signposts, roads, public transport networks) much easier. Obviously, to do this well, the information architect has to have a good understanding of how the parts relate to each other and connect.
Once the site has been built, the blueprint will be an accurate map, but we wouldn’t want to give that to the user. A classic example is the London Underground map; if it was accurate to scale it would not be usable without significant training. Instead, we present the user with a model they can mentally grasp with ease and even remember. The commuter’s map will be more or less useless when it comes to maintenance. So an administrator needs a structure he can maintain and extend, while the user needs to be presented with a model he, as a layman, can quickly understand - it could be a simplified version of the true stucture or it could be faked structure based on meta-data.
Rosenfeld and Morville compare IA closely to librarianship - cataloging, classifying, labelling information. The larger a web-site or application becomes, the more important this work is. To really be successful, the secret lies in making a labelling system that uses the audience’s (or users’) language. Behind the scenes, it can use your language, but the front-facing interface needs to speak the users’ language. A library visitor doesn’t need to understand the organisation system’s numbering or terminology, but simply to quickly locate books on, say, house plants. This is a good example of why the front-end and back-end will use different jargon and it’s important to keep technical jargon from creeping into non-technical interfaces. Having said that, chronological numbering and alphabetic sorting will be intuitive to both.
It should, of course, be obvious that user-research is indespensible in IA. What is the users’ language or languages? What conventions and system features are they used to? The only way to find out is to ask and test.
Notes
- Information Architecture for the World Wide Web, 1st Ed., 1998, Rosenfeld & Morville, p.22.
- ibid. p.11.
- ibid. p.23.