Information architecture and omnipresence
Information architecture often creates structured, domain-centric spaces. Putting a page (especially an aggregation) at the end of a link in a menu creates a ‘space’ and implies this is where the content lives. I know that sometimes, structure is good. Structure helps users. But what about when things live in multiple places, or serve multiple user needs? Is there a way to do information architecture and experience design that is less about places and more to do with user intent, desire and interest?
I know that if you’re looking for a specific piece of content, structure is the thing that will help you succeed. Information architecture is about creating order. Order is needed because there’s just so much information. If nothing was organised how would we find or use anything (check that drawer in your kitchen or hall way where you just throw stuff)? But there’s a difference between creating order by building structures for information to sit in, and the rest of what information architecture can offer. I worry (a bit) that we sometimes conflate information architecture with navigation and ‘spatial’ structure. Has location ruled the roost for too long?
ShURLy you can’t be serious?
Information architecture is about understanding content and ensuring it’s presented (described, represented and made accessible) in a way that helps users know what it is and where it is. IA is about understanding, constructing and communicating mental models. We use these models to organise content. And part of modelling things requires us to ask questions. In order to tell the user ‘what it is’ and ‘where it is’ we need to ask what it is and where it belongs. But these are two separate questions. Maybe too often we’ve reduced the question of ‘what it is’ to other (simpler) questions – what is it about? what is it for? what format is it? Then in turn, we’ve reduced our focus (or at least what we communicate of our information architecture to users) to concentrate on ‘where it lives?’
The biggest part of an IA’s job has often been creating a sitemap. We draw connected squares to represent buckets of content. We decide how many buckets to create, then describe what sort of thing should go in each bucket. We work hard at creating buckets that users recognise and understand quickly and easily. But maybe good IA is not always about segmenting the model into ‘physical’ sections – buckets.
Like with new babies, each bucket needs a name, often with an address based on this (less true for babies). Web resources rely on their addresses – they all live in a specific place. So it’s no wonder that location has been such a big part of information architecture. But our need to give things addresses, and these addresses often being hierarchical, has meant that lots of information architecture has seemed like an exercise in reduction and abstraction, rather than description.
We created folders with names. We made menus to provide access to these folders. Embedded folders often mean we created sub-navigation when things got complicated. And because so much of the structure of the web has been defined by the hierarchical folder structure of this older kind of website, information architecture has been forced to communicate in taxonomic thinking. Folders within folders have organised our content.
We’ve afforded access to information by simplifying things. We settle on a dominant paradigm on our site and carry it through to its conclusion (whether logical or otherwise). This leads to arbitrary classification of content. In good cases this arbitration is done following good IA practice, so things work nicely. But with a more complicated webby (graphy) web, context and identity cannot be conveyed through location. It’s not just about folders nowadays. Ever decreasing folders is not the only way to organise content. It’s probably not even the best way.
Identity vs. location in information architecture
Location is a technical necessity of the web. You need an address to access a resource. But I’m starting to wonder whether ‘structure and sections’ are the best way to think about the foundations of information architecture. Maybe we’ve relied too heavily on navigation – menus and sections – to communicate (and create) the structure that sets expectations and directs user journeys. The web doesn’t really create physical spaces – the folders are virtual, it’s a metaphor.
Users don’t necessarily see all the wonderful information architecture we create, nor should they. But does the reliance on location mean they can’t even experience or feel the influence of all the work we do? Users see menus. They see the names of the buckets. Too often the buckets have been the focus of attention and orientation. But the bucket-naming started with the content.
The buckets don’t matter – it’s the stuff that’s in them that counts.
Not where, but what
What if we don’t need to always abstract. While our folder structures, bucket-based abstractions and structural navigation have implied that the web is a series of carefully constrained domains, I’m not sure that the experience of the web as a whole has ever really been this ordered.
Dynamic publishing means that the content that used to fill our buckets is a bit more magic. Now, those “pages” we grew accustomed to are often dynamically generated, and can belong in multiple buckets at the same time. We have the semantic web, which isn’t so much about buckets – it’s much more about the things in them – with identifiers representing the things.
On the web, domains are locations. We register a domain and create a space (often a hierarchical one). But ‘real domains’ aren’t hierarchical, taxonomic constructions. In the real world, domains are areas of knowledge, influence or activity. What if we could create information architecture that resists the location-centric (taxonomic), technical definition of domain and started to think about other ways to represent the real world online?
By thinking structurally and physically about the web we force users to occupy spaces. But this reliance on physical metaphors puts the blinkers on our users. We’re all forced to think in terms of the structures, not the connections. A big part of God’s omniscience is his omnipresence (I know this for sure. I did A-Level Theology). What if we could break the reliance on structured space and get our users closer to God, giving them a sense of being everywhere, but with a ‘deitific focus’ to turn down ‘noise’ and focus on their interests and intentions.
Menus, the structural bits of websites that do so much to communicate the information architecture have always(?) been based on location. We’ve done domain-driven design, and created information architecture. But this architecture has then relied on spatial metaphors. What if a domain demands something else? What if the restrictions of translating a model to a physical representation does our model, content and experiences a disservice? Is there a way to create modal navigation?
Over the next few months I’m going to be playing with this idea. I want to explore whether sub-navigation can be used to represent a users’ mode or intent, rather than their location. I like the idea that true domain-driven navigation wouldn’t be a reflection of the reduction and abstraction of the model, but reflect the reality and the reason the user is interested in the content.
Most of the time users don’t care about where they are. They care about the content on the page and the tools to take them elsewhere. What if you could devise a new type of page furniture that can wrap more functionality around content?
I’m not really sure whether and how this could work. But I think exploring a way of navigation being less about places and more about about content would be a good thing. What if the area usually reserved for sub-navigation enabled users to switch modes? We could add a footer to provide additional context and onward journeys. Content lives in the middle. But things are surrounded by their context(s), and context is more than just their place in a structure.
Omnipresence is a neat trick to pull off. It’s probably also the easiest way to create a route towards omniscience. Shouldn’t our information architecture give users the chance to learn about anything (and everything) they want? Maybe untethering them from ‘places’ on the web is a way to do that.