The design of a website or computerised environment for a good user experience involves a lot more than the visual interface, says information architect Brian Lyall. It should also be rooted in logical organisation of data and metadata to meet the needs and expectations of the user.
Lyall, who is working on a new version of NZ Trade and Enterprise’s biz.org.nz business portal, which is scheduled for launch around July, says design is a matter of considering the user’s relationship with each item of content in its context.
He told a meeting of the NZ chapter of the Worldwide Institution of Software Architects (WWISA) in Wellington last week that identical links to the same item may have a variety of meanings depending on the visual context in which they appear and the operational context in which they are used.
“Context is king” rather than content, he says. Content cannot be poured into a vessel of any shape and still be reliably found by any kind of user. The first questions to ask in designing a website or other interface are “what will the user want to do?” or “what do you, the owner or other ‘stakeholder’ in the system, want the user to be able to do?”. Whether it is to find information, to contribute information or to transact business, the reasons for going to the site will influence the actions the user will need to take at a particular point in their interaction with the system.
One well-tried way of gaining insight into the way users’ minds work is to give them cards marked with the data elements to be accessed by the system and ask them to sort the cards into categories and sub-categories. The way some of them categorise might surprise architects and lead to a change of plan.
More than likely, there will be two or more different kinds of users, who may think in different ways and the way the data items relate to one another may have to be differently presented for the two groups. Many architects recommend devising “personas” supposedly describing the different kinds of user, but the usefulness of this practice is disputed, Lyall says.
Some social websites allow users to attach their own “tags” to information, in the hope that intuitive structures will emerge. Again, there are different views on the usefulness of this idea, Lyall says, and it often becomes evident that some users are better than others at categorising in a way that helps most of their fellow users and other stakeholders.
It should be borne in mind that public websites do not exist in isolation; other sites may “deep link” to pages of a site other than the home page, Lyall notes. It should therefore be obvious at any stage what the items on the page are, how they are used, where they sit in the overall structure and how to navigate to another page, particularly at higher levels in page structures that have the home page at the top.
Lyall says some of the treasured conventions, such as that belief the home page should be readable without scrolling, have been shown to be less than essential. It is all right to have a home page that needs scrolling, he says, providing this is made obvious by, for example, text that runs on below the bottom edge.
Some of the most-used websites, such as Google and Flickr, evidence a well-thought-out information architecture, Lyall says.
The improved biz.com website will be more interactive, he says, so users following a checklist of how to start a small business, for example, will be able to see items they haven’t read or completed.