Reading at work: Rosenfeld and Morville's Information Architecture for the World Wide Web, 2e
My flight to Seattle’s TechDays a couple of weeks ago ended up being delayed by about an hour, so I had enough reading time to finish the second edition of Louis Rosenfeld and Peter Morville‘s Information Architecture for the World Wide Web.
Because the book’s publication date is 2002, the various dynamic elements of AJAX-style interactions are touched upon very lightly. The book ends up arguing along the lines of “designing for the lowest fidelity is safest”, but I’m not sure that browser expectations can be traded away so quickly today. Certainly some of the example websites cited in the text now use a substantial amount of AJAX enhancements that improve site interaction.
Rosenfeld and Morville’s main point is that information architecture—the organization of information so that that information can be successfully retrieved by the various individuals that interact with a site—is hard. Having spent some time worrying about how to publish the engineering workflow around ON (and, I believe, OpenSolaris more widely) open software development, and to make sure that that publication is navigable and meaningful over time, I can only agree. The book presents a medium-scale overview of information architecture: there are “basic principles” chapters on organization, labelling, navigation, and search, as well as case studies and thoughts about practice. Various classic tradeoffs, simple usability tests, and kinds of site prototypes are concisely introduced and then expanded upon, leading to more advanced considerations.
Beyond the illustrative examples from a wide set of websites, one thing that was of particular value to me was learning about the potential role of thesauruses as a specific form of metadata, and how that metadata can be used to make search-based navigation capture a larger set of related documents. We’re also outgrowing the initial design for the opensolaris.org site, so we’ll use some of the ideas presented in the “Process and Methodology” section. The example strategy document is also a useful case study—we probably need to develop a similar document for opensolaris.org in a collaborative fashion in the website project.
The authors also identify some of the potential career forms an information architectural role might take, as well as how information architecture’s shared boundaries with implementation and graphic design are fuzzy. It seems to me that every substantial open development community needs some kind of information architecture, if only to support the classic user/developer audience split, so some of the credentials they suggest might be necessary for a corporate information architect appear elite, or unusual, for a community effort.
Available via Sun’s subscription to the O’Reilly Safari online library.