Information Architecture for the World Wide Web: Designing Large-Scale Web Sites. By Peter Morville and Louis Rosenfeld, O'Reilly Media, ISBN 978-0596527341, 528 pages.
Information architecture helps people find the information they want, whether it's a prospective student looking for a course, or staff using the intranet. It is mostly concerned with how people browse and search the institutional Web site, and with how knowledge is shared. People working in information architecture are likely to be responsible for institutional Web strategy; global links on the corporate template; the institutional search engine, site map and index; institutional analytics; portals and primary stakeholder home pages; content management and content management systems. This book is an essential reference in this field of work.
The term, information architecture, was coined by the architect Wurman  in 1976. He felt that information professionals could learn from the discipline of architecture. For example, wayfinding helps people to find their way around an unfamiliar building. Morville  and Rosenfeld  both have a library information management background and so contribute powerful metaphors, such as the thesaurus.
In this book, the context of information architecture is large-scale or enterprise Web sites which consist of multiple sub-sites and distributed authoring.
The book has 21 chapters in six parts:
The book is almost comprehensive in describing the theory and practice of information architecture. Because of the library background of the authors, some of the links between the terminology and systems are not always clear. For example, the thesaurus theory in chapter 9 is comprehensive. However, it is not until reading the MSWeb case study that it becomes evident how useful, indeed, essential a thesaurus can be to the quality of the Web presence. After reading the book, the reader will have to work out the gaps in their own organisation's resources and develop a strategy for change. The book suggests starting by asking the business and the users where they experience the most pain. If you work in-house, you probably have a good idea where this is already. But the voice of users lends credibility to the opinions of the information architect and business owners, which is why there is a strong emphasis on user needs and behaviour.
This edition is a major change and improvement from the first edition , which has less than 200 pages and just 10 chapters.
The second edition  has 461 pages in 21 chapters with identical chapter names to the third edition. The new edition benefits from an updated list of tools and software, and of essential resources. These alone could make it worth updating from the second edition. Furthermore, the book includes new sections on tagging, folksonomies and diagramming. The section on enterprise information architecture has been rewritten to reflect lessons learnt by the authors since the second edition.
This reference covers almost everything the institutional information architect needs to consider. For anyone practising or thinking of practising enterprise information architecture, it is an essential read. Krug  is recommended for user testing. If you are not an information architect but would like to find out more about the field, or if you are interested in applying information architecture on a smaller scale, a good reference is Wodtke .