Volume 1

Issue 1

March 2004

Information Architecture Ann Rockley, The Rockley Group Information Architecture of Content Management When people think about content management, they generally think about it from a systems perspective, focusing primarily on tools and technology. While it is true that content management usually requires a technological solution, it also requires that content be designed for reuse, retrieval, and delivery to meet your authors' and customers' needs. Content management requires that tools be configured to support authoring, reviewing, and publishing tasks, but first, those tasks must be designed. Designing content and the processes to create, review, and publish it is what information architecture is all about. The Information Architecture section of The Rockley Report will focus on the different aspects of information architecture for content management. This article introduces you to some of the components of information architecture that we will cover in The Rockley Report over time. Information architecture has become synonymous with information architecture for the web. However, as more organizations are adopting content management systems to manage both web and enterprise content, there is a new area of information architecture emerging—the information architecture of content management. One of the key factors for a successful content management implementation is a solid information architecture. Too often organizations implement content management without identifying the authors' needs, without looking closely at the content to determine how it could be most effectively structured to support user/customer needs, and without analyzing their current and desired content life cycle. This results in resistance to adoption, increased costs, and failure to achieve the desired results. Information architecture can make a significant contribution to the success of your content management solution. This is a view supported by Lou Rosenfeld, (www.louisrosenfeld.com), an information architecture consultant and co-author of Information Architecture for the World Wide Web: Designing Large-Scale Web Sites [1]. Rosenfeld has been instrumental in establishing the industry of information architecture for the web and points out: When it comes to making content accessible, content management and information architecture are two sides of the same coin. Authors and end users alike benefit from intelligent design and well-organized processes. [2] People like Lou Rosenfeld, Peter Morville, Christina Wodtke and others in the information architecture and information design industry have laid the groundwork for a move to information architecture for content management beyond the web. The components of information architecture There are a number of components of information architecture that are key in building a solid base for a content management implementation. They include analysis, content models, granularity, metadata, reuse and repository architectures, reuse management, and content management. We introduce you to these components in this article, and will delve into them more deeply in future issues of The Rockley Report. Analysis Good information architecture requires that you start with a thorough analysis of your organizations' needs, your current and desired content life cycle, your customers' needs, the state of your current content, and your technological requirements. During the analysis phase, you need to look at your content very closely to determine how it's put together and the types of content it contains. This will help you to determine opportunities for reuse. You also need to talk to the people who create and use the content to learn what their issues are. This will help you to determine problem areas in work processes that can be addressed in workflow. Content models One of the most critical phases of your information architecture is building the content models on which your content management strategy is based. Content modeling involves identifying and documenting the structure of your content in detail. During the content modeling phase, you determine the elements required for each information product (or output) and how each information product will be designed for optimum usability and reuse. Content models define the structure and organization of your information products, indicating which individual elements they contain, their frequency, and their usage (e.g., is an element optional or mandatory). Models become the road map for your content and are used to develop DTDs/schemas (if you are using XML), or content frameworks and templates. Granularity of content Designing the granularity of your content can sometimes be problematic. Authors typically like content very granular so they know exactly what to put into an element (e.g., overview, procedure step). Very granular content usually results from more semantic models (models with tags that indicate the meaning of the element such as “overview” instead of tags with generic names such as “body” or “para”). Highly semantic models are more problematic for style sheet designers because all unique elements require an individual style. Because semantic names by their nature are unique, all semantically-named elements require their own styles. Granularity also affects how you reuse content. Content that is too granular can be difficult to manage in your content management system, but content that is not granular enough may not be as reusable. Accordingly, CMS developers may push back on the level of granularity, opting for content that is not granular. Analysis of reusability, authoring processes, and tools is important when determining granularity and as you develop your information architecture, you will make changes to your granularity as you determine the optimum level of granularity for everyone. Metadata There are typically two types of metadata: categorization metadata and element metadata. Users tend to retrieve information based on categorization metadata, whereas authors tend to retrieve information based on element metadata. Categorization metadata is used extensively on web sites to categorize content for effective retrieval. It is also used extensively in document management to classify documents for storage. Authors, on the other hand, use element metadata to classify elements of content for reuse, retrieval, and tracking. Care should be taken to ensure that you can retrieve your elements once stored. Your ability to reuse information is only as good as your ability to find it. And if you employ systematic reuse (see Reuse architecture) your metadata must be very thorough so that the system can correctly find and populate the content into the required information products and into the required places within information products. Like granularity, metadata design also continues to develop as you refine your architecture. Reuse architecture Content can be reused within an information product, across information products, and potentially across the enterprise. Traditionally, the most common form of reuse has been opportunistic, meaning that authors make a decision whether to reuse content or not. However, opportunistic reuse is also the least efficient because it requires that authors know a reusable element exists and what it is called, then find the element and reuse it in their information product. In addition, if authors are not aware that an element already exists, they may recreate it causing multiple elements to proliferate in your content management system. This also makes it difficult to know which of the multiple elements is the definitive one. Alternatively, systematic reuse is automatic reuse. Once specific content has been identified as reusable in a specific location, it is automatically inserted (auto-populated) into the appropriate locations. Authors do not have to determine if the reusable content exists or search for, retrieve it, and insert it into the appropriate places. Systematic reuse ensures that content is automatically reused where necessary, thus reducing the burden on authors. When designing your reuse architecture, considerable analysis of information products is required to decide which elements are systematically reusable and where. Once you've decided which elements are systematically reusable, you create content and structure reuse maps as part of your reuse architecture. The content reuse maps identify where content can and should be reused and if it should be reused identically or can be used derivatively (with change). Content reuse maps are used by your content management system to programmatically (automatically) ensure that content is reused. In addition to identifying content reuse, you need to identify structural reuse as part of your reuse architecture. Structural reuse identifies where common structures are reused. For example, you might have a product description element in a brochure, but you would also have a product description element on the web. Even though those product description elements may be structurally the same, they may contain different content. Structure reuse maps are used by DTD/template developers in creating consistent structures for authors to follow. Repository architecture The repository architecture defines how you will structure your repository. For example you may have “building block” directories that include content that is frequently reused (e.g., glossary, procedures, product descriptions) and the remainder of your content stored in information product directories (e.g., all brochures) that are further organized by product. Or you may decide to organize your content by product with each of the information products as a subset of the product. You need to determine what is the most effective repository structure for your needs. Note, however, that the identified structure is not a physical file structure. Content is stored in the database, not in directories. The repository structure enables your authors to easily find information. Reuse management An area of information architecture that is frequently overlooked is that of reuse management. If authors opportunistically reuse content and create derivatives of the content, it quickly becomes difficult to identify which element is the definitive one. Your content management system will end up looking like your current file structure and you will have no clear idea of what is source content, where content is reused, and if there are multiple versions of the same piece of content. Reuse management means creating rules to manage your reusable content. The reuse rules are formalized in your content management system through workflow and in your system configuration. Content control Content control, as part of your information architecture, identifies how your content should be managed. You need to determine how content should be controlled through its life cycle and what security should be applied to it. Content control is tightly integrated with your reuse management strategy and business practices and like reuse management, it is formalized in workflow. Summary Bob Boiko (www.metatorial.com), Director of the University of Washington's iSchool Content management system evaluation lab, content management expert, and best-selling author of Content Management Bible on content management [3], sums up the discussion of information architecture and content management very well: Content management is the dynamic organization of information architecture, business management, software and network engineering, content creation, and publications development. If you don't master each of these areas, CM will fail. If you don't get them to integrate, CM will fail. Information architecture is the structuring of information for effective management and presentation. While the discipline has focused more to date on the presentation side of structure, it is now turning solidly toward management. As it does, the tight connection between content management and information architecture is becoming crystal clear. Information architects, like the building architects before them, create structures. They lay the foundations under and the frames around information. Content managers gather and dynamically deliver masses of information. Without a solid information structure at the core, a CMS effort can't get off the ground. At best, it will be hugely inefficient and at worst it will crumble under its own weight. Information architects have the skills to structure a content domain so that information can flow in a reasoned and efficient way. It flows in according to well understood rules of relevance, segmentation and tagging, and it flows out according to well understood rules of audience interest and use. So, CM needs IA. But IA needs CM as well. CM provides a wider context for IA. It makes IA not just about the best page, or even the best site, but rather about the best system behind all the pages, sites and myriad other outlets for information. CM centralizes IA in the organization. It upstreams IA toward the center of the organization's information systems infrastructure. It integrates IA with business management, software and network engineering, content creation, and it's old friend publications development toward a new concept of what it means to be an organization in the information age. ” [4] References Rosenfeld, Louis and Peter Morville. Information Architecture for the World Wide Web: Designing Large-Scale Web Sites. 2002. O'Reilly amp; Associates, 2nd Edition. Rosenfeld, L. Email interview, February 2004. Boiko, Bob. Content Management Bible. 2001. John Wiley & Sons. Boiko, B. Email interview, February 2004.