Information architecture: five things information managers need to know; as the information boom in business, public, and consumer cultures continues, information architecture and information management will become indispensable.
* Defines the field of information architecture (IA) and the role of information architect
* Describes the principles and methodology of IA
* Provides a starting point for learning more and applying these ideas to information management
Information architecture (IA) is a Relatively new field, one born in the high-tech boom of the 1990s. Given the current interest in information technology (IT) it might seem strange that IA values human brainpower so highly in the struggle to organize information rather than accepts portal-in-a-box-style solutions. Though the dot-com bubble has burst, IA lives on with important lessons for those involved in the continuing struggle to structure complex information on Web sites and intranets.
What is IA?
Information architects help build Web sites by organizing them to make it easier for people to find what they want. This also makes sites easier to manage. Often, they lack the technical skills to be a graphic designer, programmer, or system administrator. Instead, their specialty is the intellectual work needed to organize large amounts of content. This is part science and part art.
Much like an architect for physical buildings, information architects design information spaces by considering the ways they will be used and then create blueprints and detailed plans for that use. They take care to ensure that people can find their way through these structures to accomplish what they want without getting lost.
In 1996, Richard Saul Wurman coined the term information architecture. According to his definition, an information architect is:
* an individual who organizes the patterns inherent in data, making the complex clear
* the person who creates the structure or map of information that allows others to find their personal paths to knowledge
* the emerging 21st-century professional addressing the needs of the age focused on clarity, human understanding, and the science of the organization of information
Wurman's definition is helpful, not least because of the metaphor it evokes. However, many IA practitioners feel that this overall approach to IA relates more to visual design and emphasizes Wurman's own background in designing printed media.
The first edition of Lou Rosenfeld and Peter Morville's influential book Information Architecture for the World Wide Web defined the role of the information architect as someone who
* clarifies the mission and vision for the site, balancing the needs of its sponsoring organization and the needs of its audiences
* determines what content and functionality the site will contain
* specifies how users will find information in the site by defining its organization, navigation, labeling, and searching systems
* maps out how the site will accommodate change and growth over time
Effective IA works mostly behind the scenes, whereas good visual design is more discernible. Rosenfeld and Morville stress principles of library and information science as they apply to the task of creating and evaluating Web sites. All sites and information systems have information architectures--whether or not they are planned. If well planned, they should be intuitive and, therefore, largely invisible.
Dissension in the Ranks
Because the field is fairly new, agreeing on a definition has not proved easy. The community of practicing information architects never tires of debating and refining the definition. At a collaborative site devoted to IA known as IA Wiki, there is a running conversation entitled "Defining the Damn Thing" (www.iawiki.net/Defining The Damn Thing). Part of the debate focuses on the duties and job description of an information architect. The field is highly interdisciplinary, and people who are involved in IA are as likely to come from the disciplines of graphic design, technical writing, instructional design, human-computer interaction, computer science, and ethnography as from the information sciences.
It can be helpful to separate the job title of "information architect" from the activity or study of "information architecture." This is the distinction between doing IA and being an information architect. Some people embrace IA as central to what they do. Others see IA as a niche discipline or as a small portion of the overall user-centered design process. Important components include:
* Visual design: the layout and graphic design of information to maximize the effectiveness of communication
* Interaction design: planning the use of dynamic components on Web pages or in a software application so users can perform essential tasks
* User experience design: creating an overall experience based on users needs
* Usability: evaluating a system to make sure that it is easy and efficient to use and then incorporating findings and recommendations into the design process
There is also significant overlap between information design and IA in terms of goals and methods. Saul Carliner makes the distinction that IA "primarily focuses on the structure of content, especially complex Web sites" and leans toward visual design while information design "represents a focus on the fusion of content, structure, and appearance of documents." (Editor's Note: See Carliner's article "Designing Better Documents" in this issue.) In contrast, IA tends to focus more on ideas about the conceptual structure of content, and here there is influence from the information sciences.
Most professionals involved in user-centered design can at least agree that somebody needs to address IA for every Web project, whether or not they have the title "information architect." The creation of any site requires that someone apply some level of graphic design, whether or not that person is a graphic designer by trade. In the same way, producers or project managers might design the information architectures for sites. An information manager would be another obvious candidate to adopt the responsibility of doing IA without being considered a full-time information architect. It is reasonable to predict that all these related disciplines will continue to evolve, although there will continue to be some convergence, particularly in the uses of IT. In the meantime, the most important point is that all will value user-centered design even while emphasizing different methods.
Why should information management professionals care about information architecture?
Records and information management is a much more mature and established field than IA, However, both share a connection to the information sciences (e.g., representation of information, thesaurus design, and information retrieval). Information architects and information management professionals share a passion for organizing information, creating effective content management strategies, and providing efficient access to that content for users.
Both professions also are riding the waves of change brought on by the conflicting forces of the explosion of networked information and the fickle winds of economic change. Information managers and information architects currently are finding that they must justify their value in an era of tightened budgets and shrink-wrapped solutions. However, many areas of our society have increasingly ambitious expectations for creating, sharing, and managing information. Not long ago, however, just having a Web site, intranet, or electronic records management system was enough for many organizations. Yet, over time, these same organizations have expanded the scope, variety, and expectations of such initiatives. Many also have learned that "solutions" ultimately require human beings skilled in organizing information and optimizing the way software tools and systems are implemented.
As the information boom in business, public, and consumer cultures continues, it seems safe to predict that both IA and information management will become indispensable. Information managers increasingly are more involved in organizational efforts to support knowledge management, intranet development, content management, customer relations management, e-commerce, and other Web-based enterprise applications. The IA field has emerged in order to address these same issues. Examining this emerging discipline may suggest useful ways of approaching current and future work.
What are the Key Concepts of IA?
IA's focus is on the logical structure of a Web site (or other information system). When one designs or provides a critique of the information architecture of a site, there are a number of important concepts to consider.
Top-Down vs. Bottom-Up
There are two complementary approaches to designing and evaluating the information architecture of a site: top-down and bottom-up. The top-down perspective designs a site from the main gateway to secondary levels and downward. A site blueprint often helps to show the big picture of how the site as a whole is organized. Because it provides access to other pages, the site's home page receives a great deal of emphasis in this approach.
Bottom-up IA emphasizes the smaller components of a site. Often this means examining and designing lower-level pages that contain the actual content. Or, in the case of a dynamic database or XML-based site, information may be broken into even smaller chunks of reusable content also known as content objects. Examining this content in close detail and understanding how it is used often will suggest patterns and ways to bring the pieces together using appropriate metadata. Having robust metadata is particularly important for sites that plan to include personalization functionality.
It may be helpful to think of these two approaches separately when learning about them. In practice, they go hand in hand. The best information architectures are built by successfully combining top-down and bottom-up strategies.
A collection of content can be organized in many different ways. For example, arranging a collection of product literature for an intranet might be done according to:
* geographic sales territories
* target customer group
* product name
* type of product or subject
* corporate division
* date of last revision
Some of these schemes are exact, and there is a commonly agreed-upon convention. Alphabetical order or geographic location constitutes organizational schemes that anyone can easily recognize. Ambiguous schemes are often helpful but more difficult to get right. Famous ambiguous schemes include the Dewey Decimal system or the categories found on the Yahoo home page; both seek to represent a broad area of human understanding. An organizational scheme can be independent of the user interface. That is, while a site might adopt a completely new look and feel, its organizational scheme may not need to be changed much if it is well planned. To achieve this objective, an organizational scheme also should include rules and plans for how new content will be added in the future.
Navigation systems are the elements of a site that help users know where they are and what is available within that site. These are the signposts and road maps that help us travel, whether attempting to accomplish a specific goal or merely browsing. This includes navigation found on content pages throughout a Web site, such as top and side navigation bars. Often the top is used as global navigation and includes links to resources that one would want to access from anywhere in the site. Side areas often are used for local navigation, which show the user what topics are nearby within a division of the site.
Contextual navigation also is often included on a page on the side or bottom. It is used to suggest related topics and content that appear in other areas of the site. For example, if you are looking at content of the history of Ireland, contextual navigation might show you links to travel to Ireland, Irish cooking, or Irish literature.
Another example of page-level navigation is breadcrumb navigation, which shows either the position of a page within a hierarchy or the path taken by the user.
Site maps and indexes are examples of site navigation and exist alongside content pages. A site map provides an overview, via a table of contents or a graphical representation, of the structure of a site and provides access to all parts. On the other hand, a site index might place the content in alphabetical order much like indexing for printed publications. Paul Kahn and Krzystof Lenk showcase this type of navigation in their book, Mapping Web Sites.
Search is certainly an important feature of site navigation; it is a topic large enough to deserve special attention. Users demand and expect to be able to search a site, but it is not trivial to design an effective search mechanism. Search interfaces may range from a simple text box to advanced search interfaces that offer a variety of ways to select criteria and limits. Results are just as important as the interface for entering search criteria. The information presented in results should be meaningful, and the interface should offer methods of revising a search strategy so that the user does not reach a dead end. One of the best resources for learning more about search interfaces and technology is found at Search Tools (www.searchtools.com).
There are many different choices to make when implementing a search feature. An information architect can help to specify user-interface requirements for selecting a search engine from a vendor or to optimize one already in place. Often a site that relies on database content management will have a custom-built search interface closely attuned to the content and data structure. Information architects who have an information sciences and information retrieval background are especially helpful in this situation when starting from scratch--rather than installing and customizing an outside vendor's solution--is the best answer.
Labels are everywhere on a Web site. They include page titles, links on the home page, links in the navigation, entries in an index, choices in a dropdown list, product names, and names of corporate divisions. Together these form sites' labeling systems, which are most effective when they are well-planned and consistent. One rule of thumb is to make links match the titles of the pages to which they connect. Also, the style and tone of a labeling system should be closely attuned to a site's audience. For example, the words chosen to describe mutual fund information to novice investors will be different from that of a site for financial advisors.
Controlled Vocabularies, Taxonomies, and Thesauri
A controlled vocabulary, in the broadest sense, is any group of preferred labels used as a naming convention or topics identified according to a set of explicit rules, such as how they are organized, how they are applied to content, and how new ones are added. A controlled vocabulary can be as simple as a flat list of terms, such as the U.S. Postal Service's two-letter state abbreviations. They also can be large with complex relationships between terms. Taxonomy has become an industry buzzword and often is associated with portals sites on intranets and the Web. Yahoo's directory is one of the most famous examples and is closely tied to this veteran site's organization scheme. In the Web world, taxonomies are rarely as strict as the taxonomy of species of plants and animals. However, taxonomies are decidedly hierarchical and form an inverted tree that gets wider as one goes deeper.
Lately, information architects have been fascinated with the possibilities of thesauri for the Web. According to Jennifer Rowley's classic textbook, Organizing Knowledge, a thesaurus is "a compilation of words and phrases showing synonyms, hierarchical and other relationships and dependencies, the function of which is to provide a standardized vocabulary for information storage and retrieval systems." There is an ANSI/NISO standard, Z39.19-1993, which describes guidelines for constructing specialized controlled vocabularies. Thesauri feature hierarchical, associative, and equivalence relationships between terms.
Whether strictly or loosely interpreted, thesaurus structure offers intriguing possibilities for ways to organize a site and classify content. Though there are not many on the Web yet, Bitpipe.com provides a good example. This site integrates a thesaurus for computer industry terminology into its searchable database of white papers.
What sort of methodology do information architects follow?
Many information architects work as external consultants providing recommendations or joining an ad-hoc team responsible for executing a project. Others work full-time within organizations and have specialties ranging from e-commerce to knowledge management. While there are many differences in the ways that people practice IA, many agree on the overall approach evangelized by Rosenfeld and Morville, who advocate a methodology that focuses on
* Content: Information architects excel at investigating the content of a site and analyzing the patterns and relationships between chunks of information. Beyond simply immersing themselves in documents, they also pay close attention to search logs, site statistics, metadata, and controlled vocabularies.
* Users: It is critical to know about the users of a site, the categories into which they fall, and their information needs. Whenever possible, information architects adopt usability and ethnographic techniques to learn about the audience for a site. They also prefer to test IA architecture designs with users early and often.
* Context: For a design to be successful, understanding the goals, metrics, and politics of the organization that owns the site is critical. It is particularly important to find out about the technical and human resources available to build and maintain the site. To accomplish this, an information architect needs to communicate with stakeholders who are strategic managers as well as those who will be running the site on a daily basis.
IA Design Process
The IA design process follows a cyclical pattern much like other user-centered design disciplines such as printed forms design. It starts with research and recommendation. Getting familiar with users, content, and context is the main concern. Based on their research, information architects produce a number of deliverables during this phase including
* critiques of the existing site (if the project is a redesign) and/or competitor sites
* composite personas that represent the site's audiences and scenarios that encompass their common tasks and needs
* summary of the site's mission, goals, and evaluation metrics for success
* bottom-up recommendations about an approach to organizing content including document templates, metadata, and controlled vocabularies that will be needed
* high-level blueprints that present a top-down approach to how the site will be organized
The next phase, conceptual design, focuses on fleshing out the details of the direction established in the first phase. This often means producing wireframes, also known as page architectures or mockups. These are black-and-white drawings of the key pages of the site, and they are useful in several ways. The process of producing them requires taking abstract requirements and making them concrete. They communicate the design of the navigation, content, and task sequences to the rest of the team and to stakeholders. As a low-fidelity prototype, they make a great instrument for user testing before time and money are spent in development. Other deliverables produced during this phase include controlled vocabularies, document templates, content management guidelines, and metadata requirements.
Again, user testing during this phase is critical. This need not be time consuming or expensive. Steve Krug's Don't Make Me Think! contains many ideas for low-cost, high-value usability testing. However, it is important to plan for time to iterate the design based on the issues that user testing reveals.
During implementation, when the site is actually being built, IA may still have an important role to play. The information architect may need to work closely with the technical, graphic design, and editorial team members. During implementation, there are many questions to answer, and information architects are often asked to handle quality assurance. The information architect may apply controlled vocabularies and metadata schemes to the content. He or she also may provide training for other people who will be performing these activities. Implementation is also a great interval in which to perform further user testing to see if the site is meeting the metrics for usability established in the beginning. Throughout the process, it is crucial to leave enough time for testing, iterations, and review.
How can I learn more about IA?
Every day, the number of resources for learning about IA grows. Several excellent books have been written on it already, and more are on the way. Professional organizations are creating special interest groups devoted to it. There are seminars and college courses as well as masters degree concentrations. A good way to start taking advantage of what IA has to offer is to explore IA resources such as those listed here.
READ MORE ABOUT IT
Argus Center for Information Architecture (ACIA). Available at www.argus-acia.com (accessed 23 July 2002).
Boxes and Arrows. Available at www.boxesandarrows.com (accessed 23 July 2002).
IAWiki. Available at www.iawiki.net/IAwiki (accessed 23 July 2002).
Intone, Keith. "Usable Web." Available at www.usableweb.com (accessed 23 July 2002).
Nielsen Norman Group (NN/g). Available at www.nngroup.com (accessed 23 July 2002).
Search Tools. Available at www.searchtools.com (accessed 23 July 2002).
American Society for Information Science and Technology (ASIS&T). Available at www.asis.org (accessed 23 July 2002).
Usability Professional Association (UPA). Available at www.upassoc.org (accessed 23 July 2002).
University of Baltimore, School of Information Arts and Technologies. Available at http://iat.ubalt.edu/ (accessed 23 July 2002).
University of Michigan, School of Information and Library Studies. Available at www.sils.umich.edu (accessed 23 July 2002).
Kahn, Paul and Krzysztof Lenk. Mapping Web Sites. East Sussex, England: Rotovision, 2001.
Krug, Steve and Roger Black. Don't Make Me Think! A Common Sense Approach to Web Usability. Indianapolis: Que, 2000.
Nielsen, Jakob. Designing Web Usability: The Practice of Simplicity. Indianapolis: New Riders Publishing, 1999.
Reiss, Eric. Practical Information Architecture: A Hands-On Approach to Structuring Successful Websites. Boston: Addison-Wesley, 2000.
Rosenfeld, Louis and Peter Morville. Information Architecture for the World Wide web. Sebastopol, CA: O'Reilly & Associates, 1997.
Rowley, Jennifer and John Farrow. Organizing Knowledge: An Introduction to Managing Access to Information. 3rd ed. Brookfield, VT: Gower Publishing Co., 2000.
Wurman, Richard Saul. Information Architects. Zurich: Graphis Press Corp., 1996.
Chris Farnum is an information architect with Compuware Corporation. More than three years of his experience were with Argus Associates, a well-known IA group. Farnum may be reached at firstname.lastname@example.org.
|Printer friendly Cite/link Email Feedback|
|Publication:||Information Management Journal|
|Date:||Sep 1, 2002|
|Previous Article:||Lies, corruption, and document destruction: never before in the history of American business has a global corporation been destroyed by acts directly...|
|Next Article:||Designing better documents: information design professionals attempt to understand what makes documents usable and to apply that knowledge in...|