Information architecture and knowledge architecture
By Margaret Adolphus
Architecture arranges space in a logical way – for example, bedrooms are normally close together, the kitchen near the dining room. This means that parents can comfort children in the night, and food can be transported easily from oven to table. Similarly, a town that is planned groups space according to function: shopping centres, car parking for shopping, pedestrianized areas, parks, etc.
Information likewise – although it often isn't – is most helpful when it is grouped in a structured way, so that particular nuggets can be found easily. An example is the way in which traditional libraries are organized: signage indicates how floors are structured and where you can find, for example, fiction or travel.
The problem nowadays is that most information is not found in traditional analogue form, but exists digitally, somewhere on the Internet, which is growing at an intense rate and in an uncontrolled way. That is why we need information to be structured – so that we know what to look for and where.
Hence the emergence of information architecture (IA), poised to become a key discipline for the 21st century, as important as usability. It even, as of Spring 2009, has its own peer reviewed journal, Journal of Information Architecture, organized by academics from Scandinavia, Poland and Italy.
What is information architecture and what do information architects do?
Information architecture is most commonly applied to the organization of digital information – websites, intranets, online communities, etc. – so that they are easy to use, and the right information can be found by the right people.
Just as the look, feel and organization of a stately home (more bedrooms, large dining area, etc.) is different from that of an ordinary family dwelling, that of an online resource will differ according to its function. Flickr or YouTube, for example, have a very informal feel, whereas a bank or library website will be much more structured. And whereas both the latter will need to look authoritative, a library website will also need to look – well – inviting (there are no alternatives to banks, whereas there are to libraries).
Thus IA needs to make a careful study of the user's motivation and information seeking behaviour.
White (2004) traces the history of IA back to 1974, but argues that its real starting point was the consulting company Argus Associates, founded in 1994 (it collapsed when the dot.com bubble burst). The architecture metaphor was used as a way of helping people understand the importance of structure and organization in information.
Two of Argus' consultants were commissioned to write a book, Information Architecture of the World Wide Web, in the second edition of which they came up with the following four-fold definition of information architecture (Rosenfeld and Morville, 2002, quoted in White, 2004):
- the combination of organization, labelling, and navigation schemes within an information system;
- the structural design of an information system to facilitate task completion and intuitive access to content;
- the art and science of structuring and classifying websites and intranets to help people find and manage information; and
- an emerging discipline and community of practice focused on bringing principles of design and architecture to the digital landscape.
Another commonly quoted definition is:
"Information architecture is a foundation discipline describing the theory, principles, guidelines, standards, conventions and factors for managing information as a resource. It produces drawings, charts, plans, documents, designs, blueprints and templates helping everyone make efficient, effective, productive and innovative use of all types of information" (Evernden and Evernden, 2003, quoted in van der Walt and du Toit, 2006).
IA's importance derives from the fact that information is a key asset, and that whether the user is studying or employed in an organization, there is a cost to not being able to find what one needs – frustration, lower grades, poor decisions. On the other hand, the proper structuring of information can reinforce an organization's strategic objectives and help improve efficiency.
This means that information architecture is a professional discipline, which requires devotion, experience and formal training – just as does being the other sort of architect. But what does an information architect do?
An information architect works as part of a project team, employing a range of techniques from such disciplines as
- information science
- artificial intelligence
- management theory
- knowledge management
- programming, and
- object oriented technology (van der Walt and du Toit, 2006).
The following are his or her main activities (Kalbach, 2003; Maier, 2009).
- Researching the audience. It's very important to understand how the product will be used, so the information architect will make use of a variety of methods, including the results of usability tests and stakeholder interviews.
- Georgia Institute of Technology Library's website had two major redesigns: the first was concerned with providing well organized information; the second with usability for non-expert users. The initial redesign involved hiring a consultancy to create a new IA; for the second, the views of non-expert users, people with little experience of the website, were sought. Eight participants were selected from a total of 900 randomly e-mailed people. The test showed that the IA generally worked well, but the difficulty appeared to be choosing the most appropriate search interface. As a result, the website was redesigned based on the users' mental models rather than the librarians' ideas of how clients would use the site, and the "Quick catalogue search" option was replaced with a "Search and locate" landing page with search options (Jeffcoat King and Jannik, 2005).
- A conglomerate of South African companies operating in the telecommunications and information and technology area sought to consolidate its knowledge and information management. As preparation, 82 specially selected individuals were interviewed out of a total of 12,000 employees: the interviews focused on high-level information needs with specific reference to information required by the participants to manage his or her operation on a day-to-day basis; questions were sent ahead of time to give participants time to prepare (van der Walt and du Toit, 2006).
- Analysing data. Working closely with designers and developers, defining the site's primary objectives, focusing on a number of typical users, often referred to as "user personas", whose behaviour patterns, skills and attitudes are described.
- Developing the structure of the site, using site maps, site-flow diagrams, and wireframes.
- Defining the positioning of the content, and the way in which the user will interact with it.
- Carrying out usability testing.
IA and librarians
And what of librarians: is this relevant to them, or are they already information architects? According to Kalbach (2003), many information architects have a background in librarianship, and both professions seek to match the user's particular information need with a resource. The difference, however, is that information architects are concerned with all sorts of information, and usually on particular projects, whereas librarians deal on an ongoing basis with fixed format documents. The deep structuring of information by controlled vocabularies, thesauri and metadata is common to both professions.
Kalbach's views, and particularly his assertion that librarians are "not very user-centred", are probably dated, but contain some truth, and the former can certainly benefit from IA's deep concern for the user experience.
The process of information architecture
Once the information architect has a good idea of user requirements, and developed the basic site structure, he or she must set about the process of organizing content. For this, some sort of visualization software to map out the flow of information will be needed, with flow charts or dataflow diagrams. Microsoft Visio is a popular diagramming tool, another is OmniGraffle.
Figure 1. An example of information architecture (photo by Thomas Grieg)
Once the overall scope has been mapped, the next stage is to organize the content. This can be done by a "top-down" approach, looking at classifications of knowledge, or one that is "bottom up", examining the documents or resources that are to appear in the system. Whichever approach is used, it is important that the structure is dynamic so that it can encompass the ongoing creation of information.
Faceted classification and taxonomies
In order to organize information, there is a need for a system of classification, often referred to as a taxonomy.
Traditional systems of classification are usually hierarchical, treating knowledge as a tree-like structure with parent-child, subtype-supertype relationships. This means that an object can only be placed in one category, and assumes that knowledge is static and not dynamic. In reality, however, objects have more than one dimension, and new aspects of knowledge are constantly appearing which don't necessarily fit into the previously established mould.
Take for example the travel section of a library or bookshop, which is normally organized by country. Suppose, for example, that one is not concerned about where to travel but wants a particular experience, such as adventure travel or an art tour. In a physical environment, only one organization is possible, but on a website devoted to holidays, one could browse by place, and type of holiday.
Knowledge is also dynamic, so new categories will appear that don't fit the mould. Take the sciences for example, which are constantly spawning new areas of research and even disciplines.
Faceted classification, originally proposed by the Indian mathematician and librarian Ranganathan in the 1930s, overcomes these problems because it avoids hierarchical structures and permits use of multiple categories. The faceted approach takes broad, overarching concepts and develops a more refined vocabulary within those concepts. For example, in the domain of shopping, the facets could be brand name, product name, price, place, production, materials. Within that, the classification can be flat, as in brand name, or hierarchical e.g. Europe > Italy > Rome (Uddin and Janecek, 2007).
The advantages of faceted classification, which is often employed by designers of commercial websites, are that it can use terms familiar to the user, and it can insert new layers of categories, which is a great advantage where knowledge is in flux.
The recipe site, Epicurious.com, is a community-based site where users can find and share recipes. You can browse the site by a number of facets, such as dietary, course, ethnicity, ingredient, occasion (Uddin and Janecek, 2007).
Facets can be developed from a study of the domain, organized hierarchically, for example colour – black, red, etc., and stored in a relational database in a range of XML tools (see Uddin and Janecek, 2007: p. 226f, for a description of the process).
Content analysis and mapping
Faceted classification is a top-down approach, representing an ideal. Another approach looks at the entities themselves, rather than the knowledge they represent. It means starting with the documents themselves, and mapping and analysing the content, with the following stages (Batley, 2007):
- Gather together a number of representative documents, and group them according to format (paper-based, electronic, audio-visual, etc.), type (newspaper article, brochure, etc.), source (where in the organization does it come from?), subject, and where it is currently located in the existing architecture.
- For each item, ask: What is it? How can it be described? What is distinctive about it? How can it be retrieved?
- Can the item be segmented into more than one chunk (defined as "the most finely grained portion of content that merits or requires individual treatment", Rosenfeld and Morville, 2002: p. 289)?
- Map out the relationship between the chunks: for example, will they be in sequence, in the same place (co-location), linked, or sharing the same semantic characteristics?
Once the information chunks have been established, each will need to be described by metadata, which details attributes and facilitates easier retrieval.
Applications of information architecture
Much of the above applies to situations where the information comprises documents created by an enterprise. However, IA is just as applicable in a traditional library, particularly as notions of Library 2.0 have done much to encourage greater accountability to the user.
Li describes (2006; p. 399) how the "next generation" (as of 2006) of library information architecture comprises three layers: the first where the users employ all sorts of devices to access library content, the second dealing with requests on a web server, and the third a series of object relational database management systems which will manage the library's resources.
Figure 2. Web-based client/server library information architecture
Yang et al. (2009) go further by proposing a more holistic model. In contrast to the linear flow of Library 1.0:
"Library management is no longer just about literature, but rather a systems engineering approach involving literature resources, librarians and patrons, equipment and facilities, and knowledge services, any or all of which need the whole industry and practitioners' research and exploring continuously for the long-term" (Yang et al., 2009: pp. 284-285).
They call this model "subsection circulatory management", and propose the concept of service-oriented architecture. Features of this system include:
- No disjunction, as often occurs in "conventional" systems, between print and electronic: digital resources should be closely embedded in the system.
- A personalized space (My Library).
- A comprehensive integration of the various "knowledge services", including the OPAC (which should include all library resources), guides to library services, and the various services such as inter-library loan, subject guides, and reference. The object is to integrate the library management systems and user services so that the architecture reflects patron rather than librarian needs.
- A uniform authentication system so that the patron can navigate between the various resources and databases to which the library has links, without being repeatedly asked for a password.
They also describe the five layers of this architecture as the:
- system platform for the software including the operating system,
- library resources stored as data according to metadata standards such as Z39.50,
- management of the library, and
- knowledge services (Yang et al., 2009: pp. 288-289).
Chongqing University Library finalized its Library 2.0 system at the end of 2008. It can be depicted as follows:
Figure 3. Chongqing University Library's 2.0 system (from Yang et al., 2009: p. 290)
A good library information architecture will seek to unite the student with the resources he or she needs to work as easily and as seamlessly as possible. The problem with traditional systems is that the student may need to check the catalogue and then a number of databases separately: federated search, however, turns the library and its related resources into a portal which can be searched through one interface (see: Part 4. ).
The Open University Library has designed its information architecture with its time poor, distance learning students in mind. The latter can download a catalogue widget into their browser, but they can also access relevant resources from the course website. These resources include databases as well as individual items, and mean that a student can browse in a search environment that has already been filtered. This is managed by RSS feeds and tagging.
Information architecture has a particular relevance to commercial and governmental organizations: it is generally recognized that information is a valuable asset and needs to be readily available in order to facilitate rapid and accurate decision making. Enterprise information architecture should therefore dovetail the organization's strategic objectives, and seek to model the information flow in and out of the organization.
An analysis was undertaken of the information requirements of a South African telecommunications conglomerate. The aim was to establish a group corporate information system that would help senior management and, in particular, look at the inflexible financial reporting system.
Information was gathered by a series of interviews, and recommendations were made which essentially involved the creation of one, centralized application which would not interfere with the other financial systems (it was considered too costly to replace the existing multiple systems with one standard system for all, so the new one had to co-exist with the legacy systems). At month end, data would be extracted and uploaded to a central repository, and people would drill down for further information (van der Walt and du Toit, 2006: p. 91).
Information architecture and knowledge architecture
Thus IA can be used in a wide range of situations to structure both conventional items, such as books and journals, as well as less formal ones, such as the documents which are created by enterprises. Building on the foundations of IA, and extending further into the organization's knowledge store, is knowledge architecture (KA).
The difference between IA and KA is that the former deals with explicit knowledge, and the latter with tacit knowledge. Knowledge is information in a context, for example how particular circumstances affect company policy, which may change over time and in different places. It is also the tacit knowledge that people carry around in their heads: for example, an experienced sales agent may have a wealth of experience of a particular client. KA is therefore focused on people, rather than systems, and how they do their work on a day-to-day basis. It will need to employ, in addition to the conventional tools of IA discussed above, some "softer" methods, such as story telling (Reamy, 2001).
KA is also useful for capturing the "softer" aspects of management, which unlike, say, finance, do not have extensive documentation.
An important principle of knowledge management is that it should extend beyond capturing documents to capturing expertise. Hence the document management strategy should not be just about collecting and archiving existing documents, but rather about communication on an ongoing basis. There might, for example, be a directory of experts, and enabling tools such as wikis to help people share experience and capture good practice.
NASA's knowledge management road map can be depicted visually as follows (Holm et al., 2006):
Figure 4. NASA's knowledge management road map (Holm et al., 2006)
A questionnaire was administered to 103 officers from the New Zealand army as a way of getting them to recognize and rate leadership traits. The leadership architecture can be depicted thus (Pech, 2003):
Figure 5. Leadership architecture (Pech, 2003)
Information architecture is essentially about managing content and information in an online environment so that the latter is easy to navigate and the former easy to find. It is a discipline to which librarians have a great deal to offer – and it can also offer a lot to librarians.
Batley, S. (2007), "The I in information architecture: the challenge of content management", Aslib Proceedings, Vol. 59 No. 2, pp. 139-151.
Evernden, R. and Evernden, E. (2003), Information First – Integrating Knowledge and Information Architecture for Business Advantage, Elsevier, Amsterdam.
Holm, J., Olla, P., Moura, D. and Warhaut, M. (2006), "Creating architectural approaches to knowledge management: an example from the space industry", Journal of Knowledge Management, Vol. 10 No. 2, pp. 36-51.
Jeffcoat King, H. and Jannik, C.M. (2005), "Redesigning for usability: Information architecture and usability testing for Georgia Tech Library's website", OCLC Systems & Services, Vol. 21 No. 3, pp. 235-243.
Kalbach, J. (2003), "IA, therefore I am", Bulletin of the American Society for Information Science and Technology, available at http://www.asis.org/Bulletin/Feb-03/kalbach.html [accessed May 14 2009].
Li, L. (2006), "Leveraging quality web-based library user services in the digital age", Library Management, Vol. 27 No. 6/7, pp. 390-400.
Maier, A. (2009), "Complete beginner's guide to information architecture", UXBoothBlog, available at http://www.uxbooth.com/blog/complete-beginners-guide-to-information-arc… [accessed 13 May 2009].
Pech, R. (2003), "Developing a leadership knowledge architecture: a cognitive approach" Leadership and Organization Development Journal, Vol. 24 No. 1, pp. 32-42.
Reamy, T. (2001), "From information architecture to knowledge architecture", Intranets Today, available at http://www.intranetstoday.com/Articles/Default2.aspx?ArticleID=5281&Aut… [accessed May 18 2009].
Rosenfeld, L. and Morville, P. (2002), Information Architecture for the World Wide Web, O'Reilly, Sebastopol, CA.
van der Walt, P.W. and du Toit, A.S.A. (2006), "Developing a scaleable information architecture for an enterprise-wide consolidated information management platforme", Aslib Proceedings, Vol. 59 No. 1, pp. 80-96.
Uddin, M.N. and Janecek, P. (2007), "Faceted classification in web information architecture: A framework for using semantic web tools", The Electronic Library, Vol. 25 No. 2, pp. 219-233.
White, M. (2004) "Viewpoint: Information Architecture", The Electronic Library, Vol. 2 No. 3, pp. 218-219.
Yang, X., Wei, Q. and Peng, X. (2009), "System architecture of Library 2.0", The Electronic Library, Vol. 27 No. 2, pp. 283-291.