The first step in transforming data into information is to explore its organization. This simple yet crucial process can appear futile, but often you can discover something through it that you had never seen before. It is important to realize that the very organization of things affects the way we interpret and understand their separate pieces. Take any set of things: students in a classroom, financials for a company, information about a city, or animals in a zoo. How would you organize these? Which is best? Richard Saul Wurman suggests five ways to organize everything, but seven seems clearer to me. Literally everything can be organized by alphabet, location, time, continuum, number, or category. Additionally, things often can be randomly organized (in other words, by not organizing them).
Often, there are often better ways to organize data than the traditional ones that first occur to us. Each organization of the same set of data expresses different attributes and messages. It is important to experiment, reflect, and choose which organization best communicates our messages. It is also important to note that these seven ways of organizing make it easy to brainstorm, but it is up to us to choose the most appropriate way to present data to our audiences. Some of the more important ways of organizing are presented below.
Most books have alphabetical indexes because, while we may know exactly what we are looking for, we often do not know where to find it. Though the alphabet is an arbitrary sequence of symbols, indexes work well because we have been taught alphabetical sequence from an early age. It is not universally useful, as you will find if you ever try to use a phone directory in a language that uses a different alphabet than your own. Many indexes are organized alphabetically, but few products are (primarily dictionaries, encyclopedias, and other reference works). This is because few data have any great meaning based on the first letter of their labels. The alphabet actually is a continuum (from A to Z) but it is a special one for the reasons stated above.
Locations are natural ways of organizing data with important relationships with or connections to other data. How easy would it be to find the exits to an airplane if they were listed and described in text with no diagram? Many projects can benefit by locational organizations but seldom are, simply because their designers never experimented with the idea. Producing maps and diagrams is not as easy as writing text, but if you have ever used an atlas or ACCESS travel guide, you know how much better a sense of a place can be achieved when things are oriented by geographical relationships. Consider a subway map that simply lists all station stops in sequence versus one that arranges them in a representation of the city they serve. This may sound obvious (and it should be), but why do not car manuals organize parts by their location in the car, or medical books by location in the body (the one thing you are sure of when you know you aren't feeling well)?
Organizing things by time or sequence may sound obvious for bus and train schedules or historic timelines but it can be just as effective for instructions such as cooking, driving, or building. Time need not be addressed only in minutes and hours but also in days, months, years, centuries, processes, or milestones.
Any qualitative comparison can be described with a continuum. All ratings systems, whether numbers of stars or the number of RBIs of a professional baseball player, indicate a value scale. Arranging items in a continuum indicates that this value scale is the most important aspect of the data. As with any organization of data, the primary organization expresses a different message and importance than other organizations.
I categorize number systems as a separate way of arranging things. Much like alphabets, numbers are merely an arbitrary continuum (usually Base-10, since our species has ten fingers). But unlike alphabets, Base-10 numbers are much more universal because they combine in different forms due to mathematical relationships. It is common but not necessary for number organizations to be continuums. For example, the Dewey Decimal System, used for organizing books in many libraries, is not a continuum because it is a number system that does not represent any magnitude or attribute; it simply assigns numbers to categories and sub-categories.
Categories are a common organization and a reliable one, since they allow similar things to be grouped together by attributes that are considered important in some way. Defining the specific categories is crucial, as they will communicate the designer's prejudices and understandings more easily than any other organization. As with all organizations, these control the perceptions of the information.
While random or arbitrary organizations might not seem a useful way to organize things and "add value" to them, it is sometimes the best way if a challenge of some kind is involved. Consider a game where all of the pieces are arranged already or one where its sequence is already determined and carefully, logically laid out. Such a game would not be very fun to play. There may be other times where random organizations present a better experience than an orderly one and it is up to the designer to explore these possibilities and employ good judgment.
beginning | next
back to thoughts | nathan.com
Copyright 1994 Nathan Shedroff
Richard Saul Wurman is one of the most renowned information architects. His book Information Anxietyis one of the few sources of Information Design instruction. His issue of Design Quarterly, Hats, is a condensed version of his most important understandings.