According to Charles Sanders Peirce all signs can be classified into icons, symbols and indexes based on the signs relation to the referent. In this system an icon is limited to a representation that resembles its object. An index carries an actual connection to its object, such as a sign of a telephone on the door of a phone booth. A symbol in contrast has no visual connection to its object. It is an abstract sign that can only be understood through learning its meaning. (Rayan & Hubner, 2006)
In the context of interface design the icon has acquired a more general meaning where it stands for any visual representation that denotes an action, a setting or a content type. Compared to semiotics it covers all the three major phenomenological categories.
The research of interface icon classification is a field of semiotic study. There have been various studies and attempts for classifying icons. Jayson M. Webb, Paul F. Sorenson and Nic P. Lyons wrote an article about potential ways for evaluating icons in 1989. The article was titled "An empirical approach to the evaluation of icons" and it was published in the SIGHCI Bulletin in July 1989.
The article is historically significant as it presents one of the early systems for icon taxonomy and attributes for evaluation. The authors recognize three main icon categories:
- Picture – Realistic depiction of system object or function. These are most detailed and easiest to interpret and remember.
- Symbol – Emphasize critical feature by analogy or symbolism. These are simplified and most affected by context.
- Sign – No intuitive connection between icon and referent. These are abstract, simple and association must be learned.
Wang et al (2007) compare this system of Webb with others such as Rogers, who regonized four main types resemblance, exemplar, symbolic and arbitrary, or Lidwell et al's system of similar, example, symbolic and arbitrary. All-in-all they compare nine different existing systems and propose one of their own. But categories overlap and different researchers use different terminology sometimes to describe what in essence is the same thing.
Comparing these systems with that of Peirce, it almost seems the discussion is verging on semantics as much as it is on semiotics. What is needed is a more concrete and practical system. One that can be adopted by interface designers and professionals even without a very profound understanding of semiotic theory.
As a good benchmark Per Mollerup (1997) has devised a system for classifying trademarks that is a good reference for designing a practical system for an icon taxonomy. The system is presented below.
1.1) Graphic marks
1.1.1) Picture marks
22.214.171.124) Figurative marks
126.96.36.199.1) Descriptive marks
188.8.131.52.2) Metaphoric marks
184.108.40.206.3) Found marks
220.127.116.11) Non-figurative marks
1.1.2) Letter marks
18.104.22.168) Name marks
22.214.171.124.1) Proper names
126.96.36.199.2) Descriptive names
188.8.131.52.3) Metaphoric names
184.108.40.206.4) Found names
220.127.116.11.5) Artificial names
18.104.22.168.1) Initial abbreviations
22.214.171.124.1.2) Non-acronym initial abbreviations
126.96.36.199.2) Non-initial abbreviations
1.2) Non-graphic marks
The classification starting from "1.1.1) Picture marks" could be applied to icons also. To build this system Mollerup has defined the following five rules a functional taxonomy must comply with.
- It must consist of classes that are distinct. The differences between the classes must be clear so there is no room for misunderstanding to which class an item belongs to.
- The characteristics the classes are based on should be used consistently and each step in the classification should be based on a single principle of division.
- There should be no overlapping between classes. Parallel (co-ordinate) classes should be exclusive.
- Co-ordinate classes should be able to collectively cover all possible entries.
- The classes should be relevant to the purpose of the taxonomy.
Following these rules, I propose two basic classifications that comply with all the rules.
- Is the icon abstract or concrete?
- Is the icon logical or arbitrary?
These classifications leave little room for interpretation. They are also ideal in that they are practically boolean values if we reformat the evaluation question in another format: "Is this icon concrete?" (true/false).
The classification terms need to be further clarified.
Concrete = The icon is visually recognizable as an image that represents a real world object. To put it more simply, can you name what object, item or thing the icon looks like?
Abstract = The icon is not recognizable as any real object or thing. The icon is just a combination of shapes that doesn't resemble anything.
Logical = There is a clear connection between the image and the function it stands for. The icon is self-explanatory and can be interpreted with little previous knowledge.
Arbitrary = There is no logical connection between the image and what it does.
The next step is to figure out how these two classifications could relate to each other, and can they somehow be used to construct a taxonomy tree. To clarify this we must consider the following assumptions:
- An arbitrary icon can be either concrete or abstract.
- A logical icon can only be concrete. And therefore an abstract icon can not be logical.
The second assumption is the weaker one, since the interpretation of whether or not something is concrete can be very subjective. For instance – is a geometric arrow concrete or abstract?
If we choose logical vs. arbitrary as the basis of our system, we can add concrete and abstract as subclasses of the arbitrary main branch. If logical icons are always concrete, then the same division is not relevant for this main branch. If we compare these classes to those of Mollerup we can find the following equivalencies.
Figurative is the same as concrete.
Descriptive is always logical.
Metaphoric is always logical.
Found is the same as arbitrary.
Non figurative is the same as abstract.
Descriptive and metaphoric icons can never be arbitrary, instead they are directly related to logical icons. Therefore, as we can see in figure 1, these two classes can be considered as natural subclasses for the main class of logical icons.
This is my proposed basis for an icon taxonomy, the validity of which I will evaluate and compare to other existing taxonomies. In addition this system and logic leads to the following research questions that need to be followed:
- Can icons loose their logical meaning over time?
- Are logical/descriptive icons more likely to shift to the class of arbitrary/concrete icons than logical/metaphorical ones?
- Can arbitrary/abstract icons be more valid in the long run, since they leave little room for misinterpretation?
- Mollerup, P. (1997). Marks of Excellence. London. Phaidon Press Limited.
- Rayan, A., Hubner, R. (2006) Pictograms, Icons and Signs. Thames & Hudson Ltd. London.
- Wang, H-F., Hung, S-H., Liao, C-C. (2007) A survey of icon taxonomy used in the interface design. Proceedings of the ECCE 2007 Conference. 28-31. London.
- Webb, J. M., Sorenson, P. F. and Lyons, N. P. (1989). An empirical approach to the evaluation of icons. SIGCHI Bulletin. 21 (1). 87-90.