Usability: Difference between revisions

From Citizendium
Jump to navigation Jump to search
imported>Howard C. Berkowitz
m (CLI, menu, GUI; VR, special-purpose, and ubiquitous computing to be added.)
imported>Caesar Schinas
m (Robot: Changing template: TOC-right)
Line 1: Line 1:
{{subpages}}
{{subpages}}
{{TOC-right}}
{{TOC|right}}
'''Usability''' is the study of how people interact with manmade objects, both physical and electronic, in order to make it easier for the goal of the interaction to be achieved. In the computer field, usability falls under the heading of [[human-computer interaction]]. In engineering, it is considered part of [[industrial design]].
'''Usability''' is the study of how people interact with manmade objects, both physical and electronic, in order to make it easier for the goal of the interaction to be achieved. In the computer field, usability falls under the heading of [[human-computer interaction]]. In engineering, it is considered part of [[industrial design]].



Revision as of 05:28, 31 May 2009

This article is developing and not approved.
Main Article
Discussion
Related Articles  [?]
Bibliography  [?]
External Links  [?]
Citable Version  [?]
 
This editable Main Article is under development and subject to a disclaimer.

Usability is the study of how people interact with manmade objects, both physical and electronic, in order to make it easier for the goal of the interaction to be achieved. In the computer field, usability falls under the heading of human-computer interaction. In engineering, it is considered part of industrial design.

Usability looks at items from the perspective of a person trying to use them and asks: Is the user able to do what they wanted to do with it? Was it easy to figure out how to use it? Could it be made easier? What was left out?

One of the biggest components of usability is accessibility, the adaptation of products to ensure they are usable by people with various disabilities: color blindness, motion impairments, low literacy, and so forth.

Usability includes ensuring the object is usable at all, but efficiently usable by the expected user. It may or may not seem paradoxical, but there will be cases both where the designer has very little idea of the typical user, and where the design is for a very specialized user, where learning to use the object will take much effort. This is very bad if the object, such as a tax form or a public World Wide Web page is intended for the public in general. It can be quite reasonable, however, to develop specialized interfaces for cases where the characteristics of the user can be highly defined, and where a significant training time is acceptable.

While the term "user-friendly" is familiar, it really tends to mean intuitive for an untrained user. At the other end of the spectrum are interfaces intended for carefully selected users who will receive extensive training. That far end of the spectrum is characterized by what can be called "expert-friendly" interfaces. For example, the pilots of high-performance fighter aircraft have gone through extensive selection processes that consider their ability to master complex tasks, work under extreme physical and mental stress, and constantly adapt to an ever-changing situation. John Boyd characterized this as the ability to:

  • Observe
  • Orient
  • Decide
  • Act

faster than one's opponent. It can take a year or more for a selected pilot to use specialized displays and controls as if they were extensions of the pilot's body and senses. This is not a luxury available to designers of tools for the general public, such as Citizendium. Nevertheless, Citizendium has to have a selection of interfaces appropriate for general users, for authors, editors, and technical support personnel. Those interfaces need to be appropriate for each class of user, and the interface modes may be quite different.

Interface types

Some unfortunately inaccurate terminology has crept into common usage about interfaces. Among general-purpose interface types, the most basic is the command line interface (CLI). The next is the use of windowing, not to be confused with Microsoft Windows, which allows multiple concurrent CLIs to coexist on one physical display.

A next level involves various types of menu interfaces. This can be a slight extension of a CLI, such as presenting several textual choice and requiring one to be used, or it can be a full screen or window containing a form with protected explanatory fields and data entry fields.

More often than not, when users speak of graphic user interfaces (GUI);;, at a minimum, they mean a menu- (MUI) or forms-based interface. A true GUI means that some sort of non-character display -- drawings, graphs, TV pictures -- is involved in one direction of communications, or in both. That can be as simple as a web site showing my cat pictures, or a graphic input mode, as in sketching presentation slides or doing precise engineering drawings.

In CLI, MUI, and GUI, there is a clear distinction between the user and the manipulated object. Other paradigms, however, include virtual reality (VR) and ubiquitous computing (UC). In VR, the user becomes part of the interface environment. In UC, the user is unaware of interfacing to the object, but the environment includes sensors that perceive user desires transparently to the user.

Principle of Least Astonishment

There is a principle of good human interface design, which goes back many years in computer science, often called "the principle of least astonishment."[1] It is unclear who first coined the term.[2]

Astonishment in activities of daily life

Before moving to the web, think of astonishment in daily life. While national conventions vary, a person entering a darkened room usually expects the light switch to be in a consistent place. Internationally, the conventions for using coin-operated telephones can vary greatly from country to country. Frequent business travelers, irritable from a long airplane flight, frequently mention the astonishment of trying to find the location of the headlight switch, and indeed the ignition key slot (and key release) in different models of automobile.

Delay in finding light switches, or in operating emergency equipment such as fire extinguishers, can have real safety consequences. Aviation safety improves when both cockpit crews, and ground controllers talking to pilots, use very standardized conventions for giving commands. In medicine, continuing attempts to reduce errors focus on avoiding giving drugs names that are close to those of others, and taking great precautions to avoid confusing injectable drugs with different concentration for different purposes.

Astonishment in telecommunications

While the principle is perhaps most often used with respect to the usability of World Wide Web pages, it appears in many other contexts. In telephony, whether traditional or voice over Internet Protocol, unpredictable pauses and variability in speech cadence, introduced by the communications system, are the greatest problems in understandability as measured by Mean Opinion Score and other metrics of user satisfaction. [3]

Astonishment in basic computer interaction

Again going back long before the Web, another way computers annoy users in astonishment about interactive response time. Back in the days of mainframe computers, it was repeatedly found that users complaining about response time were more disturbed by similar actions taking wildly different amounts of time. A longer, but consistent, response time was far more acceptable.

Astonishment in Web design

User friendliness and expert friendliness

To be moved inline or to bibliography

  • Stanford Linear Accelerator Center study of the effect on telephony Mean Opinion Score of various sorts of impairments
  • Meredith Bricken: Virtual Worlds: No Interface to DesignTechnical Report R-90-2, Human Interface Technology Laboratory (HITL), Washington Technology Center, University of Washington, Seattle, WA [1]


References

  1. Peter Seebach (01 Aug 2001), "The cranky user: The Principle of Least Astonishment. Some tips for meeting user expectations and avoiding unpleasant surprises", Developer Works, IBM Corporation
  2. I believe I first heard it from Marv Zelkowitz of the University of Maryland, sometime in the seventies; it may also have been Ben Schneiderman. User: Howard C. Berkowitz
  3. Llana, Andy (April 1, 2004), "Changing support requirements for the wireless end user.(Internet)", Computer Technology Review