Chapter 6: User Interfaces for Mobile Media 253 can be used by specifi ed users to achieve specifi ed goals with effective- ness, effi ciency, and satisfaction in a specifi ed context of use.” (ISO 1998). The user interface should be easy to use, but how can we say that one UI is easier to use than the next? We need to evaluate the UI and obtain results that can be measured. In this section, we list some of the measurable usability factors (Dix et al. 2003, Nielsen 1994; Shneiderman 1998). Learnability refers to how long it takes a novice user to learn the commands required for performing a set of tasks. Often this is pre- sented graphically in the form of a learning curve, where the x-axis shows time and the y-axis presents the number of mastered tasks or an equivalent measure. By observing and comparing learning curves, one can claim that an application is easier to learn than another. Learnability is diffi cult to measure in a short test session. In this case, we can say that we do not measure the learning process as such, but immediate usability; for example, something that is related to how logical the structure and navigation is; or how familiar the terminology is. Also, if we perform a series of tests with the same application and users, we can study how well the users have maintained the knowledge and skills they have learned over time. Effectiveness relates to how well the user can achieve the goals, measured by the percentage of accomplished goals or the number of features used, to name but two criteria. Effi ciency determines how long it takes to perform a certain task. Typically, these are the most fundamental and most frequently per- formed tasks. This is a useful measure and easy to do since we can measure time (typically milliseconds) provided that we have defi ned the start and end points of the task in question. However, we do not necessarily have to count seconds; instead, we can count the number of mouse clicks, key presses, or navigation steps. How many and what kind of errors do users make when they are performing frequent tasks? It is error rate that we are referring to here. Counting errors and removing their sources from the design is impor- tant for user satisfaction. For example, if you have a speech recognition system with 90% accuracy, it still means that every tenth word is rec- ognized incorrectly and the user must react accordingly. Also, it may be worthwhile to study how long it takes to recover from the error. Subjective satisfaction is how pleasant the user perceives the design, depending on the attributes above, as well as visual and auditory design and styling. There are different ways for evaluating usability. Common methods include heuristic evaluations, where the experts review the application