3.3 Intelligent Output Interfaces

3.3.1 Introduction

Passing on information to the computational device is one part of the equation, but to be of use, the device also needs to be able to respond to the users. Output, in most cases, means output of visual data. Devices for "dynamic" visualization include the traditional cathode ray tube (CRT), liquid crystal display (LCD), and specialized devices like a pilot's head-up display. Printers are also very important devices for visual output, but they are substantially different from screens in that their output is staticit won't change over time except for the yellowing of paper.

In order to increase bandwidth for information reaching the user, it is important to use more channels in addition to visual output. One commonly used supplement for visual information is sound, but its true potential is often not recognized. Audible feedback can make interaction substantially more comfortable for the user , providing unambiguous information about the system state and success or failure of interaction (e.g., a button press), without putting still more load onto the visual channel.

For the same purpose, tactile feedback or even smell might do very well, with keys and mice " reacting " to input.

3.3.2 Output Purposes

If we look at the output purposes, there are as many as for the input. Output can be used, for example, to give the user the perception of the current situation in a certain process. The output device could propose a set of choices (e.g., "what to do next ?"), display the state of the system (e.g., "searching for new information...") or show the results of the information processing (e.g., "you need to turn left"). It could also be used to summarize information (e.g., "It will be warm and sunny in France"), provide detailed information (e.g., "It will rain in Paris, but not in Nice"), illustrate processes (e.g., "Enter your name and birthdate to receive your horoscope") or visualize information for easier understanding (e.g., show a picture instead of text). Probably most important, the system should provide feedback to every action a user performs ; otherwise , it becomes difficult for the user to know if and when things are going wrongfor example, when input is missing or incorrect, or the requested task cannot be executed because of external factors, such as low battery or wrong context, or that it cannot accept input at the moment, as it is still performing another task.

3.3.3 Output Techniques

Display

Probably the most commonly used output for computing devices is a visual display. Visual displays range in format, size, and color depth from large-screen displays and projectors to LCDs, mobile phones, goggles, and elevator displays. These displays can show information in various formats and media, such as scrolling texts , windows , animation, sprites (animated element), or fish-eye displays, just to name a few.

One purpose of displays is to visualize information, that is to give the user understanding of some complex phenomena. Visualizing information can either mean that a computer produces the data and gives a representation of it, or that it visualizes external data, such as statistics. Visualization requires mapping between the data and the representation. Through this mapping, the user can control the representation.

To understand representation, consider how a word processor can show different views of the text (outline view, layout view, etc.), while a spreadsheet application can offer alternative views of the data it contains.

Sounds

Sounds have become a very important output device over the past few years . Sound, in the form of synthesized sounds, such as speech, music, and natural sound, are used to convey information to the user. Sound is not restricted to alert messages, but can be used to read e-mail messages, explain data sheets, or recite the user's appointments.

Sound is especially useful when the user has limited visual scanning possibilities. This can happen, for example, in a car. Sound can also be to used to complement the visual interface. It can be used when attention is required on the screen, in alarm situations, or data sonification.

The sound synthesis technology has made some huge progress over the last few years, which makes it possible to use it for the arrangements of digital recordings, creating sentences, phrases, or word segments. Sound can either be created by the concatenation of prerecorded phrases or by synthesis-by-rule (text or phonemes to speech).

Mice, Trackball and Joysticks

For the interaction process (i.e., the control of the cursor and the haptic output), mice, trackballs, and joysticks have additional technology built in that gives feedback about the movements by the user. Force feedback joysticks are probably best known as they are very commonly used with computer games . They are also used with aircraft simulators, for example, to give the crew the exact experience of the airplane.

Printers

Printers have been used long before display, sound, mice, and multimedia output were fashionable. The printed paper provides the users not only with the result of the computation, for example, but also with a haptic object that they can move and use. Although the paperless office was envisioned a few years ago, quite the opposite happened , as people print more than ever before. The main reason is that they can see the results without having to switch on a computer again (which takes time to boot up, etc.).

The printer can display text and images in any combination making it ideal for complex information, as reading on a paper is less tiring than reading on a computer screen. Because most printers are color nowadays, it is possible to print photos in high quality, for example. The only issue we can see with printers is the fact that so far they have been a one-way output street. Changes made on a piece of paper can be re-imported into the computer only with much difficulty (via scanning, for example).

Multimodal

Multimodal output representations are composed of unimodal modalities. A comprehensive set of unimodal modalities covering the media of graphics, acoustics, and haptics are systematically generated at descending levels of abstraction until the result is a practical "designer's toolbox" of unimodal modalities. Multimodal interfaces are more natural, so they are more engaging.

The best-known multimodal output is graphics and sound together. This is often referred to as multimedia. Multimedia is technology driven and is about particular media channels that are available, such as text, graphics, animation, and video. Although different in media, they share the same modality.

So far, much research has been conducted on multimedia, and we see many different applications in place. Unfortunately, not much research has been done on multimodal output outside of multimedia.

The basic idea behind multimodal output interfaces is to take advantage of natural human communication skills. As mentioned before for the input devices, this means that human behavior is almost always multimodal and that having systems behave like this makes it easier for humans to use them. As with the input, the output can be either simultaneous or not. This means that the modalities are fused (used as one), used in sequence, or used independently.

The current interfaces fall far short of what the human can do, adding extra complexity to the human-computer interaction. With multimodal systems, much higher bandwidth is possible, as different modalities excel at different tasks and different modalities may fit better to different user groups and timings. Through multimodal interfaces, where multiple modes are used to output the same information, errors and awkward behavior can be reduced dramatically.

Multimodal output leverages the human senses and perceptual system, because humans can perceive and correlate multiple things at once by focusing on a certain subject. Appliances need to find a way to filter information correlated to a certain subject otherwise they won't be able to process the information in a timely manner, as they would need to analyse every piece of information available.

3.3.4 Output Issues

Although output devices have existed since the beginning of computing, they still are not perfect. If we look at display technologies, we can find several issues related to the screen layout, such as focus, clutter, and visual logic, which make it difficult for the user to follow the instructions or get the information presented on screen. Another issue known from the Web is that different screens have various resolutions and color depths. Many applications cannot adjust themselves to these variables , making it hard to display the information in a useful way on screen.

Another issue that has appeared on the Web widely is the problem of interpreting information through application logic. Without digging into browser wars too deeply, we just want to mention that the display of Web pages in Internet Explorer and Netscape, for example, can vary a lot, to the extent that some elements are not visible in one browser or the other. In the me-centric world, in which we will have many different displays and display technologies, it is of utmost importance to provide the information in a device-independent format that adheres to some standard like XML to ensure the correct display of information.

With sounds, we have similar problems as with the sound input devices. But now the problem is with the human ear and not necessarily with the device. In noisy environments, you may have trouble understanding what the device is telling you, or you may misinterpret the information you are hearing. Therefore, it is very important that you reconfirm the instructions in critical situations before applying them. On the other hand, devices should be aware of their surroundings and adjust themselves automatically to make sure that the user is able to understand what the device is saying.

Haptic output by joysticks, mice, and trackballs is good for the experienced , but can provide problems for the novice, meaning that new users may not know how to react to haptic output, as they are not used to this kind of feedback from electronic devices. Once they understand that haptic output works the same as for other devices, they should be able to use it quite well.

Printing has always had some issues with the paper throughput that seems to get slowly better (e.g., paper jams). Today, technology in printing is so good that you can even print high-quality output on low-price printers (which are slower and more costly to run, typically). Timing may be a problem with printouts, as a printer may be shared by several persons through the network. If you need to print something urgently and someone else is printing 5,000 pages of a document, you will have to wait until the other person finishes.

Multimodal output provides a combination of display and sound output, which normally makes it easier for the user to understand, as it is more like the real world. One problem of multimodal output can be the wrong combination of display and sound, which may confuse the user. This could be because of lags in output or because the output was not properly coordinated. Apart from these issues, all the issues that may arise from display or sound output can be a problem in multimodal output as well.

3.3.5 Dialogue Techniques

To make interfaces successful, it is necessary to implement dialogue techniques properly. The basic software architecture and techniques for interacting with humans can be specified as follows . First, it is important to specify the dialogue interaction techniques that should be implemented for a specific interface. This includes techniques like alphanumeric techniques, form filling, menu selection, icons and direct manipulation, generic functions, and natural language. Also important for a good human-computer dialogue is navigation and orientation in dialogues and the associated error management.

There are problems associated with this type of dialogue. Humans expect a real-time response, as they would from speaking to other people; therefore, it is important for an interface to give immediate feedback, even if it is only acknowledging that the input was accepted by the system. In automated systems such as agents , it is important to give feedback from time to time to make sure that the user understands why there is no result yet from the system.

Also important is the dialogue genre selection, which describes the conceptual uses to which the technical means are put. Such concepts arise in any media discipline (e.g., film, graphic design, etc.). For your interfaces, you can use interaction metaphors such as the tool metaphor, which is used quite commonly today, or the agent metaphor, which will hopefully replace many tools today. Besides the interaction metaphor, it is also possible to think about a content metaphor with elements, such as a desktop metaphor or paper document metaphor, that make it easier for the user to understand the meaning of an object. This also includes workspace models and transition management techniques.



Radical Simplicity. Transforming Computers Into Me-centric Appliances
Radical Simplicity: Transforming Computers Into Me-centric Appliances (Hewlett-Packard Press Strategic Books)
ISBN: 0131002910
EAN: 2147483647
Year: 2002
Pages: 88

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net