Unidad de Investigación ACCESO Departament de Psicologia Evolutiva i de l'Educació de la Universitat de València.
Formación, investigación y desarrollo sobre tecnología, educación y discapacidad.

Unidad Acceso > Publicaciones > Esta publicación

Presented to ICCHP 2000. International Conference on Computers Helping People with Special needs.        
Karlsruhe (Germany) 17-21 July 2000.            
http://szswww.ira.uka.de/icchp2000/wgs.htm

Analysis of disabled user requirements for
a web browsing scanning selection system

Ferrer, A.; Romero, R.; Alcantud, F.
Research Unit Acceso http://acceso.uv.es

Dpto. Psicología Evolutiva y de la Educación

Universitat de València Estudi General (Spain)

This paper studies the features to consider for the design of an alternative browsing interface for people with severe motor impairments who work with scanning selections sets to browse the web. Finally we present an example of an interface based on Switch Access to Windows which works together with Internet Explorer 4.0 in Spanish version.

1. Introduction

The advancements brought by New Technologies and, more specifically Internet, are meaning the opening of new communication possibilities for people severely handicapped in their motorical abilities. Disabled people that can barely move around, manipulate things or even talk. One of the research lines of our unit Acceso is the promotion of the use of telecommunication networks so that people severely handicapped can have access, despite their limitations, to the world of education, and later on, to a proper job [1].

Despite the existing tools, which may be more or less standard, both in hardware (alternative keyboards and mouses…), software (on-screen keyboard, authoring tools…) and assistive technology in general, people severely disabled in a physical sense still need of new adaptations that make possible or optimise their access to the opportunities offered by Internet. This user group usually needs scanning systems. The switch for selection will depend on the motor possibilities of each particular user. However, the system that the person interacts with through the switch should have a higher degree of standarisation, because it is does not dependent so much on the user but on the tasks to perform and/or the underlying program to which it communicates.

This paper studies the features to consider for the design of an alternative browsing interface for people with severe motor impairments who work with scanning selections system to browse the web.

2. Computer access systems and web browsing

To be able to determine the most effective tools we first need to clarify if the tasks that the user performs in a web browser differ significantly from other computer applications. From the point of view of the interaction between user, access device and software, there is no unique condition implied by the software being a browser. However, when we consider the activity factor which is the browsing action, there are some characteristical and relevant aspects to take into account. Let us see, for example the comparison between the use of a web browser and a word processor. In this last one the main task is to introduce information by typing on the keyboard. However, in a browser the main tasks are the searching and retrieval of information, more than its production. These differences do not mean qualitative changes in the actions demanded to the user (use of keyboard and mouse or equivalent devices), but they do influence quantitatively as they affect the frequency of use of the periferal devices. Hence, according to this example, a word processor implies more keyboard use as information input device while in a browser it is more important the use of the mouse. However, both applications will need from time to time the use of both devices, at least in their standard and more popular interaction mode, as there are also browsers than can be completely controlled using just the keyboard.

Next we must wonder if those needs can be properly addresed using standard assistive technology like keyboard and mouse emulators, or if we should better go for specific adaptations development. The use of the most common browsers implies, like any other Windows software, the use of the keyboard and mouse. We can find in the market some closed systems (Wivik, My-T-Mouse, Gus System tools...), and authoring systems (Ke:nx, Discover, SAW, Hands-Off) [3] [4] that including switch access possibilities allow us to design applications that simulate keyboard and mouse with scanning systems. These applications permit therefore the use of any computer program, including the most frequently used like web browsers. However a usability study perfomed directly with diabled users under different access conditions [2] lead us to conclude that there is no product giving a global solution. We find many useful solutions for the use of the mouse as well as high quality on-screen keyboards with multiple configuration possibilities including word prediction, macros and so on. But there is a lack of global solutions that allow the user to combine efficiently both input devices, with the exception of  products like Words+’s E Z Keys.

To this problem we must add that during browsing the mouse is intensively used. Besides browsers have their own task related actions like move back and forward between pages, refreshing the current page, going to the home page, stopping a download, entering a new URL... These actions are used so frequently that it is important that they are particularly considered in order to speed up the user access.

For these reasons it is necessary to design a browser interface that allow switch scanning system users a more efficient access to web browsing. To accomplish this goal, instead of a closed and rigid application, we chose for its development the authoring tool Switch Access to Windows (SAW) (see http://atschool.eduweb.co.uk/acecent/saw.html ). SAW has been developed by Oxford Ace Centre Advisory Trust.

It must be noted that SAW version for Windows 95/98 is currently (April 2000) being distributed as shareware. It is now in beta phase and for this reason some functions still do not work properly. The technical personnel of Ace Centre Advisory Trust are in the process of developing the final working version for this operating system.

SAW produces an open interface where modifications can be introduced to optimise the scanning application for each disabled user. SAW permits to send keystrokes to each underlying application, in this case an standard web browser, as well as a direct mouse control. It is possible to select between different selection methods, which is highly important for motor impaired users as each one has different characteristics. It is also possible to easily modify the application to work with a different browser version or model. Besides, SAW permit to interact with any Windows program. Therefore the user could finally have, not only a browsing interface, but a set of interfaces to access the whole of possibilities offered by the computer.

3. Considerations for a web browsing scanning interface

Independently of the tool used for the interface construction, an scanning browsing interface for motor impaired users should consider the following:

1-     It must cover the most  used browser functions.

2-     It must permit the input functions that the disabled user can not perform naturally. That is typing on the keyboard and mouse moving and clicking, This can be done connecting the interface with other emulating applications or embedding those functions in the interface itself.

3-     With regard to the function representations:

a)      Choose elements, words or icons, as significative as possible with regard to the function realised.

b)      If possible, add an explanatory element, to facilitate the learning of the system initially. This element should ideally be removable in a later stage to save space in the screen.

4-     Regarding buttons and other scanning elements layout:

a)      The most used functions must be highlighted first, so that they can be selected faster.

b)      The interface must respect clear organization criteria to allow the user a better comprehension and faster mastering of its use.

5-     Interface size should be minimum, but without losing visibility, because it will always be on top of the browser window, unless the window is resized (that is not maximised to avoid this efect). For this reason functions must be included to permit an easy change of the interface position on screen or even to hide and show it with a single switch press.

6-     Regarding alternative input devices (switch), the interface must admit as many options as possible to account for different user characteristics: automatic scanning, manual scanning, one or two switchs operation… Similarly time factors should be adjustable: scanning speed, dwell select option, pressing time to accept as intended selection action, time to ignore continuing press after a selection has been made…

4. Sample application for Internet Explorer 4 using SAW

Following the previous criteria we have developed a prototype based in SAW (see figue 1). Its features are as follows:

Figura 1. Overview of interface developed with SAW

1.- It includes the most used browsing tasks: move between links, previous and next page, move within the page… When an icon is selected through a switch press a combination of keystrokes is sent to the active application that in our case is Microsoft Internet Explorer version 4.0 Spanish. For example, to go to the previous page the following key stroke set is defined using SAW language: “Key <esc>,combine,alt,left ”. This means “press simultaneously Alt and Left arrow keys”, which is a key shortcut used in MS Explorer to do that function.

2.- It emulates keyboard and mouse functioning. SAW includes functions to move the mouse pointer, to alter the length of each movement, to make the movement continuous or discrete as well as to emulate the click, doucle-click and right-click.

3.- It represents the buttons in different ways, by means of letters, icons or a combination of both. Icons are the same ones used in original Explorer interface for the same functions in the toolbar.

4.- The size of the windows is closely adjusted to the number of elements contained. This size could be further reduced by eliminated the pop-up item that provide information about the interface element currently highlighted.

5.- It is open to be configured for use with different access types and time parameters in order to better adapt it to different users.

6.- The interface starts Microsoft Explorer automatically when it starts using a script function in the start setup section.

7.- When an action implies the use of another function next, the row containing that function is highlighted initially after selecting the first option. For example, the button for mouse right click takes the cursor directly to the second row where the up and down arrows are located. Another example is the the button for new URL which opens automatically the selection set for writing (see figure 2) that emulates the keyboard.

8.- There is a selection set for writing that includes in its first position the acronyms and special characters most used to write URLs, as well as a button with Internet Explorer icon to return to the main browsing screen. See figure 2 next.

Figure 2. Selection set for character typing

9.-Some functions related to cursor movement, and other like the tab key (move to next link) are defined with auto-repeatable attribute, so that there is continuous repetition of the function while the switch is pressed, without having to redo all the scanning.

10.- There is a prompt item for the beginner user to familiarise with the interface. This field shows an informative legend about the action that performs the item which is currently highlighted.

11.- Each row of elements has an item that if activated takes the cursor to the first row. This can be used if the user entered in the wrong row or after stopping the selection of an autorepeatable item, as in this last case the scanning continues indefinitely in the current row.

A summary of the functions realised by each item can be found in next table. Most of them make use of the existing key shortcuts defined in Internet Explorer

5. Acknowledgements

This work has been carried out during the project ACCESO25 of the European Social Fund Horizon Initiative with the co-financing of Universitat de València Estudi General.

6. References

[1]     ALCANTUD, F. & FERRER, A. (1999) Ayudas Técnicas para Estudiantes con Discapacidades Físicas y Sensoriales: Las Tecnologías de Ayuda. A chapter in RIVAS, F. & LOPEZ, M. Asesoramiento Vocacional de Estudiantes con Minusvalías Físicas y Sensoriales. Servei de Publicacions Universitat de Valencia Estudi General.

[2]     ALCANTUD, F.; FERRER, A. & ROMERO, R. (1999) Analysis of web pages accessibility and usability for physically disabled users: Case studies. Final report for transnational Horizon project ATIN.

[3]     ANSON, D.K. (1997) Alternative Computer Access: a Guide to Selection. IOS press.

[4]     CUNNINGHAM, C.& COOMBS, N. (1997) Information access and adaptative technology. American Council on Education, Oryx Press.