2002 Conference Proceedings

Go to previous article 
Go to next article 
Return to 2002 Table of Contents


Hans Rasmussen
The Institute for Blind and Partially Sighted
Copenhagen, Denmark
E-mail: hr@ibos.dk

In everyday situations, most blind users of Windows are on their own in a group of sighted persons. Anyone who works with Windows learn from other users by getting tips, seeing how they solve problems or through their explanations of solutions. If a blind person is to cope with this situation, he needs tools to handle it. The goal for our training at the Institute for the Blind and Partially Sighted in Copenhagen is to give the blind users the tools to get the most out of explanations and examples exchanged among colleagues.

What does the blind user need in order to reach this goal?

There are three basic things:
- A good knowledge of what Windows looks like
- A clear understanding of the Windows elements and how to handle them
- A screen reader which offers the necessary information on Windows

What mental image must the blind user have of Windows?

To give the blind user an impression of what Windows looks like, we use a few tactile reliefs covering basic important situations in Windows. The goal is to have the blind user create mental images of the screen, images that enable communication with sighted users. And I mean communication -both ways. Through the understanding of these images and his mental version of them, the blind user must be able to get the most out of sighted peoples explanations and descriptions, but also to understand how things appear on the screen, and to get the relevant information on the contents of a screen by asking questions of sighted users, so that he can correct his mental image in specific situations.

For example:
We have in our material a relief of the traditional open dialog from Word. A rather complex figure with many different items. In most programs you have this dialog, both when you are about to save and when you are about to open an item. But the dialog is often slightly different from program to program. Therefore, the blind user needs to be able to modify his mental image, if he is to be able to have a good communication with a sighted user on a specific open/close box, and he must be aware of the changing situation, when he tries to practice in the box on the basis of an explanation he has heard. This means that not all the items that he knows from the Word open dialog need to be there, they do not need to be there in the same sequence, and they might have been supplied with facilities, which he until now has only met in other dialogs.

What should the blind user know to perform adequately in Windows?

We have built our training around a standard program, which is Microsoft Word. This program has all the standard elements as well as some non-standard ones. Furthermore, this is a program that the user will work with on a daily basis. This gives us a guaranteed practice after the training.

The training consists of a step by step introduction of tasks, starting with the empty desktop, using the start button and menus, finding the application and activating it. Within the application, we practice activating menus, dialogs, elements in the dialogs and so on.

For each activated element we must make several things clear to the blind user:
- How to activate the element (what keystrokes)
- Explain what it looks like (sometimes by use of relief, sometimes by making analogies to known reliefs)
- What does it consist of (one or more basic elements, buttons, edit fields, lists ...)
- Explain the way it appears on the screen (how does it expand, what does it overlap, what happens to the rest of the screen)
- Explain what task it is there to perform (open a document, change font, close the program ...)
- Explain how to operate the element (what to do with the basic element, type in information, mark by spacebar, toggle by arrow keys ...)

All the above must be taken into consideration, but not everything can be handled at once. We normally start with the appearance of the elements on the screen. This is done without the PC, by using the tactile reliefs. Then we practice on the PC, taking the keystrokes in, explaining the actions and eventually going back to the relief or relating to it to get a mental image of the environment that we are moving in.

What information should the screen reader give?

The information you get from you screen reader must correspond to the situation that is the basis for our training. This means that the screen reader must present the information the way it is on the screen. The important thing is that the screen reader gives the blind user the exact picture of what is seen on the screen, because the basis for communication with the sighted user is what is on the screen. Remember that the goal is to give the blind users tools to get the most out of the sighted persons' eyes, even if the sighted user does not know Windows. In many situations we see that the cleverer the sighted user is at Windows, the more difficult it is to get uninterpreted information from his eyes.

The second thing the screen reader must do is to communicate content of the screen to the blind user in a way that makes relevant action possible. This means that information on each window must be available, in most without requiring further action from the user. The blind user must learn how to handle the different Windows elements, what keys to use, what results his keystrokes create, etc.

For example:
When you arrive at the open dialog of word, the screen reader should say: you are in a dialog box, you are standing in an edit field, the prompt is filename.

This information must trigger the following for the user:

Edit field: This is a field where I may type the response to the prompt Dialog box: This is a series of controls which I can reach using the tab key
Open dialog: This is a type of dialog where I can expect lists of elements to choose from, buttons to activate and perhaps buttons to toggle

From this knowledge, the blind user can move safely around the box, making the adequate decisions and actions.

In its basic function the screen reader should be transparent to the application. You should not need to use extra keystrokes to manipulate the application. Extra keystrokes should be available for repeating structural information and concrete text and items on the screen.

With structural information I mean information concerning what Windows element I am focusing on and the characteristics of this element. With other information I mean the elements as they appear on the screen, what the eye see, not what the brain interprets.

Is it really that black and white?

Above, I have described the ideal situation. That is, the situation where you work with a program where all the elements are standard elements from the user's point of view, elements that react on the correct keystrokes, with standard items, dialogs, lists, buttons and so on. We all know that this is only part of the reality for the blind user. But our experience shows that if you can teach the blind user to work with these standard programs - programs that he needs to work with on a daily basis, not training programs, then it becomes easier, when the basic actions are learned, to accept and cope with the non-standard elements that he encounters. We even see an acceptance of presentations on the screen that are completely different from the ones that sighted users encounter.

I think the reason for this acceptance is that the blind user now feels safe in most of his actions in the Windows environment, and therefore he is able to cope with the exceptions as they occur. A similar situation arises when he attempts to access certain information, at this moment mostly information from the Internet. Here, he must accept that this information is frequently presented in a way that makes the earlier one-to-one communication on the screen impossible. In this latter case, the user is persuaded by the fact that this different, and more difficult, presentation provides the best, updated information.

Our experience shows that if one trains and practices with standard Windows applications that work along the standard lines, then one gains the necessary skills for working with programs one has never encountered before, accepting and coping with minor variations from the standards, and even accepting almost a complete break away from the standard presentation on the screen.

Go to previous article 
Go to next article 
Return to 2002 Table of Contents 
Return to Table of Proceedings

Reprinted with author(s) permission. Author(s) retain copyright.