Free software provides opportunities but also challenges for disabled users

MediaLT has tested Norwegian Linux distributions and other free software with different assitive technologies. Is Linux a real alternative for blind users in office work places or in study situations? This does of course depend on the tasks they are carrying out, but as yet free software is not the best option for the average visually impaired user.

Story by: Morten Tollefsen - 04.09.2008

Background

MediaLT received financial support from FAD in spring 2008 to carry out a survey on accessibility of free software.  This project has focused on Norwegian distributions of Linux, and various other applications. Open Office for Microsoft Windows has also been tested with aids for the visually impaired. User testing has been more central to the project than literature search and thorough studies of technical documentation. To achieve optimal results within the project framework we used the following methods:

• User testing according to a specified test plan
• Web search
• Communication with disabled users

The project has been very instructive, but it is important to point out that we have not tested all free software. The most important aspect of the project has been to draw up proposals for future work.

Test plan

There are very many Linux distributions and other free software. We would not of course have been able to test everything. With this in mind the project first prepared a general test plan. The test plan limited testing to the most important  applications for study and work places. We have not given priority to entertainment, for example computer games and media players, not because this is unimportant, but as a necessary delimitation in relation to the project's scope. We have still chosen to study and test some applications that in certain contexts may be considered entertainment: for example DAISY, audio messages, browsers and e-mail. Ubuntu was used for most user tests, both on a desktop and a laptop computer.

Brief summary of the test results

This project has shown that free software, with primary focus on Linux, is an exciting option for people with disabilities. In fact a blind user can use a PC without the purchase of software! The project has also shown that work on the development of expertise, disseminating information, and software development should continue. The most important issues to be considered are listed below.

• Simple mouse and keyboard adjustments. These customizations in Linux are essentially comparable with newer Windows versions and are important for some disabled users
• Aids for the disabled (alternative pointing devices, keyboards). Many of these devices can also be used with Linux.
• DAISY-player. The standard for digital audio books (multimedia books) is DAISY. There is unfortunately no player for Linux. This should be developed. There is probably source that can be used as a starting point.
• Solutions for optical character recognition. Optical character recognition is an important tool for the visually impaired. We have not been able to investigate this in depth, but preliminary findings indicate that there is a need for continued work towards good solutions. In our testing OCR was partially unstable and/or the quality of the characters was deficient.
• Improved screen reader and Norwegian speech for Linux. Orca is the only screen reader for Linux (with Gnome). This is not as good as commercial screen readers for Windows, but for simple tasks such as writing e-mail or surfing simple sites it works more or less well. Norwegian speech is certainly not good enough but this situation can be remedied. New synthetic speech can be developed from scratch, or from earlier projects (for example NST).
• Info-book and training materials. A key measure to ensure that severely visually impaired get started with Linux is the preparation of basic and accessible learning materials (for example in the DAISY format). Learning materials will also be of great help to teachers and others. Using a screen reader is so different to the visual user interface that good education is necessary with respect to software selection, the use of screen reader functionality etc.
• Accessibility for mobility disabled persons. An understanding of this is important, and combined disabilities should also be taken into account. We suggest the development of coaches for programmers to explain how software should be designed.
• Net based competence development using online services.  This will ensure that the disabled can share expertise/experience and can probably be achieved simply by using e-mail lists. Nevertheless, we believe that someone should take responsibility for the operation of such a service in the start-up phase.
• Offers of support aimed specifically at the Disabled. Certainly one of the greatest challenges we have faced in this project is the limited knowledge generally in Norway about screen readers, on-screen keyboards, special pointing devices etc. This lack of knowledge means that end users are unable to solve technical problems.  Since experience with utilities in Linux is also limited, we suggest the setting up of a support system that can be used by individuals, teachers, IT departments etc.
• Inexpensive PCs. In order to increase the interest and expertise around Linux and free software, more disabled users are needed. One way to achieve this would be to distribute free PCs to a test group.
• Localize a free screen reader for Windows. As mentioned initially the main focus has been on software related to Linux. We have however also tested Open Office with screen readers for Windows. This did not work successfully, but we know that work in this area is ongoing. We also tested some free screen readers for Windows. These worked fairly well, and there is clearly a need for at least one of these to be localized (translated into Norwegian).

News archive