Asked why he is organizing the Accessible Computing Foundation (ACF), Jonathan Nadeau has a simple answer: “That’s what free software is all about – empowering people who have otherwise been written off.”

Nadeau knows first-hand what he is talking about. Blinded in a car accident as he started high school, he has spent the last two decades using assistive technologies – proprietary at first; free and open source for the last six years.

As Nadeau explains, the problem is not just that people with disabilities have trouble using ordinary software. The problem is that proprietary assistive software is predatory in its pricing, with many programs costing more than US$ 1,000 or more, putting them out of reach of many who who would benefit and who are often on a fixed income or unemployed.

“Most people who live on a fixed income don’t even see that kind of money in a month. And I mean, 80 percent of blind people in the United States are unemployed. So usually, the only way people who have disabilities can get this software is through government assistance or non-government organizations that will purchase a copy of the software for them. It’s really only win-win for the company, because there’s limited funding, and only a small percentage of people can actually get the copy of the assistive technology they need to run a computer.”

Already frustrated by this situation, Nadeau was inspired by a summer internship at the Free Software Foundation to begin the ACF. “Seeing how it functioned as a non-profit, that pushed me into high gear,” he says. “I still talk to a lot of people who work for the FSF, and they’ve helped me to really get the ACF going.”

The foundation has just filed for tax-exempt status in the United States and is in the process of appointing its board of directors. Besides Nadeau, the board so far includes Sina Bahram, who is doing graduate work on how to present information to the sight-impaired, and Bill Cox, a developer and entrepreneur with a long history of involvement in assistive technologies. Like Nadeau, both Bahram and Cox are blind, “which I feel is kind of important,” Nadeau says. “I want people dependent on assistive technology steering the foundation. We all understand first hand how important assistive technology is and what kind of difference it can make.”

For about a year now, the ACF has been working on the Sonar Project, a Linux distribution to support users with impaired sight and motor skills, dyslexics, and paraplegics. Fundraising and outreach in the free software community are also being planned now that the initial paperwork for the Foundation is complete.

Improving Free Software Accessibility

Asked how free software assistive technology measures up against proprietary solutions, Nadeau replies, “Right now, I would definitely say it’s comparable. I came from using the screen reader JAWS when I first switched over to Linux, and, at that time, JAWS definitely had a bit better performance. But now Orca is definitely better than JAWS was when I was using it.”

All the same, countless problems remain. For one thing, the state of assistive technology is changing so quickly that Sonar requires the most recent software possible. “You never want to to be behind, because there’s always advances being made every six months, enhancements being made, or stability being improved.”

That means that Ubuntu is becoming less and less reliable as the basis for Sonar. “Now that it’s using Unity, it doesn’t have all the latest stuff,” Nadeau explains. “Some of the pieces of Gnome are old. I believe that Nautilus the file manager is the Gnome 3.6 version, and the Gnome Control Center is 3.8. The version of Orca is newer – it’s 3.9 – but the package versions are all over the place. It’s not really good for accessibility. I’ve been testing it out, and some parts of Orca and eSpeak [a speech synthesizer] aren’t really getting along well. The voices are all chopped up and at some point Orca will just talk incoherently fast, and you have to restart it.”

Nadeau is considering making the latest Gnome applications available via Launchpad, but Nadeau is seriously considering basing Sonar on Arch Linux, which uses a rolling release system. “There were a few packages that were not in the main Arch repository, and we gave them to the maintainers, and within a day they were in the repo. We’re going to be way more flexible and customizable basing [Sonar] off Arch.”

Another consideration is that most free software assistive technologies are designed for Gnome, which may require too many resources for the older computers that many people with disabilities are using. The ACF is currently working with the LXDE and Xfce desktops, but while progress is being made, the panel and its applets remain inaccessible. Similarly, in Fluxbox, the problem is its menu. Workarounds like defining keyboard shortcuts or starting applications from the command line are possible, but not ideal for inexperienced users.

An even more basic problem is an installer that people with disabilities can use. Debian offers an accessible installer, but Nadeau judges it “Not for the faint of heart. You have to be connected to the Internet, and you have to know how to partition your hard drive. You also need to know what packages you want and things like that.” Even worse, it crashes if Orca is running at the same time. Currently, Sonar is relying on a command-line installer that does not require any selection of packages, but room for improvement is obvious.

Moreover, just because the basic desktop environment is installable does not mean that individual applications are usable. In LibreOffice, for example, Nadeau reports that “when I go to use the spellcheck, it won’t actually tell me what word it’s trying to correct. I can look at the list of words it gives me, but I don’t actually know what word is misspelled. By looking at the list of suggestions, I can usually guess what the misspelled word is, but sometimes I might get it wrong.”

Nadeau has talked with LibreOffice developers about making accessibility bugs a priority and of eventually creating an accessibility framework, but much remains to be done.

And so it goes for application after application. “Developers don’t do these things on purpose,” Nadeau says. “They just don’t take these things into consideration. Assistive technology just isn’t in the forefront of their thinking. I talk to developers and, 99 percent of the time, they say, ‘I never thought of someone using my application that couldn’t see [or] only had one good hand.’ It’s not that they don’t care – it’s just that it never crosses their minds. If you don’t use accessibility software yourself, it’s not at the forefront of your thinking. Yet a lot of applications that are inaccessible can be fixed rather quickly if they just put aside half an hour or an hour.”

Raising Awareness

Besides making practical improvements, Nadeau is also hoping to be able to raise awareness of such issues. “I want to reach out to some of the assistive application community, saying, ‘Hey, which [applications] are we using? Which ones are we having problems with? What can we improve on?’ And I guess we can start reaching out to maintainers, and even starting to go to conferences like GUADEC, and possibly giving a talk explaining why accessibility is important. I mean, we’re not trying to annoy you and make you spend more time on your software, but you can potentially have more people using your software.”

Nadeau continues, “You have to bring awareness because, even in 2013, you get people who are just blown away, asking, ‘How do you even use a computer?’ When I talk about the number of people in the world who have disabilities, they’re just blown away. There’s over a billion people in the world with some kind of disability, 350 million with some kind of vision impairment. I mean, that’s a whole United States. And if we can bring computing to these people, who knows who’s going to write the next great web application? Or the next great mobile operating system? Who knows what potential is locked up in these people who can’t even use a computer right now?”

The questions are rhetorical, but the implications as practical as they could be. Accessibility is a little-known but major issue for computing – and if the ACF is right, free software is exactly the community that can tackle it.