- About Us
The GNOME Foundation has turned its attention to accessibility for people with disabilities. To help improve both Web accessibility within GNOME and the project's long-term direction, the Mozilla Foundation is joining the GNOME advisory board, and plans to help improve integration of the XUL development platform with GNOME. Even more significantly, the GNOME Foundation, the Mozilla Foundation, Novell, Google, and Canonical are jointly sponsoring a $50,000 outreach program to help improve accessibility in GNOME.
To an extent, these announcements simply formalize what has already been happening for the last few years. GNOME has included accessibility tools such as the Orca screen reader for several versions now, and has a large and active accessibility community inside the project. Willie Walker, a Sun employer and a lead developer on Orca, who has been working on accessibility for the X Window System for 20 years, describes the current state of GNOME accessibility as "a pretty decent solution," and says it is evolving rapidly, although he acknowledges that the tools do not yet match what is available on Windows.
Similarly, Mozilla has a history of making grants to improve accessibility in free software -- including grants in 2007 to the Dojo Ajax toolkit; the Colorblind Simulator extension to help programmers view their work with a palette that simulates colorblindness; improvements to GNOME's Assistive Technology Service Provider Interface (AT-SPI), and a number of specific improvements to Orca, such as improvement of Braille output, and a number of general improvements to make the program work better with AT-SPI.
In addition, the two projects have been cooperating to ensure accessibility for Firefox 3. Walker says, "Firefox 2 was kind of accessible, but not very good. Firefox 3 was the big push to make a nice, accessible web browser. We had the Mozilla accessibility team and the GNOME accessibility team working very well together."
However, as Christopher Blizzard, an open source evangelist at Mozilla says, the announcements are "both a recognition of the past and a look to the future as well. It's a case of an area in which we have invested in the past, and GNOME wanted to do some work in it."
In addition, Blizzard says, "This is also a recognition that the two organizations care about many of the same things. Mozilla cares very deeply about an open web, making sure standards and information continues to be freely available to all, and not just through proprietary tools and platforms. GNOME cares deeply about those sides of things, and I think there's a real opportunity for us to work together to realize a common vision of what the Internet should be."
One of the changes scheduled to emerge from the new partnership is improved integration of the XUL markup language into GNOME. This change will mean that future improvements in accessibility will be easier to add to Firefox, but also that interaction with Firefox and other XUL-based applications like the Songbird media player will improve for all GNOME users.
Blizzard explains that, while XUL has been in use for several years, it is only now starting to become widely used. He expects XUL's popularity to increase dramatically in the near future, thanks in no small part to the XulRunner runtime environment. Not only will Firefox 3 be the first release of the popular browser to use XulRunner as its back end, but upcoming releases of GNOME-based distros such as Fedora and Ubuntu are expected to install XulRunner by default as well -- changes that could blur the traditional distinction between desktop and Web browser, with programs being written that are not only cross-platform, but capable of running in either desktop or browser, a development that might improve remote access.
Integration will assist and promote such developments. According to Marco Zehe, who was hired in December 2007 as quality assurance lead for accessibility at Mozilla, integration will mean that Firefox and XUL applications that run in it will resemble native GNOME applications in their layout, using GNOME themes and the native GNOME Print dialog, and sharing the same hotkeys for navigation. Some of this work is already visible in the latest betas of Firefox 3, Zehe notes.
Increased collaboration across projects and greater XUL integration are worthwhile in themselves. Unsurprisingly, however, it is in accessibility that these announcements should have their greatest effect.
According to the release, advisers will identify major accessibility tasks and bugs for the program. The program will shortly start accepting applications, and grants will be announced in late 2008. Successful applicants can earn $6000 for completing one of the tasks within six months, or $1000 for fixing five of the designated bugs.
Walker, who has been heavily involved in deciding which tasks and bugs will be tackled by the program, emphasizes that not all outstanding problems will be addressed by the program. Instead, the goal is to resolve the highest priority problems, and to provide a basis for solving less urgent problems.
"All disabilities are important," says Walker, choosing his words with care after some heated mailing list discussions. "We want everyone to have equal access to applications on our machines. With our hearts, we know that and believe that. But what we need to do is focus and target specific audiences and do a good job. Because, if we spread ourselves too thin, we're just going to do a poor job everywhere. If we can shore up the core implementations and the core features that can help other people improve the technology, then we've got a stronger platform."
According to Walker, two of the top priorities will be documentation and testing. Among other things, for end users, these priorities mean improvements in desktop magnification, as well as to the Evince document reader. For developers of individual applications, documentation means instructions of how to include accessibility considerations in their development and testing -- an important consideration, since as Walker says, referring to himself as part of the GNOME accessibility community. "We have dependencies on everything. Anything you see on the screen, we depend upon." Similarly, for release managers, these priorities means the development of a test suite that becomes part of the release cycle for each new version of GNOME.
However, because of the need to focus efforts, some specifics will be left for the future. They include tools for people with cognitive or reading impairments, whether dyslexia or limited vision, such as highlighting, the enlargement of words and phrases, easier navigation, and other aids to assist reading.
Walker is also looking for ways to address problems that are not confined to GNOME, including speech recognition and accessibility on mobile devices. While these issues may not be addressed in the current outreach program, he hopes that future funding may solve them.
Even though the current announcements do not include everything Walker believes should be done, he clearly finds the new emphasis on accessibility extremely gratifying. Looking back on his two decades of accessibility, Walker remembers, "When I first started, we would have to fight for every last piece of accessibility that we could add to anything. It was always a constant negotiation, and we just basically got what people allowed us to put in. You basically had to have a lot of interpersonal working skills in addition to engineering skills. [And] you had to go and dig into people's code and figure out how to make it more sensible."
By contrast, the recent atmosphere represented by the GNOME Foundation's announcements is almost a complete reversal.
"Now what we're seeing with this whole thing is that accessibility is being flagged out as an important thing at the board level," Walker says. "We're also seeing within the GNOME community -- and this has been happening for the past several years -- accessibility is just a fact of life. I'ts part of our normal thinking. We don't have to fight for it, people understand the need for it, and we work better together."