on the state of open-source accessibility
Over the years, great progress has been made in the accessibility of GNOME. Using Orca, I am able to navigate around GNOME applications and Firefox. This has been made possible through a collaboration between engineers at Sun and dedicated volunteers (or perhaps "a dedicated volunteer" would be more apt). Engineers working on Firefox have also spent significant amounts of time ensuring that Gecko exposes the information that access technologies such as Orca need to represent web pages. This collaboration has been a blessing for the community and has allowed people with disabilities to productively use GNOME. On January 29, however, Will Walker, the lead Orca developer, was laid off by Oracle, and so now there are no more paid Orca developers. This was a continuation of what has been an over-all trend over the last couple of years in which companies have pulled funding from open source accessibility projects. I am concerned about the potential implications for the future if some sort of funding structure is not established.
Funding is crucial to most large projects, open source or not. Large open source projects typically either have developers who are sponsored by one or more corporations or have some sort of model for generating income to support developers. Having funding guarantees that someone will be focused on a project and can act as a point person. Will has not only led Orca development but has also been acting as the de-facto GNOME accessibility lead, ensuring that the GNOME stack is accessible as a whole, reaching out to others, and educating developers when necessary about the need to consider accessibility and what they should do to ensure that their applications are accessible. If work on GNOME accessibility were to stop or slow down significantly, then what has been done will atrophy over time. For instance, Gecko's accessibility framework was not very usable prior to 1.9, and, right around the time 1.9 came out of beta, the decision was made to start migrating GNOME towards Webkit, which has needed significant work of its own to become accessible. There are many more changes for GNOME 3.0 which will affect accessibility. Bonobo is being deprecated, which is affecting AT-SPI (the software that allows programs such as Orca to inspect the user's session), gnome-speech, and gnome-mag. Work has been on-going to address each of these--AT-SPI is being ported to dbus, gnome-speech is being replaced with speech-dispatcher, and magnification will be handled in gnome-shell--but funding for speech-dispatcher and the AT-SPI port to dbus are not expected to be continued. Having adequate funding allotted to accessibility would help significantly towards ensuring that GNOME continues to be an accessible platform in the future.
There are any number of reasons that it would be desirable to ensure that GNOME remains accessible. For businesses that sell GNOME-based offerings and wish to be able to sell to the U.S. government, accessibility is a requirement under Sec. 508. It could be a project on which Linux and other Unix distributors could cooperate to their mutual advantage. The accessibility framework is not used just by access technologies but also by on-line testing frameworks such as Strongwind and LDTP. From the perspective of a person with a disability, GNOME accessibility provides the missing pieces that allow us to fully utilize a non-proprietary operating system and have the same freedom of choice enjoyed by those without disabilities. If we are applying for jobs where GNOME is used on the desktop, then we can be confident that the tools we need are available. I hope that this continues to be the case and that accessibility work continues to move forward.