"The only way accessibility could be handled is by creating a base framework (from the GTK, or QT or JSwing or any other GUI toolkits) that provides accessiblity "features to the interfaces. This means that the application developers will not need to "worry (too much) about accessibility because the "text control" will contain the features "by default*(and this is the "most-most-important-issue"*
This is exactly what we've done (with GTK+ and, originally, Swing; KDE/Qt are following suit). In reality it's still easy for developers to inadvertantly break stuff, but at least stock widgets do the right thing. However, the other requirement is that somebody (i.e. the assistive technologies) are listening to the accessibility interfaces provided. That's why the toolkits above are more accessible on the FOSS platforms (GNOME, etc.) than on the Windows platform at present - Windows ATs have historically focussed on eavesdropping on proprietary Win32 APIs, out of necessity, and have tended to focus resources on custom support for the apps with the 'largest market share'.
Now that the built-in support is maturing, things are getting better on the FOSS platforms fast, though there's a lot of catching up to do for some user groups.
Bill Haneman
Gnome Accessibility Project
As well as SpeakUp (http://www.linux-speakup.org/)
and EmacsSpeak
(http://emacspeak.sourceforge.net/)
which give voice access to the Linux console and console applications, newer Linux releases include api-based assistive technology support for applications that use gtk+, mozilla, or Java... via an interface called "AT-SPI"
(http://developer.gnome.org/doc/API/2.0/at-spi/ind ex.html)
which very much resembles (but predates) the accessibility APIs used in OSX.
Support for AT-SPI in Qt is slated for a future release of KDE/Qt.
New versions of the Gnome desktop include the built-in gnopernicus screenreader and magnifier, which supports speech, braille, and magnification
(http://www.baum.ro/gnopernicus.html), and the gok suite of dynamic onscreen keyboards too
(http://www.gok.ca./ There's also another free (as in freedom) screenreader available from ftp.gnome.org, called "orca" - it's a less full-featured offering, but it has scripting capabilities that make it interesting to hackers, and it's written in python.
There are also some speech and magnification utilities included with KDE, thanks to the "KDE Accessibility Project", though they are currently more limited in scope. When support for the AT-SPI is available for KDE apps, all the assistive technologies written to this api should interoperate nicely. I believe that there may be a talking version of konqueror already. There are also projects that provide talking plugins for Mozilla.
Since the GUI-based Linux [and Solaris:-)] accessibility technologies are still in their early days, end users are still likely to have a somewhat bumpier ride than users of established screenreaders like JAWS for Windows - but at last blind and low-vision users have significant access to the graphical Linux desktop. In particular, the web browsing experience requires a patched Mozilla for best results - Sun has produced such a version and makes periodic tarballs available.
Provided the distros recognize the value in all this, we can expect improved testing and support in upcoming Linux distributions.
The premise, that there are not many handicapped open-source developers, is not valid. I personally know of many (OK, I'm involved in accessibility work so of course I do). Just as significant are the numbers of open source developers who work on accessibility features for friends or loved ones - true, in the open source world most stuff is written to meet a need or desire of the developer or his/her friends.
To the extent that the 1-in-6 figure is correct (and it is, if you count things like old age, arthritis, and RSI) then that adds up to a lot of developers.
In the context of "Gnome Accessibility" a lot of work is being done for linux and *nix generally. At the moment it requires bits of the Gnome 2 software stack but for best success one would want to interoperate with all major toolkits and desktops.
A couple of people have said linux is a bad forum for this since it's about people scratching thei own itches... but in fact that's probably all the more reason why linux can become the premier platform for accessibility. The disabled community, even if it is 50 million strong:-), has historically been seen as a niche market without the pull to get bugfixes prioritized, and assistive technologies have had difficulty getting proprietary information needed to retrofit accessibility onto closed apps and OS's.
There are lots of disabled programmers out there, and lots of hackers with friends and loved ones who can benefit from accessibility. With an open OS, and even open source assistive technologies (there are two in development under the Gnome umbrella now), the power to improve accessibility really is being delivered to those who most benefit from doing it.
Speaking of pricey or otherwise, we have (finally) announced two free software projects for assistive technologies for linux (using the Gnome framework): an integrated screenreader/magnifier (Gnopernicus) and a set of dynamic onscreen keyboards (GOK) for single switch devices, etc.
Two pieces of info in the magnification arena: Gnome 2 will come with a simple magnification utility that can provide focus tracking, mouse-follow, and fullscreen magnification (if you have a second video card, it may be a few months before fullscreen support for single-framebuffer machines is available).
Secondly, a full-featured screen magnifier which we hope will be on par with expensive commercial offerings is now under development, with an LGPL license, as part of the Gnome Accessibility Project, called "Gnopernicus", and now available from Gnome CVS. It also provides screenreading and braille display support. It is being developed in coordination with a commercial firm with extensive experience in this area.
Actually, there are a lot of open source/free software developers who are interested in accessibility either to 'scratch their own itch' or that of a friend or loved one.
Traditionally, disabled computer users have faced a very frustrating situation - they desperately need/want improved accessibility and certain bugfixes, but there was no 'market justification'.
Add to this the fact that careers in computer technologies are well suited to many disabled people, and you get the result that many disabled programmers have been stymied by their desire to improve their situation but have not been able to because of proprietary software. Open source software (and perhaps even more so, completely free software) takes that barrier away, and for the first time the disabled community no longer must appear as supplicants begging for the fixes or technical documentation required to make software products accessible.
Historically many of the first accessibility solutions have been pioneered and developed by end users, in a situation that parallels much of the history of the free software movement. Now that accessibility is gaining a toehold in the linux GUI world via Gnome Accessibility and the nascent KDE Accessibility projects, I think we will see a sea change in the quality of accessibilty support, notwithstanding the impact of legislature such as the US Americans with Disability Act.
- Bill Haneman, Architect, Gnome Accessibility Project
Slashdot Mirror
Support for AT-SPI in Qt is slated for a future release of KDE/Qt.
New versions of the Gnome desktop include the built-in gnopernicus screenreader and magnifier, which supports speech, braille, and magnification (http://www.baum.ro/gnopernicus.html), and the gok suite of dynamic onscreen keyboards too (http://www.gok.ca./ There's also another free (as in freedom) screenreader available from ftp.gnome.org, called "orca" - it's a less full-featured offering, but it has scripting capabilities that make it interesting to hackers, and it's written in python.
There are also some speech and magnification utilities included with KDE, thanks to the "KDE Accessibility Project", though they are currently more limited in scope. When support for the AT-SPI is available for KDE apps, all the assistive technologies written to this api should interoperate nicely. I believe that there may be a talking version of konqueror already. There are also projects that provide talking plugins for Mozilla.
Since the GUI-based Linux [and Solaris
Provided the distros recognize the value in all this, we can expect improved testing and support in upcoming Linux distributions.
There is a mailing list available for early adopters of this technology: http://mail.gnome.org/mailman/listinfo/gnome-acce
sigh.
The premise, that there are not many handicapped open-source developers, is not valid. I personally know of many (OK, I'm involved in accessibility work so of course I do). Just as significant are the numbers of open source developers who work on accessibility features for friends or loved ones - true, in the open source world most stuff is written to meet a need or desire of the developer or his/her friends.
To the extent that the 1-in-6 figure is correct (and it is, if you count things like old age, arthritis, and RSI) then that adds up to a lot of developers.In the context of "Gnome Accessibility" a lot of work is being done for linux and *nix generally. At the moment it requires bits of the Gnome 2 software stack but for best success one would want to interoperate with all major toolkits and desktops.
A couple of people have said linux is a bad forum for this since it's about people scratching thei own itches... but in fact that's probably all the more reason why linux can become the premier platform for accessibility. The disabled community, even if it is 50 million strong :-), has historically been seen as a niche market without the pull to get bugfixes prioritized, and assistive technologies have had difficulty getting proprietary information needed to retrofit accessibility onto closed apps and OS's.
There are lots of disabled programmers out there, and lots of hackers with friends and loved ones who can benefit from accessibility. With an open OS, and even open source assistive technologies (there are two in development under the Gnome umbrella now), the power to improve accessibility really is being delivered to those who most benefit from doing it.
Bill Haneman, Gnome Accessibility ArchitectSpeaking of pricey or otherwise, we have (finally) announced two free software projects for assistive technologies for linux (using the Gnome framework): an integrated screenreader/magnifier (Gnopernicus) and a set of dynamic onscreen keyboards (GOK) for single switch devices, etc.
Secondly, a full-featured screen magnifier which we hope will be on par with expensive commercial offerings is now under development, with an LGPL license, as part of the Gnome Accessibility Project, called "Gnopernicus", and now available from Gnome CVS. It also provides screenreading and braille display support. It is being developed in coordination with a commercial firm with extensive experience in this area.
this reminds me, I have been too busy adding features lately to update the News page... :-)
- BillActually, there are a lot of open source/free software developers who are interested in accessibility either to 'scratch their own itch' or that of a friend or loved one.
Traditionally, disabled computer users have faced a very frustrating situation - they desperately need/want improved accessibility and certain bugfixes, but there was no 'market justification'.
Add to this the fact that careers in computer technologies are well suited to many disabled people, and you get the result that many disabled programmers have been stymied by their desire to improve their situation but have not been able to because of proprietary software. Open source software (and perhaps even more so, completely free software) takes that barrier away, and for the first time the disabled community no longer must appear as supplicants begging for the fixes or technical documentation required to make software products accessible.
Historically many of the first accessibility solutions have been pioneered and developed by end users, in a situation that parallels much of the history of the free software movement. Now that accessibility is gaining a toehold in the linux GUI world via Gnome Accessibility and the nascent KDE Accessibility projects, I think we will see a sea change in the quality of accessibilty support, notwithstanding the impact of legislature such as the US Americans with Disability Act.
- Bill Haneman, Architect, Gnome Accessibility Project