[This post was originally published on Cohost.]

The Australian Broadcasting Company recently ran a TV episode (AD-enabled) and accompanying news article featuring Mick Curran and Jamie Teh, creators of the NVDA screen-reading software for blind and visually-impaired users of Microsoft Windows.

The “Australian Story” episode tells the human side of the story, but as someone involved with Mick and Jamie at a key point in their journey, I thought it might be of interest to at least a few people to know more of what’s behind it.

Briefly, ever since personal computers became popular there’s been a problem of how to make them accessible to people who are blind or otherwise visually impaired—a problem that became especially acute with the advent of window-based UIs like that of the Apple Macintosh or Microsoft Windows. The primary solution proposed was screen-reading software, which did exactly what its name implies: taking text displayed on the computer screen and reading it to the user in a synthesized voice.

Leaving aside the technical issues with screen-reading software products (which in the early days were essentially hacks that hooked into video driver modules), they had a fatal flaw from an economic perspective: more than most software, they took a lot of work to write and (especially) keep up to date with OS and application changes. However, that development cost had to be amortized over a relatively small target user base.

The result was that commercial screen-reading software was (and is) very expensive—so expensive that its cost is typically subsidized in one way or another. This leaves out the vast majority of people in the world who need such software but who can’t afford it themselves, and don’t have a government or an insurer that will help with the cost.

As we entered the twentieth century, multiple trends impacted the small and somewhat isolated industry creating assistive technologies, including most notably the growth of the Internet and the World Wide Web and the increasing population of developers creating free and open source software. After working for a few years as a volunteer in the Mozilla project, in 2005 I went to work for the “new” Mozilla Foundation, after the bulk of the organization was split off to form the for-profit Mozilla Corporation, leaving behind a non-profit rump.¹ One of the first things brought to my attention (by Aaron Leventhal of the accessibility team at IBM) was the opportunity for the Mozilla Foundation to make a significant contribution to the future of accessibility technology, with a focus on open standards and free and open source software.

As a result of conversations with and suggestions from Aaron, the Mozilla Foundation started making some small grants in the area of web accessibility, and went public with that support in March 2006 at one of the premier conferences on assistive technology. That was a pretty successful debut for our efforts, and led to me turning my informal thoughts on what the Mozilla Foundation could do for web accessibility into a more coherent proposal.

Throughout this period we were making some relatively small grants to students and others working on various aspects of web and open source accessibility; for more details, see my reports on Mozilla Foundation grant activities for 2006 and 2007. We also had another successful appearance at the 2007 CSUN accessibility conference.

In the summer of 2007 we made an initial grant to NV Access, the newly formed nonprofit organization created to further develop the NVDA screen reader. We followed up in early 2008 with a larger grant to enable Jamie Teh to work full-time on NVDA.

I can’t remember exactly, but I think that may have been the last major accessibility work funded by the Mozilla Foundation itself. During this time the Mozilla Corporation was putting together its own accessibility team, many of whom were originally funded by the Foundation, and began funding other accessibility work as well. At the end of 2008 Microsoft also made a large grant to NV Access to allow Mick Curran to join Jamie in working on NVDA full-time.

As the last post notes, the Mozilla Foundation can thus best be thought of as providing “seed funding” for NVDA and related accessibility development. It was an investment that paid off handsomely: here we are fifteen years later and NVDA is the number two screenreader in the world, supports a host of different languages, and is used by hundreds of thousands of people, many of whom would never have been able to access the Internet or use PC applications had NVDA not been created.²

So, if you have the time, please consider watching the show about Mick’s and Jamie’s lives and work. If you don’t have the time, but do have the money, please consider donating to NV Access. (Note that the suggested amounts are in Australian dollars, which are currently worth about two-thirds of US dollars.) And if you have neither but are still on Twitter, please consider sharing some of their tweets. Thanks!

This is very much a work in progress, and I’ll be revising this over time to reflect changed circumstances and priorities. I’ve reviewed the document with a number of people involved in Mozilla accessibility efforts, and would be happy to consider further revisions based on public comments.

Finally, note that this document supplements and (to some extent) supercedes my previous posts from 2006 and 2007 on making choice and innovation accessible to all and a proposed vision and strategy for Mozilla accessibility.

David Baron - 2009-04-20 04:33

The document currently says: “Making applications accessible should be a core task of Internet and web developers, as important as improving performance and usability in general.” I think it is also worth emphasizing that the technologies of the Web should be designed so this is as little work as possible. When standard semantic markup can provide the information needed for accessibility, they should be used by assistive technology, rather than making authors have to write something else that means the same thing. The goal should be to make as much information as possible accessible to as many people as possible, not to criticize or demonize authors for their prioritization decisions.

Last week in the “Accessible” module, April 20, 2009 « Marco’s accessibility blog - 2009-04-20 09:12

[…] members, has worked on a new high-level accessibility strategy document. Frank Hecker has a blog post explaining this in greater […]

Scott Rippon (Scott.Rippon@gmail.com) - 2009-04-21 03:52

How does Firefox rate against the W3C User Agent Accessibility Guidelines 1.0 (http://www.w3.org/WAI/intro/uaag.html)? Should compliance with these guidelines be in the strategy? Is Mozilla participating in the development of the UAAG 2.0?

hecker - 2009-04-21 13:04

Scott: This is better addressed in the mozilla.dev.accessibility newsgroup (and I’ll forward your comment there). As noted, the strategy is a living document, and if there’s a consensus that UUAG compliance should be called out in the strategy then we can change it accordingly.

I’m happy to pass on the news that NV Access has now received financial support from Microsoft sufficient to allow Mick Curran (the original developer of NVDA) to also work full-time on NVDA. From my point of view this is a significant development for both NV Access itself, which has taken another step towards sustaining itself and the NVDA project for the long term, and for blind users of Windows.

Users of both OS X and Linux-based systems have available very capable screen readers as part of the base operating system, VoiceOver and Orca respectively. With NVDA Windows users also have a no-cost alternative to expensive proprietary screen readers (albeit one not bundled into the OS itself). This supports the general goal of providing a base level of no-cost high-quality assistive technology in all PCs and PC-like devices. (Accessibility for mobile phones and other mobile devices is another story, and one for another day.)

This grant also highlights an important aspect of the Mozilla Foundation’s grant program: I think we are best seen as providing the equivalent of “seed funding” for worthy organizations and individuals. For example, a number of the people we’ve funded for (relatively small) accessibility and other grants have gone on to full-time employment in their respective fields. In this case Microsoft’s funding of NVDA can be seen as a validation of our original “investment” in NV Access and the NVDA project. I hope to see other Mozilla-funded organizations graduate to the next level as well. (In this respect, see Gen Kanai’s post highlighting some favorable publicity for Project:Possibility and its founder Chris Leung.)

Steve Lee - 2008-12-19 12:35

> This grant also highlights an important aspect of the Mozilla Foundation’s grant program: I think we are best seen as providing the equivalent of “seed funding” Frank, I agree that this has been such an important aspect of the grant programme and a realisation of the Mozilla Manifesto. More I think your funding has been critical in raising awareness and growing the ecosystem in other less tangible ways, not least due to the community around MoCo accessibility. Personally I am grateful as the seed funding you provided enabled me to give up my day job and concentrate on Open Accessibility. As a freelancer I have been involved in many projects and activities that I would not otherwise have been able to (like Maavis, straight-street, Dkey, Project:Possibility and even oatsoft.). While Jambu may not yet be as polished as we would like, the ideas, lessons, and contacts made have enabled me to contribute to the wider community in many ways. I hope to figure out how best to include Jambu in this wider picture. So a big “thank you” from someone else who has been able to contribute as a result of your grants. Next looks like being an exciting one for Open Accessibility, in no small part due to ground preparation work you have done. A merry Christmas to you and all at MoFo and MoCo.

Projects for the week

Here’s a summary of what Foundation-related activities went on last week; for more information about others’ activities please see the weekly status reports published by David Boswell, Gerv Markham, and Zak Greant.

• Grants and related expenditures
  • The GNOME Foundation announced the GNOME outreach program for accessibility, which the Mozilla Foundation helped fund.
• CAs and related issues
  • I gave preliminary approval to the application from VeriSign to add a new EV-capable root (bug 402947).
  • I also worked on EV-related applications from Comodo (bug 401587), GeoTrust (bug 407168), and thawte (bug 407163).
• Public communication
  • David continued work on various www.mozilla.org-related tasks.
• Conferences and events
  • Zak spoke at a number of Norwegian universities February 25-27 on open source and Mozilla-related topics.
  • Gerv attended FOSDEM.
  • I continued work on arrangements for CSUN.
• Other
  • Gerv worked on a script to assist with the proposed Bugzilla product/component reorganization.

Upcoming events and activities

• I’m planning to be in Mountain View on March 12 and at the CSUN conference on March 13 in Los Angeles CA. I’m also planning to attend the Freedom to Connect conference March 31 and April 1, 2008 in Silver Spring MD, and (if possible) the Berkman@10 conference May 15-16 in Boston MA.
• Zak will be speaking at the PHP Quebec Conference March 12-14, and other events later in 2008.

Projects for the week

Here’s a summary of what Foundation-related activities went on last week; for more information about others’ activities please see the weekly status reports published by David Boswell, Gerv Markham, and Zak Greant.

• Grants and related expenditures
  • I’m happy to announce a major new grant from the Mozilla Foundation to NV Access to support development of the NVDA open source screen reader for Windows; the NVDA developers blog has all the details.
  • Scott Haeger completed the final milestone on his Foundation-funded project to improve rich document browsing in the Orca screen reader for Linux. For more information see the GNOME wiki pages for Firefox 3.0 ARIA widget support in Orca, Firefox 3.0 Live Region support in Orca, and integrating the Collection interface into Orca.
  • Codethink Ltd completed its Foundation-funded study on the feasibility of migrating the AT-SPI accessibility API to use D-Bus.
  • I got a grant agreement in place for another major new grant to be announced later.
• CAs and related issues
  • I reviewed the applications from KISA and WISeKey, tentatively approved them, and started the two-week public comment periods prior to final approval. For more information see bugs 335197 and 371362 respectively.
• Other
  • Gerv is recovering well from his recent surgery.
  • David moved out to California (not without some excitement).
  • I completed my migration to Google Apps, as discussed in my last status report. (However I’ve postponed my blog migration for a week or two until I have more spare time.)

Upcoming trips and events

• I’ll be attending (at least part of) the CSUN conference March 10-15 in Los Angeles CA, and the Freedom to Connect conference March 31 and April 1 in Silver Spring MD.
• Zak will be speaking at the Southern California Linux Expo February 8-10 in Los Angeles, the PHP Quebec Conference March 12-14 in Montreal, and other events later in 2008.

A reminder: We’re still accepting donations to the Mozilla Foundation through our directed giving program. Help support Bugzilla, Camino, SeaMonkey, and the Mozilla accessibility project, as well as general Foundation activities for the coming year.

Projects for the week

Here’s a summary of what Foundation-related activities went on in the last three weeks; for more information about others’ activities please see the weekly status reports published by David Boswell, Gerv Markham, and Zak Greant.

• Grants and related expenditures
  • A number of significant grant proposals were approved in December; I’ll make public announcements on these soon and will update the 2007 grants report to reflect all 2007 activities.
  • Eitan Isaacson completed work on a Mozilla Foundation-funded project to improve performance of the Orca open source screen reader; for more information see GNOME bugs 446277, 495077, 502924, and 491756, as well as Marco Zehe’s report on the overall improvements to Orca performance with Firefox 3. Eitan is now working on a new Foundation-funded project to enhance the Braille output of Orca, including adding support for contracted Braille.
• Licensing and legal
  • I looked over the proposed replacement for the CVS contributor form to determine what’s needed to complete the project.
• CAs and related issues
  • I went over the application for inclusion from TÜRKTRUST and all the public comments prior to making a final decision on approval.
• Web sites
  • David Boswell worked more on proposed changes to www.mozilla.org.
• Other
  • Gerv had surgery and is now back home recovering.
  • Zak worked on Mozilla Foundation participation in various upcoming events.

Upcoming trips and events

• I’m planning to attend the Freedom to Connect conference March 31 and April 1, 2008 in Silver Spring MD.

Projects for the week

Here’s a summary of what Foundation-related activities went on last week; for more information about others’ activities please see the weekly status reports published by David Boswell, Gerv Markham, and Zak Greant.

• Grants and related expenditures
  • Ariel Rios completed his Foundation-funded project to add the Accessibility::Collection interface to the AT-SPI accessibility API for GNOME; for more information see GNOME bug 326516.
  • Patrick Michaud produced a progress report on the Perl 6 work jointly funded by the Mozilla Foundation and the Perl Foundation.
  • Zak and I worked on some new grant proposals.
• Licensing and legal
  • Gerv worked on revising and updating the Mozilla license policy, to bring the document in line with current practice.
• CAs and related issues
  • I approved the application for inclusion from SwissSign and filed bug 407396 to have the SwissSign CA certificates added to NSS.
• Web sites
  • David Boswell worked on various changes to www.mozilla.org, including a new Mozilla Foundation home page and a new “Featured Projects” page (bug 393447).
• Miscellaneous
  • We have a new #foundation IRC channel on irc.mozilla.org for anyone wishing to contact us real-time.

Upcoming trips and events

• I’m planning to be in Mountain View on December 11 (afternoon only) and 12.

A final note: We have just over two weeks left in 2007 for you to donate to the Mozilla Foundation through our directed giving program and support Bugzilla, Camino, SeaMonkey, and the Mozilla accessibility project.

Projects for the week

Here’s a summary of what Foundation-related activities went on last week; for more information about others’ activities please see the weekly status reports published by David Boswell, Gerv Markham, and Zak Greant.

• Grants and related expenditures
  • Steve Lee provided a report on his participation in the RAatE conference. (Steve’s Jambu project has received funding from the Foundation.)
  • I received some new grant proposals for the board’s consideration.
• CAs and related issues
  • I worked on the applications for inclusion from SwissSign and TÜRKTRUST, as well as on the EV applications from VeriSign and QuoVadis.
  • I made updates to the pending list to reflect the current status of CA applications.
  • I created a wiki page to track recommended practices relating to CA applications.
• Web sites
  • David Boswell posted a proposed new Mozilla Foundation page mockup for comment.

Upcoming trips and events

• I’m planning to be in Mountain View on December 11 (afternoon only) and 12.

Projects for the week

Here’s a summary of what Foundation-related activities went on last week; for more information about others’ activities please see the weekly status reports published by David Boswell, Gerv Markham, and Zak Greant.

• Legal/IP issues
  • I’ve been responding to queries sent to licensing@mozilla.org.
  • I’m still waiting on more feedback on the proposed corporate contributors agreement.
• Grants and related activities
  • Steve Lee posted a report on the current status of the Jambu project being funded by the Foundation.
• CAs and related issues
  • I’m taking over CA-related tasks from Gerv Markham, and am now the default assignee on CA-related bugs.
  • I filed bug 399214 to start work on updating the Mozilla CA Certificate Policy to address Extended Validation (EV) certificates. (See also the discussion in m.d.t.crypto.)
• Other
  • Zak Greant attended and represented the Foundation at the OpenMind 2007 and MindTrek 2007 conferences.
  • David Boswell attended the Legal Summit for Software Freedom.

Upcoming activities

• I’ll be in Washington DC Friday evening October 19 to attend a G3Ict event.
• I’ll attending the Seneca College Free Software and Open Source Symposium (FSOSS 2007) in Toronto on October 25 and 26 (although due to my flight schedule I’ll miss most of the talks on the 25th).

Random notes

Some of you may recall the Mozilla Digital Memory Bank project being run by the Center for History and New Media (CHNM) at George Mason University; among other things they’ve published a series of interviews with various members of the Mozilla project. More recently the Firefox extension Zotero created by CHNM was included in the Firefox Campus Edition.

When the Mozilla Digital Memory Bank project was just starting I had the pleasure of meeting with a number of people at CHNM, including its founder and director, Roy Rosenzweig. I was therefore saddened to read this weekend in the Washington Post that Dr. Rosenzweig had died last Thursday. Please join me in expressing condolences to Dr. Rosenzweig’s family and the staff at CHNM.

Projects for the week

Here’s a summary of what Foundation-related activities went on last week; for more information about others’ activities please see the weekly status reports published by David Boswell, Gerv Markham, and Zak Greant.

• Legal/IP issues
  • I’m still waiting on more feedback on the proposed corporate contributors agreement.
• Grants and related activities
  • Alexander Surkov completed the second milestone on his implementation in Mozilla of the IAccessible2 accessibility API for Windows (to be used by future versions of Firefox and Firefox-compatible screen readers on Windows). The IA2ToGecko project page has more information, including links to the relevant bugs.
  • Ariel Rios completed the second milestone on his project to implement a new Collection interface for the GNOME AT-SPI accessibility API (to be used by future versions of Firefox and Firefox-compatible screen readers on Linux). For more information (and to track progress) see GNOME bug 326516.
  • Steve Lee completed the final milestone for his project to create a prototype switch-activated on-screen keyboard for Firefox; he’ll be showing this at the Mozilla accessibility summit.
• Web sites
  • David continued work on getting old www.mozilla.org content archived.

Upcoming activities

• I’m considering attending the Mozilla accessibility summit on October 5 in Boston.
• I’m also considering attending one or both days of the Seneca College Free Software and Open Source Symposium (FSOSS 2007) in Toronto on October 25 and 26.

Note that the ideas expressed in this post are my own and do not necessarily represent the official position of the Mozilla Foundation. It’s up to the Foundation board and to the new executive director to decide whether and how to implement this proposed strategy. Also note that my comments are primarily concerning the Mozilla Foundation; however the Mozilla Corporation has also taken some great steps to support accessibility (for example, hiring Tim Keenan to work full-time on accessibility QA and evangelism) and no doubt will do more in future.

Accessibility vision

The Mozilla Manifesto has as its second principle “The Internet is a global public resource that must remain open and accessible.” “Accessible” has many meanings, but in this context we interpret it to mean “accessible to people with disabilities.” This may seem a significant narrowing of the principle, but in fact both “disability” and “accessibility” can be interpreted more broadly. As David Bolter of the University of Toronto Assistive Technology Resource Centre (a Mozilla Foundation grantee) points out, disabilities can simply be viewed as mismatches between particular individuals and their environment, including the software they use.

In this sense accessibility is simply an aspect of software usability in general, and thus extends beyond the stereotypical accessibility scenario (e.g., a blind person using a screen reader) to the more general problem of matching user interfaces to the abilities of the people using them. This significantly expands the population to which the promise of the Mozilla Manifesto applies: not just blind and low-vision users, not just people with physical mobility or dexterity problems, but also people with dyslexia and other learning disabilities, people with minimal literacy skills, and so on.

The proposed “accessibility vision” for Mozilla is thus to help make the Internet an accessible resource in the sense used above, and to do so in a way consistent with the other principles of the Manifesto, including those addressing the importance of open source software, innovation, interoperability, decentralized participation, and making it possible for individuals to shape their own experiences.

Accessibility strategy

The proposed Mozilla accessibility strategy encompasses the following three elements:

• making Firefox a showcase for how to address Internet accessibility issues
• building a thriving community of developers and other contributors interested in Firefox and Mozilla accessibility issues
• working with others to extend the Firefox and Mozilla accessibility success story across all major OS/desktop platforms, including in particular open source platforms

These elements can be further expanded as discussed below.

Firefox accessibility

Over the past few years Firefox has advanced considerably with regard to its accessibility, both through changes to Firefox itself and changes to assistive technology products like the Jaws and Window-Eyes screen readers. There are still a few remaining problems here and there, and we’ll continue to work with the screen reader vendors and others to resolve such issues as they arise.

However the overall goal is not just to play catch up to Internet Explorer in terms of accessibility and compatibility with existing AT products, but rather to leapfrog IE in terms of advanced accessibility functionality when and where possible. The proposed initiatives in enhancing Firefox accessibility are:

• Enhancement of Firefox to support the WAI ARIA set of standards for accessible rich Internet applications. These standards are key to removing the accessibility barriers posed by use of AJAX-based techniques for enabling dynamic web-based content and applications.
• Implementation of the IAccessible2 accessibility API. IAccessible2 is key to moving Windows-based assistive technnology (AT) products such as screen readers onto a modern API that extends and complements the legacy MSAA API while at the same time being modeled on nonproprietary Unix/Linux accessibility APIs (including the OpenOffice.org API).
• Creating test cases, testing tools, and related QA infrastructure for the above projects.

Building a community of accessibility developers

Originally the Mozilla accessibility “community” consisted of essentially one person, Aaron Leventhal of IBM. We need to attract new developers and other contributors to the task of improving Mozilla and Firefox accessibility; we also should make a special effort to attract developers and other contributors who themselves have disabilities, in line with the goal of helping people “scratch their own itch.”

Proposed initiatives in support of this goal include:

• Using small “mini-grants” to encourage existing Mozilla developers to try their hand at accessibility-related projects.
• Encouraging Mozilla accessibility developers to interact more deeply with each other and with the broader community of accessibility developer, through sponsorship of travel to accessibility-related meetings and conferences.
• Improving the accessibility of key Mozilla project tools and infrastructure components to developers with disabilities, enabling them to participate in an equal basis in Mozilla project activities.
• Creating tools, documentation, and related resources specifically intended to help developers concerned with Firefox and Mozilla accessibility issues.

Extending accessibility across platforms

Historically assistive technology products such as screen readers have focused almost exclusively on Microsoft Windows and on major Windows-based applications such as Microsoft Office. Combined with a high learning curve for users of AT products, this has resulted in an AT market and user base that is dominated by proprietary software to a degree significantly greater than other areas of IT. This dominance also means that there has been relatively little success for open source AT software and not much of a community of open source accessibility developers.

Our goal is to break this dependence and to make open source software and development processes as important in the AT market as they are becoming in the wider IT market. Key initiatives in support of this goal include:

• Enhancement of the Linux and Mac versions of Firefox to support better accessibility on those platforms.
• Support for harmonization of accessibility APIs across platforms (e.g., harmonizing IAccessible2 for Windows with ATK/AT-SPI for Linux).
• Support for creation of open source alternatives to existing proprietary screen readers and other AT products; examples include NonVisual Desktop Access (NVDA) for Windows and ORCA for Linux/Unix.

Conclusion

The Mozilla project is well-positioned to play a leading role in Internet accessibility. It has one of the most popular and accessible Internet-related products in Firefox, it has a thriving community of people developing Firefox and Mozilla code and related add-ons, and it has the financial resources to make an ongoing long-term investment in the future of open source accessibility. My comments above are meant as a starting point for further discussion as to how Mozilla can best support the cause of making the web more accessible to all.

Why accessibility is important for Mozilla

Since I joined the Mozilla Foundation last year I’ve spent a fair amount of time working with people involved in Mozilla-related accessibility issues. Why do I consider accessibility important enough for the Mozilla Foundation to spend time and money on it? I’ve been thinking about this question and I’ve concluded that there are lots of reasons, all good but some more compelling than others.

I’ve often heard the following reasons given for promoting accessibility; I think they’re good reasons, but in various ways they leave something to be desired (as I note):

• Accessibility is important to comply with government and corporate standards. My initial encounter with accessibility issues occurred when I was working in Netscape’s government sales group and got called in to a government agency to hear complaints that Netscape Communicator was not very accessible to blind users and other users with disabilities, and in particular was not compliant with the US government’s Section 508 requirements.

  That sort of experience is one way people think of accessibility: At best it’s an annoyance you need to worry about if you’re doing government business, at worst it’s something you could be sued over. In that respect and others accessibility is not that dissimilar to security: just another cross-product issue that developers have to worry about, but often wish they didn’t.

• Making software accessible helps people with disabilities live more productive and satisfying lives. Many people working in the accessibility field either have disabilities themselves or have known people with disabilities, and use their personal experiences as inspiration for their work on accessibility-related projects. For example, in his interview Aaron Leventhal notes his encounter with a man who was blind and deaf and used a Braille display to assist him in running a non-profit organization. As Aaron put it, “The experience had quite an impact on me… [I]dealism is what brought me into the business of accessibility.”

  However, although inspiration can be very motivating I think it would be a mistake to rely too much on it. Otherwise I think the rest of us (i.e., those not working directly in the accessibility field) run the risk of falling into the “Olympics syndrome”: We watch the games on TV, watch the five-minute videos about John the speed skater or Jane the marathon runner, think “how inspiring!” and then after the games are over we promptly forget about John or Jane, leaving them to live and work in obscurity for the next four years.

• All of us will likely have a disability at some point, so it’s in our own interest to support better accessibility. For example, I’ve always worn glasses, but recently my vision has been deteriorating a bit; thus I’m very glad for the features in Firefox that allow enlarging the text on web pages. Similarly, we never know if we or a member of our families might become permanently blind, paralyzed, etc.

  There’s nothing wrong with an appeal to self-interest (indeed, I’ll repeat this tactic below); the problem is that in this case the self-interest is not necessarily that strong—when we’re young we think we’ll never grow old, when we’re healthy we don’t think about getting sick, and if we don’t have any disabilities then we discount the possibility of ever having any in the future.

Can we come up with more (and perhaps even better) reasons to support accessibility-related work? Here are some that I think are important in the context of the Mozilla project:

• Improving accessibility is key to our mission of promoting choice and innovation on the Internet. In this respect improving accessibility is like supporting multiple platforms and multiple languages: Our mission is not just to promote choice and innovation for people using Microsoft Windows XP, so we develop Mozilla-based products to support three main platforms (Windows, Mac, and Linux) and encourage ports to many others. Our mission is not just to promote choice and innovation for people who speak English, so we support localization of Mozilla-based products into as many languages as possible. Similarly, if we promote choice and innovation only for people who have good vision, who can use a mouse, who can type, and so on, then we won’t be fulfilling our mission. If we’re going to promote choice and innovation then we should promote it for everyone, without exception; otherwise we’re not living up to the ideals we set for ourselves.

• Users with disabilities could be key supporters of the Mozilla project. As I’ve previously discussed , there are several different groups who for one reason or another are part of the Firefox value network and who in their various ways support the Mozilla project. In particular the project depends on the efforts of those who help evangelize the use of Mozilla-based products, test the products and report bugs, create extensions for Firefox and other products, develop new web applications supporting Firefox, contribute to Mozilla code and documentation, and so on. In working to make Firefox and other Mozilla-based products more accessible we have an opportunity to build another group of passionate project supporters and contributors.

• Improving accessibility is key to improving access to information, communications, and applications for everyone. We sometimes tend to think of accessibility issues in a very narrow fashion, for example, as just involving getting products to work with screen readers. But in reality accessibility is a very broad topic; in fact, it’s so broad that in my opinion it’s even misleading to use the term “accessibility,” because that term is stereotypically associated with technology used by people with disabilities. The overall goal is really enabling someone to more easily and quickly find information, communicate with others, and engage in online transactions, in situations where there are constraints on the input and/or output mechanisms that can be employed. Stated that way accessibility benefits everyone: Lots of people might like to have better ways to use the web while driving a car, or hurrying through an airport, or dealing with a dog or a toddler, or doing a host of other things that don’t involve sitting in front of a traditional keyboard and computer display.

Why Mozilla is important for accessibility

Above I’ve discussed why I think accessibility is important for the Mozilla project. Now I’d like to talk about what I think the Mozilla project has to offer to the community of people involved in accessibility issues as users of assistive technologies, as creators of AT products, and so on.

• We can bring more choice and innovation to AT users. There are plenty of people with disabilities who are quite happy to use Internet Explorer, Windows, etc., but there are other people who don’t want to be restricted to a single set of products. We can provide them with more choices, just as we provide those same choices to everyone else.

• We can give AT users more ways to help themselves. Up to now AT products have been created, sold, and supported using the traditional proprietary software model. We can offer AT users not just alternative products but an alternative way of doing things, namely the open source model. This is potentially attractive both to individual users and to organizations advocating on behalf of those users. As Aaron Leventhal put it in his interview,

  Perhaps the most powerful approach is for blindness and other disability-related organizations to work together, and actually hire their own members with the technical skills to go work on the problem. After all, this is open source, so why not make it how you want rather than deal with company X’s idea of accessibility?

  Beyond just “scratching their own itch,” working on open source projects like Mozilla can also be a way for people with disabilities to build their skills and find new job opportunities.

• We can drive new and better accessibility standards for the entire industry. Firefox now has enough market share that we can work with other organizations to develop standards, and then be a key driver for those standard by implementing support for them in our code base and encouraging their implementation in AT products with which we interoperate. DHTML accessibility is just the first example. We’re also having a lot of impact on improving desktop accessibility APIs by explaining how to use them, making them more powerful, and synchronizing them to work in similar ways so that the differences among them are lessened. This will make it much easier for other future cross-platform applications to be accessible. We communicate with the developers of all the various web and desktop accessibility APIs to reduce unnecessary gaps and differences, and in this way and others help bring the whole industry together onto new and better standards, to the benefit of everyone.

• We can support the growth of a new generation of accessibility developers, especially for free and open source software. We can get talented new developers to work on accessibility in Mozilla, and they can then take what they’ve learned to other projects or to other non-accessibility areas of the Mozilla codebase.

• We can help bring the benefits of disruptive innovation to the assistive technology industry. In his interview Aaron has a lot to say about what he sees as the current “state of stagnation” of the AT industry. To be fair, I think that what Aaron sees as stagnation can also be seen as a necessary conservatism caused by the long learning curves for people to become productive with AT products, combined with AT vendors having to continually adapt to whatever changes Microsoft and other software vendors make to their products. In such an environment change is not necessarily always a welcome thing.

  However in the longer run I think that the AT industry will be affected by the same forces for change that are now affecting the software industry at large: the growing popularity of free and open source software (FOSS), the rise of web-based services as an alternative to or even replacement for desktop applications, and the move away from traditional software licensing schemes to alternative business models.

  The Mozilla project is deeply involved in all three of these trends, and as such I think that we can potentially be a force for positive change in the AT industry. To take but one example, in a more web-centric world more and more significant applications will be web-based, and adapting such applications to the user’s particular needs and abilities may be more straightforward in the long run, since the web is based on standard protocols and formats, and the nature of the web means that web applications can be modified and repurposed without necessarily having the cooperation of the application provider. (Greasemonkey is a good example of what’s possible in this domain.)

Where do we go from here?

So, accessibility issues are important to us, and we can play an important role in promoting better accessibility for the Internet and web. How specifically should we proceed? Here’s what the Mozilla Foundation has been doing thus far, and what we could be doing:

• Promote better accessibility for Firefox and other Mozilla-based products. This is our most important priority, has been the primary focus of Aaron’s and others’ work over the years, and is also the primary focus of the accessibility-related grants that the Mozilla Foundation has been making. At present we have a pretty good accessibility story for vanilla Firefox running on Windows; however our story becomes less good when we turn to non-Windows platforms like Linux and OS X, to products other than Firefox (e.g., Thunderbird), and to Firefox extensions and other Mozilla-related add-on products. We need to do more work in all these areas.

  At a platform level, our goal is for Gecko 1.9 to be the first platform that makes it possible to write true cross-platform accessible desktop applications (i.e., using XUL).

• Provide support and advocacy for efforts to make “Web 2.0” applications and services accessible. The Mozilla project has been a pioneer in this area by virtue of Firefox’s support for accessible DHTML. However there’s much more we could be doing, including evangelizing the use of accessible DHTML to web application providers and other browser developers, and helping to support creation of better documentation and tools to help web developers make AJAX-enabled applications more accessible.

  Here our goal is for developers to be able to write powerful web applications with HTML and JavaScript while still offering excellent accessibility. Such applications if written carefully will also have the additional benefit of working well on small devices using browsers like Minimo (since writing accessible web sites is very similar to writing good sites for small devices), as well as providing other benefits such as good search engine optimization.

• Help with other open source accessibility efforts. As we try to improve our own accessibility story on non-Windows platforms, we’ll find that we’re constrained by the accessibility architectures and tools on such platforms. In particular, there’s still a lot of work that needs to be done on Linux-based platforms, e.g., to provide better accessibility APIs, screenreader software, and so on. We should help support such efforts if and where it makes sense.

  The accessibility of open source software is important in particular because of the battle in Massachusetts over the Open Document Format—it’s a serious tipping point for government open source adoption. In the past open source software was considered generally inaccessible, which sent the message that customers need to turn to proprietary software companies that are legally required to make their products accessible; otherwise improvements to accessibility wouldn’t happen. Now there’s a counterexample: When people ask whether products that support ODF will ever be as accessible as Microsoft Office, Firefox is often brought up as an example of a product that can be even better in terms of accessibility.

• Help promote and sponsor other new approaches to better accessibility. As noted above, the move to web-based applications opens up some new opportunities to providing better accessibility, and it may make sense for the Mozilla Foundation to sponsor research and development work in this area. As one example, perhaps some of the work being done to provide better systems for bookmarking and site navigation might be adaptable to improving accessibility for complex web sites. (And vice versa: research into improving the accessibility of complex web sites might provide good ideas for how to improve bookmark systems and browser navigation aids.)

In summary, accessibility is important to the Mozilla Foundation and will continue to be important to us. The project has done a lot thus far, but we’ve only scratched the surface of what we might do in the future.

However I don’t want to give the impression that accessibility will be the only or even just the primary focus of the Mozilla Foundation. It just happens to be a good example of the sorts of things I think the Foundation should be involved in: activities that support the overall mission of the Foundation, complement the core product development activities of the Mozilla Corporation, and are consistent with our nonprofit nature.

For all the reasons discussed above I’ll continue to push to have the Foundation be involved in the accessibility arena, but at the same time I’d like to see us become more active in a wide range of other activities. More on this in future blog posts.

Why were we there?

We attended the CSUN conference basically because Aaron Leventhal suggested it to me several months ago, and I thought it sounded like a good idea. More specifically:

• I thought that accessibility was important, and that promoting accessibility in Mozilla-based products was a good task for the Mozilla Foundation to take on.
• Although we were already going to have indirect representation at the CSUN conference (in the form of the IBM team that’s been working on Firefox accessibility), I thought it was important for the Mozilla project to have an “official” presence to show our commitment to making our products accessible.
• Given the release of Firefox 1.5 with improved accessibility support it was a good time to be at the CSUN conference, as we’d have something concrete to promote.

Our goals for the CSUN conference were as follows:

• Promote Firefox 1.5 and its accessibility support to both end users with disabilities and organizations supporting those users.

• Promote new technologies like accessible DHTML to organizations trying to create web sites that implement advanced features (e.g., using AJAX techniques) while still remaining accessible.

• Initiate dialogues with and get feedback from end users, assistive technology vendors, and other organizations (both for-profit and nonprofit) concerned with accessibility issues.

• Start building a community of developers working on Mozilla-related accessibility features, especially those implemented using Firefox extensions.

As noted below we achieved all these goals and more.

Who was there?

Our success at the CSUN conference was due to contributions from lots of people. Here’s who did what:

• The Mozilla Foundation sponsored a booth in the CSUN exhibit area; we were one of over two hundred CSUN exhibitors spread across two hotels.
• The Mozilla Foundation also sponsored travel expenses for four student volunteers actively working on Firefox accessibility extensions and related topics:
  • Charles Chen (University of Texas). Charles has created CLC-4-TTS and Fire Vox, a JavaScript function library and accompanying Firefox extension to make Firefox a “self-voicing” browser for people and platforms that don’t have general-purpose screen reader software.
  • Hans Hillen (University of Amsterdam). Hans is working on a Firefox extension to assist with site navigation and related functions, building on his previous NavAccess work.
  • Brian Romanowski (University of Illinois at Urbana-Champaign). Brian is part of a team at UIUC creating an accessibility extension for Firefox, for both end users and web developers.
  • Doug Williams (University of North Carolina). Doug is working to improve Firefox support for Braille displays.The students did booth duty and talked to end users, attended conference sessions, and jump-started some collaborative activities around areas of common interest.
• I did booth duty and represented the Mozilla Foundation in discussions with other organizations.
• Mike Beltzner of the Mozilla Corporation did booth duty, represented the Corporation in discussions with other organizations, and was a co-presenter with Aaron Leventhal at one of the conference sessions (see next item).
• Aaron Leventhal of IBM presented a session “Firefox 1.5: Web Browsing Beyond The Status Quo” with Mike Beltzner and Glen Gordon of Freedom Scientific (creators of the popular JAWS screen reader). Aaron and Mark Pilgrim of IBM also helped out at the Mozilla Foundation booth; Aaron introduced Mike, the students, and I to key people at various assistive technology (AT) vendors, while Mark got feedback on accessibility-related issues for Firefox 2.
• Rafael Ebron of the Mozilla Corporation arranged for us to have a good-looking booth display and a nice collection of Mozilla logo items to hand out to attendees. (The Sharpie Mini markers were a particular hit.)

What happened?

In general the CSUN conference was an “over-the-top success” for the Mozilla project (to quote Aaron Leventhal). Here are some of the highlights of the conference as far as we were concerned:

• With Firefox 1.5 the Mozilla project has a good accessibility story; Firefox has now reached near-parity with Internet Explorer in terms of its support by the dominant screen readers (JAWS and Window-Eyes) and screen magnifier software (ZoomText). Both end users and organizations interested in assistive technology are noticing this, with people stopping by the booth to tell us how much they loved Firefox (even JAWS users, who historically have had problems using Firefox). The conference session on Firefox 1.5 was incredibly popular, with a standing room only audience and over a hundred people turned away at the door. (Aaron repeated the presentation later for people who missed it the first time.)
• There’s more and more support for Firefox by vendors of screen reader software, screen magnification software, makers of special-purpose assistive devices, and so on. In some cases Firefox will be included as a core part of their products and devices. (We saw some very interesting Linux-based devices, a break from traditional Windows-centric AT products.)
• As a result of the accessible DHTML work by Aaron and others at IBM and elsewhere, the Mozilla project and its corporate partners are starting to get a reputation as innovators in this space. Both Victor Tsaran of Yahoo! and Becky Gibson of IBM presented very popular CSUN sessions on DHTML/AJAX accessibility; there’s a groundswell of interest this topic, some coming from surprising places. Accessible DHTML is on track to become a very key technology, with more organizations planning to participate in the W3C Dynamic Web Content Accessibility project working in this area. As Aaron put it, “Both Firefox and DHTML accessibility are officially recognized as the future by the community now. . . . [W]e’re leading the industry to support new web standards which will help the entire web be more accessible.”
• Karen McCall, a well-known accessibility consultant and JAWS user, is the author of a new book on using JAWS 7 with Firefox. Karen told us that she prefers Firefox so much that when she trains government employees to browse the web with JAWS she recommends they ask their IT support staff to have Firefox installed in their systems.
• We are reaching out to try to build potentially productive relationships with other nonprofit organizations interested in accessibility issues, including the National Federation of the Blind, Benetech, and WebAIM. These organizations are interested in innovations like accessible DHTML, and more generally in the power of the power of the open source development model to support user-driven innovations in the accessibility space.
• Last but not least, we are starting to build a larger and more vibrant community of Mozilla/Firefox accessibility developers. Having the student volunteers attend the conference and do booth duty was invaluable in terms of promoting future collaboration between them, learning about users’ needs and wants firsthand, and getting more integrated into the overall Mozilla developer community. (In particular, I recall a conversation between Charles Chen, Mike Beltzner, and Mark Pilgrim in which Charles got to find out how Mozilla development really works.)

What happens next

Clearly given the success of our participation in this year’s CSUN conference it’s an easy decision to return next year for the 2007 conference, with a better booth location if possible (for example, next to or near the IBM booth). We’ll almost certainly sponsor volunteers again next year, albeit with some reasonable limit on the number of attendees. We’ve also been discussing a number of other things to improve the experience next year, e.g., having a CD of all software being demonstrated and a single demo laptop with all software installed, having an “accessibility hackfest” among Mozilla attendees, and so on.

That concludes my conference report. Overall I think that the impact of the Mozilla Foundation’s CSUN- and accessibility-related activities has more than justified the investment that the Foundation has made in them thus far, and I’m looking forward to working with the Mozilla Foundation board of directors to extend our efforts in this area.

P.S. Rich Schwerdtfeger of IBM has another report on the CSUN 2006 conference from an open source perspective.

I’ve tried to make this site accessible to as many people as possible; here I describe the accessibility features of this site. (This statement is based on Mark Pilgrim’s accessibility statement.) If you have any questions or comments about the accessibility of this site, feel free to email me at the address found elsewhere on this site.

Access keys

Most browsers support jumping to specific links by typing special key combinations defined on the web site. On Windows, you can press ALT + an access key; on Macintosh, you can press Control + an access key.

The home page and all archives define the following access keys:

• Access key 1: Home page
• Access key 4: Search box
• Access key 9: Feedback
• Access key 0: Accessibility statement

(Note that I didn’t define an access key to skip to the main content because the main content is already the first thing on the page.)

Standards compliance

1. I intend to ensure that all pages are Bobby AAA approved. More on that later as I complete the necessary work.
2. I intend to ensure that all pages are Section 508 compliant. More on that later as I complete the necessary work.
3. The home page and blog archives validate as HTML 4.01 Strict. (Some older pages on the site have not yet been modified to validate properly.)
4. The home page, blog archives, and other pages use structured semantic markup. For example, on pages with more than one entry H2 tags are used for individual post titles, so that JAWS users can skip to the next post with ALT+INSERT+2.

Navigation aids

1. All blog archive pages have rel=home links to aid navigation in text-only browsers and screen readers; I may add rel=previous, next, and up links in the future. (Unfortunately prev and next in particular are not simple to implement in Blosxom, the blogging system I’m using.) Mozilla users can take advantage of this feature by selecting the View menu, Show/Hide, Site Navigation Bar, Show Only As Needed (or Show Always). Opera 7 has similar functionality.
2. The home page and all archive pages include a search box (access key 4).

Links

1. Many links have title attributes which describe the link in greater detail, unless the text of the link already fully describes the target (such as the headline of an article).
2. Whever possible, links are written to make sense out of context. Many browsers (such as JAWS, Home Page Reader, Lynx, and Opera) can extract the list of links on a page and allow the user to browse the list, separately from the page.
3. Link text is never duplicated; two links with the same link text always point to the same address.
4. There are no javascript: pseudo-links. All links can be followed in any browser, even if scripting is turned off.
5. There are no links that open new windows without warning.

Images

1. With one exception (a photo for my biography page) this site does not use images at all.

Visual design

This site and all its archives use cascading style sheets for visual layout.

1. The style sheets for this site do not specify a base font size, and use relative font sizes to specify the appearance of headings and related text. Text on this site should be resizable in any browser that permits text resizing.
2. If your browser or browsing device does not support stylesheets at all, the content of each page is still readable.

References

In creating this site I made use of Mark Pilgrim’s Dive Into Accessibility book and related materials. See the book for a complete list of other references.