Volume XIV Number 1, March 2014

A Traveller's Notepad: Reflections on a Journey toward Web Accessibility

Cyndi Rowland
WebAIM; National Center on Disability and Access to Education
Utah State University
I like the dreams of the future better than the history of the past.
—Thomas Jefferson

In my roles as the founder and executive director of WebAIM and the technology director for the National Center on Disability and Access to Education (NCDAE), I have walked the journey of web accessibility with many of you for over 15 years. Together we have worked and watched as web accessibility developed and changed with each passing year. Our collective accomplishments are making the dream a reality—to live in a world where digital materials are available to all, in ways both useful and profound. The field has done great things, and together our work is the foundation for future efforts. With this said, there remains much to accomplish.

When I was asked to provide a personal and informal reflection on web accessibility over my time in the field, I felt honored to do so; there are certainly many people who have worked in this field longer than I, and other individuals and groups have certainly had greater impact. The paper that follows provides my reflection on a journey towards web accessibility in the U.S. over the past 15 years. In it, I describe changes I have seen across five key topics: (a) awareness, (b) complexity, (c) standards and laws, (d) harmonization, and (e) supports. I end each with challenges—emerging or sustained—that I believe continue to vex us as we move forward on our path towards full accessibility.

Awareness

Fifteen years ago, web accessibility was just beginning its push for visibility on a broad stage. The World Wide Web Consortium’s Web Accessibility Initiative (WAI) published the Web Content Accessibility Guidelines (WCAG) 1.0 in May of 1999, and for most web developers, it was an entirely new concept to design pages and sites that could be accessed by individuals with disabilities. For many of us, our earliest web accessibility efforts were focused on increasing awareness of the issue in the field with the recognition that nobody was going to change their practices until they saw the need. In the early days of WebAIM, which began its work in the fall of 1999, I recall web developers were quite accepting of this challenge and would often comment, “You know, I just never thought about that before.” This revelation seemed all they needed to make a personal and professional commitment to learning and using the skills required at the time.

I believe three forces combined to create enormous changes in awareness over the past 15 years:

  1. First was the impact of personal stories. Helping developers connect the dots between their practice and the impact it would have on the life or livelihood of another was a powerful force for change. There was no doubt at the time that the web was already becoming a transformative force in society and, if we did nothing, the digital divide would grow, leaving many in the dust. Individuals with disabilities stepped forward to offer their own experiences, sharing them directly or allowing others to use those personal experiences as they worked to raise awareness of the issue. “Web accessibility gives me a professional soul,” was the sort of comment many of us were fortunate to hear from developers in the late 1990s early 2000s.
  2. Second was the impact of litigation. Over time, the sheer number of complaints, and that complaints led to litigation, created its own awareness of web accessibility on a national level. News and professional outlets took front stage in awareness work as they shared these consumer complaints and court decisions. While case law has been slow in coming, one thing has been made clear over these 15 years: discrimination that blocks access to web content violates an individual’s civil rights. For example, in higher education, where WebAIM began its work, most developers now know that individuals who have disabilities need access and that this access is protected under the law.
  3. Third was the impact of our collective efforts across societal sectors. We could not have had an impact on the critical aspect of awareness had accessibility been an issue that was fought only in government, only in education, only in commerce, or only in employment. The visibility of work in one area bled over into another. As a field, we have successfully cross-pollinated visibility from our own sector to that of others. Over time this critical mass translated into awareness for many.

While the contribution of personal stories, litigation, and cross-sector work has had a positive effect on awareness for technical individuals, the awareness is not yet there for many other groups. In many places, CEOs, CIOs, CAOs, and CFOs have little awareness of the importance of web accessibility for their constituents. I can say without hesitation that most educators who create web content in education are completely unaware of the issue. As such they are unaware of their responsibility to ensure that what they create and subsequently post in online environments does not create a problem for others.

We need to ask ourselves who needs to be aware of web accessibility? Of course, the answer is complex. One part of the puzzle is anyone who creates content that can end up on the web, including most support staff (e.g., administrative assistants) and in education, all faculties. Other groups to consider are those with administrative decision-making duties and those with budget authority. Still others include those in procurement circles, human resources, and those who are charged with providing support or quality assurance.

We need to help all of these individuals understand that efforts on behalf of web accessibility must happen during the design and development phase. Old thinking of the “accommodation” model as the first line of defense—or the only appropriate vehicle for access—continues to plague us. Many well-meaning individuals are of the opinion that it is acceptable to build it, deploy it, and then fix it later if a request or complaint comes through. Allowing others to consider web accessibility to be a post-hoc issue marginalizes it for everyone.

So the work of awareness will continue and will migrate to the next groups who need it the most. If we implement the strategies used in the past (i.e., personal stories, litigation, and cross-sector work), we should be able to replicate success.

Complexity

Another theme that has seen marked change over the past 15 years is the increasing complexity of achieving accessibility. This has shifted along many dimensions, including the complexity (a) of the web itself, (b) of the personnel who need training and support, and (c) inherent in achieving system-wide change.

The web.

In the early years of web accessibility, developers used pretty straightforward methods to create accessibly. I know I am guilty of having said in the early years of WebAIM, “It’s not hard to accomplish.” By the mid 2000s, it shifted to “It’s not easy, but it’s not rocket science.” At some magic point in time some of it actually became rocket science. We fool ourselves and fail others in the field if we forget the fact that the web has become a much more complex environment, replete with dynamic content, simulated and immersive environments; and there has been an ongoing shift to mobile technologies. Each of these developments requires a greater sophistication of both technology and accessibility techniques to get it right.

That does not mean accessibility must suffer. For example, accessibility is strengthened under Accessible Rich Internet Applications (ARIA) specifications—assuming you know of and can use them. With that one sentence, I have relegated the accessibility of dynamic content to those with a sophisticated understanding of ARIA. In the last published WebAIM Screen Reader Survey (WebAIM, 2012), only 35% of the respondents (n=1,782) indicated that they believe the web has become more accessible over the previous year. This may be one reason why web accessibility continues to be a challenge. Because complexity is on the rise, it feels as if the floor is shifting beneath us. Due to the increasing complexity, we now need technical individuals (e.g., web designers, content developers, application developers) who have a deep understanding of technical accessibility and know how to implement it from the outset. Gone is the thinking that all the tools in the accessibility toolkit can be picked up with a few hours or days of self-paced tutorials on the web. The point is that while the vast majority of web content is still comprised of text and images that are easily made accessible with basic knowledge, a small percentage is extraordinarily complex and difficult to make accessible. As we acknowledge this, we acknowledge that our strategies to achieve accessibility will need to change.

Personnel needing training and support.

The complexity of web accessibility affects other personnel, too. Examples include project managers who oversee development cycles, those who participate in quality assurance work, and procurement specialists. As web and design complexity increases, staff from many different professional areas must be able to spot and respond to inadequacies.

I think we need to reconsider the universe of those who should receive training and support in web accessibility. Fifteen years ago, content creators were primarily the same technical people who received training on accessibility. That was because it was the web developers that were posting most of the web content. Then along came the templates. In education, we saw the proliferation of learning management systems (LMS), course management systems (CMS), and locally determined institutional templates for departments and units. As it became easier to post web content created with ubiquitous office tools (e.g., Word, PowerPoint), more staff members who lacked understanding of accessibility were doing so. Sadly, many of them continue to lack even basic technical understanding today. And I mean this even outside of web accessibility. For example, many faculty members and support staff in higher education today continue to be unaware of how to set and use “styles” in Word, although this feature has been around and has supported better accessibility for a very long time. It is more common to see individuals using combinations of large text, bold, or underline to create headers; or using asterisks or text numbers to create lists. Awareness of the issue (i.e., the need for semantic structure) alone will fail to create the change needed; staff also need to know techniques to implement the changes.

The point is that there is a gap between the technical skill level needed for accessibility and the actual skill level of those who are being relied upon as today’s content creators. Does the training and support available to them match their technical proficiency, or should I say, lack thereof? My answer is no, but others might disagree. While there is no doubt that many resources exist to provide information useful to those new to accessibility, I am not sure they are easily consumable by many of today’s content creators. This makes the job of providing accessibility training and technical assistance to content creators a very complex proposition indeed.

System-level change.

Fifteen years ago most of us were focused on raising awareness of accessibility with web designers and developers. It was thought that if we could influence what they do that this would be adequate. Over time we would chip away at the problem, one developer at a time. What we failed to realize in the early days was the degree to which the ecosystem around each developer would determine ultimate outcomes. For example, how many well-meaning developers had these experiences?

By the mid 2000s, the field was dialed into the need to approach web accessibility from a perspective of system change, acknowledging multiple interactive ecosystems (e.g., an example found in systems theory (2014) and an analogous example found in the five contexts of ecological systems theory (2014). The problem was acknowledged for all its complexity, with systems nested within other systems. This is the reality of the work that is ongoing today: the recognition that change is complex, and you need to identify all the interactive components and make sure you understand the contribution of each as you begin to change the system (i.e., will each buy into and support the change to come?).

We no longer consider the work of single designers or single developers when accessibility is the question. We expect that entire systems will be made accessible and that key decision-makers commit to these changes. One example can be found in higher education where everything from course registration, to employment, to the digital aspects of student, staff, and faculty life should be made accessible. This is no longer just a technical issue but an administrative one—an issue that local champions cannot entirely influence because they lack the authority to do so. Progressive institutions have administrations that are backing the work of accessibility across the enterprise. Institution-wide policies are created on web accessibility and are being implemented at all levels of campuses. If one were to only look at accessible output, one would miss the massive coordination up and down administrative channels that support a system’s decision to be accessible (e.g., for examples of this coordination see Project GOALS [NCDAE, 2013a]).

Guidelines, Standards, and Laws

In my mind, the greatest impact to the web accessibility movement over the past 15 years has been the work of guidelines, standards, and laws. This has also been an area that has seen change, or at least seen the promise of change. Of course guidelines and standards provide us with technical specifications so we know what we should do, and laws require us to use them. Because of this convergence, important strides have been made by industry, particularly in the past five years. These changes are summarized below. They have been (a) seen in the guidelines of the Web Accessibility Initiative of the W3C, (b) promised in the standards from Section 508 of the reauthorized Rehabilitation Act, and (c) incorporated into other U.S. laws supporting accessibility.

Web Content Accessibility Guidelines.

Most of you know the history here, but for someone new to accessibility, here is a brief sketch: The launch of the Web Accessibility Initiative (WAI) of the W3C occurred in 1997. They went to work creating the Web Content Accessibility Guidelines (WCAG) 1.0. By spring of 1999 it was released. The 14 guidelines and numerous checkpoints helped focus technical efforts on work known to be necessary for accessibility. Assigning each checkpoint to one of three priority levels also helped those in the field become explicit about the level of accessibility they were after (e.g., if I met Priority 1 checkpoints I was changing only those things I know are needed by many with disabilities to access web content, rather than making the less notable changes desired by some with disabilities and reflected in Priority 3). WCAG 1.0 was HTML and CSS specific. As such, it did not take too long before plans were set in motion to update WCAG to a guideline that was more technology-agnostic and based on principles of accessibility instead. Nine years after WCAG 1.0 set web accessibility on a clear path, WCAG 2.0 was published. There is no way to accurately state the impact the four principles of WCAG 2.0 (perceivable, operable, understandable, and robust) have had on the work of web accessibility today. Part of the impact of WCAG is that it is an international guideline. Moreover, the work of the WAI continues as they influence how guidelines and standards are harmonized across the global communities that must achieve accessibility.

Section 508 Standards.

Whether you are a fan of regulation or not, one fact remains clear: the promulgation, in 2000, of the Section 508 standards for the accessibility of electronic and information technology (E&IT) that is procured, developed, or used by the federal government, has provided the greatest point of leverage for web accessibility in our nation’s history. Although Section 508 covered all E&IT, it was the first to specifically name web accessibility in statute. It was the first to define what is required for web accessibility in the context of a U.S. law; even if that law is tailored for the government itself. The Section 508 technical standards largely followed the guidelines set forth by WCAG 1.0.

Tying governmental procurement to accessibility was a brilliant strategy. And while I could go into a long diatribe about the ineffectual Voluntary Product Accessibility Template (VPAT) system and lack of effectiveness to ensure accessible procurements, the big message was heard—industry needed to gear up for accessibility. They started doing this in earnest across several product lines. For those vendors that embraced a new business environment that included a 508 reality, each development cycle was providing real improvement and impact. Sad to say this was not happening in equal measure across the market.

At the same time, the field was starting to see major changes in technologies (e.g., the proliferation of dynamic content). As technologies changed, meeting the Section 508 standard could actually have a negative (or silly) impact on accessibility. For this reason, the Access Board set out to refresh the Section 508 standards. The Telecommunications and Electronic and Information Technology Advisory Committee (TEITAC) was formed in 2006. I was thrilled that the staff of NCDAE could sit on TEITAC and represent education interests throughout the process, using joint staff we shared with WebAIM. The work of TEITAC mirrored what had already occurred under the changes to WCAG 2.0, where technology-specific standards were removed in favor of principals that would better maintain relevance over time. The Access Board received TEITAC recommendations in 2008 and as of this writing (five and a half years later) the process to promulgate a refreshed 508 standard is not yet complete.

It is unfortunate that we are currently tied (by statute) to an antiquated set of standards. Moreover, it is likely that once the refreshed 508 standards are actually brought into effect, they will already be showing some age.

Laws.

We have seen slow but consistent changes, mainly to existing laws, that support web accessibility over the past 15 years. Here are just a few of the many possible examples: the Reauthorized Rehabilitation Act of 1998 that contains both Sections 504 and 508, which are so critical to web accessibility; the Individuals with Disabilities Education Act in 2004 that included the National Instructional Materials Accessibility Standard (NIMAS); and the new Twenty-First Century Communications and Video Accessibility Act of 2010. There is also promising new legislation (e.g., the Technology, Education, and Accessibility in College and Higher Education Act – TEACH Act) that would transform higher education if signed into law. Moreover, many states added laws to require that web accessibility become a part of state government too. These laws largely paralleled the federal relationship with Section 508, pulling in procurement and setting standards; many used the existing federal Section 508 to be their state accessibility standard.

However, the greatest potential to align all sectors of society to an accessible reality would be the explicit and unwavering inclusion of web accessibility as a civil right under the Americans with Disabilities Act (ADA). The Department of Justice (DOJ) is on record that they wish to clarify web accessibility’s relationship with the ADA. To be fair, the DOJ has already provided numerous comments that an accessible web is part of any covered entity for a covered purpose under the ADA. DOJ has also recognized that greater clarity would benefit everyone and began the process to revise regulations. In my opinion, this would be the most influential change to the legal landscape since Section 508 standards included E&IT in 2000. In 2010, they announced the beginning of this process with an Advanced Notice of Proposed Rulemaking (ANPRM). WebAIM joined many of your voices as together we provided comments to the ANPRM in 2011 (Rowland, 2011). While there have been several rumors of just when DOJ would publish the formal Notice of Proposed Rulemaking (NPRM) on the matter, as of this writing it has not yet occurred.

It is unfortunate that the delays of the 508 refresh and the delays in DOJ’s NPRM process are, in fact, delaying the work of web accessibility in the U.S. itself. It is my opinion that we are missing critical windows in the U.S. regulatory environment as the process unfolds. Let’s face facts—the U.S. will soon head into an election cycle and federal regulation is a hot topic. I wonder to what extent the political realities will impact the practical ones. Everyone knows that the first rule to getting out of a ditch is to stop digging. As a field, we must acknowledge that many, many people are unaware that they are even in a ditch. With each passing month that refreshed standards do not appear, and with each passing month that ADA rules can be ignored (i.e., because “web accessibility is not clear in statute”), new product lines are developed to old 508 standards or without any thought to accessibility at all. New businesses create e-commerce sites that are not accessible, and new employees find issues in the timeliness of the digital environments in which they need to succeed in their jobs. While I may sound pessimistic, I do believe that the next decade will bring a watershed of protections and supports so that everyone is clear on their roles to produce and maintain an accessible web.

Harmonization

One of the brilliant acts of the WAI, and others, over the years has been a continuous attempt to harmonize guidelines at an international level. This makes sense if we intend to play in the same sandbox. This work toward harmonization has had enormous impact on U.S. applications of technical specifications over the past 15 years. Harmonization acknowledges the interconnected nature of web accessibility; not only of web content (i.e., WCAG), but of the contribution of authoring tools to accessible content (i.e., the Authoring Tools Accessibility Guidelines [ATAG]), browsers and assistive technologies (i.e., the User Agent Accessibility Guidelines [UAAG]), and the need to ensure that both the front and back ends of markup and dynamic content are coordinated (i.e., HTML5, and the Accessible Rich Internet Applications [ARIA]). This is not work for the faint of heart. We should not be surprised that from time to time we may see issues that need to be addressed. These seeming cracks in the system are opportunities for the field to make sure we are all on the same page. Here are three examples:

  1. WCAG 2.0 is clear that, with few exceptions, all non-text content presented to users should have text alternatives (Guideline 1.1). Recently, there have been discussions around how the HTML 5 specification will support useful text alternatives. After all the time folks in the field put into items so simplistic as this, it has set off alarm bells for many.
  2. In recent forum discussions, disagreements have arisen regarding whether or not the supporting documentation for WCAG 2.0 absolutely defines success criteria. Again, the fact that the field is not on the same page shows that we may have some cracks in the system.
  3. Some assistive technology (AT) companies in the U.S. have not had harmonization to standards as a priority. Rather, they develop to resolve customer complaints, even if in a nonstandard way. While it is not a bad idea to respond to your consumers, when doing so jeopardizes harmonization to existing standards, it is a bad idea. As a field we continue to struggle with user agents (i.e., AT and tool vendors) who choose standards to which they will conform. This is creating a crazy patchwork quilt of behaviors users have come to expect (e.g., it works with this product, but not with the other). This is becoming increasingly problematic as the web becomes more complex and reliant on standardization.

These examples aside, the past 15 years have seen tremendous work toward harmonization, and more continues to get pulled in and addressed. It is imperative that as a field we are on the same page with respect to the specifics of this harmonization. Failure to do so will undo that which has already been set in place.

As artificial boundaries that do not represent the technological world continue to shrink, international harmonization will be even more important. Far too many of us are U.S.-centric and consider only issues and solutions that pertain to this country. This needs to change.

Supports

Those who engage in the work of web accessibility need two things: to understand web accessibility and their role in it, and to receive skill training and support so they can succeed. There is no doubt that over the past 15 years the field has marshaled to this challenge by developing the supports that are currently in place to help with web accessibility for technical individuals. If you are a web developer, there is a dizzying array of places you can go to get high quality information. The field has truly responded to the need to help one another. This healthy support system includes blogs, tutorials, video instruction, tools, newsletters, discussion forums, RSS feeds, and social media (e.g., Twitter and Facebook). WebAIM has been proud to be part of this important work for all 15 years. While there are always improvements to be made in technical supports to the field—technology supports still need to be hammered out for ARIA, for example—technical personnel have places they can go to get what they need, discuss difficult topics, and process disagreements that inevitably occur in the field. The same is not true for others.

As the field moves toward system change to achieve web accessibility, we need to provide training, technical assistance, and supports to many new groups where we do not have history or influence. While some resources already exist to assist some groups, they are not widely known and are therefore limited in the support they can provide. Moreover, it is critical that we understand that many of these groups are not technically minded. If I send a faculty member to a WebAIM resource on creating accessible PDFs, I had best expect their eyes to spin around as soon as they read what is needed to tag a complex table within the document. That is not to say that faculty could not learn these techniques, nor that we should eliminate complex tables from PDFs or from tutorials. I would suggest, however, that a different type of resource that is designed primarily for non-technical individuals might begin by acknowledging the complexity of the issue, and walk the user through alternatives to creating a complex table. This sort of resource may not be as appropriate for a web developer who is up to the technical task.

Groups of individuals that I believe we need to better support include:

For example the staff of NCDAE has created an array of Cheatsheets (NCDAE, 2013b) to assist non-technical individuals in higher education as they create content that will subsequently be uploaded into an LMS. The Cheatsheets were designed to be one-page reference guides for use after individuals have received instruction; it was never intended to be the instruction that we assumed already existed. However as we looked to link good instruction for non-technical individuals, we were challenged to do so. In a perfect world we would link solid instruction with our reference guide. This shows our project staff that gaps continue to exist.

Another challenge for the future includes decisions on how we prepare the next generation of professionals for the reality of web accessibility. Do we expect them to gain this content in a preservice program (e.g., technical school, college, or university)? Are we expecting this in only technical fields, or across disciplines (e.g., included in the curricula for administrative certificates, procurement officers, HR personnel)? If we don’t think it likely that they will get the information they need during preservice experiences, do we expect them to gain the knowledge and skills in professional development opportunities once they are in the field? The answer is likely “yes” to the need for both preservice and inservice opportunities. Yet, neither is adequately scaled for the job that is just around the corner (e.g., there are millions of content creators in the U.S. alone).

Finally, the field will continue to grapple with the question of professional certification. Who certifies whom? At what levels are certifications given? What do the certifications enable them to do? Finally, how, and by whom, is the certification program managed and operated? Many groups have courses or workshops but this does not provide the independent certification by a third party of the skill of another. There is not even consensus that certification should be made on the independent demonstration of skill. While there have been discussions in the field for several years, as of this writing, no U.S. plan has been released to my knowledge.

Summary

I have been thrilled to walk the journey of web accessibility with many of you. The field has seen tremendous change over the past 15 years in awareness, complexity, standards and laws, harmonization, and supports. Even with persistent and challenging issues, the field is moving in positive directions. The next 15 years will likely bring an astonishing array of new developments, and new challenges. Since the best predictor of the future is the past, as long as we continue this journey together, good things will happen.

Make no little plans. They have no magic to stir men's blood ... Make big plans; aim high in hope and work...
—Daniel Burnham

References

Ecological systems theory. (2014). Retrieved from http://en.wikipedia.org/wiki/Ecological_systems_theory

National Center on Disability and Access to Education. (2013a). Introduction to the GOALS project. Retrieved from http://ncdae.org/goals/

National Center on Disability and Access to Education. (2013b). Cheatsheets. Retrieved from http://ncdae.org/cheatsheets/

Rowland, C. (2011). ADA ANRPM Response. Retrieved from http://webaim.org/blog/ada-anprm-response/

Systems theory. (2014). Retrieved from http://en.wikipedia.org/wiki/Systems_Theory

WebAIM. (2012). Screen Reader User Survey #4 Results. Retrieved from http://webaim.org/projects/screenreadersurvey4/

Rowland, C. (2014). A Traveller's Notepad: Reflections on a journey toward web accessibility. Information Technology and Disabilities E-Journal, 14(1).