Skip to content Skip to navigation

Cultivating and Maintaining Web Accessibility Expertise and Institutional Support in Higher Education - Paul Bohman

ATHEN E-Journal Issue #2 (2007)

Paul Bohman
Technology Coordinator: Kellar Institute for Human disAbilities
George Mason University

When web content is designed with accessibility in mind, it allows an unprecedented level of access for people with disabilities to information, communication, entertainment, employment, and education. In contrast, when web content is not designed with web accessibility in mind, it creates barriers, despite the fact that web technologies already offer built-in methods and solutions to overcome most of those barriers. The vast majority of the accessibility obstacles on the web are not the result of technological failings. Usually they are the result of human error, negligence, ignorance, or, less frequently, discriminatory bias against people with disabilities. None of these excuses is acceptable within higher education environments, where discriminatory practices are both unethical and illegal. Colleges and universities must cultivate and maintain the expertise and institutional support necessary to ensure the accessibility of their web content.

Higher education institutions are complex systems -or, more accurately, complex layers and networks of systems- and there are many possible points of failure in efforts to achieve web accessibility. Holistic systems-level solutions are necessary. Only systematic, coordinated effort can result in comprehensive, sustained implementation of the best techniques and technologies. However even well designed systems can fail to achieve their goals if they stray from their central purpose. Gall's (1975) somewhat tongue-in-cheek treatise on why systems fail states wryly that "new systems generate new problems" and "systems in general work poorly or not at all" but that "a simple system, designed from scratch, sometimes works." Gall warns that "a complex system designed from scratch never works and cannot be patched up to make it work. You have to start over, beginning with a working simple solution." These cautionary notes are included here not to discourage the reader, but to cast the topic in a realistic light. Web accessibility is a real challenge for real people-both for people with disabilities and for people seeking to accommodate those with disabilities. Successful approaches to web accessibility do not always reveal themselves immediately. They usually evolve out of the day-to-day experiences of the people closest to the challenges. The issues, ideas, and suggestions in this paper have emerged over the course of several years of trying to address web accessibility within higher education settings, but this is not the final word on the matter, and does not prescribe a definitive set of easy steps to follow. It is more like a discussion, and higher education institutions would do well to continue the thread at their own institutions, discussing, accepting, adapting, or rejecting the ideas presented here as appropriate to their own particular circumstances.

Accessibility for Whom?

The categories of disabilities that affect a person's ability to use the internet include physical or motor disabilities (especially inability or difficulty using the hands), visual disabilities (blindness, low vision, color-blindness), deafness, cognitive or intellectual disabilities, seizure disorders, and combinations of these conditions (Bohman, et al, 2005). Ideally, institutions would create web content that is as near to universally accessible as possible for all people, with or without disabilities. Striving for universal accessibility is usually the most efficient approach because it eliminates the need for multiple versions of the same content. From a technical perspective, it is also often the most practical and most effective approach, though occasionally web developers encounter scenarios that require alternate parallel versions of content.

Most discussions about web accessibility in higher education focus on the needs of students, and understandably so, considering that higher education is in the business of attracting and educating students. However students are not the only ones to consider. Employees-staff and faculty-also have accessibility needs. It would be a mistake to focus on one group at the expense of the other.

Accessibility of What?

What constitutes web content? There are a few obvious answers to this question. The university's main web site is a good starting point. A click or two of the mouse (or keyboard) can take users to the web sites of different departments, colleges, offices, projects, institutes, organizations, or other academic entities within the institution. Already, this constitutes thousands, or possibly millions, of pages of content. Adding to this list are things like online courses, supplemental online materials for classroom-based courses, library databases, library subscription services, campus intranet services, employee and student records, bookstore purchasing services, and the personal web sites of students and faculty. Some of these web-based resources are simple HTML (Hyper Text Markup Language) files created by staff. Others are complex software products licensed from commercial vendors. In between these two extremes is an array of miscellaneous resources, such as word processor documents, slide shows, spreadsheets, videos, Java applets, etc. All of this is web content, and a holistic approach would seek to take it all into account.

With so much information and interactivity available now in a digital format, the sheer volume of resources that must be accessible is daunting, especially when including "legacy" (old) resources (Richards & Hanson, 2004). To make matters more difficult, higher education institutions do not have direct control over some of these resources. Institutions usually license proprietary learning management software, employee databases, bookstore "shopping cart" systems, library subscription services, web-based email services, online applications for grants and loans, and other products from third-party vendors. If these products have accessibility flaws, the institution can only hope that the vendor will eventually fix them, yet the institution is still required to ensure that it does not discriminate against people with disabilities. This type of dilemma is all too common.

Spheres of Accountability and Opportunity

A holistic, systems-level approach suggests a distributed model of web accessibility expertise. There is more than one type of accessibility expertise, and more than one group of people in need of some form of accessibility expertise. Yes, web developers need to know the technical skills of accessible web design, but the larger realm of accountability includes vision and leadership, procurement policies and procedures, hiring practices, library management, online course materials, instructional technology services, student disability services, employee intranets, the main web site, faculty and student sites, curriculum design and development, research, and other areas within higher education. Not everyone in all of these areas must learn how to create accessible web pages, but key people in each area must know enough of the right kinds of information-within their spheres of accountability and opportunity-to prevent inaccessible practices from creeping into the system. They also must know how and where to receive additional accessibility assistance when needed.

A model for cultivating and maintaining web accessibility expertise must strive to account for multiple internal and external factors. There is no one right way to do this, and, realistically speaking, perfect accessibility across all of these factors is impossible. There are too many variables both within and without the institution. Nonetheless, well-informed personnel in supportive colleges and universities can radically expand the range of possibilities for students and employees with disabilities, replacing exclusion and inconvenience with inclusion and independence. To accomplish this, an institution could assign the authority and accountability over all web accessibility efforts to one person (a sort of "web accessibility czar"), or perhaps to a committee of representative stakeholders. Another option in strong participatory systems is to decentralize accountability throughout the organization, adding appropriate web accessibility responsibilities to existing job descriptions. In some cases, institutions may choose to contract out portions of the accessibility work, though this carries the risk of a diminished sense of accountability and responsiveness within the organization. Any of these approaches, or hybrids of them, can work, as long as the institution takes the goal of accessibility seriously, and as long as there is sufficient communication between the various areas of the campus. Lax attention to the issues in at any point within this network of responsibility weakens the overall effort and leaves students and employees with disabilities vulnerable to discriminatory practices, whether intentional or not.

Meadows (1999) discusses twelve "leverage points" or places to intervene in systems to achieve the maximum effect. Though much of Meadows' work focuses on environmental systems, the leverage points also can help pinpoint areas for effective intervention in higher education. In the interest of keeping the discussion focused, only the top six leverage points are listed here, starting with the least effective and progressing toward the most effective: 6) the structure of information flows (who does and does not have access to what kinds of information, 5) the rules of the system (such as incentives, punishments, constraints), 4) the power to add, change, evolve, or self-organize system structure, 3) the goals of the system, 2) the mindset or paradigm out of which the system-its goals structure, rules, delays, parameters-arises, and 1) the power to transcend paradigms.

Vision and Leadership

Meadows' two highest-ranked leverage points have to do with the ability to recognize and transcend paradigms. True leaders cultivate these skills within themselves and others. Some leaders work in administrative and managerial roles, but others do not. Leaders can emerge from the rank and file of organizations, regardless of their titles. Anyone with a clear vision and sense of purpose can pave the way for change. Without this type of leadership at some level, accessibility efforts cannot move forward, unless perhaps forced to by a formal complaint or a lawsuit.

If no one has yet taken up the cause of accessibility on campus, leaders-to-be may find Rowland's (2004) eight-step model of reform helpful. Rowland's steps are: 1) gather baseline information, 2) gain top level support, 3) organize a web accessibility committee, 4) define a standard, 5) create an implementation plan, 6) provide training and technical support, 7) monitor conformance, and 8) remain flexible through the changes. The Web Accessibility Initiative (WAI) of the World Wide Web Consortium presents a similar set of ideas (Brewer & Horton, 2006), along with a business case for accessibility (Lawton Henry, 2005). Every institution must forge its own path, and some may find these steps to be a bit too formulaic, but they can at least serve as a starting point for further discussion and adaptation. For instance, if top-level support is not forthcoming it may be necessary-even if not particularly desirable-to forgo that step. In addition, it may not be necessary or desirable to follow the steps in the order as written, or to treat them as steps at all. Participants can pursue some of these "steps" simultaneously, rather than in any particular sequence. The recommendation to remain flexible through the changes is one to keep in mind from the beginning, and maps to Meadows' (1999) fourth leverage point: the power to add, change, evolve, or self-organize system structure.

Technical Standards and Policies

A vision for web accessibility leadership might include a policy that sets a technical standard for web accessibility, as Meadows' fifth leverage point-the rules of the system-suggests. Many college and universities have done this (WebAIM, n.d.; Schmetzke, n.d.; Information Technology Technical Assistance and Training Center, 2006; Johns Hopkins University, n.d.; Johnson & Ruppert, 2001). Unfortunately, some of these policies and pseudo-policies fall far short of their stated purpose, crafting a "bridge not quite far enough" to achieve the intended goal (Bohman, 2004). Where policies are inadequate, they are likely to produce equally inadequate results. The shortcomings vary from one policy to another, but nearly all of them have a common flaw: they fail to take into account the many ways that web content enters the college or university system. They may contain a statement to the effect that the institution is committed to achieving web accessibility, and a few of them make explicit the technical standard to adhere to (most do not), but most do show little evidence of thinking outside of the prevailing paradigms. They address part of the challenge, but do not address the pervasive and interconnected nature of web content responsibilities across college and university campuses.

The Web Accessibility Initiative (WAI) of the World Wide Web Consortium (W3C) produced the now internationally dominant set of web accessibility guidelines, including the Web Content Accessibility Guidelines (WCAG) (Chisolm, Vanderheiden, & Jacobs, 1999), User Agent Accessibility Guidelines (UAAG) (Jacobs, Gunderson, & Hansen, 2002), and Authoring Tool Accessibility Guidelines (ATAG) (Treviranus, McCathieNevile, Jacobs, & Richards, 2000). Note that this is a set of three documents. Organizations often make the mistake of ignoring all but the Web Content Accessibility Guidelines, even if they create or purchase tools that could be considered authoring tools or user agents. Also worth mentioning is the fact that there are three "levels" of conformance to these guidelines, with the first level being "essential," the second level being "important," and the third level being "beneficial." Conformance is usually easier to verify with policies that specify the required level within the guidelines. The WAI guidelines are not without their critics (Clark, 2006; Kelly, Phipps, & Swift, 2004), who assert that the guidelines are difficult to understand, too theoretical, among other complaints.

The Unites States produced an alternative set of web accessibility guidelines in association with Section 508 of the Rehabilitation Act (U.S. Access Board, 2001), as part of a larger set of guidelines for information technologies, including copiers, fax machines, phones, and a host of other devices. Although written to apply to federal government entities only, the Section 508 technical standards can serve as a model for higher education, especially in their inclusion of a wide range of information technologies. The technical standards for web accessibility within Section 508 are less ambitious than the WAI guidelines. Accessibility advocates have expressed several grievances (Bartlett, 2001; Brewer, 2006; Schroeder, Sajka, & Dinsmore, 2000), saying, for example, that the technical standards are superficial and overly minimalist, or that they exclude people with cognitive disabilities, or that competing sets of standards (i.e. WAI and Section 508) would fracture the web accessibility. These are legitimate concerns that warrant consideration. On the positive side, the Section 508 guidelines leave less room for interpretation, facilitating compliance verification. In fact, Section 508 seems to have exerted some influence on the editors of WCAG 2.0-still in draft form at the time of this writing-who specified in the requirements document that the final guidelines should "ensure that the conformance requirements are clear" (Vanderheiden, Slatin, & Chisolm, 2006).

Protecting Against Legal Liabilities

One motivation for addressing accessibility is to protect against legal liabilities. If a student or employee decides to pursue legal action for a perceived violation of civil rights, the institution finds itself in damage-control mode, which can be uncomfortable and costly. The expense of defending a college or university against a lawsuit can easily reach into the millions of dollars, and the negative publicity generated can be difficult to dissipate or repair. By then, it is too late for good intentions, and the best the institution can do is to regroup and try to make sure that does not happen again. A smarter approach is to anticipate potential problems before they occur.

The Americans with Disabilities Act (ADA) (U.S. Department of Justice, 1990) protects employees of higher education institutions in the United States from discrimination. Section 504 of the Rehabilitation Act (U.S. Department of Labor, 1973) protects students, though there is some overlap in the applicability of these laws to the two constituencies. The accessibility laws governing higher education take a reactionary approach. Institutions do not have to change anything until a student or an employee with a disability needs or requests changes. This may be convenient for the institution, but it can be quite inconvenient for the individual with a disability, who must initiate the request. Some accommodations require only "quick fixes," whereas others require a considerable amount of time and effort to accomplish. The more difficult accommodations are usually the ones most tempting to procrastinate, so institutions often ignore the situation until a request, or emergency, arises. Unfortunately, at that point, the institution must scramble to meet immediate, externally imposed needs and timelines, having forfeited the chance to work through the issues under a more forgiving set of circumstances.

Procurement Policies and Procedures

In addition to creating their own web content, colleges and universities install web-based software created by third parties, which they usually cannot manipulate to make more accessible. Once the institution begins using the product, the window of opportunity to make accessibility decisions has passed. There is currently no law explicitly requiring higher education institutions to purchase accessible products. However Section 508 sets a precedent for this type of approach. Section 508 requires that Federal government entities in the United States assess the impact of information technologies on accessibility before purchasing any of them, and are required to either purchase the most accessible product from among competing vendors, or justify in writing their reasons for purchasing less accessible products. Section 508 has exerted a tremendous amount of influence across a wide swath of the information technology industry due to the ingenious way in which it links accessibility with procurement. Many vendors have responded by increasing their commitment to accessibility as evidenced by the long list of products with VPATs, or Voluntary Product Accessibility Templates (VPATs) catalogued on the federal government's "Buy Accessible" web site (General Services Administration, n.d.), if only to increase the marketability of their products. Admittedly, some of these VPATs overstate the accessibility of the product, or mislead the consumer in other ways, and there is no formal monitoring system to evaluate VPAT claims, thus reducing their value somewhat. In any case, the preventative nature of Section 508, with its focus on procurement, can reduce current and future accessibility problems when interpreted and applied effectively as a leverage point during the procurement process.

A holistic approach would include provisions for procurement policies and procedures that support accessibility proactively, rather than reactively (National Center on Disability and Access to Education, n.d.). In terms of web technology, a procurement approach applies mainly to the purchase or licensing of third-party web-based tools and services, such as learning management software, library databases, subscriptions to online journals or abstracts, etc. Some states and/or higher education institutions have adopted the Section 508 standards for web development (WebAIM, n.d.), but in most cases they have not adopted its procurement focus, nor have they adopted the full set of Section 508 standards for other information technologies, such as stand-alone software, copiers, fax machines, telephones, etc. Considering the interdependence of many information technologies, and the extent to which nearly all digital content is potentially web-capable, it may not make sense to single out web accessibility from among other types of information technologies, especially from a procurement policy perspective. It would probably make more sense to create the equivalent of a campus "Procurement Officer," or "508 Coordinator" (508 Coordinators, n.d.), a position common in federal government agencies. This person would not handle all of the institution's procurement needs personally. Such a task would be impossible in such large organizations. Instead, she would oversee procurement operations and require accountability of all the other employees with procurement authority.

A procurement policy would not require adoption of the Section 508 technical standards for the web. An institution may decide to replicate the procurement focus of Section 508, but with a different set of guidelines, such as the WAI guidelines. However given that the WAI guidelines exclude non-web technologies (copiers, fax machines, phones, etc.), a hybrid policy incorporating elements from both sets of guidelines may be appropriate in some cases.

Once a policy or procedure of some kind links accessibility to procurement, employees involved with procurement will need the institutional support and expertise necessary to implement the policy effectively. While they may not need to be experts in every aspect or nuance of the technical issues of accessibility, they should have a solid working knowledge of the accessibility requirements specified in the policy. They should also have access to and cultivate relationships with people who can help them make informed decisions. They need to be able to research the accessibility of new products and services, soliciting feedback from individuals with disabilities whenever possible, either on campus, or through extended social and professional networks. They need to understand both the value and limitations of self-reported measures of accessibility, such as VPATs. They need to communicate their accessibility requirements to vendors before making the decision to buy products, or at least before letting the vendors know that a decision has been made. Similarly, they must communicate the institution's accessibility expectations to faculty and staff before and during the procurement process. Throughout the process, the people in procurement positions must maintain records of the research and purchasing decisions, subject to evaluation and quality control mechanisms.

On occasion, the procurement officer must also have the courage to refuse to pay for products or services with significant accessibility flaws, despite the almost certain backlash. This is a gatekeeper position that some may view negatively as stifling the purchasing freedom of employees at the institution. The person in this position must feel confident in the counter argument that she is instead protecting the freedom of people with disabilities who would otherwise be required to study and work in inaccessible circumstances created not by disabilities per se, but by the institution's inability or unwillingness to live up to the ideal of creating an inclusive environment for all.

Libraries

Libraries are portals to information. If either the portal or the information is inaccessible to people with disabilities, the library is not serving its purpose fully. Schmetzke (2003, 2001b, 2001a) has alerted library scientists to the pervasive problem of inaccessible library web sites and, more recently, of usability problems for people with disabilities on library web sites (Steward, Narendra, & Schmetzke, 2005), while noting that accessibility of these sites had improved since his first articles on the topic. At this point, the most serious accessibility concern is the inaccessibility of the library materials themselves to people with "print disabilities" (Epp, 2006), such as people who experience blindness or low vision. Printed books are useless to these individuals. One solution is to convert books to audio or Braille formats. This is usually expensive, either in terms of actual money spent or volunteer hours required, and is not sustainable or scalable.

The most viable long-term solution is to make as much content as possible available in an accessible digital format. Many journal publishers now offer electronic versions of materials online as a subscription-based service. This is a giant leap forward from the exclusively print-based libraries of yesteryear. Still, some electronic resources are nothing more than scanned paper documents, without any machine-translatable text, making them just as useless to people who rely on text-to-speech conversion software (known as "screen readers") as the original paper versions. In addition, most book publishers are loathe to offer any sort of electronic version of their books for fear that people will copy and share the content illegally, thus negatively impacting the publisher's profit margin. Their fear is understandable, but they place an enormous burden on people with print disabilities, who must find alternative ways to access the content through much more circuitous routes. Many classic and popular books in the public domain (free from copyright restraints) are available online through volunteer-supported initiatives such as Bookshare.org and Project Gutenberg (http://www.gutenberg.org/), but progress is slow and the number of volumes is still modest-reported about 100,000 in 2004-(Leventhal & Sajka, 2004) compared to the billions of printed resources. There are also repositories of digital books, often laboriously scanned and processed by college and university staff rather than the publishers, available specifically for students with disabilities through state school systems. Some publishers offer electronic versions of their content upon request, but many do not. In both cases, the burden and inconvenience weigh heavily upon individuals with print disabilities and upon those who work on their behalf.

Copyright and profits are legitimate concerns, but not at the expense of excluding a whole segment of the population. There is only one reasonable response: publishers must change their ways. Publishers must provide original digital versions of their materials through organizations that can provide the materials to people with disabilities. Electronic versions of books can be produced in HTML, Word, and text files, as well as disability-specific formats such as DAISY (Digital Audio-based Information System), BRF (Digital Braille), and NIMAS (National Instructional Materials Accessibility Standard). All of these formats can be distributed through secure services that protect the original copyright.

In a paper addressing the responsibilities of higher education institutions it may seem strange to devote space to a discussion of the shortcomings of the publishing industry. It is true that librarians do not have the authority to demand that publishers change their ways, but librarians do have somewhat of an inside track, albeit indirectly, in the publishing industry. If nothing else, librarians can leverage their role as prominent consumers within the publishing industry, to petition for easier and more complete access to accessible digital versions of published content. As publishers respond to these petitions, libraries will spend less time compensating for the shortcomings of the publishing industry, and patrons with disabilities will be able to spend more time partaking of the same freedoms that the rest of the population usually takes for granted.

Until that day, libraries in colleges and universities must still work to provide accommodations as needed. They must still ensure that physical access to the building's resources is possible in a wheelchair, that students with print disabilities receive requested ad-hoc accommodations, that the web site adheres to accessibility principles, and that purchases and acquisitions take accessibility into account to the extent possible.

Many libraries also play a direct role in ensuring the accessibility of course content. Some instructors select journal articles or other library materials as a part of the required reading, and may ask the library to make these available online through an "online library reserve" system. All too often, libraries simply scan in the file as a graphic and post it in Adobe Acrobat PDF format. The problem with this method is not the Acrobat file format, which can produce accessible files. The main problem is that no technology can directly translate a graphic into synthesized speech, as required by users with blindness. Someone must first process the page with optical character recognition (OCR) software, proofread it, and correct any errors before the content is ready for screen reader users. The relative success of OCR software conversion is dependent upon factors like the visual clarity of the original document (contrast, sharpness), the font type (fancy font faces are problematic), the orientation of the text (diagonal or sideways text is difficult or impossible), and so on. A different kind of problem arises when the document is nothing more than a copy of a copy of a copy, as is sometimes the case. The indistinct text can make reading difficult or impossible for users with low vision, and can reduce readability for all users, whether they have disabilities or not. Again, the long-term solution is for publishers to make their content available in accessible electronic formats, but when electronic versions are not available, libraries must anticipate that some students will need scanned copies of materials converted into text. The library may decide to perform this conversion in-house, or work with student disability services and/or outside contractors to accomplish this goal.

Online Course Materials

The accessibility of course materials is one of the most important, and sometimes most difficult, areas of focus within higher education. The diversity of approach, content type, materials, and medium can make this a daunting task to coordinate. The task is easiest when course materials are already in a transformable digital format, such as text, as is the case with many online courses. Even digital materials can be difficult, or at least time-consuming, to make accessible if they involve the live transmission of information (e.g. webcasts), audio requiring synchronous captions and/or transcripts, or simulations or other programmed objects or applets (e.g. objects created with JavaScript, Java, Flash, Shockwave, etc.). The technical challenges with these and other types of course content can be considerable, even for technology experts in the field, which most instructors are not. Some of them, especially in the older generation, use technology gingerly and reluctantly, if at all, to the extent that the use of any instructional technology may be a leap of faith for them. Others, especially in the younger generation, feel more at home using and creating digital instructional materials. The latter group constitutes the group that Prensky (2001) calls "digital natives," as opposed to the "digital immigrants" who did not grow up in a society permeated by digital technologies. The immense variability in technological mastery among instructors complicates accessibility coordination efforts.

Beyond this, a fundamental question emerges: to what degree are instructors responsible for the accessibility of their instructional content? Instructors know-or should know-that they can turn to the office of student disability services to help provide accommodations for students who need them, and most are willing to do so. Convincing instructors of the merits of accommodations is usually not a problem. The dilemma lies in trying to devise a system or set of protocols that addresses course content accessibility that neither overburdens instructors with unreasonable expectations nor allows them to abdicate all responsibility.

Whenever possible, instructors with the requisite disposition for technology should make their own course content accessible, especially if the instructors create the materials themselves. They are familiar with the materials and often have the easiest access to them. When this is not possible, instructors should initiate a fallback protocol in which web developers or instructional technologists provide accessibility services. The office of student disability services may or may not need to play a role in this process, depending on the type of accommodations needed. Oftentimes those who work in student disability services are more familiar with assistive technologies on the user's end (screen readers, variable height desks, audio books, alternative keyboards, etc.) than they are with the techniques for making web content accessible. This is not a problem as long as another campus unit, such as an office of instructional technology services, can fill the gap-and as long as communication between the groups is adequate. Campuses with degree-granting instructional technology programs should look to faculty and staff there to provide theoretical and technological guidance in all aspects of instructional technology, including accessibility.

Kelly et al. (2004) make the case that accessible learning is a more important goal than accessible e-learning. They propose that the learner's needs should guide instructors toward a design that considers accessibility, usability, local factors, infrastructure, and learning outcome. An over-emphasis on technical accessibility can derail meaningful learning, which would be counterproductive, even if it meets official standards for accessibility. This argument is intuitively persuasive, even as it raises questions about how to assure that this happens in the many courses across the college or university system. There is just as much danger in being overly prescriptive as there is in being overly permissive or ambiguous.

Academic Freedom

A few instructors may raise the banner of academic freedom in an attempt to rally against the requirement to make their course content accessible. As it turns out, the argument that web accessibility encroaches on academic freedom is a bit of a red herring, rooted in a distortion of what accessibility actually means. It is based on the fear that making web content accessible requires some kind of fundamental alteration to that content that would compromise its expressive or aesthetic integrity. As it turns out, nearly all of the techniques for making web content accessible leave the underlying content intact, and are invisible to users without disabilities. The concept of academic freedom implies the right to perform research or express one's self according to one's conscience, without fear of sanctions from the institution. Web accessibility does not preclude the publication or expression of any particular type of ideas, it merely expects those publications and expressions to be accessible to people with disabilities when published on the web. Nevertheless, the fear that accessibility encroaches on one's creative freedom persists in some circles, and can commandeer otherwise productive conversations, leaving the legitimate concerns of individuals with disabilities to languish unproductively in the stagnate eddies of circular debate.

To be fair, and to underscore the complexity of the matter further, there are certain types of web content that are indeed difficult or impossible to make accessible without either some kind of fundamental alteration, or, as a last resort, the creation of a separate, alternative version. Some highly interactive and visually intense instructional simulations fit into this category. Making such software accessible to users of screen readers, and/or to individuals who cannot use a mouse may weaken or destroy the pedagogical underpinnings of the design. If the act of making the content accessible would interfere with the pedagogical intent of the software, academic and innovative freedom might emerge as central concerns (Kelly et al., 2004), though this circumstance is somewhat rare. Most often, pedagogy and accessibility are completely compatible. Institutions would need to address these special circumstances on a case-by-case basis, and some cases will undoubtedly prove quite challenging from an accessibility standpoint. These exceptions cannot be ignored, but neither should they dominate the conversation about accessibility.

Curriculum Design and Development

Today's students will become tomorrow's professionals. Higher education programs have the opportunity to influence the accessibility inclinations of a rising generation of future leaders. Even programs without a technology focus can incorporate accessibility into the curriculum at some level. Students can learn to create accessible word processor documents by providing alternative text for images, using headings where appropriate (e.g. by applying heading "styles" rather than just making the text big and bold), and so on. For assignments requiring students to create web content, at a minimum, instructors can provide links to basic information about web accessibility. In information technology disciplines, such as instructional technology, computer science, business information systems, etc., students need more than minimal knowledge. They need to become well-versed in the relevant laws, standards, principles, and techniques. Some programs offer complete courses in web accessibility, such as George Mason University's instructional technology program. Others spread accessibility information throughout the courses, which can be effective when done well, though it risks a dilution of the core competencies.

Research

Colleges and universities have a history of developing web-based tools. A few prominent examples include Google, Blackboard and WebCT (now owned by Blackboard). In the past, most tools developed in higher education research settings were latecomers to the accessibility scene. As momentum has grown over the years, more higher education projects have incorporated accessibility into the design in the early stages. This is a significant step in the right direction. Institutions should continue to encourage this trend, by acknowledging successes in this area and initiating discussions on the topic, perhaps as a part of "brown bag" lectures over lunch, faculty meetings, or in other settings.

Hiring Practices

A sensible way to cultivate expertise at an institution is to hire people who already have the desired expertise. Job announcements can include web accessibility expertise as a prerequisite skill. Each type of job would include a different aspect of web accessibility expertise. Prospective procurement agents would need to understand and apply the Section 508 VPAT system, for example. Prospective web developers would need to present a portfolio of past work that demonstrates practical application of web standards and accessibility techniques. Applicants to library positions would need to know how to work with the student disability services office and/or the ADA office to assist students, faculty, and staff with disabilities and how to work with procurement agents to ensure the maximum accessibility of library content. Other job announcements would include similar requirements.

A variation on this approach would be to require recently-hired employees to complete a sort of accessibility "certification" by attending a workshop or series of workshops-depending on the level of expertise required for the particular job-and demonstrating mastery of certain key concepts and skills. There is no standardized certification assessment or licensure program for web certification, but the institution could initiate its own in-house procedure by preparing a training curriculum for use in classroom-type settings and/or online settings. Materials such as the WebAIM Guide to Web Accessibility (CD-ROM) (Bohman, et al, 2005) or books about web accessibility (Slatin & Rush, 2003; Clark, 2002a; Thatcher et al., 2002) may be of use with this type of approach.

Another commonly-overlooked technique for cultivating accessibility expertise is to hire people with disabilities. The more employees with disabilities working at a college or university, the more likely that accessibility changes will occur during the natural course of day-to-day business. For many people, accessibility is an abstract principle with little practical application. They may not have friends, family members, or co-workers with disabilities, so they have no natural incentive to think or act with accessibility in mind. Working alongside a person with disabilities may provide the impetus to change a person's paradigm and, consequently, their actions. At the very least, the presence of people with disabilities in the workplace will reveal the barriers that already exist and perhaps quicken the removal of those barriers.

The motive to provide equal employment opportunities may provide additional justification to hire more people with disabilities, whether in the context of "affirmative action" programs or otherwise. People with disabilities face some of the most daunting social and physical obstacles of all minorities. Data from the United States in 2002 (Employment and Disability Institute, n.d.) show that the unemployment rate of people with disabilities (14.1%) was more than double the unemployment rate for people without disabilities (5.8%), but these statistics only count people actively seeking work. The employment rate for non-institutionalized civilians with disabilities between the ages of 18 and 64 was a meager 20.8%, compared to 78.0% for people without disabilities. This chasmal discrepancy is due in part to the nature of disability itself: some people with disabilities cannot work, or are limited in the work they can perform. However this is only part of the picture. Disability is more than a biomedical impairment or deficiency. Disability is often a result of preventable disabling conditions within the social and built environment. An additive accrual of these disabling conditions over a lifetime can exert a powerful negative influence on the people who experience them, in terms of diminished social, educational, and professional opportunities. Providing increased employment opportunities to people with disabilities may help to reverse this historical trend.

The Main Web Site

Most examinations of web accessibility in higher education concentrate on the main web site, often to the exclusion of other types of web content. This is precisely why this paper addresses other areas first. Full-time experienced professional web designers create some, but not all, of the resources on the main web site. Many web sites-or parts of web sites-fall under the jurisdiction of part-time student employees, administrative or clerical staff, interns, or individuals for whom web content creation is only one of many other responsibilities. Some of these people are motivated and skilled to produce high-quality web content. Others have taken on their web-related duties reluctantly or prematurely, not yet having a solid foundation in the basics of web content creation. The diversity of interest and skill levels is due partially to budget constraints, but also to the mission of higher education itself, wherein student employees are counted not just as cheap labor but as apprentices and learners in need of authentic learning experiences, even if those learning experiences sometimes compromise the quality of the institution's web presence. This is a reality that higher education institutions must allow for, but it is not an excuse to encourage substandard work.

It is time to raise the bar and move past the basic accessibility barriers identified by a multitude of accessibility studies (for example, Loiacono, McCoy, & Chin, 2005; Williams & Rattray, 2005; Zeng, 2004; E. Loiacono & McCoy, 2004; Zeng & Parmanto, 2004; Thompson, Burgstahler, & Comden, 2003; Walden, Rowland, & Bohman, 2000). By now, most web developers have heard of web accessibility and know at least a little about it. The next step is to create a system that cultivates and supports the application of that knowledge.

Sustainable Knowledge Management Solutions

Creating an accessible web presence is an ongoing consideration requiring commitment and constant vigilance. Web sites are under constant revision, with new resources, news stories, changes, and updates being added on a continual basis. The sustainability of accessibility efforts depends upon an institution's ability to anticipate and monitor the accessibility impact of these changes. Just as it is impossible to eat a big enough meal to last the rest of a lifetime, it is impossible to create an accessible web site "once and for all," and then move on to other projects. Comprehensive, sustainable solutions require constancy and commitment.

Considering the large number of people needing some kind of web accessibility expertise, combined with the high turnover rate among web developers, higher education institutions should seek to maximize the "organizational memory" of web accessibility expertise over the long term. Walsh and Ungson (1991) identify six areas where organizational memory can reside: (i) the individuals who work for the organization, (ii) the organization's culture, which includes shared frameworks, (iii) standard operating procedures and practices, (iv) roles and organizational structures, and (v) the physical structure of the workplace. Huber (2001) talks of "three knowledge contexts-archived codified knowledge, archived non-codified information, and person-to-person communications." The word archived represents one of the key words in this last list. Encouraging employees to keep records of their sources of information, best-practice techniques, and interpersonal communications can help alleviate the burden of re-training new employees in high-turnover working environments. Archived sources of information can include books, articles, web sites, databases, emails, discussion forums, blogs, meeting agendas and minutes, CD-ROMs, DVDs, videos, lists of consultants, lists of the particular areas of expertise of staff members, former employee contact information, conference notes, help desk notes, and many others. Accessible digital formats offer the advantage of being searchable, but non-digital information sources-including Walsh & Ungson's (1991) "softer" sources of information such as culture, shared frameworks, roles, structures and people-can be equally useful.

Knowledge Acquisition and Capacity-Building

There are many ways to ensure that the right people learn the right information to perform the right tasks. Sometimes this is referred to as "training" employees, which can be an accurate description of some of these options. The hope is that "training" is understood as only one important component among many in a system that adequately supports knowledge acquisition and builds capacity in individuals and the organization as a whole. There is a time and a place for training events within a holistic approach. In these moments and places, it can be helpful to have a list of training ideas from which to choose. This last section describes a few ideas as a starting point.

High quality books and curricula for training web developers have existed for years in both printed (Slatin & Rush, 2003; Clark, 2002a; Thatcher et al., 2002) and electronic (Bohman, et al, 2005; J. Clark, 2002b) formats. Institutions can purchase or license these materials for employees. Electronic materials are often available in site licenses, or in some cases at no charge and without licensing fees, increasing the reach of their potential impact compared to the more limited reach of printed books. Institutions can provide these as general reference materials, or as resources in courses, workshops, or certification preparation. There are also numerous electronic discussion forums and email lists within existing professional networks. See, for example, http://accessify.com/, http://www.webaim.org/discussion/, and http://www.w3.org/WAI/IG/. Web developers are the target audience for most of these resources.

Workshops facilitate the rapid dissemination of ideas to groups of people. They can offer hands-on experience in classroom settings or computer labs, with the instructor acting as a mentor to the participant apprentices. Workshops can be one-time events, recurring events, a series of events, or some combination of these options. They can include all people on the campus with an interest or role in accessibility, or they can focus on specific groups, such as webmasters, instructors, librarians, procurement officers, and so on. Workshops can also cover specific technologies or methods of making content accessible. They can cover any aspect of web accessibility for any type of audience. Possible sponsors of workshops can include libraries, instructional technology support units, the office of student disability services, the webmaster, the institution's chief technology officer, and others. Both face-to-face and online workshops are possible.

Conferences are another option. Some institutions sponsor their own conferences and require employees to attend. Sometimes they plan these events in conjunction with other institutions or other campuses within the same institution. Conferences may or may not involve travel to other locations. Sometimes holding conferences off-campus helps to focus the attention of participants by providing an "escape" from the daily routine. Conferences require more planning and organization than workshops, but can involve more people at once, perhaps providing additional leverage, especially if the goal is to jump-start a new or recently re-invigorated push for accessibility. As with workshops, both face-to-face and online conferences are possible. Online conferences may not be as compelling or motivating as face-to-face conferences, but they can be an effective alternative if done well, especially across multiple campuses.

Invited presentations by outside experts can give weight to the proceedings of conferences or workshops because of their perceived level of expertise, and because participants often appreciate the chance to hear from people outside of the organization, where it may be difficult for web accessibility advocates to be prophets in their own land. On the other hand, outside experts are not familiar with the intricacies of the local system. They can offer advice based on their experience, but that experience may not map directly to the needs of a given organization. There is also the risk of invoking the "not invented here" resistance to proposed changes if participants do not buy into the need for them. Webmasters can hold regular meetings or mini-conferences-perhaps monthly or quarterly-to share ideas of all kinds about effective web site creation. Occasional presentations or workshops about different aspects of web accessibility can be a recurring theme without being the only theme. The advantage to this approach is that it does not separate out web accessibility as a special need for special people that must be treated separately. Web accessibility is integral to effective web site design in many ways, and not just for people with disabilities.

It may be appropriate to create campus-based electronic discussion forums, email list servs, blogs, or other forms of communication within the group. These can reside behind a password-protected intranet, or they can be made public for all to see. Most people enjoy sharing their expertise. Constant, Kiesler, & Sproull (1994), found that "information as expertise belongs to a special category of information that is part of people's identity and is self-expressive." Communication mechanisms that facilitate this kind of natural exchange can diminish the need for formal training events.

Summary

Cultivating and maintaining web accessibility expertise and institutional support in higher education is essential to providing equitable opportunities for students and employees with disabilities. As higher education institutions deepen their dependence on web-based content and interaction, the need for web accessibility grows proportionally. Despite the tendency to think of web accessibility as the exclusive purview of the university webmaster or of a small group of campus technophiles, the vast majority of university students and employees use the web on a regular basis. The responsibilities for ensuring the accessibility of web content are variously distributed throughout the organization of the campus as a whole. Different jobs and roles are responsible for different aspects of web accessibility, and should receive the training and institutional support necessary to perform their various web accessibility functions. Institutions should seek to establish flexible, comprehensive, holistic systems with the long-term goal of sustainability. Throughout the process, participants should strive to remember that the real purpose of web accessibility is not merely legal compliance. The real purpose is to create a more inclusive society in which people with disabilities enjoy the same opportunities and freedoms as the rest of the population.