1. Executive Summary

The World Wide Web is fast becoming the de facto repository of preference for on-line information, yet the technology of the Web has inadvertently created barriers for people with disabilities. The World Wide Web Consortium (W3C), housed at three internationally recognized research facilities, coordinates the evolution of the Web and has a mission to "realize the full potential of the Web." W3C intends to take a leadership role in removing these accessibility barriers. To that effect, W3C proposes the creation of an International Program Office (IPO) for coordinating five Web-related activities:

  1. Technology development. Centered on protocols and data formats, especially HTML, CSS, SGML, XML, HTTP, PICS and PEP.
  2. Development of tools. In particular, authoring tools that encourage development of content in a format that supports use by people with disabilities.
  3. Guidelines for use of the technology. Guidelines targeted at browser vendors, authoring tool vendors, and content creators.
  4. Educational outreach. For technology, tools, and guidelines to be effective, their users (the application developers, content creators, and hardware designers) must choose to employ them regularly and correctly.
  5. Research and advanced development. User interface design, novel devices, certification tools and labels are all areas where additional work is required before standardization is appropriate.

W3C proposes to combine its own membership funds with those of foundations, industrial sponsors, the U.S. government, the Canadian government, and the European Commission to fund the IPO. The IPO will have its own director and its own budget, but will be housed in and fully integrated with the W3C. In addition, the W3C staff members who coordinate the evolution of the Web protocols will work with the IPO to ensure that the evolution removes, rather than reinforces, accessibility barriers.

At a recent White House meeting called by the U.S. Government (see participant list in Annex), this International Program Office was recommended, and W3C was clearly designated as the ideal host for such a program. Five keywords were put forward to justify this choice: International, Centralized, Consensus, Predictability, and Participation.

  1. Background

The emergence of the World Wide Web has made it possible for individuals with appropriate computer and telecommunications equipment to interact as never before. The Web is the stepping stone, the infrastructure, which will pave the way for next generation interfaces. But the Web is still very early in its development cycle. Careful evolution is essential to prevent either the fragmentation of the Web due to proprietary developments or the inadvertent disenfranchisement of minorities who do not have sufficient market presence to be deemed economically significant.

The World Wide Web Consortium (W3C) was created in 1994 to coordinate evolution of the Web. W3C is a joint project of the Massachusetts Institute of Technology Laboratory for Computer Science (MIT LCS, USA), the Institut National de Recherche en Informatique et en Automatique (INRIA, Europe), and Keio University (Asia). Directed by Tim Berners-Lee (inventor of the Web), W3C’s mission is "to realize the full potential of the Web." With over 150 member companies and organizations world wide, W3C has taken responsibility for the mark-up standards HTML, CSS and XML; metadata format (PICS); negotiation (PEP); and is a major contributor to the evolution of the HTTP transport standard.

Part of the W3C's commitment to realize the full potential of the Web is to promote a high degree of usability for people with disabilities. Unfortunately, the current situation in that area is not very good and is getting worse every day as more and more people rush into the Web business without any awareness of the new limitations and frontiers they may create. No single disability population is unaffected. For example:

Worldwide, there are more than 750 million people with disabilities. A significant percentage of that population is affected by the emergence of the Web, directly or indirectly. For those without disabilities, the Web is a new technology that can help unify geographically dispersed groups. But these barriers put the Web in danger of disenfranchising people with disabilities in this emerging infrastructure.

This is a clear and compelling social argument for undertaking the work described here. But W3C’s member organizations have pointed out that helping people with disabilities isn’t just a social responsibility. It is also just plain good business: it opens new markets and creates new products. While the market for people with disabilities is 750 million people, the market for the illiterate (who will use speech-based interfaces developed for the blind) number over 2 billion. And even those without disabilities will benefit from many of the proposed changes and guidelines. When driving a car, for example, the driver may wish to browse the Web for information (movie schedules, etc.) – and drivers, as far as a browser is concerned, are both blind and physically disabled since their eyes, hands, feet, and legs are otherwise occupied!

  1. W3C: Qualifications and Current Work

Since its inception, the W3C has had an official activity area devoted to accessibility for people with disabilities, thanks to the continuous effort of Mike Paciello of the Yuri Rubinsky Insight Foundation (see http://www.w3.org/pub/WWW/Disabilities). While the members of the Consortium have supported this area, it has not yet received significant resources. It has been treated as an adjunct to other work (primarily the design of HTML and CSS) rather than as a task important in its own right. W3C believes that it is now time to address these problems, and has sent a Briefing Package to its members asking formal permission to invest heavily in this area.

The current proposal is part of that effort, since it is clear that W3C’s own resources are not sufficient to adequately address this area. Our staff expertise lies in a combination of technical design, implementation, and project management. We have experience bringing the major technical companies heading Web development together in a neutral forum and reaching consensus among the industry to

We are also international by nature, with member representatives worldwide and permanent staff in Europe, the Pacific Rim, and the U.S.

The W3C Accessibility activity statement says: "W3C's position is clear: All the protocols and languages we issue as Recommendations should meet or exceed established accessibility goals. In addition, we will actively encourage the development of Web software and content that is accessible to people with most disabilities."

To meet these goals, the W3C staff has recommended to their membership that W3C take on three roles with respect to accessibility:

  1. Act as a central point for setting accessibility goals for the Web. In this role, W3C will perform a liason role, coordinating with external organizations that represent people with disabilities to generate a widely accepted set of goals and guidelines that take into account the needs of the user community, the details of the technology, and engineering realities. W3C already fills this role in several other areas of technology, and this is a logical extension of that role.
  2. Facilitate development of tools. As the internationally acknowledged coordinator of World Wide Web evolution, the W3C acknowledges its responsibility to support the development of tools that help make Web content easily accessible for people with disabilities. As the Web user interface and infrastructure continues to evolve, the W3C will help its members (who create these tools) in their efforts to design and develop the Web in a way that considers the user needs of people with disabilities.
  3. Educate the Web content community. W3C already serves as a neutral party for distributing information about Web technology. W3C would like to extend this education role to be proactive in explaining to content producers how best to use the technology to serve the needs of people with disabilities.
While the formal process of soliciting input from the member companies is not yet complete, there has been strong support for all of these activities in the past and every indication that it will continue. In fact, W3C has chosen to act now because of strong pressure from several individual member companies to focus on this area as quickly as possible. Early efforts to raise funds from industry, including the member companies, continue to encourage this effort.


  1. International Program Office for Web Accessibility

    1. Introduction
    2. The staff of the World Wide Web Consortium (W3C) proposes the creation of an International Program Office (IPO) to motivate, fund, and coordinate work aimed at making the Web accessible to those with disabilities. W3C has asked its members to fund work in this area, but recognizes that an IPO is necessary to capitalize on its own investment as well as excellent related work done elsewhere. One important additional role of this International Program Office will be to work with content creators to ensure that the technical work in this area will be acceptable to them, that guidelines are clear and practical, and that the tools are useable.

      The Consortium’s own resources will be used to create and manage what is formally known as a W3C Project, which includes significant W3C staff resource and has proven effective in unifying the industry to solve other problems of common interest. Ordinarily, a W3C Project contains a pilot that includes significant implementation work to demonstrate feasibility and test design decisions. Instead, we propose an International Program Office to coordinate the W3C Project with other work at individual companies, content providers, national governments, and key disability organizations. The W3C believes that without this coordination and outreach effort its own proposed investment is at risk. Structurally, the W3C’s Briefing Package, sent to the members for approval by late March 1997, asks that they approve the creation of an International Program Office to be organized within W3C itself:

      The International Program Office is … a coordination body for work in this area both inside and outside W3C. The Director of the IPO [reports] to the Director of W3C… As such, the Project Manager for W3C reports to the Director of the IPO and will coordinate work on the W3C Project with external projects sponsored by the IPO. The Director of the IPO will work with the Project Manager to ensure the full participation of all the Disability Organizations that have provided their support to this project …

      We recognize that several efforts, funded by various government agencies and supported by various disability organizations, are already in place. We do not seek to supplant or replace these commendable efforts. Rather, the intent of this proposal is to complement and coordinate those projects, while funding the areas of Web accessibility development that are not currently supported. A critical objective of this proposal is to enable the IPO (and, through it, W3C and its member companies) to act as an important resource for, and partner with, governments and disability organizations worldwide whose mission statements include access to the Web for people with disabilities.

    3. Scope of the IPO's Activities
    4. To adequately address the Accessibility problem a number of organizations that do not normally work together must coordinate their activities. These include the W3C and its member organizations, government agencies and non-profit organizations dealing with people with disabilities, as well as technology companies that directly address this market segment.

      We believe that the work can be divided into five major areas, with careful cross-coordination between the organizations and companies working in each area.

      1. Technology development

This area is centered on Web protocols and data formats, especially HTML, CSS, XML, SGML, HTTP, PICS and PEP. Since the Program Office is intended to concentrate on Web (rather than general computer) accessibility, we do not expect work on physical devices, etc. The work of the W3C, since its inception, has concentrated on precisely these technologies. We see this work as the primary focus of W3C’s internally-funded Project, with the following initial work items:

      1. Development of tools
      2. The key to making the Web accessible is making sure that the content is produced in a way that keeps in mind the needs of people with disabilities. There are two kinds of tools that are critical to making this happen: authoring tools that prompt and remind content creators about the needs of people with disabilities, and tools for checking that standards for accessibility have been met.

        The International Program Office must work with vendors of tools to encourage them to take the needs of people with disabilities into account. While this is a role that W3C has played to a certain extent, W3C holds no stick with which to force members into compliance; and especially so on issues that do not relate to interoperability with other Web agents. An internationally recognized IPO will be significantly more effective than W3C itself.

        The primary initial work item here is the development of a Style Guide for tool developers: items that they should provide in their user interface to make it easy for content creators to "do the right thing" for people with disabilities. Developing this Style Guide (as well as the one mentioned in 3., below) will lead to suggestions for specific tools, which may be integrated into authoring environments, for validating accessibility of content.

      3. Guidelines for use of the technology

There are several existing guidelines for the use of HTML for people with disabilities, but these are rapidly falling behind the technology. There is confusion in the industry because there are several, incompatible, sets of guidelines. The industry needs a mechanism for generating either a single set of guidelines or (at worst) several compatible sets of guidelines, and, most importantly, for keeping the guidelines up-to-date as the technology evolves. These guidelines are not trivial - they depend on a wide range of factors, including

W3C does not, as a general rule, deal with producing style guidelines, concentrating on "mechanism, not policy" and allowing the market to shape the use of our technologies. Its own resources are spent making sure that the mechanisms exist and are adequate for a wide variety of purposes.

In the area of Accessibility, W3C would like to work with the IPO and outside groups to ensure that guidelines reflect the best uses of technology, are up to date, and are adopted and supported by the industry. Without an outside partner (like the IPO) it is hard for W3C to focus on these specific uses of its technology.

      1. Education and outreach
      2. As mentioned above, the primary issue is making sure that Web content is produced in a form accessible to people with disabilities. We do not believe that this process can be completely automatic, even given good authoring tools; it requires attention on the part of the designer to the needs of a community that is all-too-often ignored. The key to success here is a combination of tools that make it easy to do the right thing, and education that reinforces the importance of using the tools routinely and correctly.

        We see three key target groups for the educational effort. There is an immediate need to make the tool vendors aware of existing accessibility guidelines and the importance of supporting the needs of persons with disabilities. This effort is based on a combination of guidelines, social conscience, and existing accessibility regulation. Once appropriate tools are easily available the second educational effort begins – the major content providers must understand the importance of using these tools and following the established guidelines. Finally, with sufficient content available, the mainstream hardware manufacturers will need to be encouraged to understand the market need for accessible components. This will be based on their use by persons without disabilities, partly by showing business cases and partly by arguing from analogy with "curb cuts" – initially seen as an expensive requirement for a minority group, but now seen as a major service for people with shopping carts, etc.

        While we believe that this is an important aspect of the work, it is not something that falls easily within W3C's existing role. Clearly, some of this work should happen as part of the training program that comes with any Web authoring tool. But part of this work goes beyond individual tools, and is part of the traditional role of government: sensitizing the key players (content providers, in this case) to the needs of an important minority population with special needs.

        We understand that this work is also not normally part of the role of NSF projects. The importance of this education and outreach, however, cannot be ignored. We have asked the Department of Education to contribute funds through NSF to support this part of the effort. In addition, W3C has undertaken a fundraising effort that asks industry and charitable foundations to support, in particular, this part of the effort.

      3. Research and advanced development

The other areas of this proposal deal with work that has developed to the point where industry-wide standardization is reasonable. But there is a great deal of work that has not reached this point and which requires funding. Just as an example of this work, the development of automatic certification tools for Web content needs to be encouraged, as does work on making accessibility a goal for scripting interfaces to HTML display tools.

W3C participates in both advanced development and research projects, as do all of its host institutions (INRIA, Keio University, and MIT). This proposal does not request funding for work in this area. Instead, we expect that the IPO will work with traditional research and development groups to create proposals (individual or joint) for this work. If W3C’s fundraising efforts are successful, industry and charitable foundation funds may be available through the IPO to directly fund or match other funds in this area.

    1. Personnel and Schedule

The nature of this proposal is admittedly unusual, since it asks different organizations (private, public, profit-making, and non-profit) to pool their resources at an international level to address a world-wide problem with the evolving electronic infrastructure. This section describes the overall project staffing; the next section contains details about the components that we seek to fund through this proposal.

If all of the fund raising runs to schedule, the overall project will be able to begin by June 30, 1997. Based on our current expectations, some of the funds will be available earlier, allowing some work to begin as early as mid-April 1997. Based on these assumptions, the schedule is as follows:


Work Item

April 6, 1997 (6th Internat’l World Wide Web Conf.)

W3C Project begins. This work will proceed in close collaboration with the existing W3C Working Groups initially, and with the IPO when it is created.
Daniel Dardailler as project manager, supported by W3C funds and initial industrial support

mid-April 1997

W3C begins search for full-time director of the IPO, on the assumption that funding from NSF and others will be available by June 30.
W3C Project team includes 1 FTE staff person in addition to Daniel. W3C funded.

May 1997

Under Daniel’s direction, the initial meetings of the W3C Project teams (which include W3C staff and member company representatives) take place.
Member companies become active.
June 1997
IPO director named with contract to begin July 1.
Corporate sponsor funds available.

July 1997

IPO director in place, initial meetings with W3C management and staff to coordinate W3C Project plans. U.S. funding in place.
NSF, Dept. of Ed funded. Charitable foundation funds available.
September 1997
IPO director chooses subcontractors for educational outreach programs in U.S. and Europe. European funding in place. IPO director now has complete budget and control over non-W3C funds.
Initial technical specifications available from W3C Project participants.
Full funding in place. EC contributes to IPO director and to educational efforts. Graduate students available for prototyping work at MIT (under NSF) and INRIA (under EC) funding.

October 1997

Educational subcontractors ready to produce materials and programs based on initial specifications and guidelines
November 1997
First educational programs launched in U.S. and Europe as a series of monthly regional conferences for tool vendors and content providers.
July 1998
First annual report of the W3C Project and the IPO. Demonstrations of prototypes and shipping products.
December 1998
IPO-sponsored event showing tools and content created and used according to established guidelines.
July 1999
Second annual report of the IPO. Focus moves from technology to guidelines and implementation. W3C Project work moves to deployment phase.
December 1999
W3C Project work completes with demonstration of deployed tools conforming to and testing compliance with established guidelines.
July 2000
Termination of IPO initial funding cycle, final report with demonstration.


  1. Funding the IPO: NSF Component

The work proposed here is too large for funding solely from W3C’s internal funds. Instead, W3C is actively pursuing three independent lines of support: W3C internal funds, Government funds, and funds from industry and charitable foundations. Each of these proposals is self-contained, but the synergy is important:

Notice that the role of the co-Principal Investigators on this project is largely administrative: initially raising the funds and choosing the IPO director, then acting as the formal contact between the IPO and the W3C.

  1. Annexes

    1. Participants at the White House meeting Jan. 6 1997
    2. Tom Kalil, Senior Director to the National Economic Council (White House), and Mike Paciello (Yuri Rubinsky Insight Foundation) arranged this meeting. The intent was to bring together some key players in the industry and academia, and to determine the best way to coordinate existing and planned work. The result of that meeting was a decision to create an International Program Office and to house that office within the World Wide Web Consortium.

      David Capozzi, U.S. Architectural and Transportation Barriers Compliance Board

      Carl Cargill, Netscape Communications Corporation

      Audrey Choi, FCC

      Daniel Dardailler, W3C

      Larry Goldberg, WGBH

      Joseph Hardin, NCSA

      Scott Isaacs, Microsoft Corporation

      Steve Jacobs, NCR

      Tom Kalil, National Economic Council

      Murray Maloney, SoftQuad, Inc./YRIF

      Jim Miller, W3C

      Howard Moses, OSERS

      Mike Paciello, Yuri Rubinsky Insight Foundation

      Dave Raggett, W3C

      Larry Scadden, NSF

      Gary Strong, NSF

      Gregg Vanderheiden, Trace Research and Development Center

      Kate Seelman, NIDRR

    3. Legal Standards That Require Web Access

Providing additional motivation to build accessibility into the Web’s infrastructure are certain legal standards and requirements (current and proposed). Many of these requirements already exist in US and other national laws. There is work in Europe to extend the national laws into a pan-European framework that would, presumably, also be considered for adoption worldwide.

Following is a brief list of existing U.S. laws that affect the accessibility of the World Wide Web for people with disabilities:

    1. References to Existing Reports
    1. Current Players and Critical Partners

The following organizations are some of the key players in the area of web access for people with disabilities. All of these organizations and corporations are deeply involved in Web accessibility development and activities. In order to launch a successful program that is geared towards designing and developing access to the Web for people with disabilities, it is imperative that the IPO establish a collaborative relationship with them.

    1. Some Groups Representing People With Disabilities
    2. The following organizations have been contacted and made aware of this Initiative. Each has expressed their support; we continue to expand this list.
      1. General Disability
      1. Organizations for the Blind/Visually Impaired
      1. Organizations for the Deaf and Hearing Impaired
      1. Organizations for the Mobility Impaired