Tuesday, 1st August 2006
Sometimes, simply launching a website can take a huge amount of time and effort, never mind making sure it is consistent with every type of web browser and compatible with different types of add-on technologies. However, making websites accessible is an issue that anyone with responsibilities for web development and web-related policy-making should be concerned about.
For one reason, it's estimated that 10 per cent of the population has some kind of a disability -- a number likely to grow as the population ages. Web content isn't always accessible to these people as they may rely on add-on or assistive technologies to access the web, such as screen readers, magnification, voice recognition and alternative mouse devices.
Web accessibility matters to far more than 10 per cent of the population, however. It also affects anyone working with voice browsers, mobile phones, handheld devices or in a hands-free environment. Websites should be for everyone, because if they aren't, organisations cannot reach their communication goals, and individuals can be left behind. The web is how we share information, and you want to be able to share with everyone visiting your site, no matter how they access or interact with it.
Clearly, not everyone is sharing. A study of the accessibility of museum, library and archive websites in the UK revealed that 'less than half the sites audited met the most basic Web Accessibility Initiative technical accessibility guidelines' . Another study, conducted by the UK Disability Rights Commission, found that of 1,000 websites, 81 per cent failed to reach the minimum Web Content Accessibility Guidelines standards . Findings from the UK Cabinet Office assessment of government websites across Europe found 'relatively few sites that achieve even Limited Pass Level A conformance with the W3C Web Content Accessibility Guidelines (WCAG 1.0)' .
Some websites are meeting those guidelines. Examples of good design, particularly for blind and visually impaired people, are provided by the Visionary Design Awards. This campaign to encourage people to make their websites more accessible is run each year by the National Library for the Blind in the UK <http://www.nlbuk.org/>. Winners from 2005-6 include Ocado <http://www.ocado.com/>, RNID <http://www.rnid.org.uk/> and Stable Close Equine Practice <http://www.horsevet.co.uk/>.
As part of the European Commission IST programme of research to support electronic inclusion, accessible web design, development and assessment has been the focus of a Web Accessibility Benchmarking (WAB) cluster of three EU-funded projects. These projects are working to develop a harmonised European methodology for website evaluation and benchmarking, called the Unified Web Accessibility Methodology or UWEM <http://www.wabcluster.org/>. The cluster projects are The European Internet Accessibility Observatory (EIAO) <http://www.eiao.net/>, SupportEAM <http://www.support-eam.org/supporteam/default.asp> and BenToWeb <http://www.bentoweb.org/>.
This article will focus on the work of the European Internet Accessibility Observatory (EIAO).
The European Internet Accessibility Observatory
The European Internet Accessibility Observatory (EIAO) aims to make sites such as these the standard, rather than the exception. This project, launched September 2004, strives to provide policy makers with a tool to observe and improve the effect of accessibility. Web designers, editors and commissioners will have a basis for policy making, research and actions to improve accessibility to web-based content.
The Observatory will consist of the following elements:
The EIAO project is a direct result of a European Commission initiative. Since 2002, the Commission has been working on an action plan to achieve 'an "Information Society for All", promoting an inclusive digital society that provides opportunities for all and minimises the risk of exclusion' . The measures recommended by the commission include Design for All methods, and evaluation and monitoring of the accessibility of websites. These methods are based on the World Wide Web Consortium's Web Accessibility Initiative (WAI) <http://www.w3.org/WAI/>, which include the Web Content Accessibility Guidelines version 1.0, with a view to move to version 2.0. WCAG 1.0 recommendations include providing alternative text for images and graphics, meaningful descriptions for all hypertext links, logical page organisation including correct use of headings and lists, and providing titles and summaries for frames and tables.
The EIAO project is the first project we are aware of that demonstrates how large-scale benchmarking can be implemented using an open-source architecture supporting the integration of third-party evaluation tools (see Ulltveit-Moe, Gjosaeter, and Snaprud (2005) , and Snaprud, Balachandran, and Goodwin Olsen , for more information on the software architecture). The measurements are fully transparent to allow review and encourage improvement. The selected license also facilitates research collaboration with other open-source developers and encourages others to use the software for measurements, teaching, research or any other purpose.
The Observatory assesses a sample of pages from a website and presents the barriers it finds, such as missing alternative text for images or inconsistent page organisation (currently based on WCAG 0.1). Many existing web-accessibility evaluation tools can evaluate one web page and indicate detected barriers for that page. In contrast, the Observatory will present data representing an entire website or a group of websites. An evaluation of one website currently includes 75 tests on 100-200 web pages.
Results, available online, may be a brief report on the accessibility of a particular feature or a full-blown report on the position of a website's accessibility in relation to a sector, region or country. The final version of the prototype Observatory will publish monthly updated measurements from around 10,000 websites.
Results obtained from the first evaluation of five European Prime Minister's websites (ministries in Germany, Norway, Spain, the Netherlands and France), demonstrated the functionality of the Observatory and illustrated the character of its data.
They revealed information about the most-frequently encountered barriers and targeted potential regional differences. Aggregated results showed between 990 and 3,098 barriers per page. To some extent the number of barriers per page can be a result of a difference in design. For instance, one site's pages may have much more content, and thus barriers, than another.
The most frequent violation was the use of obsolete commands, known as deprecated features. For example, using an HTML tag rather than the 'font' element in cascading style sheets.
Lack of style sheets used to control layout and presentation was the second most frequent problem, as defined by the WCAG guidelines. Appropriate use of style sheets will allow users to adjust the appearance of a webpage according to his or her needs, for example, to change the font size or background colour.
Large blocks of text were the third most frequently detected deviation from the guidelines. Dividing larger blocks of text into paragraphs and organising the structure with headers improves accessibility for all users.
Other examples of barriers include pop-ups, missing text-equivalents for non-text elements, such as a text description for a picture, and improper identification of the primary natural language. Failing to comply with this can cause a speech synthesiser to pronounce the text in unintelligible ways, for instance by using an American accent with Norwegian text.
Together, these barriers may render a site impossible to navigate for a user who is visually impaired or working on a scaled-down device such as a phone.
A user-centric approach to improve the observatory
The project has adopted a user-centric approach to development because it recognises the limitations and concerns surrounding automated testing. The tools for automated data collection and dissemination will be developed and continuously improved throughout the project by feedback from users and user testing.
Involving users in the process has provided many benefits. For example, it has been possible to identify the main accessibility barriers experienced by disabled user groups and then to feed this information into the selection and development of the Web Accessibility Metrics. It has also enabled comparisons from user testing and automated testing.
Initially the difference between user and automated results was wide, prompting further development and enhancement of the aggregation methods. Now automated and user testing is more aligned, and more tests and evaluation will continue to improve the Observatory.
This has been an important exercise for the project team, who are keen to address the concerns of stakeholder groups relating to the reliability of automated testing. The team also recognises the fact that the user experience of accessibility does not always match an automated approach. Issues such as these will be addressed through ongoing development, dissemination, promotion and evaluation.
The future of accessibility
Ensuring access is widened to as many people as possible should be a high priority for anyone involved in the delivery of web-based resources, whether it be a web designer, web manager, commissioner or policy maker. Evidence has shown that there is a growing awareness among such people of the importance of accessibility (see for example ENABLED: Analysis of the ENABLED web developer survey <http://digbig.com/4mkkq>, 'Involving Users in the Development of a Web Accessibility Tool' <http://www.ariadne.ac.uk/issue44/craven/intro.html>, SupportEAM <http://www.support-eam.org/> and the recent Minister conference on e-Inclusion in Riga <http://digbig.com/4mkkp>).
The extensive EU programme of projects and activities that fall under the IST e-inclusion programme is helping to improve web accessibility. But, even with these projects, and initiatives such as the Lisbon Strategy (now i2010), the e-accessibility action and the World Wide Web Consortium, web developers often still need practical advice on how best to implement, assess, monitor and maintain accessible website.
The EIAO project is planning to extend beyond the period funded by the European Commission. This may include synergies with e-government benchmarking and training to improve the benchmarked web sites and to increase general awareness.
How you can help
The EIAO project will soon allow online access to the data warehouse via a user interface. As part of the on-going user focus, the project team <http://www.eiao.net/organisation/> will be inviting anyone who has web responsibilities (web design, development, commissioning, policy development) to try out the EIAO user interface and to complete a short online evaluation form, which will enable improvements for the next release version. Keep checking the EIAO website for details and the work of the WAB Cluster.
Currently, there are a number of valuable resources that will help people create and maintain better and more accessible web sites. These include:
 City University (2004). Accessibility of museum, library and archive websites: the MLA audit. London: Centre for Human Computer Interaction Design, City University, 2004. <http://digbig.com/4mkkw> <http://digbig.com/4mkkx>
 Disability Rights Commission (2004). The Web: access and inclusion for disabled people. A formal investigation conducted by the Disability Rights Commission. London: The Stationery Office. <http://www.drc-gb.org/PDF/2.pdf> <http://www.drc-gb.org/Docs/theweb1.rtf>
 Cabinet Office (2005) eAccessibility of public sector services in the European Union. November <http://digbig.com/4mkky>.
 European Commission (2005). Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee, and the Committee of Regions: eAccessibility. COM (2005) 425.
 Ulltveit-Moe, N., Gjosaeter T., Snaprud M., (2005) An open architecture for large scale monitoring of web accessibility., Workshop on Web Accessibility and Metamodelling, Grimstad Norway, April 15 2005.
 Snaprud M.H., N. Ulltveit-Moe, A. Balachandran Pillai, M. Goodwin Olsen: A Proposed Architecture for Large Scale Web Accessibility Assessment. 234-241, ICCHP proceedings, 2006, ISBN 3-540-36020-4.
Related FreePint links:
Related Subscription Articles:
Document the value chain, and transform the way you think about, manage and report on your product portfolio and your information service contributions to your organisation goals.
Focus on Value Chain
Risk assessment is a required process for a healthy information department. It gauges the ability of your services, team, portfolio and overall value to withstand stress.
Focus on Risk Assessment
Sorry, there seems to be a problem with Webinar and Community listings. Please let us know, by email to firstname.lastname@example.org. Thank you.
Our proven processes, resources and guidance will help your team make the shift from transaction centre to strategic asset.
Designed around the most common challenges and pain points for time- and resource-strapped information teams
Supercharge remote productivity and value
Holistic content portfolio management
Future-proof your information service
A tailored overview of our research and active discussion with your Jinfo analyst.
Measure your starting point to articulate your strengths and set priorities for future improvements. Assessments gauge risk, capacity, value and more.
Read case studies, and start the conversation:
Connect your team with the practical tools, original research and expertise to build and support information strategy in your organisation.
A Jinfo Subscription gives access to all Content (articles, reports, webinars) and Community.