SEMI-AUTOMATED TEST ENVIROMENT FOR

WEB CONTENT ACCESSIBILITY BASED ON WCAG 2.0

Kathrin Malack, Reiner Dumke, Steffen Mencke

Faculty of Computer Science, Otto-von-Guericke-University of Magdeburg, Magdeburg, Germany

Cornelius Wille

Fachhochschule Bingen, University of Applied Sciences, Bingen, Germany

Keywords: Web content, Web Content Accessibility Guidelines 2.0, WCAG, disabilities, web-based self evaluation,

test tool, WAI.

Abstract: Internet is becoming an important part of the lives of people with disabilities. Currently the Web Content

Accessibility Guidelines 2.0 (WCAG 2.0) has been published as a working draft. To reach the goal of

accessible Web content for all people, including people with disabilities, the Web content has to satisfy the

success criteria. We will present an approach for a semi-automated web based evaluation tool which helps

to fulfil the WCAG 2.0 guidelines. The tool can be used by developers to evaluate how good Web content

fulfils the success criteria. It also allows estimation of effort to change a n existing Web content to fulfil the

WCAG 2.0 guidelines.

1 INTRODUCTION

For a wide range of people with disabilities well

accessible Web content can help to make their lives

comfortable. A study in Germany shows that around

80 percent of disabled people have simple or good

knowledge of using the Internet (Hammer, 2006).

On other side a lot of Web content is only

marginally accessible and a study shows that only

three percent of the web pages have a good

accessibility (“Barrierefreiheit,” 2006).

As part of the Web Accessibility Initiative

(WAI), the W3C has developed Web Content

Accessibility Guidelines. After May 2007 the

working draft of WCAG 2.0 became available. This

paper will present a semi-automated web based tool

that helps to check how good the WCAG 2.0

guidelines and success criteria have been fulfilled by

a given web content. The tool can be used by

designer and developer of Web content to improve

their work. Web content in this context includes text,

images, sounds, video and other. The tool can also

be used to estimate the effort to change an existing

Web content with the goal to fulfil the WCAG 2.0

success criteria. Some international and national

laws and directives ask for Web content to be

accessible to a wide range of people with

disabilities.

2 DISABILITY

To describe what a disability means the World

Health Organisations (WHO) International

Classification of Functioning, Disability and Health

(IFC) can be used. The ICF is a framework for

measuring health and disability at individual and

population level and lists 9 broad domains of

functioning which can be affected (“International

Classification,” 2007). Disability includes blindness,

deafness, low vision, hearing loss, cognitive

limitations, limited movement and other.

2.1 Current Situation in Germany

In Germany the Statistisches Bundesamt provides an

official statistics about the situation of disabled

persons every two years. These statistics only

include disabled persons with a degree of disability

at least 50. At the end of 2005, around 6.7 million

people in Germany are more than 50 percent

disabled. These are 8.2 percent of Germany’s

304

Malack K., Dumke R., Mencke S. and Wille C. (2008).

SEMI-AUTOMATED TEST ENVIROMENT FOR WEB CONTENT ACCESSIBILITY BASED ON WCAG 2.0.

In Proceedings of the Fourth International Conference on Web Information Systems and Technologies, pages 304-307

DOI: 10.5220/0001526803040307

 SciTePress

population (“6,7 Millionen schwerbehinderte

Menschen,” 2006). The number of severely disabled

persons has increased. The reason is the advanced

medicine. So today people can live with diseases,

which were deadly a few years before.

2.2 Types of Handicaps

From accessibility Web content everybody has a

benefit. Of course disabled persons have the major

benefit. They meet on more barriers than non-

disabled persons.

There are different types of handicaps. In a study

from the German Federal Ministry of Family,

Seniors, Women’s and Adolescence types of

handicaps are analysed. The impairment of the

function of the internal organ is the frequently

encountered type of handicap.

From different literature and statistics we can

classify four main types of handicaps (“Gender

Datenreport,” 2005): Body-disability (67 %),

Cognitive disability (8%), Visual impairment,

blindness/colour-ametropia (5%) and Aural-demage

(4%). The values in brackets show the percentages

of this group opposite to all collected data of

severely disabled persons. The group of body-

disability is the largest group. The others are all

about the same.

It is an established fact that 83.5 % of a disability

is caused by diseases and only 4.7 % of a disability

is due to inheritance. 74 percent of severely disabled

persons are 55 years or older.

3 INTERNATIONAL

GUIDELINES

Guidelines and laws for accessibility exist on

international and national level. In March 2007, The

United Nations has published the Convention on the

Rights of Persons with Disabilities.

In the area of Web Content the WAI as a part of

the W3C has developed the Web Content

Accessibility Guidelines. These guidelines explain

how to make Web content accessible to people with

disabilities. The guidelines have met a breadth of

international acceptance and should be used by all

Web content developers. Following the guidelines

makes the Web content more accessible for all users

if whether they are disabled or not.

Under the leadership of the WAI three series of

accessibility guidelines have been developed. The

Authoring Tool Accessibility Guidelines (ATAG)

defines how authoring tools should help developers

produce accessibility Web content. The User Agent

Accessibility Guidelines (UAAG) is primarily for

developers of user agents like Web browser, media

player or assistive technologies. Core of the WCAG

is to explain Web content’s good accessibility.

Figure 1: WAI Guidelines and Techniques (“WAI

Guidelines,”2006).

Figure 1 shows that the accessibility of Web

content is not only the function of a web developer.

Also the developers of authoring and evaluation

tools as well as developer of browsers, media

players and assistive technologies have to account

for compliance with regulations.

Based on the Last Call Working Draft from 27

April 2006 the WCAG 2.0 now includes many new

aspects. Hundreds of comments have been given by

reviewers and currently it is in the state of a

Working Draft. This Working Draft (May 17. 2007)

is the basis for our test tool. The four major

accessibility principles are perceivable, operable,

understandable and robust as shown in figure 2.

These principles are subdivided in 12 guidelines and

the guidelines are further subdivided in 56 success

criteria. The 56 success criteria are organized into

three levels. Single A conformance means, that all

these criteria have to be fulfilled in order that the

Web content satisfies level A. To satisfy

conformance level AA the Web content must fulfil

all single A and all double A criteria. Level AAA is

the highest conformance level and all 56 success

criteria on the three levels have to be satisfied. The

guideline describes that a level does not mean that

some criteria are more important than other. With

the goal of a triple A conformance this is right but

indirectly a weighting is given. On the other side a

AAA conformance does not guarantee that all

people can access the Web content. For example not

SEMI-AUTOMATED TEST ENVIROMENT FOR WEB CONTENT ACCESSIBILITY BASED ON WCAG 2.0

305

all accessibility issues of people with cognitive

disability can be addressed by the guideline.

Figure 2: Overview WCAG principles and success criteria.

Based on the Barrierefreie Informationstechnik-

Verordnung (BITV) all public authorities have to

realize Web content with a high accessibility

(“Verordnung zur Schaffung,” 2002). In the next

chapter we will present an approach for a semi-

automated web based tool which will help to

evaluate Web content under the view of the WCAG

2.0.

4 REALISATION

When WCAG 2.0 has a final state, much Web

content has to be reviewed and adapted. Our web

based SET tool should give developer and designer

semi-automated support. SET tool means free

translated self evaluation and test tool.

WCAG 2.0 describes 56 criteria for creating

accessible Web content. For some of these criteria

automated test tools exist and can be used. But also

success criteria exist which are currently not

automatically testable. So our concept was to

develop a tool for a semi-automated web based test

of Web content. The tool should give descriptions

for the different criteria and reference notes on how

to test these criteria. If automated test tools exist,

they can be used. A description of which test tools

exist will be given by the SET tool. The descriptions

and overviews are for every user available.

Authenticated user additionally can use the SET tool

to evaluate Web content.

They can evaluate the success criteria step-by-

step using the description, references, examples or

automated tools. A quick navigation helps to see

which criteria have been estimated and which not. If

a criterion can not be evaluated it can be marked as

non-applicable. A quick overview like a traffic light

shows at every time which criteria are successfully

evaluated, which criteria are not fulfilled, which

criteria are non-applicable and which criteria are not

tested now.

If all 56 success criteria are tested the user can

produce a report. The report offers generated

diagrams and graphs. These diagrams and graphs

show the result and if conformance level A, level

AA or level AAA is reached. They want also show

which success criteria are not fulfilled. The non-

applicable success criteria will be evaluated like

fulfilled success criteria. In this context non-

applicable criteria means that special technologies

have not been used. For example if no images in a

Web content are used the question after all attributes

can not usefully be answered.

5 EXAMPLES

In this chapter we will present examples of Web

content of public authorities evaluated by using the

SET tool. The first screenshot in figure 6 presents a

radar plot which shows the results of the SET tool

based evaluation. That a lot of criteria can be

visualized at one view is the major benefit of the

radar plot.

Figure 3: Radar plot of level A conformance.

Because analysis is anonymous we can only explain

that the owner of the Web content is a big German

health insurance company.

Figure 3 shows that 12 of the 21 level A criteria

are successfully visualized by the outer circle. The

inner circle shows that 3 criteria are not fulfilled and

the middle circle shows the 6 non-applicable criteria.

To reach the Conformance level A the developers

WEBIST 2008 - International Conference on Web Information Systems and Technologies

306

have to change the content. The SET tool shows

which success criteria fail and it has some support

how to change the content. The radar plot in figure 4

shows the conformance level AAA result.

Figure 4: Radar plot of level AAA conformance.

Figure 5 shows that 19 success criteria from 56 are

not fulfilled and so of course the conformance level

AAA is not reached. On the other side only 21

success criteria have been fulfilled and so that are

only around 50 percent. We can say that this Web

content is not usable for people with disabilities.

Also we can identify that a large effort has to be

done to make the Web content conform to level

AAA.

For a fast overview across all three conformance

levels a bar chart is available. With one view the

user can see how well the analysed Web content

fulfils the different conformance levels. The graph in

figure 5 also gives a feeling how much effort has to

be done to satisfy a conformance level.

6 CONCLUSIONS

Based on international and national laws and

guidelines Web content with a good accessibility

becomes more and more important. In this paper we

have presented an approach for a semi automated

web based test environment. This test environment

should help developer and designer to make their

Web content more accessible. The development of

the tool has been based on the WCAG 2.0 guideline.

Using the tool gives an answer to the question if one

of the conformance levels has been reached.

Figure 5: Bar chart over all conformance level.

Additionally, the developer gets a support by

changing Web content with the goal to satisfy a

special conformance level. He also gets information

on how much effort is required for the change.

The presented tool is helpful not only during the

development of new Web content but is also useful

for calculating the effort to change existing Web

content under the view of a high accessibility.

REFERENCES

Verordnung zur Schaffung barrierefreier Informations-

technik nach dem Behindertengleichstellungsgesetz.

(2002). Retrieved October 14, 2007, from

http://bundesrecht.juris.de/bitv/index.html

Hammer, T. 2006. Surfen ohne Stolperfallen. Retrieved

October 15, 2007, from http://www.zeit.de/2006/

18/Gutes_Internet

International Classification of Functioning, Disability and

Health. (2007). Retrieved October 16, 2007, from

http://www.who.int/classifications/icf/site/

icftemplate.cfm

Barrierefreiheit ist bei den Internetauftritten

mittelständischer Betriebe in Deutschland nur

unzureichend verwirklicht. (2006). Retrieved October

11 2007 from, http://www.indeca.de/ximages/25893

_ind ecapm01.pdf

Gender Datenreport - Kommentierter Datenreport zur

Gleichstellung von Frauen und Männern in der

Bundesrepublik Deutschland /Forschungsbericht.

(2005). Retrieved October 2007 12, from

http://www.bmfsfj.de/Publikationen/

genderreport/9-Behinderung/9-5-Frauen-und-maenner-

mit-behinderung-in-der-amtlichen-statistik/9-5-3-

medizinische-dimension-der-behinderung.html

6,7 Millionen schwerbehinderte Menschen in

Deutschland. (2006). Retrieved October 18, 2007,

from http://www.destatis.de/jetspeed/portal/cms/Sites/

destatis/Internet/DE/Presse/pm/2006/07/PD06__277__

227.psml

WAI Guidelines and Techniques. (2006). Retrieved

October 20, 2007, from http://www.w3.org/WAI/

Web Content Accessibility Guidelines 2.0. (2007).

Retrieved October 21, 2007, from http://

www.w3.org/TR/WCAG20/

SEMI-AUTOMATED TEST ENVIROMENT FOR WEB CONTENT ACCESSIBILITY BASED ON WCAG 2.0

307