The mapping excercise between tests in WCAG 2.0 techniques and the new BITV-Test (to all intents and purposes, a German WCAG test) has now been completed.
The new BITV-Test that we have now published as draft is based on a new German accessibility regulation, BITV 2, which will soon come into force. It closely mirrors WCAG 2.0 so that the new BITV-Test can be seen as a German-language WCAG-Test. (Read the overview of the BITV-Test for a desription of the general accessibility evaluation approach.)
The mapping excercise between WCAG 2.0 techniques and failures and the test steps of new BITV-Test was carried out to ensure that the tests provided in the WCAG techniques are actually covered in the new BITV-Test so that related accessibility failures will be caught in testing.
The mapping results are presented in tables. The links to the steps of the BITV-Test in the right column (BITV-Test-Prüfschritt) jump straight to the part of the test procedure that correspond to the respective WCAG technique. For some techniques, additional text is provided to qualify the coverage (for the time being, these texts are all in German).
The new BITV-Test covers the conformance levels A and AA. Level AAA techniques and failures have generally been omitted (with a few exceptions where they also seem to apply to AA level success criteria, such as HTML technique H76 or failure F9).
The presentation of mapping results has one page for each of the different types of WCAG techniques, except for general techniques which are covered on two pages:
The actual procedure in the 50 checkpoints of BITV-Test often deviates from the process described in the tests at the end of WCAG techniques and failures. WCAG techniques test on an atomic level, whereas in the BITV-Test, a number of different techniques are usually covered in the same checkpoint.
A few techniques have not been included either because they have little practical relevance to the German field of application (think of testing for the directionality of text in H34 and H56); are the rather obscure like the test for alternative text for glyphs in F71; or are irrelevant in terms of practical accessibility (an example: whether or not decorative and spacer images are included as background images as required in C9 does not matter as long as images proper are given an empty
The focus on particular WCAG techniques in the tests provided at their end usually does not allow the conclusion that the corresponding success criteria listed in the section applicability have actually been met when the test is successful: they could have equally been met by using a different technique. For the purpose of testing the actual compliance to success criteria, the focus therefore has to be on a comprehensive assessment wether a particular success criterion has been met regardless of the technique used. This is the approach followed in the BITV-Test.
A number of tests prescribe the use of a suitable accessibility checker to verify whether name, role, and value of scripted interface controls is passed on to the accessibility API of the browser. Even when the check returns a positive result, this does not always mean that such interface controls are correctly exposed to older (and still widely used) user agents and assistive technologies.
Die tests included especially in checkpoint 4.1.2a Name und Rolle von Bedienelementen verfügbar (name and role of interface elements exposed) therefore mainly have a heuristic value: they check whether developers of scripted interface controls were aware of the accessibility problem they create and used WAI-ARIA at all to make these elements and widgets accessible also to screen reader users. If WAI-ARIA is not used for scripted interface elements, checkpoint 4.1.2a returns a fail, lowering the overall score.
Checking whether WAI-ARIA has been used correctly, however, is beyond the means of the current BITV-Test. First of all, WAI-ARIA is still a Candidate Recommendation, i.e., it has not been finalised. Also, its implementation across browers and AT is still uneven: Apparently, there are even deviations between the WAI-ARIA spec and the WAI-ARIA Best Practices document. In any case, the practical accessibility of WAI-ARIA cannot be reliably determined without recourse to tests with different makes of screen readers under various system environments and versions. Currently, such tests are not part of the BITV-Test for practical reasons.
If you have any thoughts on this or would like to comment, please use the comments section below. We suggest to follow @wcagtest on Twitter if you want to keep informed about the progress of the development of WCAG-Test.