WCAG 2.0 Baseline Concept

Sunday, 20th November 2005

Author: Gez Lemon

Contents

• Introduction
• Baseline Concept
• Conformance Claims
• "Relied upon", and "Used, but not Relied Upon"
• Scope of a Conformance Claim

Introduction

Guideline 11 of WCAG 1.0 states: Use W3C technologies and guidelines. WCAG 2.0 recognises that other technologies have emerged and are continuing to emerge that are beneficial to the Web, and doesn't want to discourage the use of these technologies. Instead, WCAG 2.0 is based on a set of principles that are technology independent, and capable of incorporating emerging technologies that include non-W3C technologies:

1. Content must be perceivable
2. User interface components in the content must be operable
3. Content and controls must be understandable
4. Content must be robust enough to work with current and future technologies

Rather than list a set of technologies that user-agents can reasonably be assumed to support, WCAG 2.0 introduces the concept of a baseline.

[Back to the contents]

Baseline Concept

The concept of a baseline is that the author, or some higher authority, may specify the minimum that needs to be supported by a user agent when making a conformance claim. It's important to note that the baseline consists of technologies that are assumed to be supported by user-agents, and not specific user agents. This avoids statements like, "best viewed in browser X". Developers are free to use other technologies that are not mentioned in the baseline, providing that the content is still accessible and usable by people whose user agent meets the baseline requirements, but may not support the extended technologies.

The baseline concept is obviously based around trust. It's possible for an organisation to set a baseline that includes technologies that are not widely supported, or more importantly, not widely supported by assistive technology. Under these circumstances, clients, organisations, or regulatory bodies will be required to set baselines that are appropriate for the time.

[Back to the contents]

Conformance Claims

A conformance claim is used to assert a level of conformance with WCAG 2.0, and must include the following components:

1. The date of the claim
2. The guidelines title/version: "Web Content Accessibility Guidelines 2.0"
3. The URI of the guidelines
4. The conformance level satisfied: (Level 1, Level 2, or Level 3)
5. The Baseline used to make the conformance claim, or a URI pointing to the baseline.
6. Scope of the claim

The baseline in the conformance claim may also specify the following optional components:

1. Specific technologies relied upon to create the content
2. Specific technologies used, but not relied upon to create the content
3. A list of user agents that the content has been tested with, including assistive technologies
4. Information about audience assumptions or target audience. This could include language, geographic information, etc, but must not specify physical, sensory, or cognitive requirements

After writing the following section, I asked the working group for clarification on "relied upon", and "used, but not relied upon". I've left my original thoughts, as it's likely someone else will come to the same conclusion as me, but the clarification does make sense of this issue. The clarification is given in the section, ""Relied upon", and "Used, but not Relied Upon".

As the baseline states the minimum set of technologies required for the content, I'm confused about technologies that are "relied upon", and technologies that are "used, but not relied upon". Particularly as the baseline is supposed to include a minimum set of technologies that are assumed to be available in a user agent, but other technologies may be used providing they do not impact the accessibility of the content. This seems to be a contradiction in terms, but will be important to understand in terms of evaluating the accessibility of a website.

If a conformance claim states that CSS is used in the baseline, but doesn't mention that it is relied upon, then it's reasonable to assume that the content will adhere to WCAG 2.0 when style sheets are not supported or disabled in the user agent. In which case, why mention CSS in the baseline? The definition of baseline states that the technologies listed are the minimum required to receive the content by a user agent, but other technologies may be used providing they do not affect the accessibility of the content. The same applies to other technologies, such as scripting. If scripting is used without having an impact on the accessibility of the document, why mention it in the baseline? It doesn't make sense.

It might make sense to mention something that is absolutely necessary for the baseline, like HTML, and then just use CSS, scripting, and other technologies in a way that does not affect the accessibility of the document. If these other technologies do not affect the accessibility, I don't understand the rationale of declaring them in the baseline, as they are not the minimum set of technologies required to meet the conformance claim. It only makes sense to me to mention those items that are assumed to be supported in order to make the conformance claim.

From a device independence point of view, it's always good practice to ensure that content works with certain technologies switched off or unavailable wherever possible. Robustness is an important principle for accessibility, and progressive enhancement and graceful degradation techniques help ensure that content is robust enough to work with current and future technologies. From what I thought I understood of the baseline before the "relied upon", and "used, but not relied upon" constraints were introduced, accessibility issues that might arise from technologies specified in a baseline would be caught and addressed by other guidelines in WCAG 2.0. Personally, I would have more faith that this would be the case if validity was a requirement under the robust principle, but the WCAG working group deem validity to be unimportant for robustness.

[Back to the contents]

"Relied upon", and "Used, but not Relied Upon"

The "relied upon", and "used, but not relied upon" options are included to cater for scenarios where the baseline has been set by a higher authority (government, regulatory body, organisation, etc), but not necessarily used on each URI in the website. It also caters for scenarios where a baseline has been specified for the whole website, but authors want to distinguish which URIs rely on technology specified in the baseline, and those that don't. The "relied upon" option should be specified as a proper subset of the technologies specified in the baseline. If a technology is in the baseline, but not relied upon, then the content would still meet the specified conformance claim if that technology is not supported, or switched off.

[Back to the contents]

Scope of a Conformance Claim

The scope of a conformance claim may be for a whole website, or for a portion of the website. The portion of the website may be specified as a list of URIs that are part of the conformance claim, or might be for the whole website with a list of URIs that are not part of the conformance claim. Each URI that is part of the conformance claim must meet all success criteria for the level of conformance claimed. It is not acceptable for any resource that is part of the conformance claim to scope out certain parts of the resource, such as the navigation.

[Back to the contents]

Category: Accessibility.