W3C

Techniques for User Agent Accessibility Guidelines 1.0

W3C Working Draft 6 December 1999

This version:
http://www.w3.org/WAI/UA/WD-WAI-USERAGENT-TECHS-19991206
(plain text, PostScript, PDF, gzip tar file of HTML, zip archive of HTML)
Latest version:
http://www.w3.org/WAI/UA/WAI-USERAGENT-TECHS
Previous version:
http://www.w3.org/WAI/UA/WD-WAI-USERAGENT-TECHS-19991121
Editors:
Jon Gunderson, University of Illinois at Urbana-Champaign
Ian Jacobs, W3C

Abstract

This document provides techniques for implementing the checkpoints defined in "User Agent Accessibility Guidelines 1.0". These techniques address the accessibility of user interfaces, content rendering, program interfaces, and languages such as the Hypertext Markup Language (HTML), Cascading Style Sheets (CSS) and the Synchronized Multimedia Integration Language (SMIL).

This document is part of a series of accessibility documents published by the Web Accessibility Initiative (WAI).

Status of this document

This section describes the status of this document at the time of its publication. Other documents may supersede this document. The latest status of this document series is maintained at the W3C.

This is a W3C Working Draft for review by W3C Members and other interested parties. It is a draft document and may be updated, replaced or obsoleted by other documents at any time. It is inappropriate to use W3C Working Drafts as reference material or to cite them as other than "work in progress". This is work in progress and does not imply endorsement by, or the consensus of, either W3C or Members of the WAI User Agent (UA) Working Group.

While User Agent Accessibility Guidelines 1.0 strives to be a stable document (as a W3C Recommendation), the current document is expected to evolve as technologies change and content developers discover more effective techniques for designing accessible Web sites and pages.

Please send comments about this document to the public mailing list: w3c-wai-ua@w3.org. Mailing list archives are available on the Web.

This document has been produced as part of the Web Accessibility Initiative. The goals of the WAI UA Working Group are discussed in the WAI UA charter. A list of the UA Working Group participants is available.

A list of current W3C Recommendations and other technical documents can be found at http://www.w3.org/TR.

Table of Contents


1 Introduction

This document suggests techniques for satisfying each requirement of "User Agent Accessibility Guidelines 1.0". It also includes references to other accessibility resources (such as platform-specific software accessibility guidelines) that provide additional information on how a user agent may satisfy each checkpoint. The Techniques provided are informative examples only, and other strategies may be used to meet the checkpoint as well as, or in place of, those discussed. The Techniques Document has been designed to track changes in technology and implementation solutions and is expected to be updated more frequently than the current document.

"User Agent Accessibility Guidelines 1.0" is part of a series of accessibility guidelines published by the Web Accessibility Initiative (WAI) . The series also includes "Web Content Accessibility Guidelines 1.0" [WAI-WEBCONTENT] and "Authoring Tool Accessibility Guidelines 1.0" [WAI-AUTOOLS].

1.1 Document conventions

The following editorial conventions are used throughout this document:

1.2 Priorities

Each checkpoint in this document is assigned a priority that indicates its importance for users with disabilities.

[Priority 1]
This checkpoint must be satisfied by user agents, otherwise one or more groups of users with disabilities will find it impossible to access information. Satisfying this checkpoint is a basic requirement for some individuals to be able to use the Web.
[Priority 2]
This checkpoint should be satisfied by user agents, otherwise one or more groups of users with disabilities will find it difficult to access information. Satisfying this checkpoint will remove significant barriers to Web access.
[Priority 3]
This checkpoint may be satisfied by user agents to make it easier for one or more groups of users with disabilities to access information. Satisfying this checkpoint will improve Web accessibility.

2 User Agent Accessibility Guidelines

This section lists each checkpoint of the Guidelines along with some possible techniques for satisfying it. Each checkpoint also links to larger accessibility topics where appropriate.

Guideline 1. Support input and output device-independence

Checkpoints for user interface accessibility:

1.1 Ensure that every functionality offered through the user interface is available through every input device API used by the user agent. User agents are not required to reimplement low-level functionalities (e.g., for character input or pointer motion) that are inherently bound to a particular API and most naturally accomplished with that API. [Priority 1]
Note. The device-independence required by this checkpoint applies to functionalities described by the other checkpoints in this document unless otherwise stated by individual checkpoints. This checkpoint does not require user agents to use all operating system input device APIs, only to make the software accessible through those they do use.

Techniques:

Operating system and application frameworks provide standard mechanisms for controlling application navigation for standard input devices. In the case of Windows, OS/2, the X Windows System, and MacOS, the window manger provides GUI applications with this information through the messaging queue. In the case of non-GUI applications, the compiler run-time libraries provide standard mechanisms for receiving keyboard input in the case of desktop operating systems. Should you use an application framework such as the Microsoft Foundation Classes, the framework used must support the same standard input mechanisms.

When implementing custom GUI controls do so using the standard input mechanisms defined above. Examples of not using the standard input devices are:


1.2 Use the standard input and output device APIs of the operating system. [Priority 1]
For example, do not directly manipulate the memory associated with information being rendered since screen review utilities, which monitor rendering through the standard APIs, will not work properly.

Techniques:


1.3 Ensure that the user can interact with all active elements in a device-independent manner. [Priority 1]
For example, users who are blind or have physical disabilities must be able to activate text links, the links in a client-side image map, and form controls without a pointing device. Note. This checkpoint is an important special case of checkpoint 1.1.

Techniques:

Refer to checkpoint 1.1 and checkpoint 1.5.

Users must be able to activate text links, form controls, image maps, and other active elements with mouse or keyboard or whatever input device is supported.

For client-side image maps:

Use the document object model to enable device independent activation of elements:


1.4 Ensure that every functionality offered through the user interface is available through the standard keyboard API. [Priority 1]
The keystroke-only command protocol of the user interface should be efficient enough to support production use. Functionalities include being able to show, hide, resize and move graphical viewports created by the user agent. Note. This checkpoint is an important special case of checkpoint 1.1.

Techniques:


1.5 Ensure that all messages to the user (e.g., informational messages, warnings, errors, etc.) are available through all output device APIs used by the user agent. Do not bypass the standard output APIs when rendering information (e.g., for reasons of speed, efficiency, etc.). [Priority 1]
For instance, ensure that information about how much content has been viewed is available through output device APIs. Proportional navigation bars may provide this information graphically, but the information must be available (e.g., as text) to users relying on synthesized speech or Braille output.

Techniques:

Operating system and application frameworks provide standard mechanisms for using standard output devices. In the case of common desktop operating systems such as Windows, OS/2, and MacOS, standard API are provided for writing to the display and the multimedia subsystems.

It is important to also support standard output notification of sound such as notifications found in the Windows control panel for sounds. Windows maps accessibility features to the event caused by generation of these specific system sounds. Accessibility features such as SoundSentry will flash the screen, as appropriate, in response to events that would cause these sounds to play. This enables users with hearing i to use the application.

When implementing standard output:


Guideline 2. Ensure user access to all content

Checkpoints for content accessibility:

2.1 Ensure that the user has access to all content, including alternative equivalents for content. [Priority 1]
Note. Although it is not a requirement that alternative equivalents be available at the same time as primary content, some users may benefit from simultaneous access. For instance, users with low vision may want to view images (even imperfectly) but require a text equivalent for the image to be rendered in a very large size or as speech.

Techniques:


2.2 Render content according to natural language identification. [Priority 1]
For example, render text with the appropriate font family or synthesized speech dictionary. Lay out text in the appropriate direction (right-to-left, left-to-right, etc.) Natural language may be identified by markup (e.g., the "lang" attribute in HTML 4.0 ([HTML40] section 8.1) or "xml:lang" in XML 1.0 ([XML], section 2.12), or by the HTTP Content-Language header ([RFC2616], section 14.12). Refer also to checkpoint 2.9 and checkpoint 5.4.
Note. A user agent may not support all languages. When a user agent supports a language identified by a multipart identifier (e.g., "en-us"), the user agent must indicate in its conformance claim which language it claims to support (e.g., "en" alone, "en-us" alone, or both).

Techniques:

Assistive technologies may recognize different languages and can render content according to language markup defined for a certain part of the document. For instance, a screen reader might change the pronunciation of spoken text according to the language definition. This is usually desired and done according to the capabilities of the tool. Some specialized tools might give some finer user control for the pronunciation as well.

Sometimes the user might also want to know when content contains content in other languages. The user should be able to control rendering of the language change. For instance, the user might choose to hear "language:German" when the language changes to German and "language:default" when it changes back. This may be implemented in CSS 2 with the ':before' and ':after' pseudo-elements ([CSS2], section 5.12.3)

For example, with the following definition in the stylesheet:

    [lang|=es]:before { content: "start Spanish "; }
    [lang|=es]:after { content: " end Spanish"; }

the following HTML example:

<P lang="es" class="Spanish">
 <A href="foo_esp.html" 
    hreflang="es">Esta pagina en español</A></P>

might be spoken "start Spanish _Esta pagina en espanol_ end Spanish". Refer also to information on matching attributes and attribute values useful for language matching in CSS 2 ([CSS2], section 5.8.1).

If users don't want to see or hear blocks of content in another language, allow the user to suggest hiding that content (e.g., with style sheets).


2.3 Provide time-independent access to time-dependent active elements or allow the user to control the timing of changes. [Priority 1]

Techniques:


2.4 When no text equivalent has been supplied, indicate what type of object is present. [Priority 2]

Techniques:


2.5 When a text equivalent for content is explicitly empty (i.e., an empty string), render nothing. [Priority 3]

Checkpoints for user interface accessibility:

2.6 If more than one alternative equivalent is available for content, allow the user to choose from among the alternatives. This includes the choice of viewing no alternatives. [Priority 1]
For example, if a multimedia presentation has several tracks of captions (or subtitles) available (e.g., in different languages, different levels of detail, different reading levels, etc.) allow the user to choose from among them.

Techniques:


2.7 Allow the user to specify that captions and auditory descriptions be rendered at the same time as the associated auditory and visual tracks. [Priority 1]
Note. Respect synchronization cues during rendering.

Techniques:

It is important that captions and auditory descriptions be rendered synchronously with the primary content. This ensures that users with disabilities can use the primary and equivalent content in combination. For example, if a hard-of- hearing user is watching a video and reading captions, it is important for the captions to be synchronized with the audio so that the viewer can use any residual hearing. For audio description, it is crucial that the primary audio track and the auditory description track be kept in sync to avoid having them both play at once, which would reduce the clarity of the presentation.
User agents that play SMIL ([SMIL]) presentations should take advantage of a variety of access features defined in SMIL (refer to "Accessibility Features of SMIL" [SMIL-ACCESS]). A future version of SMIL (known currently as SMIL "Boston") is in development and additional access features may be available when this specification becomes a W3C Recommendation.

As defined in SMIL 1.0, SMIL players should allow users to turn captions on and off by implementing the test attribute system-captions which takes the values "on" and "off." For example, include in the player preferences a way for users to indicate that they wish to view captions, when available. SMIL files with captions available should use the following syntax:

<textstream alt="English captions for My Favorite Movie"
            system-captions="on"
            src="closed-caps.txt"/>

In this case, when the user has requested captions, this textstream should be rendered, and when they have not it should not be rendered.

SMIL 1.0 does not provide a test attribute to control auditory descriptions. However, future versions of SMIL (including SMIL "Boston") are expected to include such a test attribute. Should SMIL "Boston" become a W3C Recommendation, developers should implement it then. A test attribute to turn auditory descriptions on and off should be implemented in parallel to the implementation of the 'system-captions' attribute. Users should be able to indicate the preference to receive auditory description, when content authors make it available, through the standard preferences setting section of user interface.

Another test attribute, 'system-overdub-or-captions' in SMIL 1.0, allows the user to choose between alternate language text or sound. This attribute specifies whether subtitles or overdub should be rendered for people who are watching a presentation where the audio may be in a language they do not understand fluently. This attribute can have two values: "overdub", which selects for substitution of one voice track for another, and "subtitle", which means that the user prefers the display of subtitles. However, this attribute should not be used to determine if users need captions. When both are available, deaf users will prefer to view captions, which contain additional information on music, sound effects, and who is speaking, which are not included in subtitles since those are intended for hearing people.

User agents that play QuickTime movies should provide the user with a way to turn on and off the different tracks embedded in the movie. Authors may use these alternate tracks to provide alternative equivalents for use by viewers with disabilities. The Apple QuickTime player currently provides this feature through the menu item "Enable Tracks."
User agents that play Microsoft Windows Media Object presentations should provide support for Synchronized Accessible Media Interchange (SAMI), a protocol for creating and displaying caption text synchronized with a multimedia presentation. Users should be given a way to indicate their preference for viewing captions. In addition, user agents which play Microsoft Windows Media Object presentations should enable viewers to turn on and off other alternative equivalents, including auditory description and alternate visual tracks.
Other video or animation formats should incorporate similar features. At a minimum, users who are blind and users who are deaf need to be able to turn on and off auditory description and and captions. The interface to set these preferences must be accessible. Information on how to author accessible tracks should be included in documentation about the media player.

2.8 If a technology allows for more than one audio track, allow the user to choose from among tracks. [Priority 1]

Techniques:


2.9 For identified but unsupported natural languages, allow the user to request notification of language changes. [Priority 3]
Note. The user should be able to request no notification of language changes.

Techniques:

A user agent should treat content language as part of contextual information. When the language changes, the user agent should either render the content in the supported language or notify the user of the language change (if configured for notification). Rendering could involve speaking in the designated language in the case of an audio browser or screen reader. If the language was not supported, the language change notification could be spoken in the default language by a screen reader or audio browser.

Language switching for blocks of content may be more helpful than inline language switching. In some language combinations, less than a few words long foreign phrases are often well-integrated in the primary language (e.g., Japanese being the primary and English being the secondary or quoted). In such situations, dynamic switching in in-line level may make the reading sound unnatural, and possibly harder to be understood.

Language information for HTML ("lang", "dir") and the Extensible Markup Language (XML) ("xml:lang") should be made available through the Document Object Model (DOM) ([DOM1], [DOM2]).

User agents may announce language changes using style sheets to generate text (refer to CSS 2 [CSS2] and XSLT [XSLT]) that indicates the change of language.


Guideline 3. Allow the user to turn off inaccessible features

In addition to the techniques below, refer also to the section on user control of style.

Checkpoints for content accessibility:

3.1 Allow the user to turn on and off rendering of background images. [Priority 1]

Techniques:


3.2 Allow the user to turn on and off rendering of background audio. [Priority 1]

Techniques:


3.3 Allow the user to turn on and off rendering of video. [Priority 1]

Techniques:


3.4 When the user agent renders audio natively, allow the user to turn on and off rendering of audio. [Priority 1]
Note. To render audio natively means that the user agent drives the speaker directly.

Techniques:


3.5 Allow the user to turn on and off animated or blinking text. [Priority 1]

Techniques:


3.6 Allow the user to turn on and off animations and blinking images. [Priority 1]

Techniques:


3.7 Allow the user to turn on and off support for scripts and applets. [Priority 1]
Note. This is particularly important for scripts that cause the screen to flicker, since people with photosensitive epilepsy can have seizures triggered by flickering or flashing particularly in the 4 to 59 flashes per second (Hertz) range.

Techniques:

Peak sensitivity to flickering or flashing occurs at 20 Hertz.
Refer to the section on script techniques

3.8 Allow the user to turn on and off rendering of images. [Priority 3]

Techniques:

Refer to techniques for checkpoint 3.1.

3.9 Allow the user to turn on and off author-specified forwards that occur after a time delay and without user intervention. [Priority 3]

Techniques:

Content refresh according to an author-specified time interval can be achieved with the following markup in HTML:

<META http-equiv="refresh" content="60">

The user agent should allow the user to disable this type of content refresh.

Although no HTML specification defines this behavior formally, some user agents support the use of the META element to refresh the current page after a specified number of seconds, with the option of replacing it by a different URI. Instead of this markup, authors should use server-side redirects (with HTTP).

User agents can provide a link to other content rather than changing the content automatically.

User agents may also prompt the user and ask whether to continue with forwards.


For example, when forwarding has been turned off, offer a static link to the target.
3.10 Allow the user to turn on and off automatic content refresh. [Priority 3]
For example, when turned off, indicate to the user that content needs refreshing and allow the user to do so through the user interface.

Guideline 4. Ensure user control over styles

In addition to the techniques below, refer also to the section on user control of style.

Checkpoints for fonts and colors:

4.1 Allow the user to control font family. [Priority 1]

Techniques:


4.2 Allow the user to control the size of text. [Priority 1]
For example, allow the user to specify a font family and style directly through the user interface. Or, allow the user to give preferences through a user style sheet. Or allow the user to magnify text.

Techniques:


4.3 Allow the user to control foreground color. [Priority 1]

Techniques:


4.4 Allow the user to control background color. [Priority 1]

Techniques:


4.5 Allow the user to control selection highlighting (e.g., foreground and background color). [Priority 1]

Techniques:


4.6 Allow the user to control focus highlighting (e.g., foreground and background color). [Priority 1]

Checkpoints for applets and animations:

4.7 Allow the user to control animation rate. [Priority 2]

Checkpoints for video.

4.8 Allow the user to control video frame rates. [Priority 1]
4.9 Allow the user to control the position of captions. [Priority 1]
4.10 Allow the user to start, stop, pause, fast forward, and rewind video. [Priority 2]

Checkpoints for audio:

4.11 Allow the user to control audio playback rate. [Priority 1]
Note. User agents may not be able to control the playback rate for some audio formats.
4.12 When the user agent renders audio natively, allow the user to control the audio volume. [Priority 2]
4.13 Allow the user to start, stop, pause, fast forward, and rewind audio. [Priority 2]

Checkpoints for synthesized speech:

4.14 Allow the user to control synthesized speech playback rate. [Priority 1]

Techniques:


4.15 Allow the user to control synthesized speech volume. [Priority 1]
4.16 Allow the user to control synthesized speech pitch, gender, and other articulation characteristics. [Priority 2]

Checkpoints for user interface accessibility:

4.17 Allow the user to select from available author and user style sheets, including no author or user style sheets. [Priority 1]
Note. The browsers default style sheet is always present.

Techniques:


4.18 Allow the user to control user agent-initiated spawned viewports. [Priority 2]
For example, in HTML, allow the user to control the process of opening a document in a new target frame or a viewport created by author-supplied scripts. In SMIL 1.0, allow the user to control viewports created with show="new". Control may involve prompting the user to confirm or cancel the viewport creation. Users may also want to control the size or position of the viewport and to be able to close the viewport (e.g., with the "back" functionality).

Techniques:

User agents may:

For example, user agents may recognize the HTML construct target="_blank" and spawn the window according to the user's preference.


Guideline 5. Observe system conventions and standard interfaces

Checkpoints for content accessibility:

5.1 Provide accessible APIs to other technologies. [Priority 1]
5.2 Conform to W3C Document Object Model specifications and export interfaces defined by those specifications. [Priority 1]
For example, refer to the DOM ([DOM1], [DOM2]). User agents should export these interfaces using available operating system conventions.

Techniques:

A Document Object Model (DOM) is an interface to a standardized tree structure representation of a document. This interface allows authors to access and modify the document with client-side scripting language (e.g., JavaScript) in a consistent manner across scripting languages. As a standard interface, a DOM makes it easier not just for authors but for assistive technology developers to extract information and render it in ways most suited to the needs of particular users. Information of particular importance to accessibility that must be available through the DOM includes:

User agents should implement W3C DOM Recommendations, including DOM Level 1 [DOM1] and DOM Level 2 [DOM2]]. The W3C Recommendation for DOM Level 1 ([DOM1]) provides access to HTML and XML document information. The DOM Level 2 ([DOM2]) is made of a set of core interfaces to create and manipulate the structure and contents of a document and a set of optional modules. These modules contain specialized interfaces dedicated to XML, HTML, an abstract view, generic stylesheets, Cascading Style Sheets, Events, traversing the document structure, and a Range object.

It is important to note that DOM is designed to be used on a server as well as a client and therefore a lot of user interface-specific information such as screen coordinate information is not relevant and not addressed by the DOM specification.

Assistive technologies also require information about browser highlight mechanisms (e.g., the selection and focus) that may not be available through the W3C DOM.

The DOM Level 1 specification states that "DOM applications may provide additional interfaces and objects not found in this specification and still be considered DOM compliant."

Note. The WAI Protocols and Formats Working Group is focusing its efforts on the DOM as the conduit from which to extract accessibility information and enhance the accessibility of a rendered document through a user agent.


Checkpoints for user interface accessibility:

5.3 Use accessibility resources and conventions of the operating system and supported programming languages, including those for plug-ins and virtual machine environments. [Priority 1]
For instance, if the user agent supports Java applets and provides a Java Virtual Machine to run them, the user agent should support the proper loading and operation of a Java native assistive technology. This assistive technology can provide access to the applet as defined by Java accessibility standards.

Techniques:

The operating system application programming interfaces (APIs) that support accessibility are designed to provide a bridge between the standard user interface supported by the operating system and alternative user interfaces developed by third-party assistive technology vendors to provide access to persons with disabilities. Applications supporting these APIs are therefore generally more compatible with third-party assistive technology.

The User Agent Accessibility Guidelines Working Group strongly recommends using and supporting APIs that improve accessibility and compatibility with third-party assistive technology. Third-party assistive technology can use the accessibility information provided by the APIs to provide an alternative user interface for various disabilities.

The following is an informative list of currently public APIs that promote accessibility:

Many operating systems have built-in accessibility features for improving the usability of the standard operating system by persons with disabilities. When designing software that runs above an underlying operating system, developers should ensure that the application:

  1. Makes use of operating system level features. See the appendix of accessibility features for some common operating systems.
  2. Inherits operating system settings related to accessibility. Pertinent settings include font and color information as well as other pieces of information discussed in this document.

Write output to and take input from standard system APIs rather than directly from hardware controls where possible. This will enable the I/O to be redirected from or to assistive technology devices - for example, screen readers and Braille devices often redirect output (or copy it) to a serial port, while many devices provide character input, or mimic mouse functionality. The use of generic APIs makes this feasible in a way that allows for interoperability of the assistive technology with a range of applications.

User agents should use standard rather than custom controls when designing user agents. Third-party assistive technology developers are more likely able to access standard controls than custom controls. If you must use custom controls, review them for accessibility and compatibility with third-party assistive technology.

For information about rapid access to Microsoft Internet Explorer's DOM through COM, refer to [BHO].


5.4 Provide programmatic read and write access to user agent functionalities and user interface controls. [Priority 1]
For example, ensure that assistive technologies have access to information about the current input configuration so that they can trigger functionalities through keyboard events, mouse events, etc. Refer also to checkpoint 5.3.
5.5 Implement selection and focus mechanisms and make the selection and focus available to users and through APIs. [Priority 1]
Refer also to checkpoint 7.1 and checkpoint 5.4. Note. This checkpoint is an important special case of checkpoint 5.4.
5.6 Provide programmatic notification of changes to content and user interface controls (including selection and focus). [Priority 1]
Refer also to checkpoint 5.3.
5.7 Provide programmatic exchange of information in a timely manner. [Priority 2]
This is important for synchronous alternative renderings and simulation of events.
5.8 Follow operating system conventions and accessibility settings. In particular, follow conventions for user interface design, default keyboard configuration, product installation, and documentation. [Priority 2]
Refer also to checkpoint 10.5.

Techniques:

Develop the UA User Interface (UI) with standard interface components per the target platform(s). Most major operating system platforms provide a series of design and usability guidelines; these should be followed when possible (see platforms below). These checklists, style guides, and human interface guidelines provide very valuable information for developing applications (e.g., UAs) for any platform/operating system/GUI.

For instance, software should use the standard interface for keyboard events rather than working around it.

Evaluate your standard interface components on the target platform against any built in operating system accessibility functions (see Appendix 8) and be sure your UA operates properly with all these functions.

For example, take caution with the following:

Some guidelines for specific platforms:

General guidelines for producing accessible software:

Follow System Conventions for loading Assistive Technologies:

User agents should follow operating system or application environment (e.g., Java) conventions for loading assistive technologies. In the case of Java applets, the browser's Java Virtual Machine should follow the Sun convention for loading an assistive technology. Writing an application that follows the Java system conventions for accessible software does not allow the applet to be accessible if an assistive technology designed for that environment cannot be run to make the applet accessible. Refer to the appendix on loading assistive technologies for DOM access for information about how an assistive technology developer can load its software into a Java Virtual Machine.


Guideline 6. Implement accessible specifications

Checkpoints for content accessibility:

6.1 Implement the accessibility features of supported specifications (markup languages, style sheet languages, metadata languages, graphics formats, etc.). [Priority 1]
Note. The Techniques Document [UA-TECHNIQUES] discusses accessibility features of W3C specifications.

Techniques:


6.2 Conform to W3C specifications when they are appropriate for a task. [Priority 2]
For instance, for markup, implement HTML 4.0 [HTML40] or XML 1.0 [XML]. For style sheets, implement CSS ([CSS1], [CSS2]). For mathematics, implement MathML [MATHML]. For synchronized multimedia, implement SMIL 1.0 [SMIL]. For access to the structure of HTML or XML documents, implement the DOM ([DOM1], [DOM2]). Refer also to checkpoint 5.2.
Note. For reasons of backward compatibility, user agents should continue to support deprecated features of specifications. The current guidelines refer to some deprecated language features that do not necessarily promote accessibility but are widely deployed.

Guideline 7. Provide navigation mechanisms

Checkpoints for user interface accessibility:

7.1 Allow the user to navigate viewports (including frames). [Priority 1]
Note. For example, when all frames of a frameset are displayed side-by-side, allow the user to navigate among them with the keyboard. Or, when frames are accessed or viewed one at a time (e.g., by a text browser or speech synthesizer), provide a list of links to other frames. Navigating into a viewport makes it the current viewport.

Techniques:


7.2 For user agents that offer a browsing history mechanism, when the user returns to a previous view, restore the point of regard in the viewport. [Priority 1]
For example, when users navigate "back" and "forth" among views, for each view they should find the viewport position where they left it.
7.3 Allow the user to navigate just among cells of a table (notably left and right within a row and up and down within a column). [Priority 1]
Note. Navigation techniques include keyboard navigation from cell to cell (e.g., using the arrow keys) and page up/down scrolling. Refer also to checkpoint 1.1 and checkpoint 5.4.

Techniques:

Refer to the section on table navigation.

7.4 Allow the user to navigate all active elements. [Priority 1]
Navigation mechanisms may range from sequential (e.g., sequential navigation) to direct (e.g., by entering link text) to searching on active elements only (e.g., based on form control text, associated labels, or form control names). Note. This checkpoint is an important special case of checkpoint 7.7.
7.5 Allow the user to navigate just among all active elements. [Priority 2]
Note. In checkpoint 7.4, navigation may include non-active elements in addition to active elements.

Techniques:

2.7.1 Sequential navigation techniques

Allow the user to sequential navigate all active elements using a single keystroke. User agents might also provide other sequential navigation mechanisms for particular element types or semantic unit. For example "Find the next table" or "Find the previous form".

It is important that application developers maintain a logical element navigation order. For instance, users may use the keyboard to navigate among elements or element groups and using the arrow keys within a group of elements. One example of a group of elements is a set of radio buttons. Users should be able to navigate to the group of buttons, then be able to select each button in the group. Similarly, allow users to navigate from table to table, but also among the cells within a given table (up, down, left, right, etc.)

2.7.2 Direct navigation techniques

Excessive use of sequential navigation can reduce the usability of software for both disabled and non-disabled users. Direct access (e.g., through keyboard shortcuts) should also be possible. Providing direct access involves:


7.6 Allow the user to search for rendered text content, including text equivalents of visual and auditory content. [Priority 2]

Note. Use operating system conventions for marking the result of a search (e.g., selection and focus).

Techniques:


7.7 Allow the user to navigate according to structure. [Priority 2]
For example, allow the user to navigate familiar elements of a document: paragraphs, tables, headers, lists, etc. Note. Use operating system conventions to indicate navigation progress (e.g., selection and focus).

Techniques:

Skipping navigation bars:

Author-supplied navigation mechanisms such as navigation bars at the top of each page may force users with screen readers or some physical disabilities to wade through numerous links on each page of a site. User agents may facilitate browsing for these users by allowing them to skip recognized navigation bars (e.g., through a configuration option). Some techniques for doing so include:

  1. Provide a functionality to jump to the first non-link content.
  2. In HTML, the MAP element may be used to mark up a navigation bar (even when there is no associated image). Thus, users might ask that MAP elements not be rendered in order to hide links inside the MAP element. Note. Starting in HTML 4.0, the MAP element allows block content, not just AREA elements.

7.8 Allow the user to configure structured navigation. [Priority 3]
For example, allow the user to navigate only paragraphs, or only headers and paragraphs, etc.

Techniques:


Guideline 8. Orient the user

Checkpoints for content accessibility:

8.1 Convey the author-specified purpose of each table and the relationships among the table cells and headers. [Priority 1]
For example, provide information about table headers, how headers relate to cells, table caption and summary information, cell position information, table dimensions, etc. Note. This checkpoint is an important special case of checkpoint 2.1.

Techniques:


8.2 Indicate whether a focused link has been marked up to indicate that following it will involve a fee. [Priority 2]
Note. This checkpoint is an important special case of checkpoint 8.3. "Common Markup for micropayment per-fee-links" [MICROPAYMENT] describes how authors may mark up micropayment information in an interoperable manner. This information may be provided through the standard user interface provided the interface is accessible. Thus, any prompt asking the user to confirm payment must be accessible.

Techniques:


8.3 Provide information to help the user decide whether to follow a focused link. [Priority 2]
Note. Useful information includes: whether the link has already been visited, whether it designates an internal anchor, the type of the target resource, the length and size of an audio or video clip that will be started, and the expected natural language of target resource.

Techniques:


Checkpoints for user interface accessibility:

8.4 Provide a mechanism for highlighting and identifying (through a standard interface where available) the current viewport, selection, and focus. [Priority 1]
Note. This includes highlighting and identifying frames. Refer also to checkpoint 9.1.

Techniques:

Refer also to the section on frame techniques

The following image illustrates the use by Opera 3.6 of a solid line border to indicate focus:

 Example of a solid line border used to indicate the focus in Opera 3.60

The following image illustrates the use of system highlight colors to indicate focus:

Example of system highlight colors used to indicate the focus by the accessible browser project

The following image illustrates the use of a dotted line border to indicate focus:

Example of dotted line border used to indicate the focus in Internet Explorer 5.0 and Netscape Navigator 4.7


8.5 Provide an outline view of a resource built from the resource's structural elements (e.g., frames, headers, lists, forms, tables, etc.) [Priority 2]
For example, for each frame in a frameset, provide a table of contents composed of headers where each entry in the table of contents links to the header in the document.

Techniques:

The following technique ideas were provided by the National Information Standards Organization [NISO]:

A "Navigation Control Center" (NCC) (NCC) resembles a traditional table of contents, but it is more. The NCC contains links to all headings at all levels in the book. In addition to the headings, links to all pages are inserted. Finally we include in the NCC links to all items that the reader may select to turn off for reading. For example, if the reader has the automatic reading of footnotes turned off, there must be a way to quickly get back to that information. For this reason, the reference to the footnote is placed in the NCC and the reader can go to the reference, understand the context for the footnote, and then read the footnote. All items that have the option of turning off automatic reading can be reached through the NCC.

Once the reader is at a desired location and wishes to begin reading, the navigation process changes. Of course, the reader may elect to read sequentially, but often some navigation is required (e.g., frequently people navigate forward or backward one word or character at a time). Moving from one sentence or paragraph at a time is also needed. This type of local navigation is different from the global navigation used to get to the location of what you want to read. It is frequently desirable to move from one block element to the next. For example, moving from a paragraph to the next block element which may be a list, blockquote, or sidebar is the normally expected mechanism for local navigation.


8.6 Allow the user to configure the outline view. [Priority 3]
For example, allow the user to control the level of detail of the outline. Refer also to checkpoint 8.5. Refer also to checkpoint 5.4.
8.7 Allow the user to configure what information about links to present. [Priority 3]
Note. Using color as the only distinguishing factor between visited and unvisited links does not suffice since color may not be perceivable by all users or rendered by all devices. Refer also to checkpoint 8.3.

Techniques:


8.8 Provide a mechanism for highlighting and identifying (through a standard interface where available) active elements. [Priority 3]
Note. User agents may satisfy this checkpoint by implementing the appropriate style sheet mechanisms, such as link highlighting.
8.9 Maintain consistent user agent behavior and default configurations between software releases. Consistency is less important than accessibility and adoption of operating system conventions. [Priority 3]
In particular, make changes conservatively to the layout of user interface controls, behavior of existing functionalities, and default keyboard configuration.

Guideline 9. Notify the user of content and viewport changes

Checkpoints for user interface accessibility:

9.1 Provide information about user agent-initiated content and viewport changes through the user interface and through APIs. [Priority 1]
For example, inform the users when a script causes a popup menu to appear.

Techniques:


9.2 Ensure that when the selection or focus changes, it is in a viewport after the change. [Priority 2]

Techniques:


9.3 Prompt the user to confirm any form submission triggered indirectly, that is by any means other than the user activating an explicit form submit control. [Priority 2]

Techniques:


For example, do not submit a form automatically when a menu option is selected, when all fields of a form have been filled out, on a mouseover event, etc.
9.4 Allow the user to configure notification preferences for common types of content and viewport changes. [Priority 3]
For example, allow the user to choose to be notified (or not) that a script has been executed, that a new viewport has been opened, that a pulldown menu has been opened, that a new frame has received focus, etc.

Techniques:


9.5 When loading content (e.g., document, video clip, audio clip, etc.) indicate what portion of the content has loaded and whether loading has stalled. [Priority 3]

Techniques:

Status information - on resource loading - should be provided in a device-independent manner. Techniques include text and non-text status indicators. Users should be able to request status information or have it rendered automatically. User agents may allow users to configure when status information should be rendered (e.g., by hiding or showing the status bar).

Screen readers may provide access on demand (e.g., through the keyboard) to the most recent status information, or to announce the new information whenever it changes.

Useful status information:

User agents may allow users to configure what status information they want rendered. Allow users to access status information on demand through a keyboard or other shortcut.

Indicate when loading has finished, for example with a percentage indication or a special message. Indication must not depend on a particular output device.


9.6 Indicate the relative position of the viewport in content (e.g., the percentage of an audio or video clip that has been played, the percentage of a Web page that has been viewed, etc.). [Priority 3]
Note. The user agent may calculate the percentage according to focus position, selection position, or viewport position, depending on how the user has been browsing.

Techniques:


Guideline 10. Allow configuration and customization

Checkpoints for user interface accessibility:

10.1 Provide information directly to the user and through APIs about the current user-specified input configuration (e.g., keyboard or voice bindings specified through the user agent's user interface). [Priority 1]

Techniques:

If the currently active configuration changes locally (e.g., a search prompt opens, changing the keyboard mapping for the duration of the prompt), alert the user. The user must also be informed of the changed (currently active) configuration. Do not rely on visual or audio cues alone to alert the user of the change. Do not limit the user to only one type of alert mechanism. Do not rely on visual cues (e.g., the underlining of an accelerator key) alone to inform the user of changes in the configuration.

Named configurations are easier to remember. This is especially important for persons with certain types of cognitive disabilities. For example, if the invocation of a search prompt changes the currently active configuration, the user may remember more easily which keystrokes are active in search mode if alerted that there is a "Search Mode". Context-sensitive help (if available) should reflect the change in mode, and a list of keybindings for the current mode should be readily available to the user.

2.10.1 Documentation of sequential navigation order

2.10.2 Documentation of keyboard shortcuts

2.10.3 Configuration of the user interface


10.2 Provide information directly to the user and through APIs about the current author-specified input configuration (e.g., keyboard bindings specified in content such as by "accesskey" in HTML 4.0). [Priority 2]

Techniques:

Distinguish the following classes of user input configurations:

In association with local (e.g., this page only) and off-default bindings, provide information about how to work around the override.

Note that user support personnel, particularly remote support personnel, will need the "departures from shipping defaults" view for orientation.

The above classes may be distinguished by displayed properties in a combined presentation as well as by filtering to present only a restricted class.

Some reserved keyboard shortcuts are listed in the appendix on accessibility features of some operating systems.

In case of conflicts between author-supplied configuration and user-supplied, operating system defaults, or user agent default configurations, here is some possible behavior:


10.3 Allow the user to change and control the input configuration. Users should be able to activate a functionality with a single-stroke (e.g., single-key, single voice command, etc.). [Priority 2]
For voice-activated browsers, allow the user to modify what voice commands activate functionalities Similarly, allow the user to modify the graphical user interface for quick access to commonly used functionalities (e.g., through buttons).

Techniques:

Provide a convenient interface for allowing users to control input configurations. For example, allow them to select from available options, rather than having them enter combinations themselves. This will speed up configuration and reduce questions to support staff later on how to configure the user agent.

User agents that allow users to customize or reconfigure mappings from keyboard, voice, etc. to user agent functionalities should allow each mapping to be accompanied by a description so that the user can understand the mapping. For example, if "Control-P" maps to a print functionality, a short description would be "Print" or "Print setup".

When using a physical keyboard, some users require single-key access, others require that keys activated in combination be physically close together, while others require that they be spaced physically far apart. When allowing users to configure keyboard access to functionalities, user agents must consider operating system conventions, author-specified shortcuts, and user preferences. The user agent's default configuration should include shortcuts for frequently performed actions and should respect operating system conventions.

To illustrate problems some users with poor motor control encounter and why single-key access is important: Put the keyboard Repeat Delay to very short and the Repeat rate to very fast (or whatever rate is bothersome). The slightest touch or accidental bump will generate a key press, illustrating the challenge key combinations pose.

User agents, to allow the user to turn on and off author-specified keyboard configurations, may offer a checkbox in the keyboard mapping dialog to that would toggle the support for author-specified keyboard configurations. In HTML 4.0, authors may specify keyboard behavior with the "tabindex" and "accesskey" attributes ([HTML40], sections 17.11.1 and 17.11.2, respectively).

Allow users to restore easily the user agent's default configuration.


10.4 Use operating system conventions to indicate the input configuration. [Priority 2]
For example, on some operating systems, if a functionality is available from a menu, the letter of the key that will activate that functionality is underlined. Note. This checkpoint is an important special case of checkpoint 5.8.

Techniques:

In some operating systems, information about shortcuts is rendered visually using an underscore under a character in a menu item or button corresponding to the shortcut key activated with an ALT+character. For menu accelerators the text in the menu item is often followed by a CNTRL+function key. These are conventions used by the Sun Java Foundations Classes [JAVA-TUT] and Microsoft Foundations Classes for Windows.

10.5 Avoid default input configurations that interfere with operating system conventions. [Priority 2]
For example, the default configuration should not include "Alt-F4" or "Control-Alt-Delete" on operating systems where that combination has special meaning to the operating system. In particular, default configurations should not interfere with the mobility access keyboard modifiers reserved for the operating system. Refer also to guideline 5.
10.6 Allow the user to configure the user agent in named profiles that may be shared (by other users or software). [Priority 2]
Users must be able to select from among available profiles or no profile (i.e., the user agent default settings).

Techniques:

Configuration profiles allow individual users to save their user agent settings and re-apply them easily. This is particularly valuable in an environment where several people may use the same machine. Profiles may include rendering preferences as well as user email address, proxy information, stylesheet preferences, etc.

The user should be able to easily transfer profiles between installations of the same user agent. One way to facilitate this is to follow applicable operating system conventions for profiles.

Users should be able to switch rapidly between profiles (or the default settings) and to set a new default profile. This is helpful when:

User agents may apply a profile when the user logs in. They may also allow users to apply settings interactively, for example by allowing them to choose from a list of named profiles in a menu.

Sample profiles (based on common usage scenarios) can assist users in the initial set up of the user agent. These profiles can serve as models and may be copied and fine-tuned to meet an individual's particular needs.

Cascading Style Sheets may be part of a source document or linked externally. Stand-alone style sheets are useful for implementing user profiles in public access computer environments where several people use the same computer. User profiles allow for convenient customization and may be shared by a group.

A text profile may be pasted in email and is useful for technical support purposes.


10.7 Provide default input configurations for frequently performed operations. [Priority 3]
Make the most frequently used commands easy to access. In particular, provide convenient mappings to functionalities that promote accessibility such as navigation of links.

Techniques:


10.8 Allow the user to configure the graphical arrangement of user interface controls. [Priority 3]

Techniques:


Guideline 11. Provide accessible product documentation and help

Checkpoints for user interface accessibility:

11.1 Provide a version of the product documentation that conforms to the Web Content Accessibility Guidelines. [Priority 1]
User agents may provide documentation in many formats, but at least one must be accessible per the Web Content Accessibility Guidelines [WAI-WEBCONTENT]. Alternative equivalents for content, navigation mechanisms, and illustrations will all help make the documentation accessible.

Techniques:

It is essential that any web-based support and/or documentation that is produced or maintained by the manufacturer of a user agent or by a sub-contractor of the user agent's developer, conform to the Web Content Accessibility Guidelines [WAI-WEBCONTENT]. This includes (but is not limited to):

  1. text equivalents of all graphics
  2. extended descriptions of screen-shots, flow-charts, etc.
  3. clear and consistent navigation and search mechanisms
  4. use of the NOFRAMES element when the support/documentation is presented in a FRAMESET
  5. serial listings of keystrokes and keybindings

Accessing documentation in familiar applications is particularly important to users with disabilities who must learn the functionalities of their tools and be able to configure them for their needs. Commonly used applications are also more likely to be compatible with assistive technology. Electronic documentation should not be provided in proprietary formats.

Where possible, allow users to download documentation in a single block (e.g., single HTML file, zip archive of several HTML and image files, etc.)

Run-time help and any Web-based help or support information, as well as the documentation distributed with the tool, must be fully accessible to persons with disabilities. Per checkpoint 1.1, the user must be able to invoke the run-time help with a simple, well documented keystroke command. It is strongly suggested that the keybinding used to invoke the UAs help system be the default "Help" keybinding for the operating system.

Users with print disabilities may need or desire documentation in alternative formats such as Braille (refer to "Braille Formats: Principles of Print to Braille Transcription 1997" [BRAILLEFORMATS]), large print, or audio tape. User agent manufacturers may provide user manuals in alternative formats. Documents in alternative formats can be created by agencies such as Recording for the Blind and Dyslexic and the National Braille Press.

User instructions should be expressed in an input device-independent manner. Provide instructions for using or configuring the user agent in a manner that can be understood by a user of any input device including a mouse or keyboard. For example, "Select the Home button on the toolbar" or "Select Home from the Go menu to return to the Home page."

Universal design means that access to features that help accessibility should be integrated into standard menus. User agents should avoid regrouping access to accessibility features into specialized menus.

Proper documentation is important to developers with disabilities, not just users with disabilities. A disabled user may be a developer using the user agent as a test bed or someone who needs critical information that can only be obtained by as direct a path to the tool's mechanics and "under-the-hood" features as possible. Detailed accessible documentation (e.g., distributed on CD-ROM) is important to allow assistive technology developers access to APIs, etc.

Ensure that product identification codes are accessible to users so they may install their software. Codes printed on product cases will not be accessible to people with visual disabilities.


11.2 Document all user agent features that promote accessibility. [Priority 1]
For example, review the documentation or help system to ensure that it discusses the functionalities addressed by the checkpoints of this document.

Techniques:

Include references to accessibility features in these parts of the documentation:

  1. Indexes. Include terms related to product accessibility in the documentation index (e.g., "accessibility", "disability" or "disabilities").
  2. Tables of Contents. Include terms related to product accessibility in the documentation table of contents (e.g., features that promote accessibility)
  3. Include instructions on how to modify all user configurable defaults and preferences (e.g, images, video, style sheets, and scripts) as specified by the documentation.
  4. Include a list of all keyboard shortcuts or other input configuration information in the accessibility section of the documentation.
  5. Document the features implemented to conform with these Guidelines.

11.3 Document the default input configuration (e.g., default keyboard bindings). [Priority 1]

Techniques:

Here is a table showing mappings between Netscape Navigator functions (or potential functions) and their keyboard shortcuts in Macintosh, Unix, and Windows versions. If a function exists in the browser but does not have a shortcut, its corresponding cell is marked with an asterisk(*). If the function does not exist, it is left blank. Note. This table lists some, but not all, functionalities and keyboard shortcuts of Netscape Navigator. It is meant to illustrate, not serve as definitive documentation.

Some entries contain links to special notes. The number in parentheses following the link is the number of the relevant note.

Linear version of of Netscape Navigator Keyboard Shortcuts.

Netscape Navigator Keyboard Shortcuts
Function Macintosh (v 4.61) Unix (v 4.51) Windows (v 4.7)
Move within a document
Scroll to next page Page Down Page Down Page Down
Scroll to previous page Page Up Page Up Page Up
Scroll to top * * Ctrl-Home
Scroll to bottom * * Ctrl-End
Move between documents
Open a new document Command+L Alt+O Ctrl+O
Stop loading a document Command+. Esc Esc
Refresh a document Command+R Alt+R Ctrl+R
Load previous document Command+[
or
Command+Left Arrow
Alt+Left Arrow Alt+Left Arrow
Load next document Command+]
or
Command+Right Arrow
Alt+Right Arrow Alt+Right Arrow
Navigate elements within a document
Move focus to next frame * * *
Move focus to previous frame * * *
Move focus to next active element (1) Tab Tab Tab
Move focus to previous active element (1) Shift+Tab Shift+Tab Shift+Tab
Find word in page Command+F Alt+F Ctrl+F
Act on HTML elements
Select a link * * Enter
Toggle a check box * * Shift or Enter
Activate radio button * * Shift
Move focus to next item in an option box * * Down Arrow or Right Arrow
Move focus to previous item in an option box * * Up Arrow or Left Arrow
Select item in an option box * * Enter
Press a button (2) Return Enter Enter
Navigate menus
Activate menu * * Alt+ the underlined letter in the menu title
Deactivate menu * Esc Esc
Move focus to next menu item * * (3) Down Arrow
Move focus to previous menu item * * (3) Up Arrow
Select menu item * underlined letter in the menu item Enter
Move focus to submenu * * (3) Right Arrow
Move focus to main menu * * (3) Left Arrow
Navigate bookmarks
View bookmarks menu * (4) * Alt+C+B
Move focus to next item in bookmarks menu Down Arrow (4) * Down Arrow
Move focus to previous item in bookmarks menu Up Arrow (4) * Up Arrow
Select item in bookmarks menu Return (4) * Enter
Add bookmark Command+D Alt+K Ctrl+D
Edit bookmarks Command+B Alt+B Ctrl+B
Delete current bookmark (5) Delete Alt+D Delete
Navigate history list
View history list Command+H Alt+H Ctrl+H
Move focus to next item in history list * * Down Arrow
Move focus to previous item in history list * * Up Arrow
Move focus to first item in history list * * Left Arrow
Select item in history list * * Enter (6)
Close history list Command+W Alt+W Ctrl+W
Define view
Increase font size (7) Shift+Command+] Alt+] Ctrl+]
Decrease font size (7) Shift+Command+[ Alt+[ Ctrl+[
Change font color * * *
Change background color * * *
Turn off author-defined style sheets * * *
Turn on user-defined style sheets (8) ? ? ?
Apply next user-defined style sheet ? ? ?
Apply previous user-defined style sheet ? ? ?
Other functionalities
Access to documentation * * *

Notes.

  1. In Windows, active elements can be links, text entry boxes, buttons, checkboxes, radio buttons, etc. In Unix and Macintosh, Tab cycles through text entry boxes only.
  2. In Windows, this works for any button, since any button can gain the focus using keyboard commands. In Unix and Macintosh, this only applies to the "Submit" button following a text entry.
  3. In Unix, the menus can not be opened with shortcut keys. However, once a menu is opened it stays opened until it is explicitly closed, which means that the menus can still be used with shortcut keys to some extent. Sometimes left and right arrows move between menus and up and down arrows move within menus, but this does not seem to work consistently, even within a single session.
  4. In Macintosh, you can not explicitly view the bookmarks menu. However, if you choose "Edit Bookmarks", which does have a keyboard shortcut, you can then navigate through the bookmarks and open bookmarked documents in the current window.
  5. To delete a bookmark you must first choose "Edit Bookmarks" and then move the focus to the bookmark you want to delete.
  6. In Windows, when you open a link from the history menu using Enter, the document opens in a new window.
  7. All three systems have menu items (and corresponding shortcut keys) meant to allow the user to change the font size. However, the menu items are consistently inactive in both Macintosh and Unix. The user seems to be able to actually change the font sizes only in Windows.
  8. It is important to allow users to set their own cascading style sheets (css). Although Netscape does currently allow the user to override the author's choice of foreground color, background color, font, and font size, it does not allow some of the advanced capabilities that make cascading style sheets so powerful. For example, a blind user may want to save a series of style sheets which show only headers, only links, etc., and then view the same page using some or all of these style sheets in order to orient himself to the contents and organization of the page before reading any of the actual content.

11.4 In a dedicated section, document all features of the user agent that promote accessibility. [Priority 2]
For example, list all pertinent features in a chapter entitled "Accessibility Features". Note that this is a more specific requirement than checkpoint 11.2.

Techniques:

When providing, in a dedicated section, documentation for all features of the user agent that promote accessibility, it is also essential that such information be clearly and obviously linked to other sections of the documentation, where appropriate. A dedicated section documenting the features of the user agent that promote accessibility, however, must not be the only method of documenting such features.


3 Accessibility Topics

This section introduces some general techniques to promote accessibility in user agent functionality. A list of assistive technologies and browsers [USERAGENTS] designed for accessibility is available at the WAI Web site.

3.1 Access to content

Users must have access to document content, however they are browsing. Content can come from:

The most basic way to give users access to content is to render the entire document in one stream, whether it be a two-dimensional graphical layout, audio stream, or line-by-line Braille stream). However, user agents should do much more to ensure that users can understand a page by:

These topics are addressed below.

3.1.1 Preserve and provide structure

Retain structure when rendering. For example, a graphical rendering of tables conveys relationships among cells. Serial renderings (e.g., to speech) must also make those relationships apparent, otherwise users will not know where a table cell ends, or a list item, etc. One technique for maintaining structure is to precede content with "header" information (upon user demand). For example, give the position of a table cell or it's associated headers. Or indicate the position of a list item within nested lists.

Provide "intelligent" structure that may not be exactly what the DTD says. For instance, in HTML, header elements do not nest, but presenting the document as hierarchical may give users a better sense of document structure. Use common idioms where known, even if they are not expressly in the DTD.

3.1.2 Allow access to selected content

In the Amaya browser [AMAYA], users may access attribute values as follows: Place the cursor at the element in question, open/swap to the structure view. You are shown list of attributes and values. Another technique: select the element (press escape in Linux), then the attributes are all available from the attributes menu. For "alt", one can also look at the alternate view, which renders text equivalents instead of images - a lynx-like view. All the views are synchronized for navigation (and for editing).

Users may want to select content based on the rendering structure alone (i.e., that amounts to selecting across element borders).

Users may want to select content based on structure (e.g., a table cell). Amaya allows users to "climb" the document tree by positioning the cursor and pressing the Escape key. Each Escape selects one node higher in the document tree, up to the root.

3.1.3 Access to alternative equivalents of content

Speech-based user agents providing accessible solutions for images should, by default, provide no information about images for which the author has provided no text equivalent. The reason for this is that the image will clutter the user's view with unusable information adding to the confusion. In the case of an speech rendering, nothing should be spoken for the image element. This user should be able to turn off this option to find out what images were inaccessible so that the content author could be contacted to correct the problem.

In the case of videos, an assistive technology should, by default, notify the user that a video exists as this will likely result in the launch of a plug-in. In the case of a video, user agents should indicate what type of video it is, accompanied by any associated alternative equivalent. User agents should prefer plug-ins that support system-specific accessibility features over those that don't.

In the case of applets, an assistive technology should, by default, notify the user that an applet exists, as this will likely result in the launch of an associated plug-in or browser specific Java Virtual Machine. In the case of an applet, the notification should include any associated alternative equivalents. This is especially important since applets typically do provide an application frame that would provide application title information.

When an applet is loaded, it should support the Java system conventions for loading an assistive technology (refer to the appendix on loading assistive technologies for DOM access). When the applet receives focus, the browser user agent should first notify the user about the applet as described in the previous paragraph and turn control over to the assistive technology that provides access to the Java applet.

Suppose an object with a preferred geometry is specified and not rendered, should the alternative equivalent be rendered in the preferred (but empty) region? What if the alternative equivalent exceeds the size of the preferred geometry? One option is to allow the user to specify through the UI whether to respect the preferred geometries or ignore them.

3.1.4 Context

In addition to providing information about content, user agents must provide contextual information. For example:

User agents can use style sheet languages such as CSS 2 [CSS2] and XSLT [XSLT] to generate contextual information.

3.2 User control of style

To ensure accessibility, users must have final control over certain renderings.

Implement CSS ([CSS1], [CSS2]) including the CSS 2 cascade order (section 6.4.1) and user style sheets. The CSS 2 cascade order ensures that user style sheets with "!important" (section 6.4.2) take precedence over author styles, giving users final control. Style sheets give authors design flexibility while offering users final control over presentation (refer also to Web Content Accessibility Guidelines 1.0 checkpoint 3.3 ([WAI-WEBCONTENT]). CSS should be implemented by user agents that implement CSS for text that it renders. CSS includes properties for audio, Braille (fixed and refreshable), screen, and print rendering, and all relevant properties for supported output media should be implemented.

Note that in the CSS cascade order, markup is given less weight than style sheet rules. Thus, an author may use both presentation markup and style sheets, and user agents that support style sheets will prefer the latter.

A user style sheet can be implemented through a user interface, which means that the user may not have to understand how to write style sheets; they are generated or the user agent acts as though they were. For an example of this, refer to the style sheets implementation of Amaya [AMAYA], which provides a GUI-based interface to create and apply internal style sheets. The same technique could be used to control a user style sheet.

For images, applets, and animations:

Background images may be controlled by the use of local style sheets, and more effectively if these can be dynamically updated. Animation rate depends on the players used. User agents that provide native rendering of animation (for example a movie player, a driver for animated GIF images, or a java machine) should enable the control of animation rates, or at least allow the user to stop, and to play frame-by-frame, as well as straight rendering. A user agent could provide control of the general timing of a presentation, combined with the ability to select from available tracks manually. An issue to bear in mind is that when animation is synchronized with audio, a user may need the ability to play the animation separately from the associated audio.

For time-based presentations:

Implement user controls to start, atop, rewind and pause presentations, and where multiple tracks are supported, to choose which tracks should be rendered. SMIL 1.0 [SMIL] provides for a number of these features. A SMIL implementation should provide for direct user control, as well as activation of the controls through a published API, for developers of assistive technologies.

For user agents rendering audio:

On selecting from among available description tracks. SMIL 1.0 [SMIL] allows users to specify captions in different languages. By setting language preferences in the SMIL player, users may access captions (or audio) in different languages.

The G2 player from Real Networks currently allows users to specify which language they prefer, which can be evaluated in a SMIL document to choose from among text or audio tracks. Currently only one language can be indicated which does not permit choosing, for example, English spoken audio with Spanish captions.

The Quicktime player currently permits turning on and off any number of tracks individually, which can include audio, video, and text.

For user agents rendering video:

Implement the CSS positioning and/or SMIL layout languages. Allow the user to freeze a presentation, manually move and resize component visual tracks (including captions, subtitles and signed translations) and to apply CSS stylesheets to text-based presentation and SVG.

For user agents rendering speech:

CSS 2 aural style sheet properties ([CSS2], section 19) can allow users to control speech rate, volume, and pitch. These can be implemented by allowing the user to write and apply a local style sheet, or can be automatically generated by means of (accessible) user controls, which should also be controllable through an API.

User interface:

3.3 Link techniques

Lynx allows the user to choose from the following options for images without "alt" supplied:

The same technique might be used when "alt" is supplied but whitespace only. However, if an image with empty or whitespace "alt" text is part of a link:

Lynx [LYNX] numbers each link and other element and provides information about the relative position of the section of the document. Position is relative to the current page and the number of the current page out of all pages. Each page usually has 24 lines.

Information about link status and other properties can be provided in an information view such as that provided by Netscape Navigator about how many and what types of elements are in a document.

User agents should not consider that all local links (to anchors in the same page) have been visited when the page has been visited.

User agents may use graphical or aural icons to indicate visited links or broken links.

Users should be able to:

3.4 List techniques

Ordered lists help non-visual users navigate. Non-visual users may "get lost" in lists, especially in nested lists and those that do not indicate the specific nest level for each list item. Until user agents provide a means to identify list context clearly (e.g., by supporting the ':before' pseudo-element in CSS2), content developers should include contextual clues in their lists.

For numbered lists, compound numbers (or letters, numbers, etc.) are more informative than simple numbers. Thus, a list numbered "1, 1.1, 1.2, 1.2.1, 1.3, 2, 2.1," provides more context than the same list without compound numbers, which might be formatted as follows:

1.
  1.
  2.
    1.
  3.
2.
  1.

and would be spoken as "1, 1, 2, 1, 2, 3, 2, 1", conveying no information about list depth.

CSS ([CSS1], [CSS2]) allow users to control number styles (for all list, not just ordered) through user style sheets.

Example.

The following CSS 2 style sheet (taken from CSS 2, section 12.5) shows how to specify compound numbers for nested lists created with either UL or OL elements. Items are numbered as "1", "1.1", "1.1.1", etc.

<STYLE type="text/css">
   UL, OL { counter-reset: item }
   LI { display: block }
   LI:before { content: counters(item, "."); counter-increment: item }
</STYLE>

End example.

Even without CSS, user agents may allow users to turn on and off contextual rendering of nested lists.

3.5 Table techniques

Tables were designed to structure relationships among data. In graphical media, tables are often rendered on a two-dimensional grid, but this is just one possible interpretation of the data. On the Web, the HTML TABLE element has been used more often than not to achieve a formatting effect ("layout tables") rather than as a way to structure true tabular data ("data tables").

Layout tables cause problems for some screen readers and when rendered, confuse users. Even data tables can be difficult to understand for users that browse in essentially one dimension, i.e. for whom tables are rendered serially. The content of any table cell that visually wraps onto more than one line can be a problem. If only one cell has content that wraps, there is less problem if it is in the last column. Large tables pose particular problems since remembering cell position and header information becomes more difficult as the table grows.

User agents facilitate browsing by providing access to specific table cells and their associated header information. How headers are associated with table cells is markup language-dependent.

Tabular navigation is required by people with visual disabilities and some types of learning disabilities to determine the content of a particular cell and spatial relationships between cells (which may convey information). If table navigation is not available users with some types of visual disabilities and learning disabilities may not be able to understand the purpose of a table or table cell.

3.5.1 Table rendering

A linear view of tables -- cells presented row by row or column by column -- can be useful, but generally only for simple tables. Where more complex structures are designed, allowing for the reading of a whole column from header downward is important as is carrying the ability to perceive which header belongs to which column or group of columns if more than one is spanned by that header. It is important for whole cells to be made available as chunks of data in a logical form. It might be that a header spans several cells so the header associated with that cell is part of the document chunk for that and each of the other cells spanned by that header. Inside the cell, order is important. It must be possible to understand what the relationships of the items in a cell are to each other.

Properly constructed data tables generally have distinct TH head cells and TD data cells. The TD cell content gains implicit identification from TH cells in the same column and/or row.

For layout tables, a user agent can assist the reader by indicating that no relationships among cells should be expected. Authors should not use TH cells just for their formatting purpose in layout tables, as those TH cells imply that some TD cells should gain meaning from the TH cell content.

When a table is "read" from the screen, the contents of multiline cells may become intermingled. For example, consider the following table:

This is the top left cell    This is the top right cell 
of the table.                of the table.

This is the bottom left      This is the bottom right 
cell of the table.           cell of the table.

If read directly from the screen, this table might be rendered as "This is the top left cell This is the top right cell", which would be confusing to the user.

A user agent should provide a means of determining the contents of cells as discrete from neighboring cells, regardless of the size and formatting of the cells. This information is made available through the DOM [DOM1]).

3.5.2 Cell rendering

Non-graphical rendering of information by a browser or an assistive technology working through a browser will generally not render more than a single cell, or a few adjacent cells at a time. Because of this, the location of a cell of interest within a large table may be difficult to determine for the users of non-graphical rendering.

In order to provide equivalent access to these users, compliant browsers should provide a means of determining the row and column coordinates of the cell having the selection via input (e.g., keyboard) commands. Additionally, to allow the user of a non-graphical rendering technology to return to a cell, the browser should allow a means of moving the selection to a cell based on its row and column coordinates.

At the time the user enters a table, or while the selection is located within a table, the user agent should allow an assistive technology to provide information to the user regarding the dimensions (in rows and columns) of the table. This information, in combination with the summary, title, and caption, can allow the user with a disability to quickly decide whether to explore the table of skip over it.

Dimensions is an appropriate term, though dimensions needn't be constants. For example a table description could read: "4 columns for 4 rows with 2 header rows. In those 2 header rows the first two columns have "colspan=2". The last two columns have a common header and two subheads. The first column, after the first two rows, contains the row headers.

Some parts of a table may have 2 dimensions, others three, others four, etc. Dimensionality higher than 2 are projected onto 2 in a table presentation.

The contents of a cell in a data table are generally only comprehensible in context (i.e., with associated header information, row/column position, neighboring cell information etc.). User agents provide users with header information and other contextual information. Techniques for rendering cells include:

3.5.3 Cell header algorithm

User agents should use the HTML 4.0 algorithm to calculate header information ([HTML40], section 11.4.3).

Since not all tables are designed with the header information, user agents should provide, as an option, a "best guess" of the header information for a cell. Note that data tables may be organized top-to-bottom, bottom-to-top, right-to-left, and left-to-right, so user agents should consider all edge rows when seeking header information.

Some repair strategies for finding header information include:

The user may choose the form and amount of this information, possibly announcing the row heads only once and then the column head or its abbreviation ("abbr") to announce the cell content.

Issues to consider:

  1. TH cells on both the left and right of the table need to be considered.
  2. For TH cells with "rowspan" set: the content of those TH cells must be considered for each of the N-1 rows below the one containing that TH content.
  3. An internal TH in a row surrounded on either side by TDs has no means to specify to which (row or column) that TH overrides what existed to its left or above it.
  4. Finding column header cells assumes they are all above the TD cell to which they apply.
  5. A TH with "colspan" set needs to be included in the list of TH for the M-1 columns to the right of the column in which the TH is found.

If the user agent is taking a guess at header information, the user agent might find two or more possibilities. Provide the user with a mechanism to review the possible choices and make a selection.

3.5.4 Table metadata

Users of screen readers or other serial access devices cannot easily glean information about a page "at a glance". This is particularly difficult when accessing two-dimensional tables and trying to determine their content. Therefore, contextual information about tables (available from author-supplied markup or generated by the user agent) is very important to making them accessible.

Text metadata about tables can come from a number of elements, attributes, the structure of the table itself, or other sources. Useful information to make available to users includes:

The user should be able to get table summary information from inside a cell. It might be helpful to provide two types of table summary information, i.e. a brief summary and a more detailed summary.

3.5.5 Table navigation

All users should be able to determine quickly the nature and purpose of a table. Examining the table visually often conveys a sense of the table contents with a quick scan of the cells. Users with blindness or low vision, users who have difficulty translating printed material, or users in an eyes-busy or speech-based environment may not able to do this. Table summary information can convey the nature of a table; in HTML, summary information for tables comes from the "summary" attribute on the TABLE element as well as the CAPTION element.

An auditory rendering agent, when the point-of-regard moves to a table, might say, "Table: Tax tables for 1998," thus identifying the nature of the table. The user could then use keyboard commands to move the selection to the next logical block of information, or use a different command to "burrow" into the table.

The "burrow" command should have an opposite "up" command, which would move the selection from an individual cell to the table as a whole, so that the user can leave a table from any cell within it, rather than navigating to the end.

If the user moves the focus up to look over the summary information, it should be possible to burrow back to the same cell.

When navigating a table that contains another table, this strategy can avoid confusion. For example, if each row of a table contained five cells, but the second row contained a 4x4 table in the third cell, a user could be disoriented when the row did not end as expected. However, when the selection moved to the third cell of the table, a compliant browser would report that this was a table, and describe its contents. The user would have the option of navigating to the fourth cell of the parent table, or burrowing into the table within this cell.

When rendering tabular information, the fact that it is tabular information should be apparent. For a graphical user agent, such information is commonly made obvious by the border attribute or by visually apparent aligned white space between columns. However, for a non-graphical agent, such information must also be made evident.

As the user agent shifts the selection to a table, it should first allow users to access summary information about the table (e.g., the CAPTION element or the "summary" attribute in HTML). Access to this information allows the user to determine whether or not to examine the contents of the table, or to move the selection to the next block of content. Users should be able to choose not to have the summary information presented, if, for example, they visit a table frequently and don't want to hear the summary information repeated each time.

In many data tables, the meaning of the contents of a cell are related to the contents of adjacent cells. For example, in a table of sales figures, the sales for the current quarter might be best understood in relation to the sales for the previous quarter, located in the adjacent cell.

In order to provide access to contextual information for individuals using non-graphical browsers, or for individuals with certain types of learning disabilities, it is necessary for the user agent to allow the selection to be moved from cell to cell, both right/left and up/down via keyboard commands. The UA should inform the user when navigation has led to a table edge.

The most direct method of performing such navigation would be via the cursor keys, though other navigation strategies might be used.

Users of graphical browsers can easily locate cells within a table that are at the intersection of a row and column of interest. To provide equivalent access to users of non-graphical browsers, equivalent means of navigation should be provided. The search function of a browser will allow the user to locate key terms within a table, but will not allow the user to find cells that are at the intersection of rows and columns of interest.

A rich set of navigation functions could include:

In some tables, cells may span more than one row or column and this affects navigation techniques. For example, if you navigate up into a cell which spans three columns, which cell above the span cell should you go into if you go up another cell? Or what happens if you are in the last cell of a row and the previous row has fewer columns?

3.5.6 Table search techniques

The header information visible in a TH cell may be abbreviated, in which case it should be user preference to see the "abbr" value if any or the full contents.

Axis information may also help the user search into confined portions of the table.

Column groups and row groups are other confining partitions of a table in which a search may be limited.

Software:

3.6 Frame techniques

Frames were originally designed for use by graphical user interfaces to allow the graphical viewport to be broken up into pieces that could change independently (e.g,. selecting an entry in a table of contents in one frame changes the contents of a second frame). However Frames can pose problems users who rely on synthesized speech, refreshable Braille, and magnified views. The ability to access frame alternatives is also important for some users with cognitive disabilities. Problems include:

To help users, user agents should:

To name frames in HTML, use:

  1. The "title" attribute on FRAME, or if not present,
  2. The "name" attribute on FRAME, or if not present,
  3. Title information of the referenced frame source (e.g., the TITLE element of the source HTML document), or
  4. Title information of the referenced long description (e.g., what "longdesc" refers to in HTML), or
  5. Frame context (e.g., "Frame 2.1.3" to indicate the path to this frame in nested framesets).

Users may also use information about the number of images and words in the frame to guess the purpose of the frame. For example, few images and few words is probably a title, more words is probably an index, many words is probably text area.

Frame structure information should be available through the DOM and appropriate accessibility interfaces. Using DOM and operating specific accessibility API to expose frame information provides one means for assistive technologies to provide alternative control of frames and rendering of frame information. The user agent should fully implement the DOM Level 1 Recommendation [DOM1] API related to frames: HTMLFrameSetElement, HTMLFrameElement, and HTMLIFrameElement.

For people with visual disabilities who enlarge text on the screen to improve readability, frames become distorted and unusable. Other users with cognitive disabilities sometimes become disoriented in complex side-by-side frame configurations. To improve access to frames, user agents should allow frames to be viewed as a list so the user can identify the number of frames and the functions of each frame. If NOFRAMES information is present it should also be rendered so the user can optionally use that view of the information.

Consider renderings of the following document:

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Frameset//EN">
<HTML lang="en">
<HEAD>
  <META http-equiv="Content-Type" 
           content="text/html; charset=iso-8859-1">
  <TITLE>Time Value of Money</TITLE>
</HEAD>

<FRAMESET COLS="*, 388">
  <FRAMESET ROWS="51, *">
    <FRAME src="sizebtn" marginheight="5" marginwidth="1" 
       name="Size buttons" title="Size buttons">
    <FRAME src="outlinec" marginheight="4" marginwidth="4" 
       name="Presentation Outline" 
           title="Presentation Outline">
  </FRAMESET>

  <FRAMESET ROWS="51, 280, *">
    <FRAME src="navbtn" marginheight="5" marginwidth="1" 
       name="Navigation buttons" 
       title="Navigation buttons">
    <FRAME src="slide001" marginheight="0" marginwidth="0" 
      name="Slide Image" title="Slide Image">
    <FRAME src="note001" name="Notes" title="Notes">
  </FRAMESET>
<NOFRAMES> 
<P>List of Presentation Slides</P>
<OL>
<LI><A HREF="slide001">Time Value of Money</A>
<LI><A HREF="slide002">Topic Overview</A>
<LI><A HREF="slide003">Terms and Short Hand</A>
<LI><A HREF="slide004">Future Value of a Single CF</A>
<LI><A HREF="slide005">Example 1: FV example:The
NBA’s new Larry Bird exception</A>
<LI><A HREF="slide006">FV Example: NBA’s Larry
Bird Exception (cont.)</A>
<LI><A HREF="slide007">SuperStar’s Contract
Breakdown</A>
<LI><A HREF="slide008">Present Value of a Single
Cash Flow</A>
<LI><A HREF="slide009">Example 2: Paying Jr, and
A-Rod</A>
<LI><A HREF="slide010">Example 3: Finding Rate of
Return or Interest Rate</A>
<LI><A HREF="slide011">Annuities</A>
<LI><A HREF="slide012">FV of Annuities</A>
<LI><A HREF="slide013">PV of Annuities</A>
<LI><A HREF="slide014">Example 4: Invest Early in
an IRA</A>
<LI><A HREF="slide015">Example 4 Solution</A>
<LI><A HREF="slide016">Example 5: Lotto Fever
</A>
<LI><A HREF="slide017">Uneven Cash Flows: Example
6:Fun with the CF function</A>
<LI><A HREF="slide018">Example 6 CF worksheet inputs</A>
<LI><A HREF="slide019">CF inputs continued</A>
<LI><A HREF="slide020">Non-Annual Interest
Compounding</A>
<LI><A HREF="slide021">Example 7: What rate are
you really paying?</A>
<LI><A HREF="slide022">Nominal to EAR Calculator</A>
<LI><A HREF="slide023">Continuous Interest Compounding</A>
<LI><A HREF="slide024">FV and PV with non-annual
interest compounding</A>
<LI><A HREF="slide025">Non-annual annuities</A>
<LI><A HREF="slide026">Example 8: Finding Monthly
Mortgage Payment</A>
<LI><A HREF="slide027">solution to Example 8</A>
</OL>
</NOFRAMES>
</FRAMESET>
</HTML>

The following illustrate how some user agents handle this frameset.

First, rendering in Microsoft Internet Explorer 5.0 on a Windows platform:

Image shows the example frameset with five frame panes rendered in Microsoft Internet Explorer 5.0

Rendering by Lynx on Linux:

                                        Time Value of Money

   FRAME: Size buttons
   FRAME: Presentation Outline
   FRAME: Navigation buttons
   FRAME: Slide Image
   FRAME: Notes

   List of Presentation Slides
    1. Time Value of Money 
    2. Topic Overview 
    3. Terms and Short Hand 
    4. Future Value of a Single CF 
    5. Example 1: FV example:The NBA's new Larry Bird exception 
    6. FV Example: NBA's Larry Bird Exception (cont.) 
    7. SuperStar's Contract Breakdown 
    8. Present Value of a Single Cash Flow 
    9. Example 2: Paying Jr, and A-Rod 
   10. Example 3: Finding Rate of Return or Interest Rate 
   11. Annuities 
   12. FV of Annuities 
   13. PV of Annuities 
   14. Example 4: Invest Early in an IRA 
   15. Example 4 Solution 
   16. Example 5: Lotto Fever 
   17. Uneven Cash Flows: Example 6:Fun with the CF function 
   18. Example 6 CF worksheet inputs 
   19. CF inputs continued 
   20. Non-Annual Interest Compounding 
   21. Example 7: What rate are you really paying? 
   22. Nominal to EAR Calculator 
   23. Continuous Interest Compounding 
   24. FV and PV with non-annual interest compounding 
   25. Non-annual annuities 
   26. Example 8: Finding Monthly Mortgage Payment 
   27. solution to Example 8 

Graphical rendering by Home Page Reader on Windows:

Image shows the example frameset with five links for each of the frame elements in IBM home page reader

Audio rendering by Home Page Reader on Windows: @@add here?@@

User agents may also indicate the number of frames in a document and which frame is the current frame via the menu bar or popup menus. Users can configure the user agent to include a FRAMES menu item in their menu bar. The menu bar makes the information highly visible to all users and is very accessible to assistive technologies. In the following snapshot, the menu bar indicates the number of frames and a check next to the name of the frame element indicates which is the current frame:

Image shows a pull down menu indicating the number of frames in a document, the labels associated with each frame, and a check mark to indicate the current frame

3.7 Form techniques

For labels explicitly associated with form controls (e.g., "for" attribute on LABEL in HTML), make available label information when the user is navigating among the form controls. This information must be provided in a device-independent manner, and the user should be able to choose from a list of mechanisms that provide access to the content of the label.

For semantic information explicitly associated with groupings of form controls (e.g., groupings of radio buttons or checkboxes contained in a FIELDSET), make available the information contained in the LEGEND defined for the FIELDSET to the user. This information must be provided in a device-independent manner, and the user should be able to choose from a list of mechanisms that provide access to the content of the LEGEND.

Provide information about the percentage of form that has already been filled out as the user moves through the form controls. This information must be provided in a device-independent manner. The user should also be able to query the user agent through a simple, well-documented mechanism (such as a keystroke or keystroke combination) to learn what percentage of the form has been completed.

Allow the user to know what percentage of a form has been completed as the user navigates the form. will help users avoid prematurely submitting an incomplete form. This is particularly important for anyone moving through a form sequentially; users who encounter a submit button usually think this means the end of a form, but it may not be. (Refer also to the technique detailing methods of providing users with orientation information about individual form controls when a form control receives focus for a more detailed discussion of this issue.)

Provide the user with orientation information about a form. Users should be able to query the user agent for:

Provide the user with orientation information about individual form controls when a form control receives focus. For example, the most basic orientation information would be to identify the form control with focus as "Field X of Y", where "Y" is the total number of fields contained in the form. This will help prevent users accessing the form serially (such as a blind user using a screen reader or someone using a voice browser over a phone) from prematurely invoking the form's submit mechanism. It is a common practice for forms (particularly those used to query search engines) to be laid out visually, so that the submit and reset buttons (if present) immediately follow a text-entry field, despite the presence of other form controls (such as radio buttons and checkboxes) within the FORM element. A user accessing such a form in a serial manner, therefore, is likely to mistake the submit button for the end of the form, and activate it, unaware that it is followed by further form controls which could -- in the example of a search engine query submission form -- prove an invaluable aid in tailoring the content being submitted via the form. Use of such orientation information (i.e., "Field X of Y" or the percentage of the form completed) will also decrease the amount of time needed to submit the form (a crucial consideration when forms are being used to facilitate bidding for online auctions) as well as reduce the frustration of the end user, who, due to the visually oriented layout of the form, is confused when the submission of the form repeatedly leads to a message such as "Form Incomplete - Use your browser's back button to return to the form".

When a document contains more than one form, form control orientation information should also include data which will identify to which form the form control with focus belongs. Notification could take the form: Form Z: Field X of Y

where "Z" identifies the form, "X" the form field with focus an "Y" the total number of form fields contained in "Form Z".

Provide more detailed orientation information pertaining to form:

Allow the user to invoke an external editor when a TEXTAREA receives focus. A user may wish to use an external editor, rather than enter text directly in a TEXTAREA for myriad reasons, including:

Provide information about the order of form controls (e.g., as specified by "tabindex" in HTML). This is important since:

Provide information about required fields. Since authors often use color changes, font styling, or a graphical symbol alone to express that a field is required, the user should be able to configure the user agent so that it alerts him that the field is required for submission of the form content. Strategies for achieving this include:

Allow the user to configure the user agent so that SELECT form fields which use the "multiple" attribute to allow the end user to select more than one OPTION can be transformed into a list of checkboxes.

Allow the user to exit a menu without selecting any option. For example, when navigating through OPTION's, have some key like the ESC key ot exit the list of OPTION's without making a selection.

Allow the user to configure the user agent sot that SELECT form fields can be transformed into a list of radio buttons.

3.7.1 Form submission techniques

Users (notably users with blindness or any user unaccustomed to online forms) do not want forms to be submitted without their consent, e.g., when submitted through scripts. In this case user agents should request confirmation before submitting the form content. Nor do they wish to be prompted before each form submission when they have explicitly submitted the form (e.g., through the standard submit button).

Inadvertently pressing the RETURN or ENTER key is quite a prevalent phenomenon among users of every level of expertise - especially those who often find it necessary to switch between user agents. Lynx, for example, uses the ENTER key within FORMs as a means of exposing drop-down (or pop-up, depending upon your point of view) SELECT menus. Thus, when one encounters a SELECT menu using Lynx, one: exposes the content of the menu by pressing the ENTER key, and then is able to navigate between OPTIONs using the up and down arrows or via Lynx's text-search feature. When one finds the appropriate OPTION, it is selected by pressing ENTER, which causes the selected item to be displayed in the SELECT menu listbox.

The problem posed by the default "submit on enter" feature of most GUI browsers is not limited to the SELECT menu problem outlined above. Lynx (as well as several other text-based browsers) uses the ENTER/RETURN key as a means of toggling several FORM controls, such as the selection of checkboxes and radio buttons.

Speech users may be frustrated and misdirected by the use of javascript and event handler controlled pseudo-forms, wherein the user is presented with a menu (in the form of a listbox in GUI browsers), and is redirected to a different viewport upon selection of an OPTION. The markup behind such pseudo-forms is a mix of javascript (in particular the "function switchpage(select)" command) and HTML FORM controls, which utilize HTML4's event handler script attributes (in particular the "onchange" event handler attribute has been defined. An example (gleaned from the document source for one Web site follows:

 <SELECT NAME="condition" onchange="switchpage(this)">

When such a menu is encountered by a individual who is using speech synthesis in conjunction with a javascript enabled user agent, his or her instinctual reaction will be to use the UA's navigation mechanism (usually the up and down arrows) to review the available OPTIONs. However, each time a new OPTION is displayed, the user is abruptly taken to a new viewport. Conversely, if one is using a user agent that does not support javascript (or has javascript support disabled), then the menu is displayed, but since there is no SUBMIT mechanism associated with it, there is no mechanism by which one can use the menu to quickly switch viewports - the apparent purpose of this type of pseudo-form. And while one can avoid having the viewport abruptly changed when encountering the menu (at least in the Windows environment) by using the ALT-LEFT-ARROW keystroke to display the menu in a drop-down list, (a) very few users know this keystroke, and (b) when one encounters a listbox on a page in an aural environment, one usually assumes that he or she is navigating a valid FORM, in which there are no unexpected side effects to perusing the contents of a SELECT menu using the arrow keys

Refer also to Web Content Accessibility Guidelines 1.0 checkpoint 6.3 [WAI-WEBCONTENT]: Content developers must ensure that pages are accessible with scripts turned off or in browsers that don't support scripts.

Some techniques to address form submission:

  1. Allow the user to configure the user agent. Choices should include: The default setting should be: "Prompt before submitting form content", so as to allow the user to decide whether or not HTML4 event handling will occur automatically.
  2. Configuration can be determined by prompting the user the first time an event handled or script-driven FORM is encountered. Choices should include: If the user chooses "Prompt Before Submitting Form Content", this prompt could be recycled in an abbreviated fashion. The prompt should include:

3.8 Script techniques

Certain elements of the document language may have associated event handlers that are triggered when certain events occur. User agents must be able to identify those elements with event handlers statically associated (i.e., associated in the document source, not in a script).

In HTML
All of the attributes beginning with the prefix "on": "onblur", "onchange", "onclick", "ondblclick", "onkeydown", "onkeypress", "onkeyup", "onload", "onmousedown", "onmousemove", "onmouseout", "onmouseover", "onmouseup", "onreset", "onselect", "onsubmit", and "onunload".

Allow the user to configure the user agent so that mouseover/mouseout events may be triggered by (and trigger) focus/blur events. Similarly, allow the user to use a key command, such as "enter" and "shift-enter" to trigger onclick and ondblclick events.

Implement DOM 2 [DOM2] events with a single activation event and provide a method for triggering that event from each supported input device or input API. These should be the same as the click events and mappings provided above (but note that a user agent which is also an editor may wish to use single click events for moving a system caret, and want to provide a different behavior to activate using the mouse). For example, Amaya [AMAYA] uses a "doAction" command for activating links and form controls, which can be triggered either by the mouse (and it is possible to set it for single-click or double-click) or by the keyboard (it is possible to set it for any key using Amaya's keyboard configuration)

Document the effects of known important scripts to give users an idea in advance of what they do. Make script source available to users so that those familiar with the scripting language may be able to understand their effects.

Note. User agents may already allow users to turn off scripts for security reasons.

3.9 Abbreviations and acronyms

User agents should make available information about abbreviation and acronym expansions. For instance, in HTML, look for abbreviations specified by the ABBR and ACRONYM elements. The expansion may be given with the "title" attribute.

To provide expansion information, user agents may:

4 Appendix: Accessibility features of some operating systems

Several of the more popular mainstream operating systems now include a common suite of built-in accessibility features that are designed to assist individuals with varying abilities. Despite operating systems differences, the built-in accessibility features use a similar naming convention and offer similar functionalities, within the limits imposed by each operating system (or particular hardware platform).

The following is a list of built-in accessibility features from several platforms:

StickyKeys
These allow users to perform a multiple simultaneous key sequence by pressing and releasing each key in sequential order. StickyKeys is designed to work with only those keys defined as modifier keys. Modifier keys are pressed in combination with other keys, to change the outcome of the second (or more) pressed keys. For example, the SHIFT key is defined as a modifier key, since it is commonly used to create upper case characters. Each operating system or hardware platform typically defines a set of keys which can act as modifier keys. The most common modifier keys include SHIFT, CONTROL, and ALTERNATE.
MouseKeys
These allow users to move the mouse cursor and activate the mouse button(s) from the keyboard.
RepeatKeys
These allow users to set how fast a key repeats (e.g., sometimes referred to as typematic rate) when the key is held pressed (e.g., Repeat Rate), and also allows control over how quickly the key starts to repeat after the key has been pressed (e.g., delay Until Repeat). Key repeating may also be eliminated.
SlowKeys
These instruct the computer not to accept a key as pressed until it has been pressed and held down for a specific user adjustable length of time.
BounceKeys
These prevent extra characters from being typed if the user bounces (e.g., tremor) on the same key when pressing or releasing it.
ToggleKeys
These provide an audible indication for the status of keys that have a toggled state (e.g., keys that maintain status after being released). The most common toggling keys include Caps Lock, Num Lock, and Scroll Lock.
SoundSentry
These monitor the operating system and applications for sounds, and attempt to provide a graphical indication when a sound is being played. Older versions of Sound Sentry may have flashed the entire display screen for example, while newer versions of SoundSentry provide the user with a selection of options, such as flashing the active window or flashing the active window caption bar.

The next three built in accessibility features are not as commonly available as the above group of features, but are included here for definition, completeness, and future compatibility.

ShowSounds
These are user setting or software switches that are available for the operating system and application (including user agents) APIs to read, to notify them that the user wishes audio information to also be presented in a graphical format.
High Contrast
These automatically change the display fonts and colors to choices which should provide for easier reading.
TimeOut
These allow the built-in accessibility features to automatically turn off if the computer is unused for a specified length of time, and is intended for use when the computer is in a public setting (e.g., library). TimeOut might also be referred to as reset or automatic reset.

The next accessibility feature listed here is not considered to be a built in accessibility feature (since it only provides an alternate input channel) and is presented here only for definition, completeness, and future compatibility.

SerialKeys
These allow a user to perform all keyboard and mouse functions from an external assistive device (such as communication aid) communicating with the computer via a serial character stream (e.g., serial port, IR port, etc.) rather than or in conjunction with, the keyboard, mouse, and other standard input devices/methods.

Microsoft Windows 95, Windows 98, and Window NT 4.0

For information about Microsoft keyboard configurations (Internet Explorer, Windows 95, Windows 98, and more), refer to documentation on keyboard assistance for Internet Explorer and MS Windows [MS-KEYBOARD].

The following accessibility features can be adjusted from the Accessibility Options Control Panel:

Additional accessibility features available in Windows 98:

Magnifier
This is a windowed, screen enlargement and enhancement program used by persons with low vision to magnify an area of the graphical display (e.g., by tracking the text cursor, focus, etc.). Magnifier can also invert the colors used by the system within the magnification window.
Accessibility Wizard
This is a setup tool intended to assist a person with making choices which setting up the accessibility features on a workstation.

Apple Macintosh Operating System

The following accessibility features can be adjusted from the Easy Access Control panel (Note: Apple convention uses a space within the accessibility feature names.)

The following accessibility features can be adjusted from the Keyboard Control Panel.

The following accessibility feature can be adjusted from the Sound or Monitors and Sound Control Panel (depends upon which version of the OS).

Additional accessibility features available for the Macintosh OS:

CloseView
This is a full screen, screen enlargement and enhancement program used by persons with low vision to magnify the information on the graphical display, and it can also change the colors used by the system.
SerialKeys
This is available as freeware from Apple and several other Web sites.

AccessX, X Keyboard Extension (XKB), and the X Window System

(Note: AccessX became a supported part of the X Window System X Server with the release of the X Keyboard Extension in version X11R6.1)

The following accessibility features can be adjusted from the AccessX graphical user interface X client on some DEC, SUN, and SGI operating systems. Other systems supporting XKB may require the user to manipulate the features via a command line parameter(s).

DOS (Disk Operating System)

The following accessibility features are available from a freeware program called AccessDOS, which is available from several Internet Web sites including IBM, Microsoft, and the Trace Center, for either PC-DOS or MS-DOS versions 3.3 or higher.

Testing UA operation with platform standards

Ensure your UA can be operated using the standard interfaces on the target platform(s). Some example tests include:

5 Appendix: Loading assistive technologies for DOM access

There are several methods for developers to accomplish this. Most of these methods fall into four categories:

  1. Launch the entire AT inside the address space of the (UA) browser
  2. Launch some part of the AT, a piece of stub code, a DLL, a Browser Helper Object [special DLL], etc., inside the address space of the (UA) browser
  3. Write your own combined UA/AT (e.g., pwWebSpeak)
  4. Out-of-process access to the DOM

These methods are ordered as developments within a rapidly changing technology with the most recent advances/methods listed first.

Loading assistive technologies for direct access to User Agent DOMs

Note. This method and the method described in the next section are very similar. What differs is the amount of, or capability of, the AT that actually gets loaded in the same process or address space as the User Agent.)

Access to application specific data across process boundaries might be costly in terms of performance. Therefore, user agents may wish to provide a mechanism to load the entire assistive technology (AT), into the process space of the application as a separate thread with direct access to the DOM.

Determining the Assistive Technologies to load

One technique is to store a reference to an assistive technology in a system registry file or, in the case of Jave, a properties file. Registry files are common among many operating system platforms.

In Windows you have the system registry file. On OS/2 you have the system.ini file and on distributed network client networks you often have a system registry server that an application running on the network client computer can query.

In Java 2, the existence of an "accessibility.properties" file causes the system event queue to examine the file for assistive technologies required for loading. If the file contains a property called "assistive_technologies", it will load all registered assistive technologies and start them on their own thread in the Java Virtual Machine that is a single process. An example entry for Java is as follows:

    assistive_technologies=com.ibm.sns.svk.AccessEngine

In Windows, a similar technique could be followed by storing the name of a Dynamic Link Library (DLL) for an assistive technology in a designated assistive technology key name, AT pair. An example entry for Windows could be as follows:

    HKEY_LOCAL_MACHINE\Software\Accessibility\DOM 
           "ScreenReader, VoiceNavigation"

Attaching the Assistive Technologies to the DOM.

Once the assistive technology is determined from the registry, any user agent on the given operating system can now determine if an assistive technology needs to be loaded with their application and load it.

On a non-Java platform, a technique to do this would be to create a separate thread with a reference to the User Agent's DOM using a Dynamic Link Library (DLL). This new thread will load the DLL and call a specified DLL entry name with a pointer to the DOM interface. The assistive technology's task will then run until such time as is necessary to end communication with the DOM.

Once loaded, the assistive technology can monitor the DOM as needed. The assistive technology has the option of communicating with a main assistive technology of its own and process the DOM as a caching mechanism for the main AT application or be used as a bridge to the DOM for the main assistive technology.

In the future, it will be necessary to provide a more comprehensive reference to the application that not only provides direct access to it's client area DOM, but also multiple DOM's that it is processing and an event model for monitoring them.

Example Technique: Java's Direct Access

Java is a working example where the direct access to application components is performed in a timely manner. Here, an assistive technology running on a separate thread monitors GUI events such as focus changes. Focus changes give the AT notification of which component object has focus. The AT can communicate directly with all components in the application by walking the parent/child hierarchy and connecting to each component's methods and monitor events directly. In this case an AT has direct access to component specific methods as well as those provided for by the Java Accessibility API. There is no reason that a DOM interface to UA components could not be provided

In Java 1.1.x, Sun's Java access utilities load an assistive by monitoring the Java awt.properties file for the presence of assistive technologies and loads them as shown in the following code example:

import java.awt.*;
import java.util.*;
      
String atNames = Toolkit.getProperty("AWT.assistive_technologies",null);
if (atNames != null) {
    StringTokenizer parser = new StringTokenizer(atNames," ,");
    String atName;
    while (parser.hasMoreTokens()) {
       atName = parser.nextToken();
       try {
          Class.forName(atName).newInstance();
       } 
       catch (ClassNotFoundException e) {
          throw new AWTError("Assistive Technology not found: " + atName);
       } 
       catch (InstantiationException e) {
          throw new AWTError("Could not instantiate Assistive" + 
                             " Technology: " + atName);
       } 
       catch (IllegalAccessException e) {
          throw new AWTError("Could not access Assistive" + 
                             " Technology: " + atName);
       } catch (Exception e) {
          throw new AWTError("Error trying to install Assistive" + 
                             " Technology: " + atName + " " + e);
       }
    }
}

In the above code example, the function Class.forName(atName).newInstance() creates a new instance of the assistive technology. The constructor for the assistive technology will then be responsible for monitoring application component objects by monitoring system events.

In the following code example, the constructor for the assistive technology "Access Engine," adds a focus change listener using Java accessibility utilities. When the assistive technology is notified of an objects gaining focus it has direct access to that object. If the Object, o, implemented a DOM interface the assistive technology would now have direct access to the DOM in the same process space as the application.

   import java.awt.*;
   import javax.accessibility.*;
   import com.sun.java.accessibility.util.*;
   import java.awt.event.FocusListener;

   class AccessEngine implements FocusListener {
      public AccessEngine() {
         //Add the AccessEngine as a focus change listener
         SwingEventMonitor.addFocusListener((FocusListener)this);
      }
      
      public void focusGained(FocusEvent theEvent) {
         // get the component object source
         Object o = theEvent.getSource();
         // check to see if this is a DOM component
         if (o instanceof DOM) {
            ...
         }
      }
      public void focusLost(FocusEvent theEvent) {
         // Do Nothing
      }
   }

In this example, the assistive technology has the option of running standalone or acting as a cache for a bridge that communicates with a main assistive technology running outside the Java virtual machine.

Loading part of the assistive technologies for direct access to User Agent DOMs

Access to application specific data across process boundaries might be costly in terms of performance. Therefore, user agents may wish to provide a mechanism to load part of the assistive technology (AT) into the process space of the application as a separate thread, with direct access to the DOM, to provide the specific functionality they require. This could consist of a piece of stub code, a DLL, a Browser Helper Object, etc. An example of how to do this follows.

Browser Helper Objects

In order to attach to a running instance of Internet Explorer 4.0, you can use a "Browser Helper Object." A "Browser Helper Object" is a DLL that will attach itself to every new instance of Internet Explorer 4.0 (only if you explicitly run iexplore.exe). You can use this feature to gain access to the object model of a particular running instance of Internet Explorer. You can also use this feature to get events from an instance of Internet Explorer 4.0. This can be tremendously helpful when many method calls need to be made to IE, as each call will be performed much more quickly than the out of process case.

There are some requirements when creating a Browser Helper Object

For more information, please check out: http://support.microsoft.com/support/kb/articles/Q179/2/30.asp.
http://msdn.microsoft.com/library/techart/bho.htm

Java Access Bridge

In order for native Windows ATs to gain access to Java applications without the creating a Java native solution Sun Microsystems provides the "Java Access Bridge." This bridge is loaded as an AT as described in section 6.1.3. The bridge uses a Java Native Invocation (JNI) to Dynamic Link Library) (DLL) communication and caching mechanism that allows a native assistive technology to gather and monitor accessibility information in the Java environment. In this environment, the AT determines that a Java application or applet is running and communicates with the Java Access Bridge DLL to process accessibility information about the application/applet running in the Java Virtual Machine.

Loading assistive technologies "as" the User Agent with access to the DOMs

Specialized user agents might also include the necessary assistive technology as part of their interface, and thus provide possibly the best of both worlds. An example would be pwWebSpeak, from The Productivity Works [PRODWORKS].

Loading assistive technologies for indirect access to User Agent DOMs

Access to application specific data across process boundaries or address space might be costly in terms of performance. However, there are other reasons to consider when accessing the User Agent DOM that might lead a developer to wish to access the DOM from their own process or memory address space. One obvious protection this method provides, is that if the User Agent application fails, it doesn't disable the user's AT as well. Another consideration would be legacy systems, where the user relies on their AT for access to other applications as well as the User Agent, and thus would have their AT loaded all the time, not just for accessing the User Agent.

There are several ways to gain access to the User Agent's DOM. Most User Agents support some kind of external interface, or act as a mini-server to other applications running on the desktop. Internet Explorer is a good example of this, as IE can behave as a component object model (COM) server to other applications. Mozilla, the open source release of Navigator also supports cross platform COM (XPCOM).

An example of using COM to access the IE Object Model can be seen in the code snippet below. This is an example of how to use COM to get a pointer to the WebBrowser2 module, which in turn allows you to get a interface/pointer to the document object, or IE DOM for the Web page in view.

   /* first, get a pointer to the WebBrowser2 control */
   if (m_pIE == NULL) {
      hr = CoCreateInstance(CLSID_InternetExplorer, 
           NULL, CLSCTX_LOCAL_SERVER, IID_IWebBrowser2, 
           (void**)&m_pIE);

      /* next, get a interface/pointer to the document in view, 
         this is an interface to the document object model (DOM)*/

      void CHelpdbDlg::Digest_Document() {
         HRESULT hr;
         if (m_pIE != NULL) {
            IDispatch* pDisp;
            hr = m_pIE->QueryInterface(IID_IDispatch, (void**) &pDisp);
            if (SUCCEEDED(hr)) {
        
               IDispatch* lDisp;
               hr = m_pIE->get_Document(&lDisp);
               if (SUCCEEDED(hr)) {
           
                   IHTMLDocument2* pHTMLDocument2;
                   hr = lDisp->QueryInterface(IID_IHTMLDocument2,
                               (void**) &pHTMLDocument2);
                   if (SUCCEEDED(hr)) {
            
                   /* with this interface/pointer, IHTMLDocument2*,
                      you can then work on the document */
                      IHTMLElementCollection* pColl;
                      hr = pHTMLDocument2->get_all(&pColl);
                      if (SUCCEEDED(hr)) {
                    
                         LONG c_elem;
                         hr = pColl->get_length(&c_elem);
                         if (SUCCEEDED(hr)) {
                            FindElements(c_elem, pColl);
                         }
                         pColl->Release();
                      }
                      pHTMLDocument2->Release();
                   }
                   lDisp->Release();
               }
               pDisp->Release();
            }
         }
      }
   }   

For more information on using COM with IE, please visit the Microsoft Web site:http://www.microsoft.com/com/default.asp

For more information on using XPCOM with Mozilla, please visit the Mozilla Web site: http://www.mozilla.org/

For a working example of the method described in 6.1.4, please visit the following web site and review HelpDB, developed as a testing tool for Web table navigation: http://trace.wisc.edu/world/web/document_access/

6 Appendix: Assistive Technology Functionalities

This informative appendix describes some functionalities for assistive technologies to promote accessibility.

Orientation

  1. Indicate the row and column dimensions of a selected table. Note. User agents should consider multidimensional tables, headers and footers, and multiple header levels. Refer also to checkpoint 5.4.
  2. Describe a selected element's position within larger structures (e.g., numerical or relative position in a document, table, list, etc.). For example: tenth link of fifty links; document header 3.4; list one of two, item 4.5; third table, three rows and four columns; current cell in third row, fourth column; etc. Allow users to get this information on demand (e.g., through a keyboard shortcut). Provide this information on the status line on demand from the user.
  3. Provide information about form structure and navigation (e.g., groups of controls, control labels, navigation order, and keyboard configuration). For instance, provide information about controls with explicitly associated labels (the "for" attribute of LABEL in HTML), about which keys activate the form controls (the "accesskey" attribute in HTML), about the sequential navigation order of the form controls (the"tabindex" attribute in HTML), and about control groups (the FIELDSET and OPTGROUP elements in HTML). Refer also to checkpoint 1.3 and checkpoint 7.4.
  4. Enable announcing of information regarding title, value, grouping, type, status and position of specific focused elements.

Metadata

Metadata of all sorts: titles, dimensions, dates, relationships, etc. promotes accessibility by providing additional context to users. Text metadata is particularly useful since it can be rendered graphically, as Braille, and as speech.

One useful form of metadata is content summary information. Provide information, for example, about the number of links, forms, tables, images, significant words, etc.

For example, this information will help a user get an impression about the purpose of each frame in a frameset. For example, if the content of a frame has many links, but few significant words, then the frame is probably an index of some kind. Content with many significant words is probably a text page. Content with only a couple of pictures and few significant words or links is probably for decoration.

Synthesized speech

Tools that work with synthesized speech do not always pronounce text correctly. Therefore, they should provide additional context so that users can understand content. Techniques include:

Refer to "Speak to Write" [SPEAK2WRITE] for information on speech recognition and accessibility.

7 Appendix: Glossary

Active element
Active elements constitute a user interface for the document. They have associated behaviors that may be activated (or "triggered") either through user interaction or through scripts. Which elements are active depends on the document language and whether the features are supported by the user agent. In HTML documents, for example, active elements include links, image maps, form controls, element instances with a value for the "longdesc" attribute, and element instances with associated scripts (event handlers) explicitly associated with them (e.g., through the various "on" attributes).
An active element's behavior may be triggered through any number of mechanisms, including the mouse, keyboard, an API, etc. The effect of activation depends on the element. For instance, when a link is activated, the user agent generally retrieves the linked resource. When a form control is activated, it may change state (e.g., check boxes) or may take user input (e.g., a text field). Activating an element with a script assigned for that particular activation mechanism (e.g., mouse down event, key press event, etc.) causes the script to be executed.
Most systems use the focus to designate the active element the user wishes to trigger.
Alternative Equivalents for Content
Since rendered content in some forms is not always accessible to users with disabilities, authors must supply alternative equivalents for content. In the context of this document, the equivalent must fulfill essentially the same function for the person with a disability (at least insofar as is feasible, given the nature of the disability and the state of technology), as the "primary" content does for the person without any disability. For example, the text "The Full Moon" might convey the same information as an image of a full moon when presented to users. Note that equivalent information focuses on fulfilling the same function. If the image is part of a link and understanding the image is crucial to guessing the link target, an equivalent must also give users an idea of the link target.
User agents must make the alternative equivalent content available to users or software that require it (in place of and/or in addition to the "primary" content). Alternative representations include text equivalents (long and short, synchronized and unsynchronized) and non-text equivalents (e.g., captions, auditory descriptions, a visual track that shows sign language translation of a written text, etc.).

The Techniques Document [UA-TECHNIQUES] describes the different mechanisms authors use to supply alternative equivalents for content. Please also consult the Web Content Accessibility Guidelines [WAI-WEBCONTENT] and the associated Techniques document [WAI-WEBCONTENT-TECHS].

Assistive Technology
In the context of this document, an assistive technology is a user agent that relies on one or more other user agents to help people with disabilities interact with a computer. For example, screen reader software is an assistive technology because it relies on browsers or other application software to enable Web access, notably for people with visual and learning disabilities.

Examples of assistive technologies that are important in the context of this document include the following:

Beyond this document, an assistive technology consists of software or hardware that has been specifically designed to assist people with disabilities in carrying out daily activities, e.g., wheelchairs, reading machines, devices for grasping, alternative computer keyboards or pointing devices, etc.)
Auditory Description
An auditory description is either a prerecorded human voice or a synthesized voice (recorded or generated on the fly) describing the key visual elements of a presentation. The auditory description is synchronized with the audio track of the presentation, usually during natural pauses in the audio track. Auditory descriptions include information about actions, body language, graphics, and scene changes.
Author Styles
Style property values that originate in documents, their associated style sheets, or come from a server are called author styles.
Captions
Captions (or sometimes "closed captions") are text transcripts that are synchronized with other auditory or visual tracks. They convey information about spoken words and non-spoken sounds such as sound effects. Captions benefit people who are deaf and hard-of-hearing, and anyone who cannot hear the audio (e.g., when in a crowded room). Captions are generally rendered graphically above, below, or superimposed over video. Note. Other terms including the word "caption" may have different meanings in this document. For instance, a "table caption" is a title for the table, often positioned graphically above or below the table. In this document, the intended meaning of "caption" will be clear from context.
Configure
To set user preferences. This may be done through the user agent's user interface, through configuration files, by scripts, etc.
Control
User control of the user agent - interface, behavior, styles, etc. - means that the user can choose preferred behavior from a set of options. For instance, control of colors means that the user can choose from available colors, within the limits offered by the operating system or user agent.
The term "control" also means "user interface component" or "form component" in this document. Which meaning is intended should be apparent from context.
Device Independence
The ability to make use of software via any input or output device API provided by the operating system and used by the user agent. User agents should follow operating system conventions and use standard APIs for device input and output.
Documentation
Documentation includes all product documentation, notably installation instructions, the help system, and all product manuals.
Documents, Elements, and Attributes
A document may be seen as a hierarchy of elements. Element types are defined by a language specification (e.g., HTML 4.0 or an XML application). Elements may include content, which generally contributes to the document's content. Elements may also have attributes that take values.
Events and scripting, Event Handler
When certain events occur (loading or unloading events, mouse press or hover events, keyboard events, etc.), user agents often perform some task (e.g., execute a script). For instance, in most user agents, when a mouse button is released over a link, the link is activated and the linked resource retrieved. User agents may also execute author-defined scripts when certain events occur. The script bound to a particular event is called an event handler. Note. The interaction of HTML, style sheets, the Document Object Model ([DOM1], [DOM2]), and scripting is commonly referred to as "Dynamic HTML" or DHTML. However, as there is no W3C specification that formally defines DHTML, this document will only refer to event handlers and scripts.
Focus, Current Focus
The user focus designates an active element in a document. A viewport has at most one focus. When several viewports co-exist, each may have a focus, but only one is active, called the current focus. The current focus is generally presented (e.g., highlighted) in a way that makes it stand out.
Highlight
Any mechanism used to emphasize selected or focused content. Visual highlight mechanisms include dotted boxes, underlining, and reverse video. Synthesized speech highlight mechanisms may include altering voice pitch or volume.
Input Configuration
Every user agent functionality available to the user is mapped to some user interface mechanism, including menus, buttons, keyboard shortcuts, voice commands. The default input configuration is the mapping the user finds after installation of the software. The documentation should tell users what functionalities are available and the user interface should remind users of the current mapping to the user interface and allow them to figure out quickly how to use the appropriate software features.
Insertion point, Current Insertion Point
The insertion point is the location where document editing takes place. The insertion point may be set by the user (e.g., by a pointing device or the keyboard editing keys) or through an application programming interface (API). A viewport has at most one logical insertion point. When several viewports co-exist, each may have an insertion point, but only one is active, called the current insertion point Note. Graphical user agents that render bidirectional text may render the insertion point at two locations on the screen. Often, the cursor is shown as split into an upper half and a lower half.
The insertion point is generally rendered specially (e.g., on the screen, by a vertical bar or similar cursor).
Native support
A user agent supports a feature natively if it does not require another piece of software (e.g., plug-in or external program) for support. Operating system features adopted as part of the user agent are considered native. However, since the user agent is responsible for the accessibility of native features, it is also considered responsible for the accessibility of adopted operating system features.

Though native support for accessibility features is encouraged, user agents must still make information available through APIs so that assistive technologies meet specialized accessibility needs.

Natural Language
Spoken, written, or signed human languages such as French, Japanese, and American Sign Language. The natural language of content may be indicated in markup (e.g., by the "lang" attribute in HTML 4.0 ([HTML40] section 8.1), the HTML 4.0 "hreflang" attribute for links ([HTML40], section 12.1.5), or by the HTTP Content-Language header ([RFC2616], section 14.12).
Properties, Values, and Defaults
A user agent renders a document by applying formatting algorithms and style information to the document's elements. Formatting depends on a number of factors, including where the document is rendered: on screen, paper, through speakers, a Braille device, a mobile device, etc. Style information (e.g., fonts, colors, voice inflection, etc.) may come from the elements themselves (e.g., certain style attributes in HTML), from style sheets, or from user agent settings. For the purposes of these guidelines, each formatting or style option is governed by a property and each property may take one value from a set of legal values. (The term "property" in this document has the meaning ascribed in CSS 2 ([CSS2], section 3). A reference to "styles" in this document means a set of style-related properties.
The value given to a property by a user agent when it is started up is called the property's default value. User agents may allow users to change default values through a variety of mechanisms (e.g., the user interface, style sheets, initialization files, etc.).
Once the user agent is running, the value of a property for a given document or part of a document may be changed from the default value. The value of the property at a given moment is called its current value. Note that changes in the current value of a property do not change its default value.
Current values may come from author styles, user styles, or user agent defaults.
Recognize
A user agent is said to recognize markup, content types, or rendering effects when it can identify (through built-in mechanisms, DTDs, style sheets, headers, etc) the information. For instance, HTML 3.2 user agents may not recognize the new elements or attributes of HTML 4.0. Similarly, a user agent may recognize blinking content specified by elements or attributes, but may not recognize that an applet is blinking. The Techniques Document [UA-TECHNIQUES] discusses some content that affects accessibility and should be recognized as such.
Rendered content
An element's rendered content is that which a user agent renders for the element. This may be what appears between the element's start and end tags, the value of an attribute (c.f. the "alt", "title", and "longdesc" attributes in HTML), or external data (e.g., the IMG element in HTML). Content may be rendered to a graphical display, to an auditory display (to a speaker as speech and non-speech sounds) or to a tactile display (Braille and haptic displays). Refer also to the description of alternative equivalents for content.
Selection, Current Selection
The user selection generally identifies a range of content (text, images, etc.) in a document. The selection may be structured (based on the document tree) or unstructured (e.g., text-based). Content may be selected through user interaction, scripts, etc. The selection may be used for a variety of purposes: for cut and paste operations, to designate a specific element in a document, to identify what a screen reader should read, etc.
The user selection may be set by the user (e.g., by a pointing device or the keyboard) or through an application programming interface (API). A viewport has at most one user selection (though the selection may be rendered graphically as discontinuous text fragments). When several viewports co-exist, each may have a user selection, but only one is active, called the current selection.
The user selection is usually presented in a way the stands out (e.g., highlighted). On the screen, the selection may be highlighted using colors, fonts, graphics, or other mechanisms. Highlighted text is often used by assistive technologies to indicate through speech or Braille output what the user wants to read. Most screen readers are sensitive to highlight colors. Assistive technologies may provide alternative presentation of the selection through speech, enlargement, or refreshable Braille display.
Text transcript
A text transcript is a text equivalent of audio information (e.g., an audio clip). It provides text for both spoken words and non-spoken sounds such as sound effects. Text transcripts make presentations accessible to people who are deaf-blind (they may be rendered as Braille) and to people who cannot play movies, animations, etc. Transcripts may be generated on the fly (e.g., by speech-to-text converters). Refer also to captions.
Spawned Viewport
Viewports that are created by the user agent process. This refers to viewports that display content and does not include, for example, messages or prompts to the user.
Standard Device APIs
Operating systems are designed to be used by default with devices such as pointing devices, keyboards, voice input, etc. The operating system (or windowing system) provides standard APIs for these devices that should be used by user agents and other software for input and output to those devices. For example, for desktop computers today, user agents are expected to use the mouse and keyboard APIs for input. For touch screen devices or mobile devices, standard input APIs may include stylus, buttons, voice, etc. The display and sound card are considered standard ouput devices for a graphical desktop computer environment and each has a standard API. Note. What is considered "standard" for a particular environment will change over time.
User-initiated and User Agent-initiated
User-initiated actions result from user input to the user agent. User Agent-initiated actions result from scripts, operating system conditions, or built-in user agent behavior.
User Agent
A user agent is an application that retrieves and renders Web resources, including text, graphics, sounds, video, images, and other objects. An user agent may require additional software to handle some types of content. For instance, a browser may run a separate program or plug-in to render sound or video. The additional software is also considered a user agent. User agents include graphical desktop browsers, multimedia players, text browsers, voice browsers, and assistive technologies such as screen readers, screen magnifiers, speech synthesizers, onscreen keyboards, and voice input software.
User Styles
Style property values that come from user interface settings, user style sheets, or other user interactions are called user styles.
Views, Viewports, Current View, and Point of Regard
User agents may handle different types of source information: documents, sound objects, video objects, etc. The user perceives the information through a viewport, which may be a window, frame, a piece of paper, a speaker, a virtual magnifying glass, etc. A viewport may contain another viewport (e.g., nested frames, plug-ins, etc.).
User agents may render the same content in a variety of ways; each rendering is called a view. For instance, a user agent may allow users to view an entire document or just a list of the document's headers. These are two different views of the document.
The view is how source information is rendered and the viewport is where it is rendered. Both the current focus and the current selection must be in the same viewport, called the current viewport. The current viewport is generally highlighted when several viewports co-exist.
Generally, viewports give users access to all rendered information, though not always at once. For example, a video player shows a certain number of frames per second, but allows the user to rewind and fast forward. A graphical browser viewport generally features scrollbars or some other paging mechanism that allows the user to bring the rendered content into the viewport.
The content currently available in the viewport is called the user's point of regard. Since users may be viewing content with browsers that render graphically, as speech, as Braille, etc., what is meant precisely by "the point of regard" may vary. It may, depending on the user agent and browsing context, refer to a two dimensional area (e.g., for graphical rendering) or a single point (e.g., for aural rendering or voice browsing). User agents should not change the point of regard unexpectedly as this can disorient users.

8 Acknowledgments

Many thanks to the following people who have contributed through review and comment: Paul Adelson, James Allan, Denis Anson, Kitch Barnicle, Harvey Bingham, Olivier Borius, Judy Brewer, Bryan Campbell, Kevin Carey, Wendy Chisholm, David Clark, Chetz Colwell, Wilson Craig, Nir Dagan, Daniel Dardailler, B. K. Delong, Neal Ewers, Geoff Freed, John Gardner, Al Gilman, Larry Goldberg, Glen Gordon, John Grotting, Markku Hakkinen, Eric Hansen, Earle Harrison, Chris Hasser, Kathy Hewitt, Philipp Hoschka, Masayasu Ishikawa, Phill Jenkins, Earl Johnson, Jan Kärrman (for help with html2ps), Leonard Kasday, George Kerscher, Peter Korn, Marja-Riitta Koivunen, Josh Krieger, Catherine Laws, Greg Lowney, Scott Luebking, William Loughborough, Napoleon Maou, Charles McCathieNevile, Peter Meijer, Karen Moses, Masafumi Nakane, Mark Novak, Charles Oppermann, Mike Paciello, David Pawson, Michael Pederson, Helen Petrie, David Poehlman, Michael Pieper, Jan Richards, Hans Riesebos, Joe Roeder, Lakespur L. Roca, Gregory Rosmaita, Lloyd Rutledge, Liam Quinn, T.V. Raman, Robert Savellis, Rich Schwerdtfeger, Constantine Stephanidis, Jim Thatcher, Jutta Treviranus, Claus Thogersen, Steve Tyler, Gregg Vanderheiden, Jaap van Lelieveld, Jon S. von Tetzchner, Willie Walker, Ben Weiss, Evan Wies, Chris Wilson, Henk Wittingen, and Tom Wlodkowski,

9 References

For the latest version of any W3C specification, please consult the list of W3C Technical Reports.

[CHARMOD]
"Character Model for the World Wide Web", M. Dürst, 25 February 1999. This W3C Working Draft is http://www.w3.org/TR/1999/WD-charmod-19990225.
[CSS1]
"CSS, level 1 Recommendation", B. Bos, H. Wium Lie, eds., 17 December 1996, revised 11 January 1999. This CSS1 Recommendation is http://www.w3.org/TR/1999/REC-CSS1-19990111.
[CSS2]
"CSS, level 2 Recommendation", B. Bos, H. Wium Lie, C. Lilley, and I. Jacobs, eds., 12 May 1998. This CSS2 Recommendation is http://www.w3.org/TR/1998/REC-CSS2-19980512.
[CSS-ACCESS]
"Accessibility Features of CSS", I. Jacobs, J. Brewer, The latest version of this W3C Note is available at http://www.w3.org/TR/CSS-access.
[DOM1]
"Document Object Model (DOM) Level 1 Specification", V. Apparao, S. Byrne, M. Champion, S. Isaacs, I. Jacobs, A. Le Hors, G. Nicol, J. Robie, R. Sutor, C. Wilson, and L. Wood, eds. The 1 October 1998 DOM Level 1 Recommendation is http://www.w3.org/TR/1998/REC-DOM-Level-1-19981001
[DOM2]
"Document Object Model (DOM) Level 2 Specification", L. Wood, A. Le Hors, V. Apparao, L. Cable, M. Champion, J. Kesselman, P. Le Hégaret, T. Pixley, J. Robie, P. Sharpe, C. Wilson, eds. The DOM2 specification is a Working Draft at the time of publication.
[HTML40]
"HTML 4.0 Recommendation", D. Raggett, A. Le Hors, and I. Jacobs, eds. The 24 April 1998 HTML 4.0 Recommendation is http://www.w3.org/TR/1998/REC-html40-19980424
[MATHML]
"Mathematical Markup Language", P. Ion and R. Miner, eds. The 7 April 1998 MathML 1.0 Recommendation is http://www.w3.org/TR/1998/REC-MathML-19980407
[MICROPAYMENT]
"Common Markup for micropayment per-fee-links", T. Michel, ed. The latest version of this W3C Working Draft is available at http://www.w3.org/TR/Micropayment-Markup.
[RFC2119]
"Key words for use in RFCs to Indicate Requirement Levels", S. Bradner, March 1997.
[RFC2616]
"Hypertext Transfer Protocol -- HTTP/1.1, J. Gettys, J. Mogul, H. Frystyk, L. Masinter, P. Leach, T. Berners-Lee, June 1999.
[SMIL]
"Synchronized Multimedia Integration Language (SMIL) 1.0 Specification", P. Hoschka, editor. The 15 June 1998 SMIL 1.0 Recommendation is http://www.w3.org/TR/1998/REC-smil-19980615
[SMIL-ACCESS]
"Accessibility Features of SMIL", M-R. Koivunen, I. Jacobs. The latest version of this W3C Note is available at http://www.w3.org/TR/SMIL-access.
[UA-CHECKLIST]
An appendix to this document lists all of the checkpoints, sorted by priority. The checklist is available in either tabular form (at http://www.w3.org/WAI/UA/WD-WAI-USERAGENT-19991206/full-checklist) or list form (at http://www.w3.org/WAI/UA/WD-WAI-USERAGENT-19991206/checkpoint-list).
[UA-TECHNIQUES]
"Techniques for User Agent Accessibility Guidelines 1.0", J. Gunderson, I. Jacobs, eds. This document explains how to implement the checkpoints defined in "User Agent Accessibility Guidelines 1.0". The draft of the Techniques Document available at the time of this document's publication is http://www.w3.org/WAI/UA/WD-WAI-USERAGENT-TECHS-19991206. The latest draft of the techniques is available at http://www.w3.org/WAI/UA/WAI-USERAGENT-TECHS/
[WAI-AUTOOLS]
"Authoring Tool Accessibility Guidelines", J. Treviranus, J. Richards, I. Jacobs, C. McCathieNevile, eds. The latest Working Draft of these guidelines for designing accessible authoring tools is available at http://www.w3.org/TR/WD-WAI-AUTOOLS/
[WAI-WEBCONTENT]
"Web Content Accessibility Guidelines 1.0", W. Chisholm, G. Vanderheiden, and I. Jacobs, eds. The 5 May 1999 Recommendation is http://www.w3.org/TR/1999/WAI-WEBCONTENT-19990505
[WAI-WEBCONTENT-TECHS]
"Techniques for Web Content Accessibility Guidelines 1.0", W. Chisholm, G. Vanderheiden, and I. Jacobs, eds. The latest version of this document is available at http://www.w3.org/TR/WAI-WEBCONTENT-TECHS
[WAI-WEBCONTENT-ERT]
"Techniques For Evaluation And Implementation Of Web Content Accessibility Guidelines, C. Ridpath. The latest version of this Working Draft is available at http://www.w3.org/WAI/ER/IG/ert/.
[XML]
"Extensible Markup Language (XML) 1.0.", T. Bray, J. Paoli, C.M. Sperberg-McQueen, eds. The 10 February 1998 XML 1.0 Recommendation is http://www.w3.org/TR/1998/REC-xml-19980210
[XSLT]
"XSL Transformations (XSLT) Version 1.0", J. Clark. The 16 November 1999 Recommendation is http://www.w3.org/TR/1999/REC-xslt-19991116.

10 Services

Note. W3C does not guarantee stability for any of the following references outside of its control. These references are included for convenience.

[ALTIFIER]
The Altifier Tool for generates "alt" text intelligently.
[AMAYA]
Amaya is W3C's testbed browser/editor.
[APPLE-HI]
Information on accessibility guidelines for Macintosh applications. Information on Apple's scripting model can be found at tn1095 and tn1164. Refer also to the Inside Macintosh chapter devoted to Interapplication Communication.
[BHO]
Browser Helper Objects: The Browser the Way You Want It, D. Esposito, January 1999.
[BRAILLEFORMATS]
"Braille Formats: Principles of Print to Braille Transcription 1997" .
[CCPP]
Composite Capability/Preference Profiles (CC/PP): A user side framework for content negotiation, F. Reynolds, J. Hjelm, S. Dawkins, and S. Singhal, eds. This W3C Note describes a general yet extensible framework for describing user preferences and device capabilities. The latest version is available at http://www.w3.org/TR/NOTE-CCPP/.
[ED-DEPT]
"Requirements for Accessible Software Design", US Department of Education, version 1.1 March 6, 1997.
[EITAAC]
"EITAAC Desktop Software standards", Electronic Information Technology Access Advisory (EITAAC) Committee.
[IBM-ACCESS]
"Software Accessibility" IBM Special Needs Systems.
[ICCCM]
"The Inter-Client communication conventions manual". A protocol for communication between clients in the X Window system.
[ICE-RAP]
"An ICE Rendezvous Mechanism for X Window System Clients", W. Walker. A description of how to use the ICE and RAP protocols for X Window clients.
[JAVA-ACCESS]
"IBM Guidelines for Writing Accessible Applications Using 100% Pure Java", R. Schwerdtfeger, IBM Special Needs Systems.
[JAVAAPI]
Information on Java Accessibility API can be found at Java Accessibility Utilities.
[JAVA-CHECKLIST]
"Java Accessibility Guidelines and Checklist". IBM Special Needs Systems.
[JAVA-TUT]
"The Java Tutorial. Trail: Creating a GUI with JFC/Swing". An online tutorial that describes how to use the Swing Java Foundation Class to build an accessible User Interface.
[LYNX]
The Lynx Browser.
[MSAA]
Information on active accessibility can be found at the Microsoft WWW site on Active Accessibility.
[MS-ENABLE]
Information on accessibility guidelines for Windows applications.
[MS-KEYBOARD]
Information on keyboard assistance for Internet Explorer and MS Windows.
[MS-SOFTWARE]
"The Microsoft Windows Guidelines for Accessible Software Design". Note. This page summarizes the guidelines and includes links to the full guidelines in various formats (including plain text).
[NISO]
National Information Standards Organization. One activity pursued by this organization concerns Digital Talking Books. Refer to the "Digital Talking Book Features List" draft for more information.
[NOTES-ACCESS]
"Lotus Notes Accessibility Guidelines" IBM Special Needs Systems.
[PRODWORKS]
The Productivity Works.
[SPEAK2WRITE]
Speak to Write is a site about using speech recognition to promote accessibility.
[SUN-DESIGN]
"Designing for Accessibility", Eric Bergman and Earl Johnson. This paper discusses specific disabilities including those related to hearing, vision, and cognitive function.
[SUN-HCI]
"Towards Accessible Human-Computer Interaction", Eric Bergman, Earl Johnson, Sun Microsytems 1995. A substantial paper, with a valuable print bibliography.
[TRACE-REF]
"Application Software Design Guidelines" compiled by G. Vanderheiden. A thorough reference work.
[UNICODE]
The Unicode Consortium. "The Unicode Standard, Version 3.0", Reading, MA, Addison-Wesley Developers Press, 2000. ISBN 0-201-61633-5. Refer also to http://www.unicode.org/unicode/standard/versions/.
[USERAGENTS]
List of Alternative Web Browsers. This list is maintained by WAI.
[WHAT-IS]
"What is Accessible Software", James W. Thatcher, Ph.D., IBM, 1997. This paper gives a short example-based introduction to the difference between software that is accessible, and software that can be used by some assistive technologies.
[XGUIDELINES]
Information on accessibility guidelines for Unix and X Window applications. The Open Group has various guides that explain the Motif and Common Desktop Environment (CDE) with topics like how users interact with Motif/CDE applications and how to customize these environments. Note. In X, the terms client and server are used differently from their use when discussing the Web.