(home : resources: alternative formats)

• Alternatives for Visual Presentations
• Alternatives for PDF and RTF Formats
• Alternatives for Multimedia
• Alternative Text Navigation

Alternative Formats

Failing to provide alternative formats is the most prevalent access barrier found on the Web. The most common barrier is images that do not include a text equivalent alternative, which makes information contained within the images inaccessible to a person who is blind. In addition to accommodating a blind person who can not see images, there are other reasons why text alternatives must be used. For example, sites should provide text content for visitors using older technology or text-based browsers, or content for those browsing with the image display option turned off. Without a text alternative, any information conveyed through images is unavailable to these users.

Practical Recommendations for Providing Alternative Formats

1. Alternatives for Visual Presentations

The ALT or LONGDESC attributes should be included for most page components, including images, image maps, animations, programmed objects (e.g.. applets), audio and video, or complex frames or tables (generically these components are referred to here as objects). The ALT attribute is used where a short description is sufficient. The LONGDESC attribute is used where more detail is required to convey the meaning contained in the object being described. A good test of an object's description is to imagine describing it to someone over the phone. Will they be able to accurately describe to another person who has not seen the object, the meaning that the object conveys?

The LONGDESC attribute was introduced in HTML 4.0, though is not widely supported as yet. More commonly used are d-links. D-links (definition or description links) are often seen on accessibility conscious sites as a "d" positioned immediately following a page object, such as a complex image. For developers who would rather not have "d's" scattered throughout their pages, d-links may be created using invisible 1 x 1 gifs with the ALT text "d-link". Clicking or following a d-link takes the user to an anchor on another page where a long description of the object is provided. All object descriptions for a site can be contained on a single page. Currently, LONGDESC should be used in a fashion similar to d-links, entering the location of the description as the LONGDESC value, but also linking the object itself to the same location, as seen in the markup below. When LONGDESC is more widely supported, d-links and objects linked to their descriptions can be eliminated.

In descriptions.php:

Often, a description of a complex object can be included within the content of the page in which it appears. For example, it is standard practice to refer to figures or tables within the text of a document. Expanding these references to include a brief description of the object will assist both sighted and non-sighted users in comprehending its content. A similar strategy can be used for other complex objects such as applets and other programmatic objects.

2. Alternatives for PDF and RTF Formats

Many sites include reformatted documents originally created in a word processor or other text or image processing software. A simple method of providing these documents to visitors is to create Portable Document Format (PDF) files. A PDF file is an image of a document, and as such is not accessible to a screen reader or text reader. Often Rich Text Format (RTF) is used as well. RTF files are accessible to most screen reading and text reading programs, provided the user has software that will display RTF formatted files.

Tools are available for converting these file formats into universally accessible formats. Many word processing programs that are capable of creating RTF files, are also capable of creating HTML files, the preferred accessible format. Adobe, the developers of Acrobat Reader, used to display PDF files, offers tools that will convert PDF files into HTML files. Conversion tools are a relatively new technology and they often produce imperfect HTML documents. If these tools are used, a developer should review the HTML they generate to ensure the coding is correct, and to ensure that new barriers have not been created during the conversion process. Since screen readers often use HTML formatting to interpret the content of a page, errors in page markup can create access barriers.

Accessibility Tools for PDF Documents

3. Alternatives for Multimedia

Multimedia objects are often inaccessible to people with disabilities, depending on the media used. Animation or video clips will be inaccessible to a blind visitor, audio or sound tracks accompanying animation or video clips will be inaccessible to deaf users. In some cases a screen reader will not recognize that a multimedia component is embedded in a page. Some multimedia authoring tools now provide methods for including alternative text descriptions for visual components, and captioned text for audio components.

In the same way LONGDESC or d-links are used to provide an alternative format for complex images and similar objects, LONGDESC of d-links can be used to describe multimedia components as well. Often a transcript of a multimedia piece will provide a sufficiently accessible format for both blind and deaf users. In some cases a transcript can be supplemented with context information provided in a long description, such as describing the surroundings in a video clip, or perhaps tone of voice being used in an audio feed.

See an example of a Macromedia Flash movie with closed captioning, played in Real Player (requires Real Player version 7+). A transcript is also included with this piece.

See an example of a keyboard accessible Flash movie, with audio, and a d-link/transcript alternative (requires Macromedia Flash Player).

4. Alternative Text Navigation

Although most current screen reading technology will read ALT text included with images and other visual objects, there are still those that will not. Until all assistive technology consistently interprets ALT text, redundant text should be used as well, particularly for navigation mechanisms. Image maps, for example, should include in addition to clickable "hot areas" with ALT text, a set of text links equivalent to the links of the map areas. You'll notice on pages of the Web-savvy site, the text links at the bottom of the page are equivalent to those in the button bar across the top. Also notice that each of the text links is separated by a non-linked character, in this case a "pipe". Without non-linked character between links, some screen readers will string together adjacent links into what sounds like one continuous link text. Separating links with a non-linked character places a pause between each, allowing the listener to distinguish between links.