introduction
Text-only Web techniques are an important set of tools in every Web developer’s repertoire. When fully utilized, they allow developers to render multiple types of information, in various organizations, into linear text. Examples of these techniques include text descriptions or transcripts of non-text media; the ability to reorganize tabular text into linear information; and label placement for Web forms. Using text-only Web techniques does not imply the exclusion of other media types from a Web site. Rather, text-only Web techniques allow for media richness by providing clear, speedy and intuitive access for many types of users. Although the term “text only” commonly arises in discussions of assistive technology, text-only techniques provide benefits for all Web users. This article will discuss some of the motivations for developers to use text-only techniques, including adherence to World Wide Web consortium standards. It also will describe the three primary technical strategies for creating a text-equivalent version of a Web page.
advantages of text-only web sites
The goal of most Web sites is to provide information, whether for free or for a fee. In today’s society, most of our information resides in text format, but the media-rich nature of the Web is increasing the number of images, audio and video clips, and animations. There are several reasons a developer might choose to provide a text equivalent for these non-text media. Since most of our systems were originally built to accommodate text, text format is one of the most portable document formats. Text files easily transfer between different pieces of proprietary software, and a variety of handheld devices, such as pen scanners, depend on text format. Other clients that make use of text-only equivalents include screen readers, dynamic braille displays, screen magnifiers and graphical browsers with adaptive technology such as Amaya (Vatton, 2003). Perhaps the most demanding text-only client is a text-only Web browser. These can process media-rich pages, but can have difficulty rendering the non-text elements into text format. If text equivalents are not provided, the user is left with incomplete information. Perhaps the most famous text-only browser is LYNX, which has existed in open-release form since 1995 (Lynx Developers, 2001). LYNX is fully documented and includes guidelines for writing LYNX-friendly Web pages. A special configuration of LYNX, called BLYNX, makes it particularly effective for blind and visually impaired users; more information about BLYNX can be found at http://leb.net/blinux/blynx/ (Rosmaita, 1997).
Another reason developers might want to provide text equivalents is to accommodate users’ equipment. Users who experience slow connection times may use a text-only browser to access information quickly. Since one of the great advantages of the Internet is its global connectivity, developers need to consider connections from developing countries, public schools and libraries, and home-based users who rely on dial-up connections.
Others appreciate the harmony between text and assistive technology. Developers for public institutions are required by law to consider the needs of assistive technology users (28 C.F.R. 35.160). Assistive technology, which includes such items as screen readers or Braille output devices, handles text quite well, making text-only pages a good design solution. Due to the variety of users’ abilities and equipment, using text-only techniques will not fulfill all the requirements of accessible Web design, but they are an important component to ensure access to all users. In addition, developers may wish to create pages accessible to mobile wireless devices such as cell phones, cars and handheld personal computers. For all these reasons, concentrating the bulk of information on a Web site in text format assists a wide variety of users with access.
accessibility standards
One of the primary groups involved in setting Web accessibility standards is the World Wide Web Consortium (W3C). The W3C provides three overall guidelines for the provision of text-only versions. First, they urge the separation of structure from presentation. While the structure of the page concerns the logical organization of information, the presentation refers to such issues as font, color and images (W3C, 2003). The W3C consortium is also encouraging developers to separate structure and presentation by recommending new versions of HTML, the newest versions of which are called XHTML (W3C, 2003b).
The second relevant W3C guideline is to create documents that do not rely on one type of hardware. As discussed throughout this chapter, pages should be usable by people without mice, with small screens, low-resolution screens, black-and-white screens, and with only voice or text output. In addition to the technologies covered by the W3C guidelines, developers may wish to consider users of mobile, wireless technology, with consideration of monochrome displays and small viewing areas.
Finally, the W3C urges developers to provide accessible equivalents. Text can be rendered in ways that are available to almost all browsing devices and accessible to almost all users. Therefore, if a developer chooses to create a media-rich page, it is important to make the information available in another format, particularly making it usable by someone who is visually or hearing impaired. According to W3C, developers will have successfully provided an accessible text equivalent for all non-text content if:
1. All non-text content is explicitly associated with a text equivalent (images have alt-text, movies have collated text transcripts, animations have descriptions, interactive scripts have a functional equivalent such as a form, audio files have a text transcript).
2. The text equivalent fulfills the same function and conveys the same information as the non-text content.
3. Where it is not possible to describe the non-text content in words, or for text to provide the same function as the non-text content, a label identifying the content is provided (Chisholm, White & Vanderheiden, 2001).
One easy test W3C recommends is to imagine reading the document aloud over the telephone, noting which page elements require description to make the information meaningful to the listener (Chisholm, Vanderheiden & Jacobs, 1999). For full documentation of the W3C standards for Web accessibility, refer to the Web Accessibility Initiative (WAI) home page, www.w3.org/WAI/.
technical strategies
Currently, three primary technical strategies exist for creating a text-equivalent version of a Web page. These include creating an entire reproduction of the Web site in a text-only format; using available tools to convert “regular” pages to text-only equivalents “on the fly”; and providing text equivalents throughout a Web site where information is conveyed in media other than text.
For certain Web sites or applications, one might consider making a text-only reproduction of a Web site. This has obvious disadvantages for large Web sites, but may be a simple solution for others. For especially straightforward Web pages, developers might consider making the original Web site itself fully text-only. The advantages to this are having only one version of the Web site that fulfils all the text-only criteria. These pages also load quickly and easily transform to alternate formats. The disadvantages include little or no creative freedom in designing the layout of the page for visual users. An example of this is the Lynx home page, http://lynx.browser.org/. Although it is plain, it conveys the information equally well to all users.
A solution that addresses most of the above disadvantages of reproducing a Web site while still providing text-only access is to use a scripting program to transform Web pages “on the fly” from media-rich pages into text-only Web pages. One example of this can be found at the BBC Web site, www.bbc.co.uk/. Clicking on the link for the text-only version transforms the page on the fly. On the text-only version, for example, it is clear it has just been transformed because the current weather is still current. The program used to accomplish the transformation is called Betsie, a freeware program that can be downloaded and modified to fit the needs of individual Web sites and applications (British Broadcasting Corp., 1999).
Even without scripting, one can avoid reproducing a Web site and still use graphics and other media by providing text equivalents when necessary. This allows developers more creative freedom with layout and graphics, but requires development and maintenance of all the text equivalents. The equivalent pieces of text would need to be updated every time the “main” version of the Web site was updated. There are several techniques for providing text-only equivalents on the Web for various media types.
Images, Bullets and Image Maps
Providing a text equivalent for images and image-like objects is syntactically easy, but may take some semantic thought. The W3C guideline reads, “Text equivalents must be written so that they convey all essential content” (Chisholm, Vanderheiden, & Jacobs, 1999). With images, the most important change is to add an alt attribute to the <IMG> tag, which provides a description of the image. Complex or lengthy text equivalents can be provided with the “longdesc” attribute (Chisholm, Vanderheiden & Jacobs, 2000). The developer may even use both “alt” and “longdesc” to provide options for familiar and first-time readers. Developers should provide alt tags for regions of image maps and may also wish to provide redundant text links (for example, at the bottom of the image map) to the same URLs found in the image map (Chisholm, Vanderheiden & Jacobs, 2000). If bulleted lists use images as bullets but convey no meaning, the alt tag suggested by the W3C is alt=”*” (Chisholm, Vanderheiden & Jacobs, 2000). If, however, the bullets convey meaning – for example, if red spinning globes are used for new items in a list and yellow spinning globes are used for old items – the alt tags should convey that information: alt = “New:” and alt = “Old:”.
Applets and programmatic objects
For the OBJECT element used to insert applets, developers can provide text equivalents between the opening and closing <OBJECT> tags. If the APPLET element is used (now deprecated), developers may provide an alt attribute within the <APPLET> tag in addition to content between the opening and closing <APPLET> tags (Chisholm, Vanderheiden & Jacobs, 2000).
Frames
Frames, which split a Web page into different visual sections, present several difficulties relating to access. The reason the Web page appears in different sections when using frames is because there are actually separate Web pages for each “frame,” which are organized by a parent document in a visually sensible arrangement on the screen. These frames can be thought of as separate windows, which means developers need to provide information about the actual organization of the page. This can be done with the <FRAME> attribute “title” to provide a description of what information the frame contains. Further descriptions for frames can be provided through the “longdesc” attribute, as with images (Chisholm, Vanderheiden & Jacobs, 2000). Note that if the content of the frame changes – for example, if a navigation bar is replaced by a different or expanded navigation bar as the user moves through the site – equivalence is lost, since the description of the frame is no longer accurate.
video and Audio
Text equivalents can be provided for audio and video elements with text transcripts, captions and, for video, auditory descriptions with transcripts. Transcripts should include non-spoken sounds, such as sound effects. Captions are text transcripts for the audio track of video and/or audio presentations synchronized with the video and/or audio tracks:
“Captions are generally rendered visually by being superimposed over the video, which benefits people who are deaf and hard of hearing, and anyone who cannot hear the audio (e.g., when in a crowded room). A collated text transcript combines (collates) captions with text descriptions of video information (descriptions of the actions, body language, graphics, and scene changes of the video track)” (Chisholm, Vanderheiden & Jacobs, 1999).
In addition to making visual presentations accessible to the visually impaired, transcripts and captions make the information available to search engines.
Tables
The W3C guidelines for providing equivalents for tables are quite lengthy (Chisholm, Vanderheiden & Jacobs, 2000). This is largely because some screen readers read straight across lines of tables, regardless of table data cell breaks. When using tables, developers should provide a caption and summary of the information contained in the table. The “caption” attribute is a title describing the table contents, while the “summary” attribute is intended to be a longer and more complete description. Developers should also identify row and column information so that data cells and header cells can be associated correctly, as specified by the W3C (HTML Techniques, 1999).
Links
Although hyperlinks are usually text, there is an HTML attribute for the <A> tag, “title” that allows developers to specify a value of the “title” attribute that clearly and accurately describes the target of the link (Ch-isholm, Vanderheiden, & Jacobs, 2000). This allows the developer freedom to choose link text that makes sense, but also describe the link target. Although the use of “click here” should be limited, one could use the “title” attribute to explain what will happen if the users “click there.”
Forms
Forms can be rendered into text equivalents more easily if form elements are grouped. To this end, developers may use the FIELDSET element and label those units with the LEGEND element. Also, labels for form elements should be placed before the form element to which they correspond. For an order form, the label “First Name:” should appear before the input box where the user should type his/her first name (Chisholm, Vanderheiden & Jacobs, 2000).
New techniques for providing alternate versions or presentations of Web pages, including text-only versions, surround the use of Extensible Markup Language (XML). XML pursues the idea of having a single format for a wide range of uses. With XML, users have the opportunity to display information in a manner that best suits their equipment. For a detailed explanation of XML, visit the W3C XML Web pages (www.w3.org/XML/).
conclusion
New output devices and browser features using text-format data will continue to be invented, limited only by human creativity. Several examples currently in development are voice-driven navigation systems (Is-Sound Corp., 2001), image-to-sound converters (Meijer, 2004) and e-mail based Web browsers (Secret & Sasse, 2004; PageGetter.com, 2000). These technologies are just a few examples of reasons to adhere to standards such as those set by W3C. Linear text is currently the most common medium used by humans to communicate. By understanding how to provide text versions of other media, developers will allow the most users to be successful in accessing information, no matter what access devices they choose.
key terms
Applet: A computer program that is portable between operating systems and requires only minimal memory to run, often written in the Java programming language.
Assistive Technology: Any of a number of devices designed to bridge between standard technology interfaces and people with different abilities.
Caption: A short textual description used to summarize a picture, table or other non-text information.
Form Element: Any part of an HTML form, including input boxes, check boxes, pull-down menus, submit or reset buttons, or radio options.
Hypertext Markup Language (HTML): A set of tags used to structure text and multimedia documents and to set up hypertext links between documents, most commonly on the World Wide Web.
Media Rich: Adjective describing the use of nontext media, such as images, sound or video.
Text Equivalent: An alternative version of a document that provides the same functions and conveys the same information as the media-rich version, but without using non-text media.
Web Accessibility Initiative: An initiative of the W3C promoting the use of Web techniques to provide equal access to information, regardless of ability.
World Wide Web Consortium (W3C): An international organization devoted to the development of interoperable technologies such as specifications, guidelines, software and tools, to lead the Web to its full potential.
