header image

Towards Barrier-free Reading in the Digital Age: Electronic Texts for the Visually Impaired

Robert Ree, FIS2309, Design of Electronic Text


Introduction

The use of electronic media for the act of reading continues to evolve. While efforts are being exerted towards technological innovation, increased attention is also being paid to the people for whom electronic documents and interfaces are designed. In this “human-centered” spirit, the following is an overview of the current implications of electronic text for a particularly vulnerable subset of the user population: the visually impaired.

It is first necessary to clarify what is meant by the terms electronic text and visually impaired. Dillon (2004) suggests that electronic texts (‘e-texts’) are fundamentally characterized by “how information technology supports the activity known as ‘reading’” (para. 2). This definition encompasses digitized versions of print, electronic document formats such as PDF, and e-books or e-journals designed to be read online or through specific hardware interfaces. In addition to ‘text,’ the above formats also commonly include elements such as images, graphics, hypertext, navigational buttons, and various reading tools such as the ability to look up words or phrases.

While many e-texts are available online, websites are not automatically included in Dillon’s characterization because they may be intended more for advertising or entertainment than for ‘reading’ per se. However, there is a substantial body of work that studies the relationships between the visually impaired and online resources. Therefore, for the purpose of this discussion, websites that provide a substantial amount of information in written form are also encompassed by the term ‘e-text’.

The term “visually impaired” is taken to refer to “all those who have a seeing disability that cannot be corrected by glasses” (Hopkins, 2000, as cited in Craven, 2003, para. 5). This definition encompasses a range of conditions, from those who are completely blind, to those who may be able to make out spatial forms but are unable to resolve the visual field to a functioning level of detail, to those who cannot discern certain colors. It is thus important to acknowledge that visual impairment is not an ‘either/or’ condition, but rather, a matter of degree.

Vision: Another Digital Divide?

In designing systems for human use, it is usually necessary to generalize about the characteristics of a population, adopting ‘average’ or ‘normal’ standards for attributes such as size and physical ability. The design of electronic media is no exception; methods of presenting information are predominantly visual, the underlying assumption being that users possess the faculty of sight (Craven, 2003). Yet a significant minority does not possess the ability to see at the levels for which many visual presentations are designed. These individuals are thus at risk of marginalization and exclusion from new developments in electronic reading, particularly as new media and digital resources grow in availability and ubiquity. Therefore, as has been similarly argued for other groups such as the financially disadvantaged, there looms the threat of another so-called ‘digital divide’ (Hackett, Parmanto, & Zeng, 2005) based on one’s ability to see.

There are current societal forces which provide impetus for seeking correction to this visual digital divide. One is demographic shift; the proportion of elderly in the population is growing, largely attributable to the aging ‘baby boom’ generation. As the average age of society increases, so will the number of people with diminished vision, for as Mogk and Goodrich (2004) point out, “the majority of those with low vision are in the latter third of their lives” (p. 589). This age group will continue to utilize electronic resources for a variety of reasons, from information seeking to reading for pleasure (Mogk & Goodrich, 2004); the question is how effectively these resources will meet the demand.

Along with aging, another societal trend worthy of concern is the increasing use of electronic text in education. Grammenos et al. (2007), in reporting their work in electronic textbook development, underscore that “access to educational content is critical to visually impaired students” (p. 107). They contend that both Braille and audio formats for textbooks have “practical problems due to their physical instantiation” and that “the delivery of textbooks in an interactive form is apparently a more promising approach…” (p. 107). If measures are not taken to ensure that visually impaired students possess the ability to access and effectively use emerging electronic learning materials, the visual digital divide may jeopardize equal opportunities for education, a threat most felt by another vulnerable group within society: the young.

Assistive Technology

A number of technological innovations assist the visually impaired to engage in the act of reading. For those with low vision, various magnifying devices are employed to enhance text visibility (Mogk & Goodrich, 2004). For those with severely-limited to no vision, Braille is a system which translates text to raised symbols which a reader interprets through touch. In the electronic realm, audio books (also known colloquially as ‘books on tape’ or ‘talking books’) have offered access to written works by means of a recording of the prose read aloud. The above methods have been in practice for several decades and were designed to aid the visually impaired in accessing and using printed text.

Today, the emergence of text representations in personal computing requires that previous assistive methods be adjusted or altogether different approaches be invented. For example, Adobe Reader and Microsoft Reader offer plug-ins with embedded speech synthesizers as well as other features which provide audible feedback on user interaction such as mouse movement and keystrokes (Grammenos et al., 2007). To assist the visually impaired in adapting to screens and monitors, various screen-scanning technologies known as ‘screen readers’ extract information from the graphical user interface and translate the contents into either auditory or Braille forms (Hacket et al., 2005). Screen reader technology is advancing rapidly, the most popular platforms being JAWS by Freedom Scientific, Hal by Dolphin, and Window-Eyes by GW Micro (Grammenos, Savidis, Georgalis, Bourddenas, & Stephanidis, 2007). In addition to screen readers, another emerging technology is the non-visual browser, a stand-alone desktop application that translates the HTML code of online information rather than the screen display (Chen, Tremaine, Lutz, Chung, & Lacsina, 2006). Because non-visual browsers decipher source code, they have the power to offer much greater interaction with features such as reading speed, navigation, hyperlinks and error messages (Chen et al., 2006).

In addition to desktop computers, mobile devices that assist the visually impaired in reading electronic text are also emerging. For example, Japanese researchers have developed a unique mobile phone which translates Braille into audio by means of an embedded camera (Zhang & Yoshino, 2008). In other more mainstream efforts, non-visual browsers are also being developed to run on standard personal digital assistants (PDAs); Chen et al. (2006) champion this approach, hypothesizing that “…the very mobility of these devices could serve as a superb aid for blind users who could then carry large amounts of useful navigation and access information that might be relevant to their mobility” (p. 4). Indeed, one of the main attractions of e-books over regular print is their inherent portability, making them very useful for travel. Mobile audio and tactile-based readers allow the visually impaired to also share in this convenience, one more measure in narrowing the visual digital divide.

Accessible Design Standards

Various standards and guidelines have been introduced to aid in the design of electronic information that is accessible to the visually impaired, similar to ‘barrier-free’ specifications in the design of environments for the physically disabled.

Accessibility … means that informa tion has been made available for almost everyone, including persons with disabilities. This accessibility may be direct or through the use of assistive technologies … Low vision users might require a large font with a sharp contrast between background and foreground color, whereas colour-blind users may need to have colour-transmittted information translated into distinguishable shades of grey. Blind users may be accessing webpages using a screen reader… [which translates] the text displayed on the computer screen into synthesized speech (Hacket et al., 2005, p. 409).

In observing both past and current work in this area, accessibility design standards appear to fall into the following categories: 1) those that enhance legibility for the user, and 2) those that facilitate the use of screen reading technology.

In the first category, screen legibility for individuals who possess some degree of vision may be enhanced through text formatting practices. An example of early research in this area is that of Fine and Pelli (1995), who examined the effects of contrast, character size, and luminance on reading rates for individuals with low vision. More recently, Arditi and Cho (2007) studied whether text case (all-caps) played a significant role in increasing text legibility for the visually impaired. Interestingly, while both studies found certain positive correlations, they also concluded that the degree of derived benefit depended largely on individual cases. Perhaps this is why customizability, rather than the prescription of specific text formats, features prominently in the web design guidelines listed in the Clear Print booklet published by the Royal National Institute for the Blind (RNIB) in the UK (Evett & Brown, 2005). While this document includes sweeping directives such as “provide a site map,” and “ensure easy navigation,” (p. 459) the RNIB guidelines also recommend that visually impaired users be offered the freedom to set background color, text size and font style according to preference. The static nature of print offers no such adjustability. One might even venture that the ability to customize electronic reading settings to one’s own particular ‘way of seeing’ is a benefit to all users, not only those with diminished vision.

The second category of standards for accessible design involves formatting and structuring electronic content to facilitate the functionality of screen reading technology. In this area, to help information designers gauge the accessibility of their documents or websites, several sets of prescriptive guidelines have been developed. The most prominent is the World Wide Web Consortium’s (W3C’s) Web Accessibility Initiative (WAI), which publishes Web Content Accessibility Guidelines (WCAG) (Hacket et al., 2005; Murphy et al., 2007). Another related project is Section 508, which sets accessibility standards for websites servicing the US government (Murphy et al., 2007). In the more specialized area of e-books, the Digital Accessible Information System (DAISY) (Grammenos et al., 2007) is a standardized format which, by virtue of “an explicit and well-defined structure” (p. 108), offers particularly seamless integration with associated reading devices. As an example of academic research in the area of effective screen reader use, Leporini and Paterno (2004) propose a comprehensive matrix of standards formulated in three categories: 1) effectiveness criteria, which cover elements that directly influence the success of a user in accomplishing desired tasks; 2) efficiency criteria, which prescribe the finding of desired information quickly and easily; and 3) satisfaction criteria, which focus on making the experience of the interface a pleasant one. (Leporini and Paterno’s model holds many conceptual similarities to Dillon’s (2004) “TIME” framework for e-text design, which provides guidelines broken down into tasks, information modeling, manipulation, and ergonomics.) Thus, for designers mandated to create accessible electronic documents, established systems and resources are readily available.

Yet despite these efforts, awareness of accessibility issues among designers remains low (Murphy et al., 2007), coinciding with low percentages of ‘acceptable’ websites in accessibility compliance evaluations (Hackett et al., 2005). A reason for this may be the current trend in information design toward increasing amounts complexity and multimedia content:

Features that add complexity to webpages are often the barriers to accessibility … Multimedia requires users to rely on more than one sense, usually vision and hearing, to fully comprehend the information being presented (Hackett et al., 2005, p. 408).

Murphy, Kuber, McAllister, Strain, and Yu (2007) also observe that “assistive devices inadequately handle graphics, moving images, frames, tables, use of scripting languages and streaming audio and video” (p. 79). Although some work is beginning to be carried out in developing specialized multimedia readers for the visually impaired such as the aiBrowser (Miyashi, Sato, Takagi, & Asakawa, 2007), it is evident that incorporation of multimedia in electronic information delivery represents one of the greater challenges for visual accessibility in the near future.

Appropriate Mental Models

A recurring theme in the literature on the use of electronic interfaces by the visually impaired is the challenge of conveying to the user the overall structure of the document, a quality that is generally accomplished through visual cues. As the second component in his TIME framework for e-texts, Dillon (2004) identifies the importance of the information model:

Readers possess (from experience), acquire (while using) and utilize a representation of the document’s structure that may be termed the mental model of the text or information space. Such models allow readers to identify likely locations for information within the document, to predict the typical contents of a document, to know the level of detail likely to be found and to appreciate the similarities between documents (p. 138).

Various iterations of Dillon’s concept have found particular relevance in current developments in visually impaired electronic reading. In the design of an accessible e-textbook technology called the Starlight Platform, Grammenos et al. (2007) seek improvement of existing platforms on this basis, stating that “…in order to effectively use them, s/he must possess an accurate mental model and a good understanding of their visual structure and layout” (p. 108). Similarly, in their research on use by the visually impaired, Murphy et al. (2007) observe:

the mental model of the perceived through a screen reader is very different to the model formed by sighted users using a graphical interface. Most users in this study said that they would like to have more spatial information presented to them (p. 89).

The problem of insufficient model representation stems from the linear mode in which screen readers typically function. Information is translated in the order that it appears; left columns of text must be read prior to right columns, and so on (Yesilada, Harper, Goble, & Stevens, 2004). While this serial manner of reading may suit certain types of texts such as novels which progress ‘from start to finish,’ it does not suit other information sources such as textbooks, where ‘skimming’ of the information may be valuable, and navigation may rely heavily on tables of contents and indexes. Translation difficulties are compounded for sources with complex graphic layouts in which information is divided into discrete zones or “chunks” (Yesilada et al., 2004), as this encourages reading different parts of the screen in an order chosen by the viewer.

A particularly compelling framework for building effective mental models for visually impaired readers is put forth by Yesilada et al. (2004), who adopt a strong metaphorical link between the structure of electronic documents and the structure of physical environments. Accessibility is equated with mobility:

Accessibility, and in particular mobility, is… about how easy it is for a traveler to complete a success- ful journey … [S]creen readers can- not see and understand the structural semantics implicitly encoded in the page so the mobility support is limited and fragile (p. 447).

Rather than be carried along a defined path from one set place to another, Yesilada et al. (2004) advocate the experience of reading in the electronic realm as a self-directed exploration. Thus, the mental model for accessible electronic documents is a “model of travel” (p. 448), which the authors formulate into an encoding system called the Travel Ontology. This framework takes on the difficult task of identifying a coherent vocabulary capable of describing the experiential and decision-making aspects of an e-text ‘journey.’ With the vocabulary articulated, the text may then be represented to a visually impaired user in a more spatial, navigable way.

While the detailed implementation of the Travel Ontology is beyond the scope of this discussion, what is important about it is that it represents perhaps the most definitive paradigm shift away from the mental model of linear screen reading, which has long been the norm for the visually impaired. By regarding the electronic reading interface as the control panel for a journey, and as much more than a digital version of print, it attempts to bring to the visually impaired many of the dynamic, convenient, and interactive aspects of electronic text so enjoyed by sighted readers.

Conclusion

It is evident from this discussion that many barriers to the effective access and use of electronic text by visually impaired users remain. The visual and increasingly complex and media-rich nature of electronic interfaces poses the threat of a digital divide based on sight. In response, concerted efforts are being made to provide visual accessibility through both technological innovation as well as accessible design practices. However, work that is post-hoc in nature (fixing problems after they arise), such as the use of serial screen readers, may prove useful in the short-term but hold less promise for the future. Instead, more compelling approaches rethink the problem conceptually, by reformulating and building more appropriate mental models for visually impaired reading in the electronic realm. By doing so, it is hoped that the use of information technology for reading will be transformed into an experience of spatial discovery, an exploratory journey rather than set path.

Works Cited

Arditi, A. & Cho, J. (2007). Letter case and text legibility in normal and low vision. Vision Research, 47(19), 2499-2505.

Chen, X., Tremaine, M., Lutz, R., Chung, J., & Lacsina, P. (2006). AudioBrowser: A mobile browsable information access for the visually impaired. Universal Access in the Information Society, 5(1), 4-22.

Craven, J. (2003). Access to electronic resources by visually impaired people. Information Research, 8(4), 454.

Dillon, A. (2004). Designing usable electronic text: Ergonomic aspects of human information usage, 2nd ed. London: Taylor & Francis.

Evett, L. & Brown, D. (2005). Text formats and web design for visually impaired and dyslexic readers - clear text for all. Interacting with Computers, 17(4), 453-572.

Fine, E. M. & Peli, E. (1995). Enhancement of text for the visually impaired. Journal of the Optical Society of America, 12(7), 1439-1447.

Grammenos, D., Savidis, A., Georgalis, Y., Bourddenas, T., & Stephanidis, C. (2007). Dual educational electronic textbooks: The starlight platform. Proceedings of the 9th International ACM SIGACCESS Conference on Computers and Accessibility (pp. 107-114). New York, NY: Association for Computing Machinery Inc.

Hackett, S., Parmanto, B., & Zeng, X. (2005). A retrospective look at website accessibility over time. Behaviour Information Technology, 24(6), 407-417.

Leporini, B. & Paternò, F. (2004). Increasing usability when interacting through screen readers. Universal Access in the Information Society, 3(1), 57-70.

Miyashita, H., Sato, D., Takagi, H., & Asakawa, C. (2007). AiBrowser for multimedia: Introducing multimedia content accessibility for visually impaired users. Proceedings of the 9th International ACM SIGACCESS Conference on Computers and Accessibility (pp. 91-98). New York, NY: Association for Computing Machinery Inc.

Mogk, L. & Goodrich, G. (2004). The history and future of low vision services in the United States. Journal of Visual Impairment Blindness, 98(10), 585-600.

Murphy, E., Kuber, R., McAllister, G., Strain, P., & Yu, W. (2008). An empirical investigation into the difficulties experienced by visually impaired users. Universal Access in the Information Society, 7(1-2), 79-91.

Yesilada, Y., Harper, S., Goble, C., & Stevens, R. (2004). Screen readers cannot see - ontology based semantic annotation for visually impaired web travellers. Lecture Notes in Computer Science, 3140, 445-457.

Zhang, S. & Yoshino (2008). A Braille recognition system by the mobile phone with embedded camera. Second International Conference on Innovative Computing, Information and Control. Inst. of Elec. and Elec. Eng. Computer Society.