Yeliz Yesilada

Accessibility: a Web engineering approach

Currently, the vast majority of web sites do not support accessibility for visually impaired users. Usually, these users have to rely on screen readers: applications that sequentially read the content of a web page in audio. Unfortunately, screen readers are not able to detect the meaning of the different page objects, and thus the implicit semantic knowledge conveyed in the presentation of the page is lost. One approach described in literature to tackle this problem, is the Dante approach, which allows semantic annotation of web pages to provide screen readers with extra (semantic) knowledge to better facilitate the audio presentation of a web page. Until now, such annotations were done manually, and failed for dynamic pages. In this paper, we combine the Dante approach with a web design method, WSDM, to fully automate the generation of the semantic annotation for visually impaired users. To do so, the semantic knowledge gathered during the design process is exploited, and the annotations are generated as a by-product of the design process, requiring no extra effort from the designer.

Evaluating DANTE: Semantic transcoding for visually disabled users

The importance of the World Wide Web for information dissemination is indisputable. However, the dominance of visual design on the Web leaves visually disabled people at a disadvantage. Although assistive technologies, such as screen readers, usually provide basic access to information, the richness of the Web experience is still often lost. In particular, traversing the Web becomes a complicated task since the richness of visual objects presented to their sighted counterparts are neither appropriate nor accessible to visually disabled users. To address this problem, we have proposed an approach called Dante in which Web pages are annotated with semantic information to make their traversal properties explicit. Dante supports usage of different annotation techniques and as a proof-of-concept in this article, pages are annotated manually which when transcoded become rich. We first introduce Dante and then present a user evaluation which compares how visually disabled users perform certain travel-related tasks on original and transcoded versions of Web pages. We discuss the evaluation methodology in detail and present our findings, which provide useful insights into the transcoding process. Our evaluation shows that, in tests with users, document objects transcoded with Dante have a tendency to be much easier for visually disabled users to interact with when traversing Web pages.

Screen readers cannot see: Ontology based Semantic annotation for visually impaired Web travellers

Travelling upon the Web is difficult for visually impaired users since the Web pages are designed for visual interaction [6]. Visually impaired users usually use screen readers to access the Web in audio. However, unlike sighted users, screen readers cannot see the implicit structural and navigational knowledge encoded within the visual presentation of Web pages. Therefore, in a visually impaired user’s environment, objects that support travel are missing or inaccessible. Our approach to remedy this is to annotate pages with an ontology, the Travel Ontology, that aims to encapsulate rich structural and navigational knowledge about these objects. We use Semantic Web technologies to make such knowledge explicit and computationally accessible. Our semi-automated tool, Dante identifies travel objects on Web pages, annotates them appropriately with the Travel Ontology and uses this to transform the pages to enhance the travel support. Thus Dante uses the Travel Ontology to enhance the travel experience of visually impaired users. This paper introduces the Travel Ontology, the annotation pipeline used in the annotation part of Dante and some transformation scenarios to illustrate how the annotations are used to guide the transformation of Web pages.

A foundation for tool based mobility support for visually impaired web users

Users make journeys through the Web. Web travel encompasses the tasks of orientation and navigation, the environment and the purpose of the journey. The ease of travel, its mobility, varies from page to page and site to site. For visually impaired users, in particular, mobility is reduced; the objects that support travel are inaccessible or missing altogether. Web development tools need to include support to increase mobility. We present a framework for finding and classifying travel objects within Web pages. The evaluation carried out has shown that this framework supports a systematic and consistent method for assessing travel upon the Web. We propose that such a framework can provide the foundation for a semi-automated tool for the support of travel upon the Web.

Rendering tables in audio: the interaction of structure and reading styles

Tables remain a persistent problem for visually impaired people using screen readers. Tables are complex structures that are widely used for different purposes such as spatial layout or data summarisation. The multi-dimensional nature of tables challenges the linear interaction styles typically supported by screen readers. To read a table, a user needs to maintain coherency of, and interact with more than one dimension. In this paper, we first characterise why tables are useful in print, but difficult to read in the audio. We present a survey of the relationship between table structure, intention and the reading styles employed to use the content of tables. We then present two different approaches for interacting with tables non-visually. These approaches are designed to support the characteristics of tables that make them such a popular and useful means of conveying information. The first approach provides a small table browser called EVITA (Enabling Visually Impaired Table Access), whose aim is to enable non-visual table browsing and reading in an analogous manner to the print medium. The second approach provides a table lineariser to transform tables into a form such that they can be easily read by screen readers.

What input errors do you experience? Typing and pointing errors of mobile Web users

Small devices such as personal digital assistants (PDAs) are widely used to access the World Wide Web (Web). However, accessing the Web from small devices is affected by poor interface bandwidth, such as small keyboards and limited pointing devices. There is little empirical work investigating the input difficulties caused by such insufficient facilities, however, anecdotal evidence suggests that there is a link between able-bodied users of the mobile Web and motor impaired users of the Web on desktop computers. This being the case we could transfer the solutions which already exists for motor impaired users into the mobile Web and vice versa. This paper presents a user study that investigates the input errors of mobile Web users in both typing and pointing. The study identifies six types of typing errors and three types of pointing errors shared between our two user domains. We find that mobile Web users often confuse the different characters located on the same key, press keys that are adjacent to the target key, and miss certain key presses. When using a stylus, they also click in the wrong places, slide the stylus during multiple clicks, and make errors when dragging. Our results confirm that despite using different input devices, mobile Web users share common problems with motor impaired desktop users; and we therefore surmise that it will be beneficial to transfer available solutions between these user domains in order to address their common problems.

Testability and validity of WCAG 2.0: the expertise effect

Web Content Accessibility Guidelines 2.0 (WCAG 2.0) require that success criteria be tested by human inspection. Further, testability of WCAG 2.0 criteria is achieved if 80% of knowledgeable inspectors agree that the criteria has been met or not. In this paper we investigate the very core WCAG 2.0, being their ability to determine web content accessibility conformance. We conducted an empirical study to ascertain the testability of WCAG 2.0 success criteria when experts and non-experts evaluated four relatively complex web pages; and the differences between the two. Further, we discuss the validity of the evaluations generated by these inspectors and look at the differences in validity due to expertise. In summary, our study, comprising 22 experts and 27 non-experts, shows that approximately 50% of success criteria fail to meet the 80% agreement threshold; experts produce 20% false positives and miss 32% of the true problems. We also compared the performance of experts against that of non-experts and found that agreement for the non-experts dropped by 6%, false positives reach 42% and false negatives 49%. This suggests that in many cases WCAG 2.0 conformance cannot be tested by human inspection to a level where it is believed that at least 80% of knowledgeable human evaluators would agree on the conclusion. Why experts fail to meet the 80% threshold and what can be done to help achieve this level are the subjects of further investigation.

How much does expertise matter?: a barrier walkthrough study with experts and non-experts

Manual accessibility evaluation plays an important role in validating the accessibility of Web pages. This role has become increasingly critical with the advent of the Web Content Accessibility Guidelines (WCAG) 2.0 and their reliance on user evaluation to validate certain conformance measures. However, the role of expertise, in such evaluations, is unknown and has not previously been studied. This paper sets out to investigate the interplay between expert and non-expert evaluation by conducting a Barrier Walkthrough (BW) study with 19 expert and 51 non-expert judges. The BW method provides an evaluation framework that can be used to manually assess the accessibility of Web pages for different user groups including motor impaired, hearing impaired, low vision, cognitive impaired, etc. We conclude that the level of expertise is an important factor in the quality of accessibility evaluation of Web pages. Expert judges spent significantly less time than non-experts; rated themselves as more productive and confident than non-experts; and ranked and rated pages differently against each type of disability. Finally, both effectiveness and reliability of the expert judges are significantly higher than non-expert judges.

Middleware to expand context and preview in hypertext

Movement, or mobility, is key to the accessibility, design, and usability of many hypermedia resources (websites); and key to good mobility is context and preview by probing. This is especially the case for visually impaired users when a hypertext anchor is inaccurately described or is described out of context. This means confusion and disorientation. Mobility is similarly reduced when the link target of the anchor has no relationship to the expected information present on the hypertext node (web-page). We suggest that confident movement with purpose, ease, and accuracy can only be achieved when complete contextual information and an accurate description of the proposed destination (preview) are available. Our past work (1) deriving mobility heuristics from mobility models, (2) transforming web-pages based on these heuristics, and(3) building tools to analyse and access these transformed pages; has shown us that a tool to expand context and preview would be useful. In this paper we describe the development of such a middleware tool to automatically and dynamically annotate web-pages with additional context information present within the page, and preview information present within hypertext link destinations found on the page.

The Expertise Effect on Web Accessibility Evaluation Methods

Web accessibility means that disabled people can effectively perceive, understand, navigate, and interact with the web. Web accessibility evaluation methods are needed to validate the accessibility of web pages. However, the role of subjectivity and of expertise in such methods is unknown and has not previously been studied. This article investigates the effect of expertise in web accessibility evaluation methods by conducting a Barrier Walkthrough (BW) study with 19 expert and 57 nonexpert judges. The BW method is an evaluation method that can be used to manually assess the accessibility of web pages for different user groups such as motor impaired, low vision, blind, and mobile users. Our results show that expertise matters, and even though the effect of expertise varies depending on the metric used to measure quality, the level of expertise is an important factor in the quality of accessibility evaluation of web pages. In brief, when pages are evaluated with nonexperts, we observe a drop in validity and reliability. We also observe a negative monotonic relationship between number of judges and reproducibility: more evaluators mean more diverse outputs. After five experts, reproducibility stabilizes, but this is not the case with nonexperts. The ability to detect all the problems increases with the number of judges: With 3 experts all problems can be found, but for such a level 14 nonexperts are needed. Even though our data show that experts rated pages differently, the difference is quite small. Finally, compared to nonexperts, experts spent much less time and the variability among them is smaller, they were significantly more confident, and they rated themselves as being more productive. The article discusses practical implications regarding how BW results should be interpreted, how to recruit evaluators, and what happens when more than one evaluator is hired. Supplemental materials are available for this article. Go to the publishers online edition of Human–Computer Interaction for statistical details and additional measures for this article.