UNIVERSAL USABILITY IN
PRACTICE
Introduction | Background | Guidelines and Recommendations | Examples of Successful Websites | Conclusion | Resources
Ser Nam Lim
Department of Computer Science
University of
Maryland, College Park, MD 20742 USA
sernam@cs.umd.edu
April 18,
2001
Motivation is synonyms to stimulus, which means something that rouses the mind or spirits or incites to activity. (Merriam-Webster Collegiate Thesaurus [3]) Low motivation users will then be users who lack stimulus to visit a website. These users could be homeless, with no access to the Internet, have low income [4,5] or jobless. More often than not, these users are also illiterate. We will first address both low education and consequently low motivation users in term of their common denominator - functional illiteracy. We will also address another group of low motivation users, who is similarly unmotivated but may not be totally illiterate. This second group of users can often be elderly who have lost touch with technology, professionals in disciplines that is not computer related, homemakers etc. These users most probably are able to read and write but have lost their touch with technologies, which in our case is the World Wide Web related technologies.
In general, website designers could identify the profiles that are pertinent to the functions of their websites. By profiles, we mean the level of motivation that is required for different websites. Identification of the profiles help designers to understand the amount of motivation that they need to instill in users [26]. Listed below is the seven profiles for websites based on a matrix of high information, high motivation to low information, low motivation:
Profile 1: No Expectations - very low need to motivate users or deliver content - e.g. personal home pages.
Profile 2: All Motivation - high need to motivate users to view site - e.g. promotional commercial sites.
Profile 3: All Content - high need to deliver content where site may be the only provider of this information or users are highly motivated to use site already- e.g. search engines or research results.
Profile 4: High Motivation - a need to provide some content along with motivating factors, at same time users must be able to distinguish between content and glitz.
Profile 5: High Content - need for content outweighs the need for high motivational factors but an attractive site is necessary - e.g. government agencies, universities.
Profile 6: Mixed Elements and Profile 7: Great Expectations - in both these profiles the need to deliver specific content and the need to motivate users to a specific response are highly interdependent and interrelated e.g. commercial catalogue sites, sites devoted to charitable or political causes, or educational sites.
1. Audio and speech compression and transmission tools: These are abundant and greatly improve the quality and practicality of delivering audio over the Internet. They are used in multimedia pages, Internet radio stations and Internet phones. They are a necessary component of any proposed system, but they serve only to add general audio capabilities.
2. Specialized software is required to apply these capabilities to the function of text-to-audible-speech conversion.
3. Specific aids for visually impaired: An example of specialized software mentioned above, these tools can read parts of a page aloud, but the page is extensively stripped and reformatted to suit the needs of visually impaired persons. The result is not suitable for those who can benefit from visual information or as an all-purpose (literate/illiterate) tool.
4. Speech synthesis and text readers: Speech synthesis is a promising means to obtain efficient delivery of audio, and is expected to become, but is not yet, available for Web-page applications as are the audio compression technologies mentioned above. Speech synthesis is used in conjunction with general text reader software.
5. Text readers can read a chosen portion of text on the screen. With some customization of the speech synthesis pronunciation, they would be useful as a first approximation to the proposed system, since no further development would be required to use them to read a page. However, their use would be limiting in the long term, as they are also a little too complex to use.
On the users' side, users with low education and motivation, who are functionally illiterate, can gain access to the Internet by learning how to read and write through continuous education. The American Library Association (ALA), the National Center for Family Literacy (NCFL), the National Institute for Literacy (NIFL), Reading is Fundamental (RIF) are among the literacy organizations for continuous education. Extensive research and development has been carried out to utilize the electronic technology in improving the illiteracy problems in America. Voice dependent learning programs are widely researched and available currently for adult learning [7]. There is also so-called assistive technology that aims to provide user-friendly interface to our user group. The functionally illiterate users will first have to have access to a computer system where assistive technologies are already available in order to visit these web sites. Trainings have to be conducted before the computer system can be used.
In the broadest sense, assistive technology is any technology, which enables someone to do something they otherwise couldn't [11,13]. The Individuals with Disabilities Education Act (IDEA), the federal special education law, provides the following legal definition of an assistive technology device: "Any item, piece of equipment, or product system... that is used to increase, maintain, or improve functional capabilities of individuals with disabilities." [12]. Current emphasis on research and development for assistive technology are focused on the physically disabled but it also renders Internet access for our group of users. For low education and motivation users, the type of assistive technology applicable would be, for examples: Scanning/Reading/Comprehension Systems, Talking Word Processors and Word Prediction, Screen Readers, CCTV and Voice Recognition [13]. These devices minimize or totally remove the need to read and write from the users and hence render accessibility to our users.
For lowly motivated but literate users, the following technologies are important:
1. Voice over IP (VOIP) technologies are important for providing online human help. Current VOIP technologies have reached a somewhat mature stage, providing good quality for audio transmission.
2. Some web sites allow users to navigate with voice via a telephone [8,9]. By using Interactive Voice Response (IVR) systems (based on proprietary hardware and software, and hard-coded dialogues and requires its own content development and maintenance infrastructure) or Voice XML-based systems (requires the creation and maintenance of separate content for voice and visual sites, in practice, applications must often be augmented with JavaScript code in order to obtain desired level of functionality) voice navigation is possible.
3. VocalPoint Technologies came out with two software applications (VoiceBrowser and VoiceASP), which promise efficient voice access to any Web site and delivers high quality voice solutions over the telephone [9].
Guidelines that are pertinent to both kinds of users are as follow:
1. Use of self-explanatory icons in a direct manipulation structure. For example, in a study that involves children with low literacy in India, the children were able to understand that when the pointer becomes a hand, where the pointer is there is a link [27]. Providing a rich graphical user interface is important for our users. A good example would be an illiterate car buyer understanding that a link with a picture of car as being relevant.
2. Ease of navigation should always be enforced in the design of a website. More often than not, this means that the page layout should be kept simple, menu structure should be kept shallow, links should be well highlighted with appropriate alternate tags and presence of permanent navigation aids that are often used. Simple language should also be used.
3. Online help should always be available. Whenever possible, human support via VOIP might be provided. For low motivation users, an online human voice that is able to answer non-structured questions serves to remove fear of making mistakes and other apprehension associated with new technologies. When such advanced online support is not possible, online documentation should always be available.
4. Cater for the provision of online communities for novice users. This could include mailing list, online support group etc. Low motivation users are able to feel more confident when there are similar users who can share with them experiences.
5. Ease of memorability is vital for low motivation users. This is because it will instill confidence in the users. Whenever possible, major upheavals of websites are not recommended so that users are able to get use to the existing structure of the websites. When the users are able to anticipate the surfing experience, they will have less apprehension and more confidence, thereby generating more pleasant experiences.
6. Speed of navigation is also another major consideration for users. When the connection is slow, lowly motivated users would experience an increase in anxiety, even believing that they could have done something wrong since most of them are strangers with technologies. Moreover, more often than not, our users already have previous bad experiences with technologies and slow connections are definitely going to add to those negative experiences. Designers should try to prevent overloading websites with excessive images, embedded files etc.
7. Font sizes should be larger in the case of elderly users. Many of these elderly users are lowly motivated to use the web not only because they are out of touch with technologies but also because of their failing eyesight. Providing a visually soothing website will definitely be attractive to this particular group of low motivation users.
8. Provision for user feedback is necessary so that website designers are able to improve their websites and generate positive experiences.
Guidelines that are pertinent to illiterate users are as follow:
1. Voice output features involve the use of pre-recorded text or speech generation technologies. This is important in translating text to speech so that our users are able to understand the content.
2. Highlighting capabilities allow the translation of speech to text for a certain portion of the content. This is important for our users to choose the content that is important to them.
3. As stated above, translation of speech is important to users that speak different languages.
4. Voice input features are required for users' interaction with the system. Instead of just receiving information via voice output of the content, it is equally important for the users to be able to perform tasks like filling a form, perform search etc.
5. It is also viable that the users are able to perform tasks like playback of a speech, rewinding of a speech, pausing a speech etc.
6. Voice-enabled content without change, meaning that there should be consistency in providing voice-enabled content. For example, every links should be able to generate a voice on mouse over.
7. Add minimal enhancements to provide full voice interaction. It is viable to provide added voice features for our users. However, caution should be practiced to prevent major upheaval of the original website just to accommodate voice features for our users. There should not be too much difference between the original system and the enhanced system.
8. Add the enhancements so that they interact with existing mechanisms, use existing standards and use the same enhancements across markup languages. This is to ensure portability.
General guidelines discussed here are targeted at rendering accessibility mainly for disabled people with emphasis on design to accommodate adaptive technology. The concept behind the W3C/WAI guidelines' is to promote accessibility by ensuring graceful transformation and making content understandable and navigable [14]. The Trace Center at the University of Wisconsin-Madison, however, focuses on four [15] major objectives:
1. Ensuring that all information presented by or through the device can be perceived (even if all sensory channels are not available to the individual). For illiterate users, the literacy channel is not available, hence the need for voice output features. For the other group, perception of information could be hindered by apprehension and fear. Guidelines above serve to reduce anxiety and hence improve the perception level.
2. Ensuring that the device is operable by the user (even if they are operating under constraints). Guidelines given above on user feedback and voice input features ensure that the users are able to interact with the system, thus fulfilling this objective.
3. Facilitating the ability of the individual to navigate through the information and controls (even if they are operating with constraints). This objective would be met too based on our above guidelines.
4. Facilitating their ability to understand the content. This is similar to
(1).
1. HTML for Adaptive technology - HTML, when used correctly, will enable assistive technology (e.g.: screen reader) to translate the information. Keyboard access should be provided for all application features, care should be taken to avoid conflicts with keyboard accessibility features. This means that our users should be able to perform voice control via keyboard access, navigate websites with keypress etc.
2. Page Layout - Consistent Page Layout and Length and Ability to return to Table of Content at will. Frequently used functions should not be placed deep in a menu structure. It is important in our case that the audio/video features should be always present in the page that illiterate users are in. For the other group, it is important that navigation aids are always present e.g. navigation bar.
3. Clear Links - Icons and graphics should be self explanatory when being used as a link. Speech should be clear and well formed to ensure that illiterate users are able to understand. Text link should also be in simple terms.
4. Alternate Text Graphic - Although this might not be applicable to illiterate users, it would be a good idea to change this to display an audible speech whenever our users point over a link.
5. Online Forms - As mentioned above, voice input features should be provided for illiterate users. Feedback forms should also be available whenever necessary.
6. Validating Pages - Always test pages with a variety of browsers and
validation tools (e.g.: Bobby from CAST)
1. Giving users a sense of the alternatives available to them is the focus of
designing effective navigation guides within a document. The content should
corresponds to the spatial, temporal, and topical
structure of the real
event, giving a metaphorical familiarity. In addition, each page was structured
following the same content template. This design lets the user determine
interest in the topic, go directly to their
preferred media for browsing the
event, or change to another topic with at most two links. The important issues
to remember include listing the subheadings within the page, allowing for easy
navigation within the page, being consistent in the placement of tools and
information, and allowing users browse information while downloading occurs.
2. Assisting users in navigating between documents involves giving the user a sense of where they are in the document hyperspace. The most important aspect of this is to use common metaphors that give feedback to the user regarding where they have been and what else is available. For example, [28] proposed a design that has different metaphors that allows the users to choose a preference that is intuitive to them. Figure 1 shows the navigation design of the spatial map and time schedule map proposed in [28].
Figure 1: Spatial map navigation tool proposed in [28]
Figure 2: Time schedule navigation tool proposed in [28]
3. Memorability includes using page names and URL's that make the resource easy to find. Because of the path dependent nature of content access (content linked to other content) users of the web must encounter content where they expect it--or they will not find it. A good search engine might help this problem--but in many situations a user must know what they want before they can find it with a search engine. This dilemma - that one must know what they want before they know that they want it - limits the user of the search engine to finding only what they expect and does not take advantage of the semantic architecture of the World Wide Web. Hence, there is a need to integrate the flow of the information domain into the flow of the website.
4. The last factor that must be considered is ease of comprehension. The best
examples of how neglecting this factor can inhibit user performance are those
pages where these human factors of document design are ignored. For example,
some of the "halls of shame" on the web. These pages use background and text
combinations that don't contrast enough to make rapid visual identification
possible, image maps
that don't give the user feedback regarding where they
have been, multiple font sizes and styles which limit the speed at which users
can read text, poorly designed icons or instructions, and poorly worded text.
The users of these pages can spend a significant amount of time trying to
understand which alternatives they are being presented with rather than deciding
if they have found the content they were searching for.
5. The key user activities that need to be supported by consistency in
document design are those related to comprehension and navigation. The document
designer and system administrator must determine which user activities relate to
their content because the specific activities that need to be supported might
vary depending on the interface, content, or system. If each and every page is a
new adventure for the
user, then some of the time they spend browsing each
page will be wasted trying to determine where the author has hidden the
information. A page designed for clarity and consistency, on the other hand,
will minimize this problem.
6. Images cause problems because they take a long time to transfer and this increases user frustration. This time can be disproportionately large compared to the time spent looking at the information on the page, particularly if the user is located behind a slow modem link. And while some browsers allow the user to view the text of a page before the images finish transferring this feature is not available on all browsers. By using small "thumbnail" images, delay in transmission can be delayed.
For illiterate users, the specific guidelines are as follow:
1. The XML language is viable for creating a system that is voice-enabled. XML has capabilities to provide customize solutions including voice XML. In addition, voice XML is able to support voice navigation operations like rewinding of speech, pausing of speech etc.
2. When users pass their mouse pointers over the words on the page (or touches them using a touch screen), they will be audibly pronounced by the computer. The functionality will be added through the use of two technologies: Browser programming and speech production. Browser programming refers to the use of built-in scripting languages (JavaScript) and code (Java applets) to add interactivity and special functions. In this case, it would be used to make the text sensitive to the mouse movements and pass information to the speech production unit. Since JavaScript and Java are open and ubiquitous standards, they represent the most cost-effective and efficient way to do the special web pages.
3. Speech synthesis is chosen as the mostly likely means to produce the audible output. The alternative is prerecorded words. Speech synthesis is far more efficient than recorded sound. The efficiency is an important issue in web system design, and refers to the amount of data that must be transferred. Recorded sounds (a recording of each word on the page) might take so long to reach the user (through a slow connection to the internet) that they render the system unusable. Another important consideration is the cost of recording and storage of all the possible words, especially if a number of alternative languages are to be provided. That said speech synthesis has the disadvantage of inaccurate pronunciation. It is also not quite ready for implementation in the browser environment as discussed below. In order to achieve speech synthesized web pages, two components are required: Speech synthesis software and a speech synthesis browser plug in. Speech synthesis software is chosen over speech synthesis hardware as being cheaper while providing quality adequate and appropriate to a prototype or exploratory project. If high quality hardware synthesis can be justified it can be exchanged for the software component at a later stage. The software runs in the native OS environment (DOS/Windows, MacOS, or UNIX). There are freely/widely available software packages for all three environments.
4. Custom browser vs. custom server. For a customizable system, the only alternative to the server-based approach is a custom-written web browser to be obtained and used by every person who wishes to have illiterate access. The development effort for such a project is far greater than for a proxy server, since browsers must run on a variety of platforms and must keep up to date with web technology changes (see note on Java below). There is really no competing with Netscape and Microsoft in this arena. Systems like the "speaking browser for the blind", pwWebSpeak was probably feasible only because it made no attempt to preserve normal visual browser functions. It extracts the text from a page, formats it for voice playback and discards almost all visual information.
5. Augmented screen reader vs. augmented browser. Rather than look for a browser plug in that integrates it with speech synthesis software, it is possible to consider modifying or enhancing a synthesizer or screen reader package. For example Creative TextAssist can be used to read each word in a web page by correctly operating the mouse and the application. The actions required are tedious however (user must highlight words manually). Any solution of this nature would be likely to involve a number of piecemeal workarounds. It would not have the elegance and integration of a synthesis-aware browser. Additional functionality would require updating an ensemble of small and large programs, which would need to be redistributed to every user. This is in contrast to the centralized nature of a system consisting of a browser (augmented with plug in) and enhanced pages created by a literacy server. Typically the plug in would supply initially to the user, and later enhancements to the system would not likely require updates to the plug in.
6. Translation. By changing the speech-production dictionary, a simple single-word language translation facility may be implemented. Pointing at an English word would cause an utterance of a native-language translation of the word. This technique can also be used to localize the accent of the English pronunciation, which can be in wide variance with the generic computerized American pronunciation and may considerably impede comprehension.
7. Visual augmentation of pages. Users are going to need help not only reading words but using the browser software, the buttons and menus of which usually rely on text to explain themselves. This problem can be addressed by adding, "talking" buttons for the most common browser functions directly onto each page. This can be done with straightforward browser programming.
8. Custom educational content. It will be possible (likely very necessary) to provide custom web pages designed as tutorials and references for the illiterate users. Being custom created pages they will incorporate the ability to read themselves aloud and interactively guide the users into learning how to use the system. Such content is currently in development at m-powa.
1. The Study Place: http://www.thestudyplace.org/welcome.taf
(Organization: CyberStep).
The Study Place is an Internet tool design to aid
adult learning in English, practice Reading, learn Math and get Job Skills. This
site is designed for both the teachers and students. In their welcome page, the
Student and Teacher Quicktour programs have audio reading by word spoken by an
"Instructor" on the screen. Users can Repeat or move to the Next page with ease.
The step by step instruction is highlighted by a magnifying glass in addition to
the voice/audio instruction.
2. SPeech-On-The-web (SPOTw): http://www.speechsolutions.com/
(Organization: Speech Solutions Inc.).
The product is a highly versatile
developer's kit that allows web site developers to easily add RealTime
Conversational Speech Recognition to any web page. Simply put, SPOTw? enables
web page designers to add voice recognition to any web site. The site will talk
and listen. Their website now features SPOTw?. All you need to evaluate it is a
SAPI compliant speech engine that supports text to speech and speech to text,
Internet Explorer 4.0 and up or Netscape 4.0 and up, a sound card that is at
least half duplex, a set of speakers or a headset, and a microphone.
However, those of you without a speech engine can still view our web site as you
would any other: SPOTw? will not interfere with normal web page interaction.
3. Streaming Audio Web Page Creation for Language Learning: http:/www.esl-lab.com/online/
(Organization: Randall's ESL Cyber Listening Lab, University of Utah, Last edit
date: April 1999).
By using RealAudio Player, users can listen to the whole
presentation. The option of using the non-embedded player version is available
if users experience any problems with the audio or if users do not have needed
plug-in to view the embedded player. With the embedded version, the Player
controls are in the browser window, and the audio will stop automatically when
users leave the page. Although the presentation is basically linear in nature,
visitors can also jump to the sections they are most interested in and listen to
that portion only.
4. Telesensory has software for Windows and DOS that allows, amongst other features related to blind access, audio reading by character, word, line or window. The website is http://www.lvproducts.com/telesensory.htm (Organization: LVProducts, Last edit date: April 2001).
5. Mac-based software speech synthesis: Apple has a freely available software speech synthesizer which, together with an Mac-based Netscape plug in called Talker by MVP solutions, allows web pages to read themselves aloud.
6. The Swiss Eidganossische Technische Hochschute (ETH) has a web service for blind users. With no special software required, the ETH site is used as an intermediate processor of web pages. Any web page is delivered after being fed through the site and reformatted for use with the user's existing voice-synthesized screen reading package.
7. Productivity Works have a speaker program, pwWebSpeak designed specifically to read web pages - that is, it read the HTML and thus understands more than English text. Again the emphasis is on aiding the visually impaired, so some of the intelligence is likely unnecessary. Their product however would be a good starting point for developing a similar one tailored to aid literacy. There is also an upgrade version called WebReader, which is tailored to dyslexics and partially sighted and work with Microsoft Internet Explorer (thus supplying fuller featured web-browsing). Foreign language (European) versions are being developed.
8. Text-to-speech software provided is as follows:
·
Arctic - Makers of WinVision, Business Vision
· GW Micro
- Makers of Vocal-Eyes, Window-Eyes (has visual word highlighting feature)
· Biolink - Makers of Protalk
·
Henter-Joyce - Makers of JAWS
· IBM - Makers of IBM
Screen Reader (tailored for use with IBM software - e.g. with Web Explorer,
has voice
changes on hyperlinks).
9. A System Design for Human Factors Studies of Speech-Enabled Web Browsing [19]. A conceptual design has been developed with a complete software system divided into three modules: Interface Design, Basic Navigation and Enhanced Navigation. Currently, this system is still undergoing usability testing.
10. Voice Enabling the Internet [20]. Pertaining to using voice in the Internet, Conversa Web, created by Conversa, allow the surfers to surf the web with ones' voice. The major operations include synthetic and natural voice output and the addition of grammar based input. Including tags for a synthesized voice offers a low-bandwidth means of interacting with users. For grammars, the group supports transmission of recognition results both through a URL ("voice" forms) and within the page using JavaScript. A JavaScript-based speech object supports dynamic integration of TTS, recorded audio, and grammars within the page. Conversa has additionally constructed simulators for WML where they use the same approach to voice-enhance WML content.
(II) Low Education users
(a) Technical research on improving the accessibility issues for users with disabilities are strongly encouraged by government [24].
In general, Government agencies, private corporations and foundations should conduct research on information technology access issues and the development of solutions. People involved in all aspects of the development of multimedia and other information technology should be educated about access issues. Support is needed for research examining the technological and social issues surrounding access to multimedia and other information technology by persons with disabilities and the benefits of such access to individuals who are not considered disabled.
Mainstream technology and multimedia developers, producers of assistive technology, disability organizations, and research institutions should strengthen partnerships aimed at developing access and usability solutions. Efforts to improve access and usability through assistive technology should be improved. Innovative methods to instruct those currently involved in the information technology and multimedia industry about access issues and solutions must be developed.
(b) Research Needs in Input / Output Technologies [25]
The usage of input
and output technologies is crucial in rendering accessibilities to users with
low motivation and low education. Research advances in both the hardware and
software systems in this area should not be neglected.
1. Speech Synthesis Efforts - are needed to automate the production of other components of speech synthesis, to facilitate the production of synthesizers in many languages from a single architecture
2. Speech Input/Recognition - More research are needed in human speech recognition performance as it is still at least an order of magnitude better than that of machines
3. Auditory Displays - Good matching to human hearing is needed by computing the right sound and getting it to each ear in a properly weighted way
4. Haptic and Tactile Displays - To provide high-quality haptic interaction over many degrees of freedom will continuously create many research challenges in mechanism design, actuator design, and control
5. Multimodal Interfaces - The ability to integrate information across modalities is essential for accurate and robust comprehension of language by machines and to enable machines to communicate effectively with people. Basic research is needed into the science of understanding how humans use multiple modalities.
6. VOIP technologies - There is a need to further enhance the quality of voice transmitted over the network.
2. Table II-6 from Falling Through the Net: Toward Digital Inclusion October 2000, Index of Charts and Tables http://www.ntia.doc.gov/ntiahome/fttn00/chartscontents.html (Organization: National Telecommunications and Information Administration, Last edit date: October 2000)
3. Merriam-Webster Online - The Language Center http://www.m-w.com/ (Organization: Merriam-Webster, Last edit date: 2001)
4. Internet and Computer Usage by Low-Income Groups by Jeff Carver. Department of Computer Science University of Maryland, College Park, MD 20742 USA http://www.otal.umd.edu/UUGuide/carver/ (Author: Jeff Carver, Last edit date: April 19, 2000)
5. Electronic Access to Services For Low-Income Population by Adam Porter. Department of Computer Science University of Maryland, College Park, MD 20742 USA http://www.nap.edu/readingroom/books/screen/14.html (Organization: National Research Council, National Information Infrastructure, Last edit date: 1997)
6. UniversalUsability.Org - Pushing Human-Computer Interaction Research to Empower Every Citizen By Dr. Ben Shneiderman. Introduction & Definition http://www.universalusability.org/definition/introduction.html (Organization: UniversalUsability.Org, Last edit date: 2000)
7. "Ask Verizon Read" at Verizon Literacy Network and CyberStep - Learning Anytime Anywhere Any Pace at http://www.cyberstep.org/ (Organization: CyberStep), The Study Place http://www.thestudyplace.org/welcome.taf (Organization: CyberStep), http://www.ala.org/ (Organization: American Library Association), http://www.otal.umd.edu/uupractice/low_ed/www.famlit.org (Organization: National Center for Family Literacy), http://www.otal.umd.edu/uupractice/low_ed/www.nifl.gov/lincs (Organization: National Institute for Literacy), http://www.otal.umd.edu/uupractice/low_ed/www.rif.org (Organization: Reading is Fundamental), Diversity University - Internet Based Tools For Education http://www.du.org/duinc/inettool.html#video (Organization: Diversity University, last modified date: 16Oct 1997)
8. IBM vaults into Voice Internet - Thu, 22 Jun 2000 13:32:37 GMT by Reuters Zdnet UK http://www.zdnet.co.uk/news/2000/24/ns-16170.html (Organization: IBM, Last edit date: 22 Jun 2000), VocalPoint Technologies Announces Partnership with CSELT/Telecom Italia Global Telecom Company Licenses VocalPoint's Voice-Enabled Browsing Technology SAN FRANCISCO, California and TORINO, Italy, April 4, 2000, http://www.vocalpoint.com/press_telecom042000pf.html, (Organization: VocalPoint Last edit date: April 4 2000)
9. VocalPoint Technologies - Infrastructure for voice-enabling the Web http://www.vocalpoint.com/index.html (Organization: VocalPoint Last edit date: 2000)
10. 'Hello, Internet?' on E-Business, Washington Post dated 10/17/2000 By Peter S Goodman http://www.bizreport.com/ebiz/2000/10/20001017-1.htm (Author: Peter S Goodman, Last edit date: 17 Oct 2000)
11. The Source for Java Technology - Assistive Technology Defined http://java.sun.com/products/jfc/accessibility.html (Organization: Sun Microsystem, Publication updated on: 10-Apr-2000)
12. Parents, Let's Unite for Kids (PLUK) in cooperation with The Federation for Children with Special Needs EDITOR: Katharin A. Kelker, Ed.D. PRODUCTION: Roger Holt, ATP ILLUSTRATION: Karen Moses http://www.pluk.org/AT1.html#13 (Organization: Parents, Let's Unite for Kids, Last edit date: 1997)
13. Assistive Technology - Don't be Disabled and Dependent, Be Able and Independent http://www.assistive-technology.com/ (Organization: Assistive Technology, Last edit date: 1999)
14. http://www.w3.org/TR/WAI-WEBCONTENT/ (Organization: World Wide Web Consortium, Recommendation date: 5-May-1999).
15. http://trace.wisc.edu/docs/fundamental_princ_and_priority_acmcuu2000/index.htm Fundamental Principles and Priority Setting for Universal Usability (Author: Gregg Vanderheiden, Ph.D).
16. "Advances in Human-Computer Interaction" Volume 5, Jakob Nielsen, Editor, Copyright 1995. It is reprinted with permission from Ablex Publishing Corporation, 355 Chestnut Street, Norwood, New Jersey 97648. (Author: Jakob Nielsen, Last edit date: 1995)
17. http://www.ala.org/editions/openstacks/insidethecovers/mates/mates_toc.html Adaptive Technology for the Internet: Making Electronic Resources Accessible to All by Barbara T. Mates (Author: Barbara T. Mates, Last edit date: 20-Jul-2000).
18. http://www.w3j.com/5/s3.paciello.html Advancing HTML: Style and Substance Volume 2, Issue 1 (Winter 1997) Technical Papers - W3C Journal - People with Disabilities Can't Access the Web! Mike Paciello (Author: Mike Paciello, Last edit date: 1997)
19. A System Design for Human Factors Studies of Speech-Enabled Web Browsing by L.J. Adams, R.I. Damper, S. Harnad and W. Hall Department of Electronics and Computer Science University of Southampton Southampton SO17 1BJ, UK http://www.kbs.twi.tudelft.nl/Research/Proceedings/ids99/papers/files/3.pdf (Authors: L.J. Adams, R.I. Damper, S. Harnad and W. Hall, Last edit date: June 1999)
20. Voice Enabling the Internet Authors: Charles Hemphill, Michael Robin, Conversa. http://www.w3.org/2000/09/Papers/Conversa.html (Authors: Charles Hemphill, Michael Robin)
21. Adaptive Technology Center - Indiana University Bloomington http://www.indiana.edu/~iubdrh/convers2.html (Organization: Indiana University Bloomington, updated date: 23 March 2000)
22. http://www.cs.purdue.edu/homes/chenh/WAP/ConversaWeb_review.html (Organization: University of Purdue, updated date: June 2000)
23. Universal Design of Websites and WebCT Platforms - Checklist for Testing Your Web Pages for Accessibility from Georgia State University http://www.gsu.edu/~wwwdls/show_case/Presentations/universal_design.html (Organization: Georgia State University)
24. Access To Multimedia Technology By People With Sensory Disabilities March 13, 1998 National Council on Disability 1331 F Street, NW, Suite 1050 Washington, DC 20004-1107 http://www.ncd.gov/newsroom/publications/sensory.html#5 (Organization: National Council on Disability, Last edit date: March 13 1998)
25. More Than Screen Deep Chapter 3 Input/Output Technologies: Current Status And Research Needs http://www.nap.edu/readingroom/books/screen/3.html (Organization: National Research Council, National Information Infrastructure, Last edit date: 1997)
26. http://www.usask.ca/education/coursework/skaalid/site/function.htm- Author: Bonnie Skaalid, Publication date: 1999
27. http://www.ksg.harvard.edu/iip/stp305/Fall2000/krishana.PDF - Author: Ronny Krishana, Publication date: Dec 18, 2000
28. http://www.microsoft.com/usability/webconf/fuller/fuller.htm- Author: Rodney Fuller, Johannes J. de Graaff, Publication date: Oct, 1996