→ Who’s who at W3C → Dom’s home page
This paper was submitted as part of the proceedings of the first AEGIS conference in 2010.
We propose to look at the differences and convergences between making a Web site accessible and making it mobile-friendly, and to explore how the two domains might help each in improving the life of many more users.
The methods and techniques used to make a Web site accessible to persons with disabilities are now rather well-known : the publication end of 2008 as a W3C Recommendation of the second version of the Web Content Accessibility Guidelines  marks the maturity of this field where work started as early as 1997.
Among the arguments that were often cited as a reason to embrace accessibility in the design of a Web site was the predicate that making a Web site accessible would make it more device independent, and thus more usable e.g. on mobile devices.
In the past few years, the usage of the Web from mobile devices has grown exponentially, enabled by improvements made both to their hardware and software, and facilitated by some of the guidance given to Web developers to make their Web site more mobile friendly.
In that perspective, we propose to look at the differences and convergences between making a Web site accessible and making it mobile-friendly, and to explore how the two domains might help each in improving the life of many more users.
Taken from a high level, there are good reasons why accessibility and adaptation to mobile devices on the Web are strongly related.
Indeed, in both cases, the end user has to rely more strongly on computer-derived assistance to be able to interpret Web content. In the case of accessibility, the content needs to provide hooks that enable additional modalities help overcoming the user’s disabilities; in the case of mobile-friendliness, these hooks are needed to help overcome the device’s limitations.
More concretely, many of the mobile devices’ limitations match a number of well-known disabilities:
Shared Web Experiences: Barriers Common to Mobile Device Users and People with Disabilities  explores these similarities in details.
The World Wide Web Consortium (W3C) is the main organization responsible for defining standards in the field of Web technologies.
Beyond its technical standards, W3C has also been a very active actor in the development of guidelines and best practices with the goal to allow an always greater number of persons and devices to make an effective use of the Web.
As part of this mission, W3C started developing in 1997 the first version of the Web Content Accessibility Guidelines (WCAG), with the goal of defining a set of well-known techniques to help making Web content accessible to persons with disabilities. These first accessibility guidelines were released as a W3C standard in May 1999 .
When in 2005 W3C started working on similar guidelines for helping Web developers create content that would work well on a large number of mobile devices, among the number of relevant existing advices, the accessibility guidelines proved to be an invaluable source of inspiration: given the number of similarities between the needs of mobile users and persons with disabilities, a number of the advices to make Web content accessible proved to almost directly applicable to making Web content mobile friendly.
And indeed, among the 60 mobile Web best practices that W3C released as part of its standard in July 2008 , roughly half of them can be traced back to the guidelines in WCAG1.
In parallel to its work on the mobile Web best practices, W3C also developed a new version of its Web Content Accessibility Guidelines, WCAG 2.0. This new version, released as a standard in December 2008 , provides an update to the proper ways to make content accessible based on the lessons learned since WCAG 1.0, but it also comes within a much better defined framework to establish whether a given Web site conforms with the guidelines or not.
Due to this change of approach, the differences between the mobile Web best practices and WCAG 2.0 are more important than they were with the previous version of the accessibility guidelines.
While there are still clearly a lot of convergences highlighted between making a Web site accessible and mobile-friendly, the strict overlap between a Web site that conforms to WCAG 2.0 and one that complies with the mobile Web best practices is actually fairly limited: only 6 of the mobile Web best practices directly map into WCAG 2.0 success criteria, as analyzed in From MWBP to WCAG 2.0: Making content that meets Mobile Web Best Practices also meet Web Content Accessibility Guidelines 2.0 , and conversely, only 7 WCAG 2.0 success criteria map directly to mobile Web best practices .
Beyond the somewhat artificial similarities and differences between making a Web site conformance to WCAG 2 and making one conforming to the mobile Web best practices, the recommended approaches used to deploy accessible Web content and mobile-friendly Web content remain quite different.
On the one hand, in most cases, Web accessibility proponents recommend integrating accessibility features within the main Web site, rather than creating a separate Web site specifically for people with disabilities.
There are indeed good reasons for avoiding this:
On the other hand, a large number of deployment of mobile-friendly Web content is made through separate Web sites. This can be made on a simple mobile/non-mobile basis, but is also often made on a device-class basis (e.g. feature phone / smartphone / desktop, or mobile without touch screen / mobile with touch screen / desktop), and even sometimes on a device-per-device basis, using content adaptation systems to customize the content sent to the devices based on their known capabilities and limitations.
Why is this workable in the mobile world, when it is not recommended for accessibility? A few explanations can be brought to this:
Overall, there are indeed a number of convergences between making a Web site accessible and making it mobile-friendly: these convergences show for instance that getting trained for one of the aspects would help understanding and approaching the other aspect. They also support the notion that in many cases, a Web site that takes one of the aspects into account will show some improvements with regard to the other aspect.
Nevertheless, the two domains have also their own set of requirements, that are better addressed by their own specific technical solutions, as they are also often brought by different business approaches.
Having looked at the current state of deployment of accessible and mobile-friendly Web content, we propose now to look at the likely synergies brought by current and future developments in both domains.
Both accessibility and mobile devices have brought opportunities for new technological developments, and in many ways these developments can bring benefits from one domain to the other.
While there remain a number of rather contentious aspects in HTML5 with regard to accessibility , the new version of HTML also provides new capabilities that will improve the experience of the Web in both domains:
These semantic annotations can also prove to be very useful on mobile devices, where the default presentation of some of these interface elements could be usefully adapted to the more limited screen space; a WAI ARIA-mobile browser could for instance choose to use a dedicated user interface for a grid widget.
Another example of the synergies between accessibility and mobile Web usage can be found with screen readers: screen-reading technologies have been at the core of numerous assistive tools, to compensate vision-related disabilities. These technologies are seeing a much broader deployment with the development of mobile Web access, for at least two reasons:
This means that mobile devices benefit from the existing work on screen-reading technologies thanks to the existing accessibility work in this area, but also that assistive technologies will benefit from the larger set of targeted users and larger set of implementations.
For similar reasons, voice-based input — useful to compensate motricity disabilities among other things — sees a renewed interest in the mobile context . They can indeed advantageously replace keyboard input which on mobile devices is always difficult due to the reduced size of the keyboard, often inadequate when the user is on the move, and impossible when these devices reach illiterate populations.
A new field that mobile devices bring to accessibility relate to the new kind of sensors they provide, and how they can be used to build rich Web applications. If done right, these Web applications can open new doors for persons with disabilities.
The latest generation of mobile devices combine high computing capabilities, mobility, and a large number of sensors that create new interaction methods:
These sensors, when usable with assistive technologies, open new doors for persons with disabilities: touch screens can help spatialization even for low vision or blind users, geolocation can guide users in unknown places, cameras and microphones can be used to analyze the surrounding environment or receive commands from the user, etc.
As an example, A. Seraphin, a person with very low vision, writes in a blog post  how the iPhone with its touch screen, its camera and its mobility has created a new way for him to interact with his environment, creating ways to gain access to information and feelings that were not available to him before.
While most people associate the usage of these advanced device capabilities to interactions made through phone applications, a growing number of these capabilities are made available to Web developers, through the browser.
It is indeed already possible for a Web site to ask the user to share its current position through the W3C Geolocation API , and other APIs to interact with all the capabilities listed above are in development, for instance as part of the W3C Device APIs and Policy Working Group .
These APIs, and the additional capabilities provided by HTML5 and related technologies to make Web sites work off-line, save data locally, export data to other applications, induce that the differences between a native phone application and a mobile Web application are diminishing rapidly.
The W3C mobile Web application best practices  provide advices on how to build effective and usable Web-based applications for mobile devices.
These developments also create new opportunities: mobile Web applications are easier to develop, maintain and deploy than native applications, making it more likely that a larger number of applications that respond to specific accessibility needs come to their users, offering innovative usage of these sensors, on top of the existing solid guidance on accessible development with Web technologies.
These opportunities are not without risks either: more than ever, these new technologies need to be developed with accessibility in mind, and mobile Web developers would very likely benefit from guidance on how to use these technologies in an inclusive approach.
But with a sustained effort and proper outreach, mobile Web applications could open the door to a new generation of assistive approaches: accessible Web sites have allowed many persons with disabilities to interact with the virtual world better than they could with the real world, accessible mobile Web applications could help make part of the real world more accessible than it used to be.
The research leading to these results has received funding from the European Union's Seventh Framework Programme (FP7/2007-2013) under grant agreement n°257800 - Mobile Web Application (MobiWebApp).