Affordable web accessibility: a case for cheaper ARIA

When a commercial entity designs a product, one of the major considerations is the cost of production. No matter how amazing a product idea is, how many features the product has, or how noble the cause is, most products will fail if they are too expensive for the target market or too expensive to compete in the market. Any website is, of course, a product; and web accessibility is often perceived as an added feature that costs extra money to develop. In this paper, we make an argument that the greatest weakness of WAI-ARIA, an accessibility specification for web developers, is the cost of its implementation. We argue that this cost, rather than specific technical constraints, may be the main culprit of the poor accessibility of many websites, except for the most popular ones, and of the slow progress in WAI-ARIA support in screen readers. We contend that the accessibility of the Web could be greatly improved if an affordable version of WAI-ARIA were to emerge. This can be achieved by: (a) extending the specification to increase its utility for a wider range of potential users, which will allow for economies of scale, and (b) refining the specification to make it both more cost effective for web development and more open to innovation in user-agent development. In this paper, we discuss accessible web development in the context of cost and identify potential areas of improvement.