“John hit the ball”: Should Simple Language Be Mandatory for Web Accessibility?

W3C WAI “Easy to Read” (e2r) Work

The W3C/WAI Research and Development Working Group (RDWG) is planning an online symposium on “Easy to Read” (e2r) language in Web Pages/Applications“ (e2r Web). The closing date for submissions (which can be up t0 1,000 words) is 24 September 12 October 2012. The symposium itself will take place on 3 December 2012.

The Easy to Read activity page provides an introduction to this work:

Providing information in a way that can be understood by the majority of users is an essential aspect of accessibility for people with disabilities. This includes rules, guidelines, and recommendations for authoring text, structuring information, enriching content with images and multimedia and designing layout to meet these requirements.

and goes on to describe how:

Easy to Read today is first of all driven by day to day practice of translating information (on demand). More research is needed to better understand the needs of the users, to analyze and compare the different approaches, to come to a common definition, and to propose a way forward in providing more comprehensive access to language on the Web.

It provides a list of potentially useful tools and methods for measuring readability:

• Flesch Reading Ease
• Flesch-Kincaid Grade Level
• Gunning Fog Index
• Wiener Sachtextformel
• Simple Measure Of Gobbledygook (SMOG)
• Gunning fog index (FOG)

The aim of this work is to address the needs of people with disabilities:

• People with cognitive disabilities related to functionalities such as
  • Memory
  • Problem solving (conceptualizing, planning, sequencing, reasoning and judging thoughts and actions)
  • Attention (e.g. Attention deficit hyperactivity disorder – ADHD) and awareness
  • Reading, linguistic, and verbal comprehension (e.g. Dyslexia)
  • Visual Comprehension
  • Mental health disabilities
• People with low language skills including people who are not fluent in a language
• Hearing Impaired and Deaf People

Early Work in this Area

When I saw this announcement it reminded me of early W3A WAI work in this area. Back in March 2004 an early draft of the WCAG 2.0 guidelines for Web accessibility provided the following guideline:

Guideline 3.1 Ensure that the meaning of content can be determined.

and went on to describe level 3 success criteria which could demonstrate that this guideline had been achieved:

• Syntax
  • Using the simplest sentence forms consistent with the purpose of the content
    • For example, the simplest sentence-form for English consists of Subject-Verb-Object, as in John hit the ball or The Web site conforms to WCAG 2.0.
  • Using bulleted or numbered lists instead of paragraphs that contain long series of words or phrases separated by commas.
• Nouns, noun-phrases, and pronouns
  • Using single nouns or short noun-phrases.
  • Making clear pronoun references and references to earlier points in the document

Yes, if that version of the WCAG guidelines had been implemented if you wished your Web site to conform with WCAG Level 3 you would have had to ensure that you avoided complex sentences!

Conformance with Level 3 guidelines were intended to Web resources “accessible to more people with all or particular types of disability“. The guidelines explained how “A conformance claim of “WCAG 2.0 AAA” can be made if all level 1, level 2, and all level 3 success criteria for all guidelines have been met.”

Such guidelines would be helpful for people with cognitive disabilities: those with Asperger’s syndrome, for example, find it difficult to understand metaphors such as “It’s raining cats and dogs“. The guidelines seem to have been developed by those who wished to implement the vision of “universal accessibility“. But I think we can see that seeking to address accessibility in this fashion is flawed.

Dangers of Such Work

I have to admit that I would be worried if the Easy to Read research activities were to lead to enhancements to the WCAG guidelines. Under the current WAI model, full conformance to WCAG, together with ATAG and UAAG guidelines is supposed to lead to universal accessibility. There is also an assumption that universal accessibility is a desired goal.

But is this really the case? The early drafts of WCAG 2.0 guidelines suggested that “John hit the ball” conformed with the goal of ensuring that the meaning of the content can be determined. Would WCAG 2.0 checking tools flag “the ball was hit by John” as an accessibility error, meaning that the Web page could not achieve the highest accessibility rating? And what about my favourite sports headline: “Super Caley Go Ballistic Celtic Are Atrocious” – a headline which brings a smile if Mary Poppins was part of your cultural background and you recognise Celtic as a football team, but which is clearly not universally accessible.

I would welcome research into ways in which styles of writing can enhance the accessibility of the content to people with disabilities. My concern would be if such research were to be incorporated into future versions of WCAG guidelines – especially if WCAG conformance is mandated in legislation, as is the case in some countries. But rather than failing to carry out such research, I feel the main challenge for WAI is to re-evaluate its underlining model based on the triumvirate of standards and its commitment to ensuring that Web resources are universally accessible – this might be a great soundbite, but in reality may be an unachievable – and even undesirable – goal. After all ‘universal accessibility’ doesn’t appear to  allow for any contextualisation and an important aspect of accessibility must surely be the context of use. What do you think?

---

Twitter conversation from: [Topsy] – [SocialMention] – [WhosTalkin]