The ghost of accessibility past - the rise of specialist screen reader solutions

In a WCAG 2.0 world, it’s easy to forget sometimes how large the leap was from the previous standard to the current one. While I’m sure that for many of the people who worked on WCAG 2.0 during the seven year process of creating it, it must have seemed like a lifetime, but for most practitioners WCAG 2.0 has only been with us for four short years.

Yet a recent project by the New York Times suggests that perhaps WCAG 1.0 still lurks in the shadows, and its implications remain a hot topic of discussion.

The first time I really understood that WCAG 1.0 had some significant issues was in 2003 when I interviewed Jonathan Hassell, the Manager for Standards and Guidelines in the New Media division of the BBC, as part of my PhD research.

Before the interview, I had assumed that WCAG 1.0, at that time four years old, was absolutely essential in ensuring accessibility. While to some degree Hassell agreed, he did point out that WCAG 1.0 wasn’t holding up so well against some then-emerging technologies, and in one key area relating to screen reader users they had actually changed their approach away from WCAG 1.0.

 The Priority 1 requirement that they had changed their focus on was:

11.4 If, after best efforts, you cannot create an accessible page, provide a link to an alternative page that uses W3C technologies, is accessible, has equivalent information (or functionality), and is updated as often as the inaccessible (original) page.

Hassell explained that in the mid to late 1990s the BBC used to focus heavily on ensuring they effectively had two websites: their main website, which they tried to make as accessible as possible, and the text-only version for screen readers and people with low vision.

Initially, it was thought that this would make everyone happy. Presumably this was also the thinking of the initial WCAG working group who put this into WCAG 1.0.

However, Hassell explained that after a few years they realised that there were some key issues around this policy. The first problem was that the text-only version didn’t get updated much due to developers forgetting it was there. Given that up-to-date news was an integral part of the BBC’s website, this complaint was taken very seriously.

The second main issue was from people who had low vision: they found the main website difficult to use in places, but found the text-only website too limiting for their needs. The argument was that while people with low vision couldn’t see much, they still wanted to at least have the opportunity to see the more multimedia-rich content.

It didn’t take too long for the BBC to change its direction by largely scrapping their text-only website and focusing heavily on making their main website accessible. This was done despite the team not knowing how to make some key emerging technologies accessible. The approach proved very successful: the BBC is highly regarded in accessibility circles as one of the pioneers of website accessibility in practice, and remains one of the most accessible media-rich websites today.

Fast forward to mid-2012, and the New York Times has launched an initiative around the idea that it would be helpful to blind and vision impaired users to have a specific area of the website that provides content specifically designed to be compatible with screen readers.

The article’s New Audience: Screen Readers states:

“It is currently very difficult for sight-impaired readers to navigate Using assistive technology tools to ‘read’ our section fronts and articles is an exercise in frustration: screen readers run through our many header and navigation items and ads before getting to the main content of the page, article text is interrupted by lists of article tools and multimedia piece titles, and the screen reader literally jumps around the page reading information that doesn’t always flow logically or in sequence. We already provide (via the National Federation for the Blind) ‘newspaper by phone,’ a text-to-speech reader that is accessed nearly 30,000 times a month. How many people might we be able to reach if our website was truly accessible? With, we’re hoping to find out.”

The solution proposed by the New York Times is the creation of a specific section on their website, ‘/accessible’, which reorganises the structure of pages to provide more screen reader-friendly content.

So this poses an interesting question: is the New York Times initiative a throwback to the days of WCAG 1.0 where screen reader-specific websites are necessary or is it just the case that it’s too lazy to implement WCAG 2.0?

The New York Times has some support in its approach due to research into accessibility guidelines conducted by the University of York. The research, which focused primarily on experienced JAWS users, tested a series of websites that claimed WCAG 2.0 compliance using Internet Explorer. The results in the abstract state that “…even when the guidelines are implemented on websites there is little indication that people with disabilities will encounter fewer problems.”

As you can imagine, such a claim has sparked significant discussion in the accessibility community including here at MAA. Further discussion at the SitePoint forum and WebAIM forum highlight a number of different views ranging from the research being completely ridiculous through to proven justification that screen reader users are largely neglected and everything in between.

In my view, both as a person who works in web accessibility and as an individual who is legally blind, I think it is important to acknowledge that there are some practical realities highlighted here.

With so many different screen readers available across so many different devices, it is highly unlikely that all users will have the same experience. A website optimised for Firefox and NVDA, for example, is unlikely to deliver the same experience for a user of JAWS with Internet Explorer or Safari with VoiceOver, as discussed in my previous columns.

While most of the time I personally rely more on screen magnification, the times when I have used a screen reader has made me appreciate the experience can be challenging and inconsistent at times, even on a website that is largely WCAG 2.0 compliant.

It’s important to remember that initiatives such as that of the New York Times are clearly trying to help. Good intentions are at work here and I’m sure there are a number of screen reader users who are applauding it. However, I strongly disagree with a number of notions raised in the research and the approach of the New York Times, due not so much to the perceived need raised in the research but to the conclusions that have been drawn.

In the W3C’s WCAG 2.0 At a Glance, some of the guidelines that are particularly relevant to the research include the need for text alternatives, ensuring that content can be accessed by assistive technologies without losing meaning, helping users find content, and ensuring that content appears and operates in predictable ways.

To my thinking, if a website’s developers have truly implemented WCAG 2.0 requirements and ensured that all of these things are addressed, screen readers should have a good experience in understanding and navigating a website. While the experience wouldn’t be the same for everyone and individual products may still have some quirks, users should ultimately be able to achieve the required tasks.

Rather than conclude that WCAG 2.0 is ineffective for screen readers or that it doesn’t address key usability issues, I would argue that the research suggests that perhaps the sample websites that claim WCAG 2.0 compliance still have some work to do.

When the guidelines themselves infer the need for screen reader testing, it seems a little unfair to blame the guidelines if screen readers don’t work. The W3C is renowned for its comprehensiveness when it comes to information on implementation techniques. If more work were done in ensuring that websites were truly standards compliant, many of the issues raised in the research around usability, understanding, and screen reader compatibility would be resolved. Perhaps even more importantly, though, is how the expectation for a specialist screen reader solution translates to the real world.

There are many different disability groups who face accessibility issues, and it is important that all people with disabilities receive access to the web. While people who are blind or vision impaired may face specific screen reader issues, it is both onerous and impractical to expect ICT professionals to implement multiple solutions for different disability groups.

It is for this reason that I’m concerned about the direction of the New York Times.  As demonstrated by the BBC, it will become very hard to keep the specialist solution going, especially as other disability groups start to question why so much effort is being put into the needs of one disability group to the detriment of time and money being spent on others.

The New York Times may also want to consider that, given the wide variety of screen readers available, it is still likely that not all screen readers will read the specialist pages in the same way, resulting in the same issues as the rest of the website. For those who genuinely believe that WCAG 2.0 isn’t meeting the needs of a particular group, I’d strongly encourage them to put their time and effort into participating in W3C working groups to resolve the issues there, so that everyone can benefit as the web continues to evolve.

With HTML5 rapidly evolving and hope rising of a day when web browsers will unite around a common standard, there is also hope that screen reader users, and other users of assistive technologies, will receive an easy, consistent experience.

However, for this to happen WCAG 2.0 still needs to be at the forefront of any web development project, with incorporation of the guidelines into daily work practices.

While specialist solutions seem to be making a comeback, there are good reasons as to why this was retired from WCAG 1.0 and I’m optimistic that we’re on the cusp of getting the real issue resolved, which is giving ICT professionals the tools and opportunity to continue taking WCAG 2.0 seriously.


Dr Scott Hollier represents Media Access Australia on the W3C Advisory Committee and publishes the W3C Column monthly.

Top of page