Weblogs: Web Accessibility

Marks and Spencer named as inaccessible

Friday, August 15, 2003

Note: The range of testing revolved around accessibility of a website in terms of response times, page sizes, broken links, missing images, not real accessibility.

The DRC's investigation into the accessibility of UK websites alleviated worries by stating that their intent wasn't to name and shame UK companies for their abysmal accessibility levels of their websites. This misses the most effective incentive to making business websites accessible: the avoidance of bad PR.

That doesn't stop companies flogging their tools from publishing a list of their findings. SiteMorse is one example - an automated testing tool checking links, images, W3 standards and accessibility. Their findings list Marks & Spencer as the worst performing website of FTSE100 companies.

The top performing website according to their tool is Bunzl which we've noticed as rather odd. The website requires both Javascript and Flash before it will consider letting you in. Hardly an accessible quality, especially when the alternative text supplied is nothing more than a browser upgrade message. This website reacts exactly the same way as Marks & Spencer - redirecting users to an upgrade / downgrade/ sell-your-soul pages instead of letting them access the content.

The list of clients using SiteMorse are also equally interesting: NHS, Abbey National, Marks & Spencer, BHP Billiton, Royal Bank of Scotland, Logica. Marks & Sparks are using their tool but still come last? Interesting.

The main point we need to draw from this is that companies will be peddling their accessibility-related wares by naming and shaming companies into signing up. This one is an excellent example of the drawbacks of automated accessibility testing - it can never replace human experience, and should not be relied on - as is the case here - to claim a website is accessibile.

Yes, Marks & Spencer need to do some accessibility work, there is no question about that. And the "top accessible website" also have a good deal of work to make their website accessible too.

I've sent an email requesting the report, and I await it with anticipation. Hopefully they've qualified that testing in an automated tool does not confirm a website as being accessible.

Update 15 August: Read through the requestable document, only one accessibility checkpoint was tested, WCAG guideline 6.3 about ensuring content was accessible without scripts and applets. Over 20% of the websites tested failed this checkpoint (and Dixons was offline during the test, so its performance was not included).

Related reading

[ Weblog | Categories and feeds | 2011 | 2010 | 2009 | 2008 | 2007 | 2006 | 2005 | 2004 | 2003 | 2002 ]