I see what you mean about certain findings implying that pages should be designed for JAWS, but I hope that our introductory piece to the series makes it clear that the reviews relate to the (very individual) views of a JAWS user. I agree – what works for him, will not be applicable to all.

I also strongly agree with your sentiment that “Pages should be designed to be as universally accessible as possible, not targeted at the moving target of different technologies.” – this was something that came across in a lot of interviews with accessibility designers and experts I spoke to beforehand, and something I’ve emphasised on the site’s blog.

I’m really grateful for your thoughts – you’ve managed to articulate a jumble of issues in my mind better than I could!

Thanks for the response, I wasn’t trying to be critical of your effort, just cautionary about what people take from this.

The issues are probably fairly common amongst screen readers users, but the cautionary aspect is what you actually do about them. In many cases people (site owners) jump to fix those issues one person brought up, not realising they may be hampering others using different technologies.

I hope that in your upcoming or concluding articles, you can make reference to the official guidelines, which make it easier to cater for all people (except perhaps people with cognitive issues).

Accessibility requirements can often seem conflicting between different user groups, but there is usually a way to satisfy all groups without resorting to different site versions.

When you say things like well-designed for use by JAWS., it implies that people should be designing for JAWs. That is fundamentally wrong.

Pages should be designed to be as universally accessible as possible, not targeted at the moving target of different technologies.

Regarding the age of access technologies, it is true that people do not tend to upgrade quickly, partly due to cost, but also due to the considerable amount of learning needed for each new version.

However, you would get significantly different results on even a 3/4 year old version, as after 3.7 Freedom Scientific made much more effort with regards to web accessibility in JAWs. Several of the issues brought up in the articles would not have been issues in even version 6, let alone 8 or 9.

When someone says something like “JAWS will only read text, so I imagine there are quite a lot of pictures on this site, which is less helpful to me as a screen reader user.” This shows a lack of knowledge about their own technology, as JAWs will quite happily read alternative texts provided on images, assuming the site has included them. There is little a site owner or developer can do in this case.

As the author of the article in this post, I just wanted to say thanks for your thoughts about the scope of our study.

You’ve pinpointed some of my concerns as the study was being conducted – namely that this it was extremely subjective.

We’ve tried to flag this up in the articles as much as possible, and given the time and resources I would like to have extended the study along the lines suggested.

While it is subjective, our experts told us that many of the problems John encountered were a common experience for screen reader users, so hopefully some wider conclusions can be drawn in this respect.

Regarding John’s equipment – it is an old version, but again many usability and accessibility specialists said this was an issue in accessibility itself: JAWS is expensive, blocking many users from having the latest version.

I agree that to address accessibility issues on a website based on the findings of one user would be damaging – The Guardian’s chief technical strategist makes some interesting points in his response to our article on Guardian.co.uk to this effect.

I think your distinction between the why and the how is a very accurate one, and I don’t think we, as journalists, are necessarily qualified to provide the how. I do hope, however, that the sites we reviewed and contacted might start looking into the how themselves, if they’re not already doing so.

Thanks again.