I really like the web. I like it because it is the only medium that’s designed to really work for everyone. In my conference talks I often show my enthusiasm for the web with an example of an illustration of circles that I created using web technology.
The illustration can be seen by everybody who has eye sight and a browser. It can even adapt to different contexts, like screen sizes.
But there are other contexts the web can deal with as well. For instance, not all browsers support the same set of features. Some older browsers don’t support generating rounded corners. In this case you could consider to present the user with a textual representation of the image instead.
And this brings us to yet another context: the web can be used by people who can see, but it can also be used by people who cannot see. These people may depend on assistive technology that reads what’s on the screen.
I really like the web because of this incredible feature: the web can be used by everybody, regardless of the software and hardware you use.1
It would be true in practice as well, if everybody who creates things for the web was a web technology expert, a user interaction expert, and an inclusive designer at the same time. Creating things that really work for everybody is complicated
If this expert knowledge was required in order to put anything on the web we would both make and break the web at the same time. On the one hand it would result in completely accessible interfaces, which would fix the web for people like Simon Dogger, who is blind. It would fix it from a user’s perspective. But it would break the web for Simon as well, but this time from a maker’s perspective: Since he is no web expert himself, he wouldn’t be allowed to publish anything on it.
The use of heading levels is a good illustration of the conflict of interests between users and makers. Heading levels can help blind users in understanding the structure of a web page. They can ask their screen reader to turn these headings into something that resembles a table of contents of the page. But this feature only really works if heading levels are used properly.
Structuring heading levels is complicated. On many, if not most websites, document structure is not accurate at all.
Here’s a video of a screen reader showing all headings on an ill-structured page. It basically shows a seemingly endless, randomly numbered list.
The random numbering, and the excessive use of headings turns this otherwise very useful feature into a useless exercise.
Misusing headings levels like this is common practice, even on larger websites. It stems from a lack of understanding. Which is possible because you are allowed to publish on the web even if you don’t understand the basics.
In the past there have been discussions about restricting who is allowed to publish on the web. For instance the discussion about certification for web professionals in 2007.2 And more recently Mike Monteiro wrote a piece in which he argues that design should be a protected profession, just like architecture and law.3
I find what-if exercises about only allowing certified experts to publish on the web to be only mildly interesting. The conclusion would inevitably be that the web would cease to be the web. I am more interested in ways to make the web more accessible in its current form, without changing the principles of the web itself. The web is a chaotic, complicated medium and it only gets more chaotic and complicated because everybody is allowed to publish on it. I am not going to try and change that.
My research has focused on other directions of possible change. By promoting a more critical design attitude within the web community for instance. By having a critical look at the design tools and some assistive technology. By using an inverted, more focused approach to inclusivity. And by having fun.
In the next chapter I will talk about the weirdness of some of the defaults we have to work with on the web.
Peter-Paul Koch et al. Guild, part 1 - certification. Blog post and comments. 2007. www.quirksmode.org/
Mike Monteiro. Design’s Lost Generation. 2018. Blog Post. medium.com/