In my personal blogging, I seem to have hit a low point recently. After weeks where I’ve been busy at work, but doing nothing that was really worth talking about, I’m now in a position where I’m doing interesting stuff, but even busier, so I have even less time to talk about it even if I wanted to.
On Tuesday, the TMA kindly invited me to their day-long seminar on Online Marketing. This was because, in one of the morning sessions, Kath Moonan from AbilityNet was going to talk about building accessibile websites, and wanted to use The Stage as an example.
Thankfully, she was pretty complimentary about it. In an illuminating and humorous talk, she described how there was a trinity of elements that are important to a holistic view of good web accessibility:
- Content: Clear, easy to understand, jargon-free text.
- Design: Clear, high-contrast text, with all images described in their
alt attributes, etc.
- Markup: Semantically structured HTML, with proper use of headings, lists, etc.
In the Venn diagram she used to illustrate how these concepts linked, Kath was kind enough to place The Stage in the intersection of all three - “sitting,” she said, “like Buddha, in the middle, contemplating the Zen-like joy of having an accessible website”. Nice.
Of course, that doesn’t mean we should rest easy. There are plenty of things that we can, and should, do better. Kath brought up some pointers that I was already aware of, and am in the process of rolling out into our revamped pages, but there were plenty that were fresh concepts to me. I hadn’t, for instance, considered whether our Hamlet cartoons should have longdesc descriptions of the cartoon’s plot. That’ll be one that will take a while to implement: we currently don’t have the scope to store any metadata on each cartoon. However, adding metadata —specifically, tagging each strip so that we can direct people from, say, news stories to relevant strips, and back again — is already on our future projects list, so adding additional metadata can be incorporated into the development process.
Also, I hadn’t really appreciated that low-contrast elements are less likely to affect visually impaired users (who may well use screen readers, customised style sheets or other methods to tweak websites to make them easier to use) than they are people with dyslexia or other learning difficulties. Talking to my boss, he recalled a report that suggested there was a higher level of dyslexia among the arts community than in other professions, so that’s certainly something we shall be looking at.
Luckily, we have a new web developer, James, starting next week, so I won’t be facing issues like this on my own anymore. We’ve already discussed the need to start adopting a formal testing framework, rather than the ad hoc system I’ve been lucky enough to get away with to date. To be honest, that’s one reason why I find Rails so appealing, because it makes it difficult to justify not formally testing everything. As we move over, I’ll be doing my best to ensure that accessibility testing can, as much as practicable, be included in all our automated tests — although, of course, automated testing is never going to be enough.
One of the areas we were pulled up on was markup on our advertising. Unfortunately, I think this is one area where we’re unlikely to be able to make much headway. The vast majority of our advertising is pulled in from third party websites — we insert a piece of supplied Javascript, which in turn generates, and inserts into the page, HTML markup that, in many cases, is woeful. Now, we may be a big fish within the UK arts community, but in the scheme of Google’s AdSense publishers (to name just one example) I suspect we’re small fry. It won’t hurt to ask, but I suspect we won’t have much luck getting stuff changed.