I’ve spent some time over the past few months thinking about how I craft the content I publish for the web. Specifically regarding my use of language when writing. In one certain context — not to suggest my writing is free from more problems in others — it’s not as inclusive as it should be.
I’m referring to how a screen reader user experiences the words I write. And with my limited use of the technology, I’ve taken note of something quite specific. If you use a screen reader to speak my words, I’m not sure you, as a listener, will get all of the “subtleties” (case in point) of my intent.
Using the example I cited immediately above, precisely how is a screen reader user supposed to know I’ve put the word “subtleties” in quotation marks? Just typing quotation marks before and after the word isn’t enough to make a screen reader speak them.
A solution?
The obvious solution that immediately jumps out at me was marking up my content properly. That’s what the <q> Hyper Text Mark-up Language (HTML) element is for, right?
Once upon a time there were issues surrounding their use. Relating to support. Explicitly with regard to the :before and :after pseudo elements and Internet Explorer. This 2006 A List Apart article lays out getting around this issue with IE (notice the date of that article?).
These support issues aside, which likely aren’t a concern any longer — I haven’t worried too much about this, frankly, they aren’t the reason I still don’t use the <q> element in my HTML — despite what the article states about screen readers, it would seem support is still lacking. At least on the broader scale. After all it’s rare theory actually reflects reality, unfortunately.
But what about context?
There is another issue a work here. A more pressing issue, I think. It’s the issue of context. A screen reader user may be made aware of the presence of quotes, but is simply making a user aware of their presence enough to convey a greater meaning the visual cue is meant to convey that an audio only experience will express? I’m not sure it will. It’s subject to speculation.
As an aside, while I’m thinking about it, same goes for emphasizing text and making words appear important. And I won’t forget to mention abbreviations without titles, citations, code snippets, etc. There are reasons for semantics.
Anyway, I think the situation of drawing more attention to quoted text can be better addressed through a finer use of language. By placing the words “so called”, for example, prior to the word, in the example I used above, would provide all users with more information in which to better interpret what I’m saying.
So using said example — employing both the <q> (in a manner which the above A List Apart article detailed) and, which I’ll go ahead and term now, “the context helpers” I outlined above — you end up with “I’m not sure you, as a listener, will get all of the so called
Is it acceptable. No, I don’t think it necessarily is. But I do think it’s better. Especially considering in the tests I performed thus far (in VoiceOver, and much more testing is pending) marking up my quotations with the ‘subtleties’
of my intent.”<q> element has not made any difference. But, this way, I can rely on my added words to provide the listener with some further context. Every little bit helps.
Use of language
I’ve asked this before, and I quote, “why isn’t anything as simple as the utility demands?” My feeling I need to ask it again only serves to instil my broader point. Language is a curious beast, ain’t it? There are so many helpful ways to use it. And I really do need to improve my use of it.