Spider Trax

Be Accessible, Don’t Meet Guidelines

Some time ago, Jack Pickard published a very interesting piece in which he questioned whether too much emphasis was being placed on sites failing 1 or 2 WAI checkpoints.

Having had this article drawn to my attention only recently, I felt he had some very good points to make and that, overall, it is exactly this kind of web accessibility discussion we need. Lip service to any set of binary checkpoints is Bad and, far from undermining the accessibility arguments, playing Devil’s Advocate from time to time can only strengthen - not weaken - the case for all sites to be as accessible as possible.

That’s not to say that I agreed with every point he made and I wanted to focus on two specific issues he raised.

Quotations

3.7 Mark up quotations. Do not use quotation markup for formatting effects such as indentation.WCAG 1.0

I disagree with Jack here. I do think that marking up quotations correctly does enhance accessibility. As I understand it, screen reader software can often be configured so that a marked-up quotation is spoken using a very different voice to the default voice used with standard web content. I suppose that the visual analogy would be to use italics or colour to highlight a phrase although, in theory, this is what quotations marks are supposed to do.

It seems to me that, when you are absorbing information via audio scanning and trying to simultaneously build a mental model of a page, this additional audio colour can greatly enhance overall comprehension. And, since it’s something that primarily affects the visually impaired, that makes it an accessibility issue.

Identifying Language Changes

4.1 Clearly identify changes in the natural language of the documents text and any text equivalents.WCAG 1.0

With regard to identifying language changes, perhaps we need to stop and ask why we are doing this. Then we need to be clear regarding who will benefit - either now or in the future?

At this point, I think it does need to be said that this checkpoint does have significant validity for sites aimed at traditionally bilingual cultures. Some Canadian sites, for example, will offer both English and French versions and it would be reasonable to assume that screen reader users from within that culture would be likely to have both English and French synthesisers at their disposal. But that is a far cry from identifying a single French word or phrase on an otherwise English page - especially when the word, although derived from French, is an accepted part of natural English. Café, for example.

That said, it is true that there are words in different languages that have identical spellings but whose pronunciation and meaning are very different. So it would logically follow that, in such cases, we should indicate which language should be used to provide meaning. Yes?

However, putting aside pure logic for a moment, who benefits?

In theory, screen reader software could use data provided by the lang attribute to correctly render non-native words but this assumes the user has access to the relevant synthesiser - which may not be a safe assumption. Even if the user can hear the word pronounced correctly, pronounciation alone does not guarantee comprehension and comprehension is an important part of overall practical accessibility.

Meanwhile what about non-screen reader users? In such situations, being able to see the word does not guarantee that the viewer can pronounce or understand the word - let alone be aware that there are different meanings in different languages. So I find myself wondering why standard graphical web browsers do not handle identified language changes in the same manner as, say, abbreviations, by default. Furthermore, if it affects the wider audience rather than impacting only on disabled user groups then, frankly, I don’t see how it can be classed as an accessibility issue.

In summary, I feel that this particular guideline is really the result of pure theory coupled with a (perhaps over optimistic?) view of how the Web and, specifically user agents, might develop. It may not have been really thought through initially and, as a result, the concept has never been been developed to its logical endpoint.

So we’re left with a guideline that, with the exception of whole web pages on bilingual sites (e.g. Welsh and English) may not have had any positive effect and may only be still in existence because no one ever stopped and asked “Why are we doing this?“.

Top