The Roadmap for Accessible Rich Internet Applications (ARIA) outlines the technologies to map controls, AJAX live regions, and events to accessibility APIs, including custom controls used for Rich Internet Applications.
It is a much needed document, and it sparked off a lot of ideas for me about where scripting and accessibility can and will need to go.
However, I do have a few questions. I won’t call them critisisms, because I’m stepping out of my usual sphere of knowledge (it refers to quite a bit of XHTML2 & DOM scripting). This post just draws attention to things I (and I think other people) could do with some further explanation on.
Early developer input
- Produce Developer Guidelines during the
process- This is a critical mistake made by people creating a new accessibility model. Developers must be on board early on so that they may contribute feedback and start producing workable solutions early.
Totally agreed, but much of this roadmap seems dependant on at least XHTML 1.1, which is still unsupported by the majority browser (Internet Explorer 6, or 7 in a few weeks). This seems especially strange when This specification was created based on an analysis of the accessibility properties defined in MSAA (Microsoft Active Accessibility) and ATK.
The MSAA is being dropped with Windows XP.
If the roadmap could make use of XHTML 1.0, perhaps with a custom (but steady and standard) DTD, it might take off quicker.
Strange use of some terms
There were a couple of terms that I either didn’t know, or didn’t understand their use in this context.
Qname came up without a link or explanation: Each RDF role may be referenced using a qname from Host XML markup such as XHTML
. Googling brought up Qualified name, but I think it could do with a glossary entry.
Enumerate I understood as counting things, but it doesn’t seem suitable for this use: An impaired user wishing to enumerate the available actions that may be performed on a page element.
It may be correct, but wouldn’t ‘discover’ or something else that isn’t a field of mathematics be easier to understand?
A look too far ahead?
Section 4.5 says Up to this point, the roadmap has discussed addressing today’s XHTML markup as supported by today’s desktop browsers.
Perhaps I’ve mis-read it, but all the examples have been of XHTML 1.1 or 2.0, which unless the mime type restrictions are dropped, is not compatible with the current majority browser or it’s upcoming replacement. If you can’t make use of the techniques in HTML 4.01 or XHTML 1.0 strict and keep valid markup, the prime audience of standards-aware developers will be reluctant to try them.
Anyway, that’s enough moaning, some of these points could be moot if IBM’s proposal for Embedding Accessibility Role and State Metadata in HTML Documents covers current browsers.
I’ll read that through in more detail. I have a few ideas sparked off by this, so I’ll either contribute or shut up!
Technorati Tags:
Perhaps they meant “evaluate” and not “enumerate”. Hmm…
I think ARIA is taking baby steps. I think they want to do it right and taking the time to think ahead – after all, it is a roadmap. 🙂
We can’t help that IE7 and less doesn’t support XHTML as application/xhtml+xml, but we can hope that support will come in the future. You have to bear in mind as well, that (as far as I know) Firefox 1.5 is the only browser with some support for roles and states at the moment. Section 4.6 of the ARIA through to 4.11 Implementations makes useful reading.
Agreed, I don’t mean to sound critical, these are minor points/questions in the scheme of things, and I will be digging through the links & current implementations soon.
I don’t think you sound critical – questions are good. 🙂