- From: Wayne Dick <waynedick@knowbility.org>
- Date: Wed, 19 Feb 2014 13:58:36 -0800
- To: Lucy Greco <lgreco@berkeley.edu>, w3c-wai-ig@w3.org
Hello Lucy, Well I was going to end this thread but your question is interesting. First, NVDA will read off lots of styles, but I turned off all of that because I couldn't concentrate. (To turn on reading font styles in NVDA do-- menu: preferences: document formatting: report font attributes) Mostly I was thinking of visual semantics that would be more useful for partial sight. However scanning from DFN element to DFN element could be a good search technique on professional literature. What I am really thinking of ultimately is something like an ARIA role for style level semantics, something like a "style guide" role. It would not define a behavior like existing ARIA roles. Instead it would describe the relationship of visual style to meaning. Such a role could have many states like "APH", "MLA" or Associated Press style guide to mention a few. All such style guides would have various professional text structures that could be included as descendent roles. Some of these style semantic objects already exist in HTML elements like headings and lists at the block level, and CITE and DFN at the text level. However, most style guides for professional reading have a much richer semantic structure like the components of bibliographic entries. Rather than expand HTML 5 with more elements, and impossible task, why not include richer document semantics using an ARIA like interface. Here is an example of the semantic detail a style guide role could contain. In a bibliographic element for an article there is an entry for the publication that contains it. A containing publication role would be a searchable structure if it was part of an ARIA hierarchy. Maybe special landmark types would be all that was needed. The point is this. Professional documents for reading have many semantic elements that are generally represented by visual style. These are difficult to perceive and use for people with partial sight and invisible to people with no sight. I got my PhD in Mathematics at UCSD with congenital partial sight and it was not easy. Electronically accessible research articles with all the semantic cues available to fully sighted readers would have made research much easier. Then I could have just concentrated on the mathematics without having to fight the medium. With adequate semantic markers in professional articles, people with screen readers could use these markers one way, and people who needed text customization could use them another way. People with dyslexia might find another way to use them altogether. The point is, if they were there, we could use them for our own learning purposes. Well that is where I am going. I'll say more at CSUN. Wayne
Received on Wednesday, 19 February 2014 21:59:07 UTC