Re: How long /hard is it to install the personlization.

Hi Alistairre ARIA:  there were abstract concepts and supported states that were based on different abstraction , and the roles were made by including them in an owl taxonomy. We hoped RDF would get AT or browser support and people could create new roles based on pulling together abstract concepts. 


1.3.1 is not so useful unless we  get rid of : "or are available in text" 



All the best

Lisa Seeman

LinkedIn, Twitter





---- On Tue, 27 Jun 2017 17:15:15 +0300 Alastair Campbell<acampbell@nomensa.com> wrote ---- 

      Hi Lisa,
  
 > I wrote the first draft of ARIA. Let me assure you we were basing it to make sure that we were trying to address new patterns as much as possible. 
  
 I don’t understand, are you saying things like progressbar, spinbutton, scrollbar, tabs, menu etc. were ‘new’ patterns?
  
 Unless there’s been a lot of mis-communication I thought the point was it codified known software patterns into HTML? I.e. the patterns were not new, but the accessible implantation without an alternative was new.
  
 What I was saying was that it would help to find current patterns in websites that map to the semantics you want to add. I think quite a few will be covered under 1.3.1 (for those which are visually apparent) or 4.1.2 (restricted to user interface components). 
  
 For example, hiding extraneous information could align with 1.3.1, ARIA landmarks and <main> weren’t reliable in 2008 but they are now. There was a long argument about whether landmarks were covered before we started 2.1, which arguably we should bring up again now we are in 2.1.
 (Ahem, David McDonald, what do you think about that?) 
  
 If there is a reliable pattern (e.g. use of main), then it is *relatively* easy to incorporate, if not in 1.3.1 then in 4.1.2.
  
 If we had a HTML technique for “Identify the primary content of the page with main”, that is something we can fail against in audits.
  
 Adding icons is for user-interface components (things within navigation if I understand the aim), might align with 4.1.2, if not perhaps it needs a new SC, but it would make sense to have a focused SC for that purpose rather than an SC intended to bring in a separate spec.
  
 What I’m fairly sure will not work is trying to specify what ‘core’ content / functionality is. “Authors” simply don’t know in too many cases, and it is even harder for testers.
  
 Taking the approach of saying “something” is specified as non-core will punish simple sites which do not have extraneous stuff on the page.
  
 I think the closest WCAG 2.0 has is ‘essential’, for which the basic test is “can you remove it?”. Authors (site owners / designers etc) get that. I can’t see how to define ‘core’ beyond that. (Note that advertising is a key feature this concept should work for. If the site wouldn’t exist without advertising then it is essential to the website, but not to the user!)
  
 Cheers,
  
 -Alastair