Re: [orca-list] links list

[Jason White <jason jasonjgw net>]

Alex Midence <alex midence gmail com> wrote:
 
> On 12/6/12, Jason White <jason jasonjgw net> wrote:
> > On the other hand, I use searching heavily (including the ' key which
> > searches
> > only for links).
>
> Searching takes a lot of typing.  

Incremental searching, i.e., moving the focus to matches as each character is
typed, requires much less typing, and that's how I would handle your scenario
(which I haven't quoted here for the sake of brevity).

I would also use heading navigation and the usual tab sequence, but I think
being able to search for words incrementally, matching only links, is the real
efficiency gain.

If headings and other structural components weren't specified in proper HTML
then I would either get that fixed or obtain a script that would automatically
apply Aria roles to the relevant elements on the page. Google's Axsjax tool is
a good example of what I mean. I think there's scope for a crowd-sourced
solution that would let users and volunteers create scripts that improve the
accessibility of Web pages and applications. I don't know to what extent the
scripting can be automated. For example, it might be possible to write it so
that in many cases, a user can just mark text as, say, a heading, and the tool
will try to generate rules to match it later on. The general point is that
there is plenty of scope for browser-based solutions that transform and add
Aria to Web pages and we'll probably see more of them over time. These scripts
can also add access keys and transform the page in other ways.

I can also envisage service providers doing scripting on behalf of clients. If
I were working with a complex Web application heavily then it would be worth
getting the scripting done to improve efficiency still further, but my
fundamental suggestion is that I don't see a need for a list of links in your
scenario as described - I might be missing something important however.