Re: Should WCAG explicitly talk about *mainstream* assistive technologies?

Interesting thought, leads me to wonder what determines when a user agent
(or the new hypothetical AT) crosses a threshold to be deemed "mainstream".
Is it some value in the worldwide marketplace, a regional market
penetration value, or even more local, a value in a closed environment?
Would the new exotic AT be considered mainstream because it offers unique
features/functionality designed to address the needs of people with a
specific cognitive impairment in a way no other AT provides, even though
only a tiny fraction of people use it?

If "mainstream" cannot be further defined in WCAG, then the consumer of
WCAG would need to define it (recall the grief of "sufficient" in WCAG
1.0's "sufficient contrast"). Moreover, how would "mainstream AT" be
measured when statistics on AT usage essentially do not exist.

In my roughly 15 years of experience, when undefined words like
"mainstream" appear in WCAG, the opportunity for distracting debate is
created at the point when people are trying to apply WCAG. One word in WCAG
can disrupt and derail efforts that were otherwise successfully proceeding
with the usual level of resistance.

On the other hand, the authors of WCAG are trying to be mindful of the
critically important need to avoid requiring content authors to support
every and any AT or user agent. The WCAG Conformance language uses the
defined term "relied upon'. Perhaps substituting "relied upon" for
"mainstream" would avoid the possibility for debate while preserving the
need to avoid requiring support for everything.

That said, if the new exotic AT conforms to UAAG, perhaps a need to code
specifically for it would not exist, nullifying the need to qualify AT with
"mainstream". If UAAG conformance does not exist, then the lack of
accessibility falls to the AT, not on the WCAG conforming authored content.

On Fri, Jul 15, 2016 at 6:17 AM, Patrick H. Lauke <redux@splintered.co.uk>
wrote:

> In the current definition of assistive technology
> https://www.w3.org/TR/WCAG20/#atdef the concept of user agent is
> qualified with "mainstream", e.g.
>
> "hardware and/or software that acts as a user agent, or along with a
> mainstream user agent"
>
> "Note 2: Assistive technologies often communicate data and messages with
> mainstream user agents by using and monitoring APIs."
>
> Would it also make sense to qualify the actual "assisitive technologies"
> with "mainstream"? In the current wording, it seems that any AT, even a
> completely non-standard one (for instance, if somebody invented a
> completely new AT that doesn't act/behave like any other common AT, and
> crucially does not interface with mainstream user agents or the OS in a
> standardised way), would still be covered, meaning that in theory authors
> would need to ensure their content works correctly (so testing, potentially
> coding specifically for that exotic AT that doesn't work like any other
> similar AT) against it in order for their content to be considered
> accessibility supported?
>
> Am I misreading/misunderstanding this, or is it an actual potential
> problem?
>
> Would adding "mainstream" in front of "assistive technologies" in most of
> that definition address this potential problem?
>
> Thoughts (and inevitable lengthy discussion) welcome,
>
> P
> --
> Patrick H. Lauke
>
> www.splintered.co.uk | https://github.com/patrickhlauke
> http://flickr.com/photos/redux/ | http://redux.deviantart.com
> twitter: @patrick_h_lauke | skype: patrick_h_lauke
>
>


-- 
Regards,
Kurt Mattes
Accessibility Program Manager
Deque Systems
610-368-1539