Re: Audience Based Validator User Interface (ISSUE-206)

Hi Henri,

You wrote [1]:
> I don't really see the point of Laura's proposal.

As I read it, the point of Ted's proposal is to give engineers of
large web applications precedent over authors trying to do the right
thing and end-user requirements for accessible web content. That is
backward. It is also contrary to the priority of constituencies design
principle. The point of my proposal is to rectify that matter and set
things in proper order. As the W3C Validator documentation states,
"Validating Web documents is an important step which can dramatically
help improving and ensuring their quality..." [2]. It provides a
teachable moment, to whit: "Validation helps teach good practices"
[3]. Authors who are tring to catch errors should by default continue
to receive errors as they always have to help them in that task.
Hiding or suppressing errors from authors by default is counter
productive as it defeats their whole effort.

As I just said to Mike in the first email in this thread, The crux of
the matter has always been that two validator user groups 1.) authors
2.) engineers of large web applications have different goals. Good
authors want to catch errors so that they can fix them. Engineers of
large Web applications want to suppress errors that are beyond their
control so it doesn't reflect poorly on their product.

After thinking hard about this issue since 2008, I am wondering if an
audience based validator user interface might have the possibility to
satisfy both constituencies. Check my email to Mike and the audience
mockup at:
http://www.d.umn.edu/~lcarlson/research/206/byaudience.html
What do you think, Henri?

Another important aim of my proposal is to try to help improve
accessibility in the future. Check:
http://ur1.ca/9vryd
Thanks to Steve I just added more info to that section.

The only other way I can think of to solve ISSUE-206 besides creating
an incomplete attribute and the audience based UI is to change the
restriction on HTML5 authoring tools from "HTML5 authoring tools MUST
NOT emit documents that do not conform to HTML5" to "HTML5 authoring
tools SHOULD NOT emit documents that do not conform to HTML5". The
allowed exceptions would be content author supplied attribute values,
which are engineers control.

Best Regards,
Laura
[1] http://lists.w3.org/Archives/Public/public-html/2012Aug/0031.html
[2] http://validator.w3.org/about.html
[3] http://validator.w3.org/docs/why.html#learning


-- 
Laura L. Carlson

Received on Saturday, 4 August 2012 20:03:28 UTC