Re: Thoughts on testing DPUB-AAM and DPUB-ARIA

Note that either the textarea into which publishers paste data OR the
script that iterates over a collection of input files would generate the
same sort of output that would in turn generate the test report.

I don't mind which solution we use.  Just pointing out the pseudo-obvious.

On Fri, Nov 4, 2016 at 9:32 AM, Ivan Herman <ivan@w3.org> wrote:

>
> On 4 Nov 2016, at 15:26, Shane McCarron <shane@spec-ops.io> wrote:
>
> I was doing some work on DPUB testing this morning, and I wanted to
> capture some ideas here to see if they resonate with anyone else.  In no
> particular order they are:
>
>
>    - We can easily use WPT to exercise a11y api mappings with the new
>    ATTA stuff [1]. This will work on major platforms.  It will not (yet) work
>    on proprietary platforms, but we don't have mappings for those anyway!  I
>    think that this is all that is needed to provide testing of dpub-aam and
>    get that over the CR hurdle.
>
> Brilliant.
>
>
>    - We could use WPT and a simple interface to check dpub-aria documents
>    for role attribute usage.  This would be a purely manual check in that a
>    user would need to put content into a text area then run the test.  This
>    would allow basically anyone who can access the internet to run their own
>    tests.
>    - Another way would be to write a simple scripts to run over input and
>    extract role attribute value usage.  This would NOT be part of the WPT
>    environment, but would probably be useful if someone wanted to process a
>    ton of HTML documents.
>
> I am tempted to go for the second option here, because my gut feeling is
> it may make it easier to generate a decent and readable report of the
> testing, with names and references to the users, etc. The report should be
> easy to read for the Director…
>
>
>    - We could probably figure out a way to automate the manual test
>    option above, but it would require some sort of instrumentation on the part
>    of the test users (e.g., a web server with a folder full of files that it
>    could retrieve).
>
> I do not think we should go there.
>
> Ivan
>
> Thoughts?  Other ideas?
>
> [1] https://spec-ops.github.io/atta-api/index.html
>
> --
> Shane McCarron
> Projects Manager, Spec-Ops
>
>
>
> ----
> Ivan Herman, W3C
> Digital Publishing Technical Lead
> Home: http://www.w3.org/People/Ivan/
> mobile: +31-641044153
> ORCID ID: http://orcid.org/0000-0003-0782-2704
>
>
>
>
>


-- 
Shane McCarron
Projects Manager, Spec-Ops