Re: Minutes for ARIA telecon on 9 December

I wrote:

> Of these, seven fail, namely, 627 - 631, although testing of those 
> cases is sparse -- I'll try to find time to fill in the ATK/AT-SPI 
> column for these seven.

There is another wrinkle.  These test cases fail not because step 2b is 
problematic, but because in this sub-case, the spec improperly)] 
discusses embedding an menu control.  It should have been embedding an 
menu *button*.  That was a goof, and will be addressed in ARIA 1.1; see 
ACTION-1154, "Re move 'If the embedded control is a menu, use the text 
alternative of the chosen menu item.'from Name Computation, as it would 
be a menu button or popup button, not the menu itself." 
(https://www.w3.org/WAI/PF/Group/track/actions/1154).

Also, test case 631 involves an embedded input type="file".  I believe 
we decided that this was invalid since a file input does not render as a 
simple atomic control and/or is rendered differently on different 
browsers.  FF, for example, renders it as a "Browse ..." button followed 
by static text that is the file name of the file chosen.

I don't think these issues, which are specific to one sub-clause of step 
2b, make the entire step at risk.

-- 
;;;;joseph.


'A: After all, it isn't rocket science.'
'K: Right. It's merely computer science.'
              - J. D. Klaun -

Received on Monday, 9 December 2013 18:38:39 UTC