Reported by spanchang02 on 2013-09-27 16:33
The NVDA+T key reads the page title which is very good for non sighted users to determine "where am I?".
It will be very very useful if this was enhanced to make NVDA expose the nearest h tag content (where there is one) when the user's focus is on a UI element or even on other text in browse mode.

1. This will make NVDA support WCAG 2 Technique H80
   http://www.w3.org/TR/WCAG20-TECHS/H80.html
2. If this behavior is permitted on non focusable content, it will certainly help users but also make life for developers easy when complex data tables have to be marked up: consider a table where rows are grouped like the ones on:
   http://mars.dequecloud.com/demo/Census_2013.htm
   The second table is the same as the first but does not use headers-id method but just TH with scope attribute for header cells. The row group headers for Age, Race, etc. are marked up as h3. So when an NVDA user is navigating the table, he will have the column and row header cells announced of course. But he can query as needed (using NVDA key + T) to determine in which group of rows he is currently in.

So the contents of the h<> tag should be read before the page title preferably with a one or two second pause or a beep sound as a separator.
That will be great!
Thanks a lot,
Sailesh