RE: aria-level a required property for role="heading" or a supported property with an RFC SHOULD for authors

I thought it might be useful to gather feedback from screen reader users, since they are the ones most affected by this proposal. With the caveat that this is both quick and informal, I posted a simple question to members of the British Computer Association of the Blind (BCAB) discussion forum [1].

 

The question was this:

 

Which of the following scenarios would you prefer:

1. A page with multiple headings, where a screen reader would announce every

heading but not the heading level.

2. A page with multiple headings, where a screen reader would announce every

heading and its level, but where the heading levels may or may not be

accurate.

 

To date 20 people have responded. Eight have chosen #1, and 12 have chosen #2.

 

The comments and explanatory notes people also supplied can be found by reading through the thread [1]. Many felt that it was a “Hobson’s choice” – in other words that it was difficult to choose, but so far it seems that more people would prefer to have levels even if there is a risk they may not be accurate.

 

It’s also worth noting that all three screen reader users on the TF call this afternoon, in addition to at least one more screen reader user on this thread, all favour the option to make aria-level a required attribute – although for reasons that stray as much into the implementation as the usability of headings.

 

Léonie.

 

[1] https://lists.bcab.org.uk/pipermail/bcab/2015-June/044351.html 

 

 

 

 

Received on Friday, 19 June 2015 00:13:28 UTC