Re: [css3-lists] [css3-speech] Purpose of the module

[Snipping most of your message, which I agree with, to talk about the
one part that I don't.]

On Thu, May 19, 2011 at 11:29 AM, Belov, Charles
<Charles.Belov@sfmta.com> wrote:
> Tab Atkins Jr. wrote on Tuesday, April 19, 2011 2:33 PM
> Without commenting specifically on number styles such as -financial,
> my take on the intended purpose of the Lists module is that it is
> intended to do the following:
>
> - Marking content semantically as a list.  This would also allow
> someone using a screen reader to skip over the list.

This is very specifically *not* a goal of the Lists module, or indeed
any CSS module.  CSS is a presentation language for transforming
meaningful content into a styled box-tree for display.  It knows
nothing of semantics, and will never influence the semantics of the
source language.


> And the following need I have previously raised, which apparently
> conflicts with the idea that CSS-generated content is not supposed to
> be semantic.
>
> - Allowing for the author to specify whether the list numbers
> are semantic, that is, whether a screen reader can safely read "1,
> 2, 3" or whether it needs to read the numbering as rendered.  I
> have previously mentioned that the SFMTA has legal notices where "A,
> B, C" must be read by the screen reader as "A, B, C" as that is how
> the items would be called.
>
> As you or others have pointed out, or as I have inferred, I could
> simply type:
>
> <ul style="list-style-type:none;">
> <li>A. First item</li>
> <li>B. Second item</li>
> <li>C. Third item</li>
> </ul>
>
> but then I would lose formatting independence of the number and the
> remainder, and the screen reader might read "1, A. First item" which
> is still not what I want.
>
> While I realize that I could achieve the formatting independence with
> <li><span class="marker">A.</span><span class="listitem">First
> item</span></li>, in real life that would be such a time sink that
> I would keep the simpler code and put up with the formatting issues.
>
> But if it is not suitable for the Lists module, then perhaps the Speech
> module would need an property as to whether or not to read the list
> sequence number, so that the listener would only hear "A. First item"
> for <li>A. First item</li>.

Lists now defines a way to specify that the contents of an element
should be used as the list marker.  Theoretically, a screenreader
could key off of this information and read that element as the marker
rather than an auto-generated marker.  If that's not acceptable, then
yes, Speech should specify a control like that.

~TJ

Received on Thursday, 19 May 2011 19:05:51 UTC