I'm reading http://www.w3.org/TR/html401/struct/tables.html#adef-abbr and http://www.w3.org/TR/2005/WD-WCAG20-HTML-TECHS-20050630/#datatables_abbr and wonder about going the other way around.
In small tables like Calendars, you use abbreviations in the first place, because there simply isn't room to use the full words. I thought abbr could be used to make the full word be spoken for AT instead of sounding weird.
<th colspan="7" scope="colgroup">november 2010</th>
<th abbr="zondag" scope="col">zo</th>
<th abbr="mandag" scope="col">ma</th>
<th abbr="dinsdag" scope="col">di</th>
<th abbr="woensdag" scope="col">wo</th>
<th abbr="donderdag" scope="col">do</th>
<th abbr="vrijdag" scope="col">vr</th>
<th abbr="zaterdag" scope="col">za</th>
So in order to get "Monday 1" instead of "ma 1" I've got abbr. What comes to mind:
The first time through, the actual th is (may be, depends on reader/settings) read out, and then the abbr versions. Defeats the purpose.
Second, I'm assuming there's a visual precedent (there is) but not an audio precedent (I don't know if people are used to "wo" as meaning "woensdag" (wednesday)).
I'm wondering if there's any discussion or precedent of using abbr (or something else) for tables who have the opposite-from-usual problem: the headers are already shortened, and the full meaning should be spoken at least the first time.
There is mention of other-way-around in the W3C specs:
User agents must render either the contents of the cell or the value of the abbr attribute. For visual media, the latter may be appropriate when there is insufficient space to render the full contents of the cell.
On WCAG, the assumption is the header may be too long, and repeated listening of a long header is not desirable, so the abbr text can be read out instead (after at least one instance of the full header being read out).