[whatwg] Default scope for table headers

[Nicholas Shanks]

On 1 October 2012 10:21, Michael[tm] Smith <mike at w3.org> wrote:

> Don't look to that document for any information about default UA behavior,
> or anything at all about UA processing behavior. I tried to make that very
> clear in the abstract and intro for that document.

Sorry, I never saw that:
https://encrypted.google.com/search?q=html5+default+header+scope

for me, returned:

Coding An HTML 5 Layout From Scratch | Smashing Coding
coding.smashingmagazine.com/.../designing-a-html-5-layout-from-sc...

HTML th scope Attribute
www.w3schools.com/tags/att_th_scope.asp <- this domain is on my block
list, yet still shows up?!

th – table header cell - HTML5
www.w3.org/TR/html-markup/th.html <- so I clicked here :-)

td – table cell - HTML5
www.w3.org/TR/html-markup/td.html


> I think what you want is already define[d] in the actual HTML spec:

So it is. Bad Google, no Cookie headers for you.

>   http://www.whatwg.org/specs/web-apps/current-work/multipage/tabular-data.html#attr-th-scope
>   http://www.whatwg.org/specs/web-apps/current-work/multipage/tabular-data.html#internal-algorithm-for-scanning-and-assigning-header-cells
>
> But if that's missing something you should file an HTML spec bug.

I got here by reading through old posts at 456bereastreet and found this one:
http://www.456bereastreet.com/archive/200910/use_the_th_element_to_specify_row_and_column_headers_in_data_tables/

It struck me that the scope attributes should have default values for
such a simple table.

Having read the algorithm you linked to, and realising I had already
read it several moons/aeons ago, I see they do different things but
towards a similar end goal. The HTML5 algorithm defines how a UA
associates headers with a particular data cell, e.g. for speaking
aloud. My algorithm changes the default value the scope attribute
(from which the cell headers can then be computed, via the existing
algorithm).

I believe applying the HTML5-specced algorithm to the table I used as
an illustration would result in the same headers being used for each
cell as if my own algorithm was used (though I have not done a
thorough verification), leaving just these questions:

Are there use cases where the value of the scope attribute matters
other than as an intermediary for computing the headers applicable to
each cell? If not, are there use cases where either the data cell
headers have not yet been computed or they are unavailable (perhaps
while Javascript DOM tree walking?) where access to the scope
attribute would be helpful?

If either of these can be answered in the affirmative, then I believe
a bug could be raised. Anyone care to chime in?

-- 
Nicholas.