W3C home > Mailing lists > Public > public-html@w3.org > June 2007

Re: headers attribute

From: Leif Halvard Silli <lhs@malform.no>
Date: Mon, 04 Jun 2007 01:40:27 +0200
Message-ID: <214fd05e6e177162647a944c4c949ae7@10013.local>
To: public-html@w3.org

On 2007-06-03 19:02:04 +0200 Sander Tekelenburg <st@isoc.nl> wrote:

> At 17:28 +0200 UTC, on 2007-06-02, Leif Halvard Silli wrote:
>>> Can you imagine a situation where the screen reader first present the SCOPE
>> view of who the header cells are, and then the HEADERS view of who they are?
>> That would be utterly confusing for users.
> 
> I'm not so sure. For one, as you say, depending on whether the user looks at
> a header or a data cell, *either* scope or headers will be useful. The more
> complex a table, the more useful it will be to not only be able to first get
> an overview with the help of scope, but to then still have headers as an aide
> when you're descending into the data cells.

For investigating which cells a header cell is the header for, it would seem logical that the UA could make use SCOPE=. An extra use case. As a ::scope pseudo-element (I took note, Anne) would be an extra use case. But appart from the extras, SCOPE is supposed to be used for the same use case as HEADERS is. One typical use of HEADERS ought to be to specify exception to what SCOPE specifies. When HEADERS is used to to specify exceptions from SCOPE, then it would be logical to think that the info from HEADERS  replaces the info which SCOPE has stored for that particular cell. But perhaps not necessarily. 

> For another, isn't it the UA's job to serve the user? If, in a given case,
> providing both scope and headers would indeed be confusing to users, it would
> seem to me it is the UA's job to solve that -- magicly choose to use either
> one, or let the user configure which to use. Better yet, provide acces to
> both in a manner that is  useful, not confusing.

I guess you are right that it could be up to the user agent to make sense of it. Perhaps they really are two different «layers». As such, in a table that is fully covered with SCOPE, in make no sense to duplicate the SCOPE info with a HEADERS attribute in each cell. (It is another matter if we are discussing author requirements. An author requirement could be that since so few AT UAs support scope, we must use HEADERS as well - or instead.)

Btw, the HTML5 proposal does not only remove HEADERS. It also puts restrictions on SCOPE: In HTML4, both SCOPE and HEADERS could be used for/in TD cells as well as TH cells, since the internal structure of a table can be more complited than it visually seems to be.

And does anyone know what happens if I put SCOPE on a TH within the THEAD element. And then add SCOPE to the correpsonding TH in same column in the TFOOT element, in HTML5 (or in HTML4)? A cell in this column would then be scoped from top and from bottom. In the code, it would come after the SCOPE in the THEAD.
-- 
leif halvard silli
Received on Sunday, 3 June 2007 23:40:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 9 May 2012 00:16:00 GMT