W3C home > Mailing lists > Public > www-html@w3.org > July 2005

Re: Why don't td and th have display/value pairs?

From: David Woolley <david@djwhome.demon.co.uk>
Date: Mon, 4 Jul 2005 21:01:50 +0100 (BST)
Message-Id: <200507042001.j64K1o306788@djwhome.demon.co.uk>
To: www-html@w3.org

> How is a UA supposed to determine if two identical string values (like
> names) are in fact the same semantic value? Wouldn't it be better if
> there was a mechanism that allowed table cells to identify their
> semantic value independant of their display value?

In case anyone missed the point, this is based on a proposal, in another
thread, to abolish rowspan and colspan.

I would say a fundamental reason for not abolishing them is that
HTML is a markup language, i.e. it takes the plain text of a document
and adds additional structural clues, not a relational data format.

Also, because it is a document format, not a database GUI interface
language, many documents will not contain unique codes for each
logically distinct value, so the author would have to assign 
arbitrary codes in order to make the associations between logically
equivalent cells.

There is also the issue that domains overlap, so codes would have to
encode the domain, which they would not have to do in the database
itself, for database derived tables.
Received on Monday, 4 July 2005 21:37:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:16:03 GMT