Re: id attribute in HTML 4.01

"Russell O'Connor" <roconnor@Math.Berkeley.EDU> wrote:

> > This is due to historical reason that the 'name' attribute was used
> > for anchors, and 'name' is of type CDATA, which is case-sensitive.
> > HTML 4 allowed the 'id' attribute (which is case-insensitive) as
> > an another way to create an anchor and defined that the 'id' and
> > 'name' attributes share the same name space.  Section 12.2.1 was
> > written in order to cope with this obvious contradiction.
> 
> Ah.  This is highly unfortunate.

Indeed.

> Of course the correct answer is to require uppercase for all fragment
> identifiers.

Since (at least some of major) existing user agents at the time HTML 4.0
was developed recognized fragment identifiers case-sensitively, that
option was unrealistic, unfortunately.

We do aware that this is weird, so XHTML 1.0 deprecated the 'name'
attribute, and XHTML Basic/1.1 removed the 'name' attribute, though,
ID is no longer case-insensitive in X(HT)ML.

Regards,
-- 
Masayasu Ishikawa / mimasa@w3.org
W3C - World Wide Web Consortium