- From: Bert Bos <bert@w3.org>
- Date: Mon, 30 Jun 1997 18:35:33 +0200
- To: w3c-sgml-wg@w3.org
Rick Jelliffe wrote: > > > !> From: Martin Bryan <mtbryan@sgml.u-net.com> > > !> For the time being I would suggest that XML simply removes ID as a class of > !> names and recommends that users use of a CDATA called id, and applies an > !> XML-specific semantic requiring uniqueness of values for this attribute > !> within a document. > > I think "id" would be a bad element name, since it goes against the > HyTime practise of "id" being "ID". I have used quite a few DTDs that > use id as CDATA, and it is frustrating and works against my > expectations. > > I suggest, if we must have them > > XML-ID CDATA > XML-IDREF CDATA I prefer "ID". "XML-ID" is too long. The name "ID" is only reserved when it is an attribute. I don't think we need to say anything about elements that are named "ID," whether they go against expectations or not. "IDREF" is not necessary, "HREF" does the job already. > > I don't think we can have XML-IDS and XML-IDREFs without constraing the > CDATA in some way. I still think it is a bad idea, because it allows > obfuscated markup, which is a great sin. Sure, it may clarify some > things too. I don't think we'll need IDS, and IDREFS are handled with xlinks: <xlink><x href="#1"/><x href="#2"/><x href="#3"/>... (A bit verbose, I admit, but it avoids having to reserve any special delimiters.) Bert
Received on Monday, 30 June 1997 12:35:37 UTC