Attributes vs. text content (Was Re: RFC: Attributes andXML-RPC)

Marcus Carr mrc at
Thu Sep 23 20:33:09 BST 1999

Mark Nutter wrote:

> This is the reason, in a nutshell, why XML needs attributes -- XML (and
> SGML) DTD's don't have any mechanism for specifying default/enumerated
> values for PCDATA content.  But now, suppose we had some kind of standard
> schema mechanism that allowed us to say things like "Content must be a
> date" or "Content must be numeric" or "Default value is 'Left'" and so
> on.  If we could constrain the character data contained within an element
> in the same way(s) we constrain attributes, would we still need attributes?

I think that we would. Attributes have never provided enough control to alleviate the
situation you describe, so semantic checks on elements, attributes and the relationships
between fragments of data have been the norm for as long as SGML has been around. You would
need to get much further than just match the constaints provided by attributes before I could
see any reason to get rid of them. My understanding is that this is what schemas will excel at
- semantics checking. But then if you're working through such a ubiquitous model, why would
you care if something existed as an element or an attribute...

> Disclaimer:  I know various schema proposals are on the table, but I'm not
> familiar with any of them.

Me either.


Marcus Carr                      email:  mrc at
Allette Systems (Australia)      www:
"Everything should be made as simple as possible, but not simpler."
       - Einstein

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at
Archived as: and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at

More information about the Xml-dev mailing list