Resource Description Format and XML-Data

Rick Jelliffe ricko at
Fri Jan 30 06:52:09 GMT 1998

> From: Tim Bray <tbray at>
> At 11:31 PM 29/01/98 +1100, Rick Jelliffe wrote:
> >In another forum the RDF people agreed that they *could* use the standard
> >DTD syntax to markup the information they wanted. 
> This statement is incorrect. 

For my evidence I refer to Ramanathan Guha (Netscape rep on RDF WG)

and Ralph Swick (W3C rep on RDF WG)

These two are the editors of the RDF parts. Their comments 
are unequivocable. 

I am not allowed to quote the drafts in public, under the archive 
rules from W3C. However, I can quote what I wrote which prompted 
their acknowledgement:

  # As far as I could find so far, the only difference from 
  # draft RDF that is required to allow me to use a standard
  # XML markup declaration is this: to shuffle the element type 
  # name of property elements into an attribute (here "RDF:Name"), 
  # and to use "RDF:Property" as the element type name instead.

> Inflammatory statements like this are defensible only when they're
> accurate.  This one is not.

See above. This is the third time we have had this discussion now Tim.

Let me say it again. RDF can use standard XML declarations if they want to.
They don't want to. That is their choice, and I wish them well. People should
not get the idea that RDF's choice is evidence supporting the necessity 
of things like XML-data (to the extent that this may occur).

> It is just as easy to conclude from the evidence that XML-Data is in
> direct competition with RDF and RDF-Schema.  I honestly don't know what
> the understanding of the authors of XML-Data is as regards things like
> RDF, and so far, nobody has said. -T.

Yes, this is a good point. But XML-data goes a lot further than RDF, both
for better and worse, IMHO. Please don't get me wrong that I don't support
XML-data and RDF 100% for what they say they are trying to do. But I 
think the implementation trade-offs that in them are skewed showing 
a bias against SGML declarations which I think is wrong-headed. 

In fact, the use of a regular expression syntax in content
models is the single best abstraction in SGML: to replace it with a million
elements somehwere in an instance, as XML-data currently does, 
seems to me to be getting rid of the only thing that programmers 
have an education to be comfortable with immediately! 
An idea verging on the bizarre.

XML-DEV is a forum where developers get their understanding of what the
markup community feels are the strengths and weaknesses of XML. And also
to gauge which way the wind is blowing for technical strategy. This is
why I have posted here; developers should be aware that many people have
opinions which are repeatedly given but which there is little evidence 
for when examined. I am sure I am often in this category. 

Personally, I believe that intuitions are highly credible when given by
a person of great expertise, especially for these complex matters where
it may be that the issues can never be articulated convincingly. Tim is 
such a person of great expertise. The reason why RDF cannot use standard
DTDs has not been articulated convincingly. But I think we should take
his intuition in this very seriously, and discuss it with the seriousness
such broad and confusing charges against the standard declarations warrant.

If I am being inflamatory, I apologize to all to the extent I may have 
unintentionally seemed rude. But, since I was not wrong, I cannot apologize
for that, much as I'd like to.

Rick Jelliffe

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at
Archived as:
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