[
Lists Home |
Date Index |
Thread Index
]
Arjun Ray scripsit:
> John Cowan <cowan@mercury.ccil.org> wrote:
> | The RDF/XML syntax is designed to minimize the Hamming distance between
> | any fairly well designed data-oriented XML and valid RDF.
>
> An example demonstrating this assertion would be enlightening. Right now,
> I'm not sure I understand how "Hamming distance" applies.
See http://www.xml.com/pub/a/2002/10/30/rdf-friendly.html . I simply
mean that it is intended to be fairly easy to evolve one's document type
to make it RDF-compatible.
> All the more so because I don't find any evidence of "design" in RDF/XML
> syntax: on the contrary it's all too obvious that some people were in an
> inordinate rush to start tossing taggery into Netploder.
Well, it looks designed to me, although there are some deficiencies,
most of which have been rectified by the new WD.
> When it - naturally, later - came to light that DTDs probably wouldn't
> "allow" such happy-go-taggy insouciance, the tack to take became, "Well,
> this DTD stuff is no damn good." BTDT.
DTDs are inherently context-insensitive, and the only way to achieve the
useful approximation required was to have a great deal of context
sensitivity.
> Yes. and that's why I find the claim of maximum flexibility dubious.
If RDF/XML had been designed simply as a syntax for capturing the RDF
knowledge model, like XTM, it would be a good deal more rigid and "meta"
than it is. But that wasn't the idea.
--
John Cowan jcowan@reutershealth.com http://www.ccil.org/~cowan
Is it not written, "That which is written, is written"?
|
