- From: Dan Brickley <danbri@w3.org>
- Date: Wed, 31 May 2000 17:45:28 -0400 (EDT)
- To: xml-uri@w3.org
On Wed, 31 May 2000, Simon St.Laurent wrote: > Okay, so the RDF WG decided to use XML... > > [Also, the RDF spec doesn't make a strong commitment to XML. From the > Introduction: > RDF>It is also important to understand that this > RDF>XML syntax is only one possible syntax for > RDF>RDF and that alternate ways to represent > RDF>the same RDF data model may emerge. > ] Just wanted to respond to this aside for now; hopefully reply to the rest later. The RDF model & syntax specification says this for good reason: by making a clear distinction between the underlying RDF data model and the initial RDF-in-XML syntax we make room for new ways of interpreting various kinds of XML as RDF. Far from being a half-hearted commitment to XML, I read this as anticipating a number of developments that couldn't be specified in the RDF spec. For example, consider the mapping of Xlink typed links into RDF, the use of XSLT or annotated schemas to extract RDF models from 'mainstream' XML, or RDF interpretations of other compatible graph-serialisation syntaxes (eg. SOAP). RDF also had a charter to provide a metadata model that provides a common system across the wide variety of pre-XML syntaxes, files formats etc in use on the Web (eg. PNG, MP3 etc), not to mention the various other places data can live (eg. LDAP, RDBMS etc). So separating the abstract model from our default XML syntax was something of a necessity. RDF implementors tend to take XML's ubiquity for granted -- the attitude seems to be "well, obviously everyone is going to use XML for everything ever, that's wonderful, but what happens after that then...?". For example, I can learn the same fragment of information from any of the RDF-mappable data sources I listed above, sometimes in XML, sometimes not. The syntaxes may vary but the RDF representation can remain constant. This turns out to be a rather useful feature. IMHO, RDF's contribution to the XML picture seems to be as a (partial) strategy for dealing with the enormous success of XML, by providing a common data interpretation strategy de-coupled from any particular XML syntax. When the RDF M&S spec takes care to distinguish RDF's data model from one particular syntax, it is not through lack of enthusiasm for XML, but because this layering is a practical way of coping with the heterogeneity of the Web (pre- and post- XML). Dan -- mailto:danbri@w3.org
Received on Wednesday, 31 May 2000 17:45:30 UTC