- From: Butler, Mark <Mark_Butler@hplb.hpl.hp.com>
- Date: Wed, 7 May 2003 17:44:54 +0100
- To: "'Dave Beckett'" <dave.beckett@bristol.ac.uk>
- Cc: www-rdf-comments@w3.org
Hi Dave Please see comments inline. > The RDF Core WG has considered your last call comment captured in > http://www.w3.org/2001/sw/RDFCore/20030123-issues/#xmlsch-12 > and decided > http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2003Apr/0361.html > to postpone it. > > The main points we felt you raised in this comment are: > > 1) RDF/XML > - doesn't match the RDF graph model well > - many ways to write things (elements, attributes, > attribute values , ...) > - cannot write a W3C XML Schema, Relax NG schema, XML 1.0 DTD > - "not convienient" to use XSLT, use XQuery, other XML tools Not quite. Sorry this is my fault so I will explain my concerns more fully. It seems to me the chief difficulty with using RDF/XML is its inherent incompatibility with existing XML tools due to the fact there are multiple possible serialisations of the same model. Some standards based on RDF use "cut-down" versions of RDF/XML, which reduce the number of possible serialisations and hence make it more ammenable to XML tools. An alternative approach is use XML and transform it to RDF/XML when required via XSLT (e.g. the MPV normalized metadata format [3]). However such "solutions" are not ideal and in my experience discouraged by W3C team members. Secondly parsing RDF/XML is less efficient than XML even when using specialized RDF/XML parsers due to the multiple serialisations [10]. Thirdly I think the striped syntax used in RDF/XML is unnecessarily complicated as it deviates from the way people commonly encode information in XML. In my opinion, people would find RDF/XML easier to adopt if it used an unstriped serialisation which would be closer common usage XML. Other people have also made the same point [11]. These issues have been widely discussed many times before [4] [5] [6] [7] [8] [9]. It has been suggested that the complexity of RDF/XML is not an issue as it can be hidden from users in a similar way to WYSIWYG HTML editors. However I note this only solves the third problem. > Although we note, most of the above XML technologies mentioned above > are successfully used with RDF/XML. I am afraid this is not my experience when using XSLT and XML Schema with RDF/XML as I describe in depth in [1] and [2]. If you have evidence to the contary, please can you point me at it? > We know and could give you more problems. However we felt we > couldn't fix it all due to the charter constraint: > [[The RDF Core WG is neither chartered to develop a new RDF > syntax, ...]] > -- http://www.w3.org/2001/sw/RDFCoreWGCharter I appreciate and have sympathy with your charter constraint. This would be fine if the W3C viewed RDF as "under development". However the problem with waiting to "fix this in the next version" is that the W3C team is strongly advocating that people should use RDF/XML in preference to XML to be "semantic web ready". We already have a number of groups using different forms of RDF/XML (OWL, CC/PP, UAProf, EARL, RSS) and other groups being encouraged to use RDF (P3P, WG glossaries and issue lists, HTML metadata, web services). Therefore I think we need to address this sooner rather than later. I would encourage the W3C team to actively consider these comments, this charter constraint, their desire to encourage the deployment of the semantic web to see if there is not some way of addressing these concerns in a timely manner. > We will add this issue to the RDFCore postponed issues list at: > > http://www.w3.org/2000/03/rdf-tracking/#rdfms-validating-embedded-rdf > > Please reply to this email, copying www-rdf-comments@w3.org indicating > whether this decision is acceptable. I would like the comments in this email recorded in postponed issues list. Thanks. [1] http://www-uk.hpl.hp.com/people/marbut/capClass.htm [2] http://www.hpl.hp.com/techreports/2002/HPL-2002-268.pdf [3] http://www.idealliance.org/papers/xml02/dx_xml02/papers/06-05-03/06-05-03.ht ml [4] http://lists.w3.org/Archives/Public/www-rdf-interest/2000Aug/0097.html [5] http://www.textuality.com/xml/RPV.html [6] http://www-db.stanford.edu/~melnik/rdf/syntax.html [7] http://www-db.stanford.edu/~melnik/rdf/fusion.html [8] http://www.w3.org/DesignIssues/Syntax [9] http://www.xml.com/pub/a/2000/08/30/instantrdf/ [10] http://www.dsv.su.se/~johank/publications/others/ManuelPalacio/palaciomsc.pd f [11] http://www-dbs.cs.uni-sb.de/lehre/ss03/xml-seminar/Material/PS02.pdf Dr Mark H. Butler Research Scientist HP Labs Bristol mark-h_butler@hp.com Internet: http://www-uk.hpl.hp.com/people/marbut/
Received on Wednesday, 7 May 2003 12:45:19 UTC