W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > August 2003

RE: pfps-06 hold off?

From: <Patrick.Stickler@nokia.com>
Date: Wed, 27 Aug 2003 14:44:48 +0300
Message-ID: <A03E60B17132A84F9B4BB5EEDE57957B5FBC4C@trebe006.europe.nokia.com>
To: <jos.deroo@agfa.com>
Cc: <bwm@hplb.hpl.hp.com>, <phayes@ihmc.us>, <w3c-rdfcore-wg@w3.org>, <Patrick.Stickler@nokia.com>


Are the .NET and Xerces APIs responsible for the *creation*
of the invalid lexical forms?

I.e., is there really a problem with using common XML tools
to scrape XML content and generate valid RDF?

It seems that the real problem is simply that RDF applications
that base their RDF/XML validation on tools that operate
using XML criteria are falling short of the line.

Please, let's keep distinct the generation of versus validation
of RDF/XML.

I would be very surprised if any of the mentioned tools are 
actually generating invalid lexical forms.

Applications transferring data from an XML to an RDF context
could likely use such tools to normalize lexical forms before
creating any RDF/XML or interacting with an RDF graph API.

Can you clarify whether .NET is generating invalid lexical
forms, or simply failing to identify them.

Thanks,

Patrick

> -----Original Message-----
> From: ext Jos De_Roo [mailto:jos.deroo@agfa.com]
> Sent: 27 August, 2003 14:26
> To: Stickler Patrick (NMP/Tampere)
> Cc: bwm@hplb.hpl.hp.com; Stickler Patrick (NMP/Tampere); 
> phayes@ihmc.us;
> w3c-rdfcore-wg@w3.org
> Subject: RE: pfps-06 hold off?
> 
> 
> 
> Patrick - it's not only Xerces. Also all of the .NET support I'm
> aware of is in that way, e.g. for
> http://www.w3.org/2000/10/rdf-tests/rdfcore/xmlsch-02/Manifest
> we always get
> 
> <http://www.w3.org/2000/10/rdf-tests/rdfcore/xmlsch-02/Manifes
> t.rdf#whitespace-facet-4>
>   <#proofFound_NegativeEntailmentTest_RDF> <#FAIL>.
> <http://www.w3.org/2000/10/rdf-tests/rdfcore/xmlsch-02/Manifes
> t.rdf#whitespace-facet-3>
>   <#proofFound_PositiveEntailmentTest_RDF> <#PASS>.
> <http://www.w3.org/2000/10/rdf-tests/rdfcore/xmlsch-02/Manifes
> t.rdf#whitespace-facet-2>
>   <#proofFound_NegativeEntailmentTest_RDF> <#FAIL>.
> <http://www.w3.org/2000/10/rdf-tests/rdfcore/xmlsch-02/Manifes
> t.rdf#whitespace-facet-1>
>   <#proofFound_NegativeEntailmentTest_RDF> <#FAIL>.
> wether we use MS .NET, pnet, mono (or Xerces).
> 
> 
> I've done an attempt to stop this behaviour but gave up as
> it was becoming a mess in conjunction with the fact that
> in our impl the lexical value and/or the datatype uri could
> be variables in our implementation (e.g. " 33 "^^?D).
> I'm trusting those assemblies/libraries/jars and don't
> see a cost/benefit argument in alternatives.
> 
> 
> --
> Jos De Roo, AGFA http://www.agfa.com/w3c/jdroo/
> 
> 
>                                                               
>                                                               
>             
>                       <Patrick.Stickler@                      
>                                                               
>             
>                       nokia.com>                To:       
> <Patrick.Stickler@nokia.com>, <phayes@ihmc.us>, 
> <bwm@hplb.hpl.hp.com>         
>                       Sent by:                  cc:       
> <w3c-rdfcore-wg@w3.org>, <Patrick.Stickler@nokia.com>         
>                 
>                       w3c-rdfcore-wg-req        Subject:  RE: 
> pfps-06 hold off?                                             
>             
>                       uest@w3.org                             
>                                                               
>             
>                                                               
>                                                               
>             
>                                                               
>                                                               
>             
>                       2003-08-27 11:44                        
>                                                               
>             
>                       AM                                      
>                                                               
>             
>                                                               
>                                                               
>             
>                                                               
>                                                               
>             
> 
> 
> 
> 
> 
> 
> Just to make a particular point: Xerces, and similar tools
> applied to XML Schema simple datatypes, are intended for the
> interpretation of lexical forms in an *XML Context*.
> 
> If those tools are not suitable for interpretation of
> lexical forms in an *RDF Context* then they are simply
> not suitable, period, end of story, game over, move on.
> 
> That's not to say that I am not sympathetic to RDF implementors
> looking for the most straightforward way to add support for
> XML Schema datatyped literals, but using the shortcomings
> of an XML tool when applied in a context it was never intended
> to be used in as a justification for bastardizing our design
> is completely unacceptable to me.
> 
> It's the tail wagging the dog, big time.
> 
> Both the XML Schema specs and our present drafts are crystal
> clear about what the lexical space of XML Schema simple types
> are and what the L2V mapping involves in an RDF context.
> 
> Unless someone presents me with overwhelming justification,
> I will strongly oppose the inclusion of whitespace processing as
> a part of the formal L2V mapping for RDF datatyping.
> 
> That said, I really don't understand the problem some folks
> have with the MAY we introduced into the latest draft. We're
> not saying that a lexical form is valid if it requires whitespace
> processing, only that an implementation is free to recover
> from such *errors* gracefully if it feels that the lexical form
> can be safely coerced to a value dispite the illegal whitespace.
> 
> Maybe we need to state explicitly the fact that a lexical form
> requiring whitespace processing (or any preprocessing of any kind)
> is invalid, if some folks seem confused about that.
> 
> Or perhaps we simply need to "punt" the issue and remove all test
> cases involving XML Schema datatypes, so that implementations
> can pass all test cases yet still remain free to employ smoke
> and mirrors so they can waffle about with Xerces to process
> technically invalid but still usable lexical forms.
> 
> But adding whitespace processing to the formal L2V mapping
> just because Xerces is not RDF-savvy is just plain dumb.
> 
> [Apologies if my views offend anyone]
> 
> Patrick
> 
> 
> > -----Original Message-----
> > From: Stickler Patrick (NMP/Tampere)
> > Sent: 27 August, 2003 12:28
> > To: 'ext pat hayes'; Brian McBride
> > Cc: w3c-rdfcore-wg@w3.org; Stickler Patrick (NMP/Tampere)
> > Subject: RE: pfps-06 hold off?
> >
> >
> >
> >
> > > -----Original Message-----
> > > From: ext pat hayes [mailto:phayes@ihmc.us]
> > > Sent: 27 August, 2003 04:16
> > > To: Brian McBride
> > > Cc: w3c-rdfcore-wg@w3.org
> > > Subject: Re: pfps-06 hold off?
> > >
> > >
> > >
> > > >Pat,
> > > >
> > > >I'm wondering whether we should hold off your following up
> > > pfps on pfps-06 as:
> > > >
> > > >   1) the xml schema lex 2 val mapping may be about to change
> >
> > Honestly, Brian, I'm wondering how this could happen. We do
> > not define the XML Schema L2V mapping, and the XML Schema
> > specs are quite clear that the L2V mapping does *not* include
> > whitespace processing, so I remain very puzzled at your
> > suggestion that this could change.
> >
> > All that we could do ourselves would be to say that the RDF
> > L2V mapping, for XML Schema datatypes, includes the whitespace
> > processing, but such a position creates such blatant dependencies
> > and other nastiness in our design that simply thinking about
> > such a thing happening makes my ass start to twitch.
> >
> > Can you please, if possible, clarify what basis you have for
> > suggesting that the XML Schema L2V mapping might change, or
> > that the RDF L2V mapping would not be the same for XML Schema
> > datatypes as defined by XML Schema?
> >
> > The few comments that we have recieved from implementors regarding
> > the looseness of the Xerces implementation does not IMO even
> > begin to justify any such changes.
> >
> > Thanks,
> >
> > Patrick
> >
> 
> 
> 
> 
> 
> 
> 
Received on Wednesday, 27 August 2003 07:48:06 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:59:42 EDT