[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [office] 2009-01-08 Single ODF 1.2 metadata proposal
Hi Dennis, many thanks for your feed-back. I am currently working on an minor update of the draft incorporating your prior mail. Before I sent an update, please let me answer this mail... Dennis E. Hamilton schrieb: > Svante, > > Thank you for this combined proposal. > > I have two concerns. > > First, I wonder whether we are making a simple problem hard. I appreciate > the care with which the metadata.rdf file is used to provide guidance to > external applications as well as ODF implementations to locate the separate > Zip items that add free-standing RDF (that is, external to the package items > that are subject-matter of the RDF) to the ODF package. But some aspects of > this seem rather complex. > It would be useful if there were a schema rather > than just the ontology (since I don't think that is a substitute for the > schema, operating at a different level of abstraction, so-to-speak). > > Indeed RDF and OWL is not easy at the beginning. RDF/XML is one serialization of RDF, the XML is not bound to a schema. For instance one RDF graph can be serialized in multiple manners, making it impossible to embrace in an XML schema. This is the reason XML techniques as XPath to not easily work on RDF/XML. But that problem was solved by RDF applications and their parsers. Perhaps it helps you that I will sent later this evening a bundle with an ODF test file (unzipped), the GRDDL XSLT file and expected output in RDF/XML and N3. This should be the first step to help the OIC in the attempt that ODF applications extract similar RDF graphs from an ODF document and give us a base for RDF regression tests to be shared. > Secondly, I am concerned that there is not enough information to understand > how exactly those inter-dependent aspects of RDF/XML, RDFa, OWL, XML 1.0, > xml:id 1.0, and the URI/IRI rfs that are pertinent to the metadata proposal > are profiled against the special requirements and conditions that apply in > the ODF document structure. > > I am not comfortable enough with that complex of interdependencies to > suggest that the difficulties are fatal in any way. I would hope that there > be some allowance for adjustments that may need to be made in the prose as > part of overall integration into a polished ODF 1.2 committee draft and in > careful review against the specifications that are referenced as somehow > related. > > Beside the interaction around XML 1.0, xml:id 1.0 and the use of ID > attributed types, which I will continue to address separately, here are two > examples of the kinds of thing that are part of this second concern: > > 1. The namespace prefix "xhtml" is assumed to be introduced but the > 09-01-18-MetaData-Change-Request.odt uses it without saying what the > namespace is (or I missed it). I assume that it is > "http://www.w3.org/1999/xhtml"; as mentioned in the "Adaption of W3C RDFa > standard" proposal, > <http://wiki.oasis-open.org/office/Change_Proposal_for_ODF_1.2_Metadata_-_Ad > aption_of_W3C_RDFa_standard>. > You seem to missed it. It is written in section 1.5.5 Adding new Namespaces (ODF Schema) > 1.1 I think it would have been better *not* to use "xhtml" as the > illustrative prefix in the ODF 1.2 specification, because we do not mean to > appeal to the entire XHML namespace. In that respect, assuming (for > illustrative purposes) something like > xmlns:rdfa="http://www.w3.org/1999/xhtml"; might be more useful in > emphasizing that we are using a very specific and limited profile of XHTML > markup (in fact, only a small number of attributes, nothing about XHTML > elements, roots, etc.). We are still using the XHTML namespace, of course, > but the use of a different prefix is useful in warning readers not to assume > too much. > Using the prefix means using XML nodes from a certain vocabulary / standard. Although the prefix is arbitrary, I am a friend of mnemonic prefixes and xhtml is the best prefix I can think of that URL. I do not see a problem here and would suggest to stay with the obvious prefix. > 1.2 A greater technical concern (rather than one about exposition) that I > have is that using the XHTML namespace, regardless of the prefix used, might > not accomplish the intended technical purpose and the effort required to > investigage that. I notice in the RDFa specification > (http://www.w3.org/TR/2008/REC-rdfa-syntax-20081014/) that most of the XHTML > attribute names are used without any namespace prefix. This suggests that > there is no namespace for those attributes, since the default namespace does > not apply to attributes in XML 1.0 (nor in XHTML 1.x). I thought maybe the > unexplained notation @content, @about, @property, @datatype, etc., might > signify some solution to this, but it is necessary to refer to the XHTML > specification, not just the RDFa specification for resolution of that. In > the XHTML Modularization W3C Recommendation 8 October 2008 (single file: > http://www.w3.org/TR/2008/REC-xhtml-modularization-20081008/xhtml-modulariza > tion.html), section 3.1(5), > <http://www.w3.org/TR/2008/REC-xhtml-modularization-20081008/xhtml-modulariz > ation.html#s_conform_document_type>, does state that those attribute names > are also defined as local names of the XHTML namespace although the XHTML > Modularization specification is inapplicable to our use of the > RDFa-augmented XHTML attributes. In the XHTML 1.0 specification > (http://www.w3.org/TR/2002/REC-xhtml1-20020801/) there is no mention of > attributes with regard to the use of XHTML namespaces (i.e., in section > 3.1.2), and there is no indication that the no-namespace attribute names of > XHTML are also local names of the XHTML namespace. > Steven Pemberton the chair of the W3C XHTML WG suggested to me this approach during the TPAC in Cannes last year. "Each of the attributes defined in an XHTML attribute collection is available for use when their corresponding module is included in an XHTML Host Language or an XHTML Integration Set. In such a situation, the attributes are available for use in the definition of elements that are NOT in the XHTML namespace when they are referenced using their namespace-qualified identifier (e.g., xhtml:class). The semantics of the attributes remain the same regardless of whether they are referenced using their qualified identifier or not. It is an error to use an XHTML namespace-qualified attribute on elements from the XHTML Namespace." http://www.w3.org/TR/xhtml-modularization/abstract_modules.html#s_commonatts And RDFa defines an attribute collection: "9.2. Metainformation Attributes Collection The following attributes are included in the attribute collection, and take values in the associated datatype:" http://www.w3.org/TR/rdfa-syntax/#col_Metainformation Hope this helps! > 1.3 BOTTOM LINE: "@" is apparently an undocumented short-hand to refer to > an attribute name in the RDFa specification (it is not used in the XHTML > specification or the XHTML Modularization specification) It is an XPath shortcut: http://www.w3.org/TR/xpath#path-abbrev > and the only > suggestion that the XHTML namespace contains the XHTML attribute names as > local names with the significance they have as (extended) attributes in > XHMTL+RDFa is the statement at 3.1(5) in the Modularization specification. > Although the XHTML Namespace URL has a resource that can be dereferenced, it > is no help. I have not checked the other links in that document to see if > there is anything more specific about this case. We should not be made to > work so hard to be able to make careful review of the RDF metadata material > in the ODF specification. > As I said, I already counter-checked this with the W3C and give some note about the namespace in my first mail to the TC: http://lists.oasis-open.org/archives/office/200811/msg00006.html If you believe that is not sufficient, I am happy to give you any information you desire. > 2. The idea that the ODF Document must have a physically-resolvable URI goes > too far, especially when the example is of a location on the hard drive of a > computer for which there is no web-resolvable URI in evidence. > I believe you think that if you make an URL based on your hard disc would not work. Someone always has to make URLs in the context of his 'network' / 'world' / 'domain'. This might be a computer, a LAN, the Internet or something else. If I identify all documents on my computer, it is fine as long I make statements in this domain and do not leave it (mix it with other computer domain). In a similar manner Internet index are being created and they work fine. > 2.1 The difference between (copies of) document instances and some > understood URI that can be used to refer to features of any such (unaltered > copy of an) instance needs to be allowed for (and this is often recognized > in RDF and demonstrated in W3C RDFa examples, especially but not limited to > the reliance on Dublin Core concepts). Examples of such features are an ISBN > number, the name of the author, bibliographic information about it, > bibliographic information about something mentioned or discussed in it, > assertions about what someone says about it, assertions about what is being > said in the document about what someone says about something, such as in a > book review or literary critique, etc. > Perhaps you are confusing the need of an URL (which comes along with the persistence of the format (office document), remember ODF is the persistent file format of Office applications) with the first appearance of a lack of an URN. Please note that it is still possible to provide an URN in addition with the URL (using owl:equal) URL are the base of a web and that's what the Semantic Web should be about. > 2.2 The use of URIs as concept identifiers that are never meant to be > resolved is critical in the working of RDF and is the underpinning of the > identification of relations and predicates and properties in RDF. In many > respects, the document instance is a different entity than a particular > location where a copy of the instance might be located in a physical > repository or on the Internet. It seems clear to me that these prevalent > uses of RDF (illustrate in examples of RDFa) should not be constrained for > RDF annotation of ODF documents. > Yes, that is what we (me included) had previously thought so. The creators of RDF, the Semantic Web Interest (SWI) Group, Dan Brickley (W3C SWI chair) and especially Tim Berners Lee, who took an hour to convince me after my presentation, that I was wrong. Convinced me of something different. Again in a certain part you are right, but you are completely wrong in ignoring the advantage of an URL. If someone is annotating (adding metadata) to a network resource, the identifier should be the location to make the web possible. I would suggest we have a call or I get you together with some RDF guys. Think it is a little late to challenge the basics, unfortunately you were not part of the Metadata SC. > 2.3. The inter-dependence between the use of CURIEs and the rules for > URIs in ODF packages need to be explained very carefully. I am assuming > that the ODF 1.2 package rule (section 2.6) for interpreting a URI does > apply after any CURIE expansion to a URI takes place, especially when the > expansion is to a relative URI. This interdependence does need to be > accounted for, however. The interdependence also needs to work properly > where, via the CURIE mechanism or not, the resulting URI has a lead string > (or Base URL) that signifies the package but is not an absolute URL that can > be used to access the package part and fragment being mentioned. On > investigation into RDFa, I notice that reliance on xhtml:base attributes > could have been very useful (and would also need to be reconciled in Package > section 2.6). There are some open questions in all of this, and I trust > there will be an opportunity to resolve them. > The mechanism defined in the schema, section 1.2.3 New RDFa compact URI (CURIE) datatype. Saying CURIE are being overtaken from the RDFa standard for the xhtml attributes. I do not understand the problem you draft here, but if you believe something is left unsaid, please provide a wording or correction. Thank you for taking the time for a such deep review. Regards, Svante > - Dennis > > -----Original Message----- > From: Svante.Schubert@Sun.COM [mailto:Svante.Schubert@Sun.COM] > http://lists.oasis-open.org/archives/office/200901/msg00064.html > Sent: Thursday, January 08, 2009 10:16 > To: OASIS Office > Cc: Svante Schubert > Subject: Re: [office] Single ODF 1.2 metadata proposal > > Dear TC members, > > Obviously I had skipped one previous revision, but uploaded now a > correction [4]. I apologize for any trouble caused. > > As quick summary of this proposal, which is only an enhancement of the > already agreed RDF based metadata proposal. > > It was triggered from the feedback of the W3C: > 1) we overtook the now existing RDFa standard. Enable us to use compact > URIs, which for example will save us a lot of space in large spreadsheets. > 2) we provide a GRDDL file to make our metadata easier extractable for > none ODF application (e.g. RDF crawlers) > 3) we dropped boilerplate in connecting the ODF XML world with the RDF > world by simply using relative URLs, making resources behind the IRIs > accessible (like in a web) and better human readable. > > Therefore this proposal is pretty much straight forward and I feel > really delighted that we have made this round-trip of refactoring. > > PS: BTW XMP is based on RDF and can refer to any ODF resources described > by the new RDF framework. There is no problem. > > Regards, > Svante > > [4] > http://www.oasis-open.org/committees/download.php/30585/09-01-08-ODF-Metadat > a-Change-Request.odt > > Svante Schubert wrote: > >> Three further remarks: >> >> 1) I have created and uploaded a W3C GRDDL file (the XSL stylesheet to >> create RDF/XML from RDFa and ODF elements from the meta.xml) [1] >> 2) The namespaces in the proposal have been adapted to the OASIS >> policies pointed out by Mary [2] >> 3) During work on the GRDDL stylesheet it came clear that the RDFa >> properties xhtml:content and xhtml:datatype are exclusive. >> This and the namespaces have been fixed in an updated draft [3]. >> >> Regards, >> Svante >> >> [1] >> >> > http://www.oasis-open.org/apps/org/workgroup/office-metadata/download.php/30 > 575/odf2rdf.xsl > >> [2] http://lists.oasis-open.org/archives/office/200812/msg00178.html >> [3] >> >> > http://www.oasis-open.org/apps/org/workgroup/office-metadata/download.php/30 > 574/09-01-07-ODF-Metadata-Change-Request.odt > >> Svante Schubert wrote: >> >>> Dear TC members, >>> >>> I would like to ask the TC to vote on this proposal on the next TC >>> call. If you have any further questions regarding the proposal, >>> please send them to me prior to the call. >>> >>> Regards, >>> Svante >>> >>> Svante Schubert wrote: >>> >>>> Dear TC members, >>>> >>>> from my feedback of the W3C Semantic Web Interest group I would like >>>> to combine the previous three existing proposal parts to a single >>>> metadata proposal, consisting of three parts: >>>> >>>> Adaption of W3C RDFa standard >>>> >>>> > <http://wiki.oasis-open.org/office/Change_Proposal_for_ODF_1.2_Metadata_-_Ad > aption_of_W3C_RDFa_standard> > >>>> Previously RDFa was not a W3C Recommendation and could not be reused >>>> in our spec, now we are able to adapt the official standard. >>>> >>>> Usage of W3C GRDDL standard >>>> >>>> > <http://wiki.oasis-open.org/office/Change_Proposal_for_ODF_1.2_Metadata_-_Us > age_of_W3C_GRDDL_standard> > >>>> Meant for non ODF applications to extract the RDF graph from our >>>> content.xml, styles.xml and meta.xml streams via a referenced XSL >>>> stylesheet. >>>> Similar adapted by XHTML files using RDFa standard. The location of >>>> the XSL stylesheet is resolved via the URL of the root namespace >>>> (ie. XHTML use case) or via a GRDDL attribute in the root element >>>> (i.e. ODF use case as namespace is an URN and no URL). >>>> Therefore the only effort for ODF applications is a new root element >>>> attribute, the effect enormous. >>>> (After proposal acceptance I would create the XSL stylesheets and >>>> place it on an OASIS server). >>>> >>>> Adaption of RDF mapping to relative URLs >>>> >>>> > <http://wiki.oasis-open.org/office/Change_Proposal_for_ODF_1.2_Metadata_-_Ad > aption_of_RDF_mapping_to_relative_URLs> > >>>> Currently in the new RDF mapping file (manifest.rdf) an arbitrary >>>> URN is being invented for every ODF element described in the RDF >>>> graph. But this creates troublesome URL alias. Instead the standard >>>> will require that a resource (ODF element) will be mapped directly >>>> to its relative URL. >>>> >>>> >>>> Currently I am finalizing the overall proposal. In case I will be >>>> ready this evening, I would set it on the agenda of the next TC call. >>>> >>>> PS: The previous announced ODF 1.2 proposal of fragment identifier >>>> for URLs to ODF mime types is being withdrawn by myself, due to the >>>> amount of proposals that have been requested to be considered for >>>> ODF 1.2. >>>> >>>> Regards, >>>> Svante >>>> -- >>>> Sun Microsystems GmbH Svante Schubert >>>> Nagelsweg 55 Software Engineer >>>> 20097 Hamburg StarOffice / OpenOffice.org Development >>>> Germany Phone: +49(0)40 236 46 500 >>>> http://www.sun.com Svante.Schubert@sun.com >>>> >>>> Sitz der Gesellschaft: >>>> Sun Microsystems GmbH, Sonnenallee 1, D-85551 Kirchheim-Heimstetten >>>> Amtsgericht Muenchen: HRB 161028 >>>> Geschaeftsfuehrer: Wolfgang Engels, Dr. Roland Boemer >>>> Vorsitzender des Aufsichtsrates: Martin Haering >>>> > > --------------------------------------------------------------------- > To unsubscribe from this mail list, you must leave the OASIS TC that > generates this mail. Follow this link to all your TCs in OASIS at: > https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php > >
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]