XMLWordPrintable

    Details

    • Proposal:
      Hide

      I have no idea how this is actually intended to work and where the applicable OWL axioms and facts are to be found. It would appear that the structure of a metadata.rdf file should be straigntforward since it is a manifest of rdf files, but any simplicity for that purpose is impossible to discern from this material. I am also concerned that

      1. The nomenclature between URIs as used in RDF and OWL are not clarified in the text of Part 3, and especially in the use of CURIEs to express the same URIs that arte actually used in the OWL file.

      2. It is disconcerting that the OWL file declares ODF-document related classes, not classes about the Package level of abstraction. This has me think that we have the wrong OWL file. That is a guess.

      3. Since the metadata.rdf file is presumably an RDF file with root element <rdf:RDF> it would seem desirable to be specific about some of this, what the conformance conditions on such a file are (and whether it is sufficient to claim conformance to the OWL-expressed ontology, and whether conformance to an ontology is enough to establish what would be a conformand metadata.rdf file.

      4. I hate to ask for an example, but I think a non-normative example of a metadata.rdf file for only the parts of a package that Part 3 establishes would be an useful exercise for demonstrating that this makes any sense at all, and also where it belongs among the ODF 1.2 parts.

      Show
      I have no idea how this is actually intended to work and where the applicable OWL axioms and facts are to be found. It would appear that the structure of a metadata.rdf file should be straigntforward since it is a manifest of rdf files, but any simplicity for that purpose is impossible to discern from this material. I am also concerned that 1. The nomenclature between URIs as used in RDF and OWL are not clarified in the text of Part 3, and especially in the use of CURIEs to express the same URIs that arte actually used in the OWL file. 2. It is disconcerting that the OWL file declares ODF-document related classes, not classes about the Package level of abstraction. This has me think that we have the wrong OWL file. That is a guess. 3. Since the metadata.rdf file is presumably an RDF file with root element <rdf:RDF> it would seem desirable to be specific about some of this, what the conformance conditions on such a file are (and whether it is sufficient to claim conformance to the OWL-expressed ontology, and whether conformance to an ontology is enough to establish what would be a conformand metadata.rdf file. 4. I hate to ask for an example, but I think a non-normative example of a metadata.rdf file for only the parts of a package that Part 3 establishes would be an useful exercise for demonstrating that this makes any sense at all, and also where it belongs among the ODF 1.2 parts.
    • Resolution:
      Hide

      Add pkg namespace to table in section 1.5 with the following description:

      OWL classes and properties contained in metadata manifest files.

      Show
      Add pkg namespace to table in section 1.5 with the following description: OWL classes and properties contained in metadata manifest files.

      Description

      1. The OWL RDF Metadata Manifest Ontology declares namespace prefix bindings via root <rdf:RDF> element declarations

      xmlns:pkg="http://docs.oasis-open.org/ns/office/1.2/meta/pkg#"
      xmlns:odf="http://docs.oasis-open.org/ns/office/1.2/meta/odf#"

      1.1 These are nowhere used in the Ontology file as prefixes in PrefixedName constructions.

      1.2 These are nowhere used in the value of RDF attributes as prefixes in Compact URI (CURIE) forms.

      1.3 There is one use of the same URI as that bound to prefix odl as an rdf:about attribute URI value.

      1.4 There is one use of the same URI as that bound to prefix pkg as an rdf:resource attribute URI value (apparently in error).

      1.5 There are occurrences of URIs that are concatenations of the odf-bound namespace URI that could also be expressed via the CURIE forms odf:ContentFile, odf:StylesFile, odf:Element, odf:prefix, and odf:suffix.

      1.6 There are occurrences of URIs that are concatenations of the pkg-bound namespace URI that could also be expressed via the CURIE forms pkg:File and pkg:Element.

      1.7 There is one occurrence of rdf:about="urn:oasis:names:tc:opendocument:xmlns:text:1.0meta-field" which could conceivably be expressed in CURIE form "text:meta-field" given xmlns:text="urn:oasis:names:tc:opendocument:xmlns:text:1.0", a representative binding for the Text Namespace in ODF 1.0, IS 26300, ODF 1.1, and the committee drafts of ODF 1.2 Part 1

      2. . Part 3 cd01 Section 5 is the only portion of this specification that refers to classes and properties defined for use in expressing content in metadata content files. It is asserted that a number of these are defined and there are brief sections about them.

      2.1 Taken as CURIE values, the following are NOT defined in the OWL ontology file or anything reachable from that file

      pkg:Document (this is used as an expanded CURIE, but not defined)
      pkg:File
      pkg:MetadataFile
      pkg:Element (this is used as an expanded CURIE, but not defined)
      pkg:hasPart
      pkg:mimeType

      2.2 Taken as CURIE values, The following are "declared" in the OWL ontology but not mentioned in the body of ODF 1.2 Part 3 cd01:

      odf:ContentFile
      odf:StylesFile
      odf:Element
      odf:prefix
      odf:suffix
      text:meta-field (by anology to the other cases)

      2.3 No namespace binding for pkg (nor for odf nor text) is mentioned in section 5 or in the section 1.5 or in the brief description in section 3.6.

      2.4 There is no description of these PrefixedNames as element or attribute names or as CURIEd URIs or as referencing the OWL classes and properties as they are described in the OWL Metadata Manifest Ontology. That is, there is nothing explicit in Part 3 that makes specific the intended associations among the terms (corrected as necessary) used in the section 6 and the declarations (corrected as necessry) in the OWL file.

        Attachments

          Activity

            People

            • Assignee:
              michael.brauer Michael Brauer (Inactive)
              Reporter:
              orcmid Dennis Hamilton (Inactive)
            • Watchers:
              1 Start watching this issue

              Dates

              • Created:
                Updated:
                Resolved: