Gleaning Resource Descriptions from Dialects of Languages (GRDDL)

Abstract

GRDDL is a mechanism for Gleaning Resource Descriptions from Dialects of Languages. This GRDDL specification introduces markup based on existing standards for declaring that an XML document includes data compatible with the Resource Description Framework (RDF) and for linking to algorithms (typically represented in XSLT), for extracting this data from the document.

The markup includes a namespace-qualified attribute for use in general-purpose XML documents and a profile-qualified link relationship for use in valid XHTML documents. The GRDDL mechanism also allows an XML namespace document (or XHTML profile document) to declare that every document associated with that namespace (or profile) includes gleanable data and for linking to an algorithm for gleaning the data.

A corresponding GRDDL Use Case Working Draft provides motivating examples. A GRDDL Primer demonstrates the mechanism on XHTML documents which include widely-deployed dialects known as microformats. A GRDDL Test Cases document illustrates specific issues in this design and provides materials to aid in test-driven development of GRDDL-aware agents.

Status of This Document

This section describes the status of this document at the time of its publication. Other documents may supersede this document. A list of current W3C publications and the latest revision of this technical report can be found in the W3C technical reports index at http://www.w3.org/TR/.

This is a W3C Recommendation.

This document has been reviewed by W3C Members, by software developers, and by other W3C groups and interested parties, and is endorsed by the Director as a W3C Recommendation. It is a stable document and may be used as reference material or cited from another document. W3C's role in making the Recommendation is to draw attention to the specification and to promote its widespread deployment. This enhances the functionality and interoperability of the Web.

Comments on this document should be sent to public-grddl-comments@w3.org, a mailing list with a public archive.

This document was produced by GRDDL Working Group, which is part of the W3C Semantic Web Activity. The first release of this document as a Working Draft was 24 Oct 2006 and the Working Group has addressed a number of comments received and issues since then. Normative assertions are marked up in this way.

The Working Group's implementation report demonstrates that the goals for interoperable implementations, set in the May 2007 Candidate Recommendation draft of this document, were achieved.

GRDDL is intended to contribute to addressing Web Architecture issues such as RDFinXHTML-35, namespaceDocument-8, and xmlFunctions-34 as well as issues postponed by the RDF Core working group such as rdfms-validating-embedded-rdf and faq-html-compliance. In particular, the GRDDL Working Group has postponed issue-faithful-infoset, and anticipates that the resolution of TAG issue xmlFunctions-34 will provide further clarification and guidance.

This document was produced by a group operating under the 5 February 2004 W3C Patent Policy. W3C maintains a public list of any patent disclosures made in connection with the deliverables of the group; that page also includes instructions for disclosing a patent. An individual who has actual knowledge of a patent which the individual believes contains Essential Claim(s) must disclose the information in accordance with section 6 of the W3C Patent Policy.

The issues appendix that used to be part of this draft has been moved to a Working Group issues list; specifically: issue-whichlangs, issue-output-formats, issue-base-param, issue-tx-element, issue-html-nsdoc, issue-faithful-infoset, issue-mt-ns, issue-conformance-labels, issue-http-header-links

Table of Contents

1. Introduction
2. Adding GRDDL to well-formed XML
3. GRDDL for XML Namespaces
4. Using GRDDL with valid XHTML
5. GRDDL for HTML Profiles
6. GRDDL Transformations
7. GRDDL-Aware Agents
8. Security Considerations
9. The GRDDL Vocabulary
10. References

• Appendix: Transformations for Styling versus data extraction
• Appendix: Base IRI considerations
• Appendix: Acknowledgements and Change History

• Appendix: About the Mechanical Rules

1. Introduction: Data and Documents

There are many domain-specific languages ("dialects") used in practice among the many XML documents on the web. There are dialects of XHTML, XML and RDF that are used to represent everything from poetry to prose, purchase orders to invoices, spreadsheets to databases, schemas to scripts, and linked lists to ontologies.

While this breadth of expression is quite liberating, inspiring new dialects to represent information, it can be a barrier to understanding across different domains or fields. How, for example, does software discover the author of a poem, a spreadsheet and an ontology? And how can software determine whether authors of each are in fact the same?

The following are examples of how the same musical work might be described in different XML dialects:

iTunes Music Library

<key>Artist</key>
  <string>The Jimi Hendrix Experience</string>
<key>Album</key>
  <string>Are You Experienced?</string>

Audioscrobbler

<album>
    <artist mbid="">The Jimi Hendrix Experience</artist>
    <name>Are You Experienced?</name>
...
</album>

Atom

<entry ... >
<title>Are You Experienced?</title>
<author>
<name>The Jimi Hendrix Experience</name>
</author>
...
</entry>

Open Office

<office:document-meta ... >
<office:meta>
<dc:title>Are You Experienced?</dc:title>
  <meta:initial-creator>
  The Jimi Hendrix Experience
  </meta:initial-creator>
<dc:creator>The Jimi Hendrix Experience</dc:creator>
</office:meta>
</office:document-meta>

Although the examples above are obviously encodings of the same information, there remains no clear mechanism through which computer software might be able to determine this connection.

Resource Descriptions

The Resource Description Framework[RDFC04] provides a standard for making statements about resources in the form of a subject-predicate-object expression. One way to represent the fact "Are You Experienced?'s artist is The Jimi Hendrix Experience" in RDF would be as a triple whose subject is Are You Experienced, whose predicate is "has artist," and whose object is The Jimi Hendrix Experience. The predicate, "has artist" expresses a relationship between the subject (Are You Experienced?) and the object (The Jimi Hendrix Experience). Using URIs to uniquely identify the album, the artist and even the relationship would facilitate software design because not everyone knows The Jimi Hendrix Experience or even spells its name consistently.

Here's the information contained in the XML fragments above, this time expressed as RDF:

<rdf:RDF
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:dc="http://purl.org/dc/elements/1.1/">

  <rdf:Description rdf:about=
"http://musicbrainz.org/mm-2.1/album/6b050dcf-7ab1-456d-9e1b-c3c41c18eed2">
    <dc:title>Are You Experienced?</dc:title>
    <foaf:maker>
      <foaf:Agent rdf:about=
  "http://musicbrainz.org/mm-2.1/artist/33b3c323-77c2-417c-a5b4-af7e6a111cc9">
        <foaf:name>The Jimi Hendrix Experience</foaf:name>
      </foaf:Agent>
    </foaf:maker>

  </rdf:Description>
</rdf:RDF>

Both the entities (subject and object resources) and relationships (predicates) are identified using unambiguous URIs.

Note that GRDDL follows HTML 4, RDF, and XML Schema in using Internationalized Resource Identifiers, i.e. IRIs[RFC3987]. While in informal usage, this specification uses the more familiar term URI interchangeably with the recently standardized term IRI, the formal rules use the relevant terms precisely.

The publishers of the XML above could also provide the same data in RDF using RDF/XML or one of the other RDF syntaxes. GRDDL provides a relatively inexpensive mechanism for bootstrapping RDF content from uniform XML dialects, shifting the burden from formulating RDF to creating transformation algorithms specifically for each dialect.

GRDDL works by associating transformations for an individual document, either through direct inclusion of references or indirectly through profile and namespace documents. Content authors can nominate the transformations for producing RDF from their content and use GRDDL to refer to them.

2. Adding GRDDL to well-formed XML

The general form of associating a GRDDL transformation link with a well-formed XML document is adding to the root element a grddl namespace declaration and a grddl:transformation attribute whose value is an IRI reference, or list of IRI references, that refer to executable scripts or programs which are expected to transform the source document into RDF. This method is suitable for use with any XML dialects that can accomodate an extra namespace-qualified attribute on the root element.

For example, this XML document, located at http://www.w3.org/2001/sw/grddl-wg/td/titleauthor.html, is linked to two GRDDL transformations:

<html xmlns="http://www.w3.org/1999/xhtml"
      xmlns:grddl='http://www.w3.org/2003/g/data-view#'
      grddl:transformation="glean_title.xsl
			http://www.w3.org/2001/sw/grddl-wg/td/getAuthor.xsl"
 >
<head>
<title>Are You Experienced?</title>
[...]
</html>

1. It is linked to the transformation identified by http://www.w3.org/2001/sw/grddl-wg/td/getAuthor.xsl.
2. To resolve the relative URI reference glean_title.xsl to absolute form, we use the base URI of this XML element, http://www.w3.org/2001/sw/grddl-wg/td/titleauthor.html. Then this document is also linked to the GRDDL transformation identified by the absolute form, http://www.w3.org/2001/sw/grddl-wg/td/glean_title.xsl.

extracting title and author information

(svg)

As you will see in later sections, there are other ways to add GRDDL to HTML documents, especially designed to leverage HTML's existing capabilities and thereby overcome constraints imposed by the XML DTDs for some dialects of HTML. See Using GRDDL with valid XHTML and GRDDL for HTML Profiles.

The formal specification of this markup is given below. An informative mechanical version of each rule is given with the premise and the conclusion written as SPARQL graph patterns[SPARQL]. See the Mechanical Rules appendix for namespace prefix bindings and further explanation. These are included for those readers who find them helpful. Other readers are encouraged to ignore them.

/*/@*[local-name()="transformation"
  and namespace-uri()=
    "http://www.w3.org/2003/g/data-view#"]

(?N "/*") gspec:xpath ?E.
(?N """/*/@*[local-name()="transformation" and
    namespace-uri()=
    "http://www.w3.org/2003/g/data-view#"]""")
   gspec:xpath [ fn:string ?V].
?V fn:normalize-space ?Vnorm.
(?Vnorm "[ \t\r\n]+") fn:tokenize [
  list:member ?REF ].
?E fn:base-uri ?BASE.
(?REF ?BASE) fn:resolve-uri ?TXURI.
?TX log:uri ?TXURI.

?N grddl:transformation ?TX.

The glean_title.xsl transformation computes the following RDF/XML document, given the XML document above as input:

<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
   xmlns:dc="http://purl.org/dc/elements/1.1/">
  <rdf:Description rdf:about="">
    <dc:title>Are You Experienced?</dc:title>
  </rdf:Description>
</rdf:RDF>

The graph serialized by that document is a GRDDL result of the resource identified by http://www.w3.org/2001/sw/grddl-wg/td/titleauthor.html. Note that this serialization of the graph contains a relative URI reference (in the value of the rdf:about attribute). The base IRI for interpretting relative IRI references in a serialization of a graph produced by a GRDDL transformation is the base IRI of the source document.

The glean_title.xsl resource specifies a function from XPath document nodes to RDF/XML documents, and hence to RDF graphs; this function is called the transformation property of the XSLT document. See the GRDDL Transformations section for more details.

The general rule for using GRDDL with well-formed XML is:

?IR log:uri [ fn:doc ?R ].
?R grddl:transformation [ grddl:transformationProperty ?TP ].
?R ?TP ?G.

?IR grddl:result ?G .

The titleauthor.html resource has another GRDDL result via the getAuthor.xsl transformation. These results can be merged together into another result, by this rule:

?IR grddl:result ?F, ?G.
(?F ?G) log:conjunction ?H.

?IR grddl:result ?H.

3. Using GRDDL with XML Namespace Documents

Transformations can be associated not only with individual documents but also with whole dialects that share an XML namespace. Any resource available for retrieval from a namespace URI is a namespace document (cf. section 4.5.4. Namespace documents in [WEBARCH]). For example, a namespace document may have an XML Schema representation or an RDF Schema representation, or perhaps both, using content negotiation.

To associate a GRDDL transformation with a whole dialect, include a grddl:namespaceTransformation property in a GRDDL result of the namespace document.

For example, consider this privacy policy written in P3Q, a contrived analog to P3P[P3P]:

<POLICIES xmlns="http://www.w3.org/2004/01/rdxh/p3q-ns-example">
	<EXPIRY max-age="604800"/>
...

The namespace document for P3Q relates the grokP3Q.xsl transformation to all P3Q documents:

<rdf:RDF
  xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
  xmlns:dataview="http://www.w3.org/2003/g/data-view#">
 <rdf:Description rdf:about="http://www.w3.org/2004/01/rdxh/p3q-ns-example">
   <dataview:namespaceTransformation
       rdf:resource="http://www.w3.org/2004/01/rdxh/grokP3Q.xsl"/>
 </rdf:Description>
</rdf:RDF>

That is: every document whose root namespace name is ...p3q-ns-example has grokP3Q.xsl as a GRDDL transformation implicitly, as illustrated in this figure:

transformation applied to namespace
(svg)

Some namespace documents, such as the XHTML namespace document http://www.w3.org/1999/xhtml have very many references to them. If GRDDL-aware agents were to retrieve these documents every time they processed a document referring to them, the origin servers of those documents could become overloaded. GRDDL-aware agents therefore should not retrieve such documents on every reference and should retain some cache or local memory of the transformations those documents indicate should be applied. To avoid misrepresentation of published information, GRDDL-aware agents should ensure that this local memory is up to date and should support user options to configure or disable the cache. See also section section 3.1. Using a URI to Access a Resource of [WEBARCH].

The general case of namespace transformations is:

?NSDOC log:uri ?NS;
   grddl:result [
     log:includes [
       rdf:subject ?NSDOC;
       rdf:predicate grddl:namespaceTransformation;
       rdf:object ?TX]].
?IR log:uri [ fn:doc ?NODE].
(?NODE "/*") gspec:xpath ?E.
?E fn:namespace-uri ?NS.

?NODE grddl:transformation ?TX.

Note that as a base case, the result of parsing an RDF/XML document is a GRDDL result of that document:

?IR log:uri [ fn:doc [ gspec:rdfParse ?G ] ].

?IR grddl:result ?G.

Note that while an application/rdf+xml media type is one indication that a document is RDF/XML, section 7.2.1 Grammar start of [RDFX] leaves open "other means" by which an RDF/XML document may be identified. For the purposes of the rule above, a root element whose local name is RDF and whose namespace URI is http://www.w3.org/1999/02/22-rdf-syntax-ns# is such a means. For a case in point, see the grddlonrdf-xmlmediatype test case.

Example: Using GRDDL with an XML Schema namespace document

A namespace transformation link may be discoverable by transforming the namespace document itself. Note that this means that namespace documents need not be written in RDF/XML directly.

Consider a purchase order that has a namespace document represented in XML Schema, where the XML Schema bears a data-view:transformation attribute licensing extraction of statements that include namespaceTransformation statements:

<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema"
            xmlns="http:.../Order-1.0"
            targetNamespace="http:.../Order-1.0"
            version="1.0"
            ...
            xmlns:data-view="http://www.w3.org/2003/g/data-view#"
            data-view:transformation="http://www.w3.org/2003/g/embeddedRDF.xsl" >
    <xsd:element name="Order" type="OrderType">
    <xsd:annotation 
      <xsd:documentation>This element is the root element.</xsd:documentation>
    </xsd:annotation>
                 ...
  <xsd:annotation>
    <xsd:appinfo>
      <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
	<rdf:Description rdf:about="http://www.w3.org/2003/g/po-ex">
	  <data-view:namespaceTransformation
	      rdf:resource="grokPO.xsl" />
	</rdf:Description>
      </rdf:RDF>
    </xsd:appinfo>
  </xsd:annotation>
...

Every purchase order using that schema as a namespace document is linked to the grokPO.xsl transformation, as illustrated below:

using GRDDL with an XML Schema

(svg)

4. Using GRDDL with valid XHTML

To accomodate the DTD-based syntax of XHTML[XHTML], which precludes using attributes from foreign namespaces, we use http://www.w3.org/2003/g/data-view as a metadata profile (cf. section 7.4.4.3 Meta data profiles of [HTML4]).

The general form of adding a GRDDL assertion to a valid XHTML document is by specifying the GRDDL profile in the profile attribute of the head element, and transformation as the value of the rel attribute of a link or a element whose href attribute value is an IRI reference that refers to an executable script or program which is expected to transform the source document into RDF. This method is suitable for use with valid XHTML documents which are constrained by an XML DTD.

An example Dublin Core META transformation

For example, this document follows the conventions of [RFC2731], and it explicitly uses the GRDDL profile and links to an XSLT transformation to RDF/XML to signal that the transformation is a faithful rendition:

<html xmlns="http://www.w3.org/1999/xhtml">
  <head profile="http://www.w3.org/2003/g/data-view">
    <title>Some Document</title>

    <link rel="transformation"
       href="http://www.w3.org/2000/06/dc-extract/dc-extract.xsl" />
    <meta name="DC.Subject"
       content="ADAM; Simple Search; Index+; prototype" />
    ...
  </head>
  ...
</html>

The figure below shows the source document, the dc-extract.xsl transformation, and the GRDDL result:

Decoding HTML meta-data to RDF

(svg)

This is what the data looks like in RDF/XML:

<rdf:RDF xmlns:dc="http://purl.org/dc/elements/1.1/"
         xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
  <rdf:Description rdf:about="">
    <dc:subject>ADAM; Simple Search; Index+; prototype</dc:subject>
  </rdf:Description>
</rdf:RDF>

Multiple transformations in XHTML

An XHTML document may conform to a number of dialects simultaneously and link to more than one GRDDL transformation. However, since the href attribute of the link and a elements accept only a single IRI reference, multiple instances of these elements must be used to assert multiple links:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
       "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head profile="http://www.w3.org/2003/g/data-view">
  <title>Joe Lambda's Home page [an example of RDF in XHTML]</title>

  <link rel="transformation" href="http://www.w3.org/2003/12/rdf-in-xhtml-xslts/grokFOAF.xsl" />
  <link rel="transformation" href="http://www.w3.org/2003/12/rdf-in-xhtml-xslts/grokCC.xsl" />
  <link rel="transformation" href="http://www.w3.org/2003/12/rdf-in-xhtml-xslts/grokGeoURL.xsl" />
...

multiple transformations

(svg)

?N gspec:profileName "http://www.w3.org/2003/g/data-view".
(?N
""".//*[namespace-uri()="http://www.w3.org/1999/xhtml" and
        (local-name() = "a"
         or local-name() = "link")"""
) gspec:xpath ?E.
(?E "@rel") gspec:xpath [ fn:string [
   fn:normalize-space ?E_REL ]].
(?E_REL "[ \t\r\n]+") fn:tokenize [
 list:member "transformation" ].
(?E "@href") gspec:xpath [ fn:string ?T_REF ].
?E gspec:htmlBase ?BASE.
(?T_REF ?BASE) fn:resolve-uri ?TURI.
?T log:uri ?TURI.

?N grddl:transformation ?T.

(?N
 """
*[local-name()="html" and
  namespace-uri()="http://www.w3.org/1999/xhtml"] /
 *[local-name()="head" and
   namespace-uri()="http://www.w3.org/1999/xhtml"]""")
 gspec:xpath ?E.
(?E "@profile") gspec:xpath [ fn:string ?V ].
?E fn:base-uri ?BASE.
?V fn:normalize-space ?Vnorm.
(?Vnorm "[ \t\r\n]+") fn:tokenize [  list:member ?P_REF ].
(?P_REF ?BASE) fn:resolve-uri ?PROFID.

?N gspec:profileName ?PROFID.

5. GRDDL for HTML Profiles

XHTML provides the profile mechanism to link to the meaning of properties and the set of legal values for those properties. As with namespace documents, a profile document can effectively be written using XHTML with embedded RDF statements and a GRDDL transformation to extract the definition of terms that are applicable. Those terms can then be used in an XHTML document to convey profile-dependent meaning. As discussed in Using GRDDL with valid XHTML, the GRDDL profile can be used with XHTML documents to apply GRDDL semantics over link elements where the value of rel attribute is transformation. This very powerful and flexible mechanism integrates well with microformat profiles[MF-RDF-FAQ] which overlay the normally semantically-poor HTML markup.

The following diagram illustrates an XFN document[XFN], friends.html associated with the grokXFN.xsl transformation indirectly via an XFN profile.

indirection via profile

(svg)

Adding a GRDDL profileTransformation assertion to a profile document is much like adding a namespaceTransformation assertion to a namespace document. For a dialect defined by a valid XHTML profile documents, add profile="http://www.w3.org/2003/g/data-view" to the head element and make a link of type profileTransformation to the transformation of the dialect.

The general rule is:

?PDOC log:uri ?PNAME;
   grddl:result [
     log:includes [
       rdf:subject ?PDOC;
       rdf:predicate grddl:profileTransformation;
       rdf:object ?TX]].
?IR log:uri [ fn:doc ?NODE].
?NODE gspec:profileName ?PNAME.

?NODE grddl:transformation ?TX.

6. GRDDL Transformations

As noted above, each GRDDL transformation specifies a transformation property, a function from XPath document nodes to RDF graphs. This function need not be total; it may have a domain smaller than all XML document nodes. For example, use of xsl:message with terminate="yes" may be used to signal that the input is outside the domain of the transformation.

Developers of transformations should make available representations in widely-supported formats. XSLT version 1[XSLT1] is the format most widely supported by GRDDL-aware agents as of this writing, though though XSLT2[XSLT2] deployment is increasing. While technically Javascript, C, or virtually any other programming language may be used to express transformations for GRDDL, XSLT is specifically designed to express XML to XML transformations and has some good safety characteristics; XQuery has similar characteristics to XSLT, though use of XQuery in GRDDL implementation is less widely deployed at the time of this writing.

?RDFXML gspec:rdfParse ?G.
(?TXNODE ?R) gspec:resultTree ?RDFXML.
?TXDOC grddl:transformationProperty ?TP;
  log:uri [fn:doc ?TXNODE].

?R ?TP ?G

The rule above covers the case of a transformation property that relates an XPath document node to an RDF graph via an RDF/XML document. Transformations may use other, unspecified, mechanisms. For example, see test #atomttl1, in which the the media-type attribute of the xsl:output element bears a "text/rdf+n3" value to indicate a media type other than "application/rdf+xml". GRDDL agents that can process such a media type can then produce an RDF graph in accordance with the media type. Non-XSLT transforms may indicate the RDF graph in some other, unspecified, fashion.

7. GRDDL-Aware Agents

A GRDDL-aware agent is a software module that computes GRDDL results of information resources.

For example, a SPARQL query service might use a GRDDL-aware agent for collecting RDF data. Or a Web browser might serve as a GRDDL-aware agent for the purpose of collecting calendar and contact data. The appropriate policy, for which results to compute and when, is likely to involve waiting for a signal from user more in the Web browser case than in the query service case.

Note that discovery by namespace or profile document is recursive; Loops in the profile/namespace structure should be detected in order to avoid infinite recursion.

[00:00.000 - client connection from 127.0.0.1:39645]
GET http://www.w3.org/2003/g/po-doc.xml HTTP/1.1
Host: www.w3.org
Accept: application/rdf+xml,application/xml,text/xml,application/xhtml+xml,text/html

[00:00.055 - server connected]
HTTP/1.1 200 OK
Last-Modified: Tue, 07 Dec 2004 22:59:02 GMT
Content-Length: 1302
Content-Type: application/xml; qs=0.9

<purchaseOrder orderDate="1999-10-20"
   xmlns="http://www.w3.org/2003/g/po-ex">
   <shipTo country="US">
      <name>Alice Smith</name>
      <street>123 Maple Street</street>
...

[00:00.000 - client connection from 127.0.0.1:39647]
GET http://www.w3.org/2003/g/po-ex HTTP/1.1
Host: www.w3.org
Accept: application/rdf+xml,application/xml,text/xml,application/xhtml+xml,text/html

[00:00.051 - server connected]
HTTP/1.1 200 OK
Content-Location: po-ex.xsd
Last-Modified: Tue, 07 Dec 2004 23:18:25 GMT
Content-Length: 2624
Content-Type: application/xml; qs=0.9

<xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
        xmlns:po="http://www.w3.org/2003/g/po-ex"
        targetNamespace="http://www.w3.org/2003/g/po-ex"
        elementFormDefault="qualified"
        attributeFormDefault="unqualified"

   xmlns:data-view="http://www.w3.org/2003/g/data-view#" 
   data-view:transformation="http://www.w3.org/2003/g/embeddedRDF.xsl"
  >

  <xs:annotation>
    <xs:appinfo>
      <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
        <rdf:Description rdf:about="http://www.w3.org/2003/g/po-ex">
          <data-view:namespaceTransformation
              rdf:resource="grokPO.xsl" />
        </rdf:Description>
      </rdf:RDF>
    </xs:appinfo>
  </xs:annotation>
...

00:00.000 - client connection from 127.0.0.1:39649]
GET http://www.w3.org/2003/g/embeddedRDF.xsl HTTP/1.1
Host: www.w3.org
Accept: application/xml

[00:00.054 - server connected]
HTTP/1.1 200 OK
Last-Modified: Wed, 23 Mar 2005 18:49:12 GMT
Content-Length: 797
Content-Type: application/xml; qs=0.9

<xsl:transform
    version="1.0"
    xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
...

<rdf:RDF
   xmlns:data-view="http://www.w3.org/2003/g/data-view#"
   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>
  <rdf:Description rdf:about="http://www.w3.org/2003/g/po-ex">
    <data-view:namespaceTransformation rdf:resource="http://www.w3.org/2003/g/grokPO.xsl"/>
  </rdf:Description>
</rdf:RDF>

[00:00.000 - client connection from 127.0.0.1:39651]
GET http://www.w3.org/2001/XMLSchema HTTP/1.1
Host: www.w3.org
Accept: application/rdf+xml,application/xml,text/xml,application/xhtml+xml,text/html
If-modified-since: Fri, 16 Dec 2005 14:19:38 GMT

[00:00.047 - server connected]
HTTP/1.1 304 Not Modified
Content-Location: XMLSchema.html
Expires: Wed, 07 Feb 2007 15:09:29 GMT
Cache-Control: max-age=21600
Vary: negotiate, accept, accept-charset

[00:00.000 - client connection from 127.0.0.1:39653]
GET http://www.w3.org/2003/g/grokPO.xsl HTTP/1.1
Host: www.w3.org
Accept: application/xml

[00:00.048 - server connected]
HTTP/1.1 200 OK
Last-Modified: Tue, 07 Dec 2004 23:33:28 GMT
Content-Length: 1739
Content-Type: application/xml; qs=0.9

<xsl:transform
    version="1.0"
    xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:po="http://www.w3.org/2003/g/po-ex"
    xmlns:poF="http://www.w3.org/2003/g/po-ex#"
    >

<xsl:output method="xml" indent="yes" />

<div xmlns="http://www.w3.org/1999/xhtml">
<h1>grokPO.xsl -- interpret purchase order format as RDF</h1>
...

<rdf:RDF
   xmlns:poF="http://www.w3.org/2003/g/po-ex#"
   xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
>
  <rdf:Description rdf:nodeID="hOhqYGhx9">
    <poF:city>Mill Valley</poF:city>
    <poF:state>CA</poF:state>
    <poF:zip>90952</poF:zip>
    <poF:street>123 Maple Street</poF:street>
    <poF:name>Alice Smith</poF:name>
  </rdf:Description>
...

8. Security considerations

The execution of general-purpose programming languages as interpreters for transformations exposes serious security risks. Designers of GRDDL-aware agents are advised to guard against simply sending GRDDL transformations to "off-the-shelf" interpreters. While it is usually safe to pass documents from trusted sources through a GRDDL transformation, implementors should consider all of the following before adding the ability to execute arbitrary GRDDL transformations linked from arbitrary Web documents.

GRDDL, like many Web technologies, fundamentally relies on the dereferencing of URIs. Writers of GRDDL transformations are advised against employing URL operations which are potentially dangerous, because these operations are more likely to be unavailable in secure GRDDL implementations. Software executing GRDDL transformations are advised to either completely disable all potentially dangerous URL operations or take special care not to delegate any special authority to their operation. In particular, operations to read or write URLs are more safely executed with the privileges associated with an untrusted party, rather than the current user. Such disabling and/or checking should be done completely outside of the reach of the transformation language itself; care should be taken to insure that no method exists for re-enabling full-function versions of these operators.

The remainder of this section outlines some, though probably not all, of the possible problems with the execution of GRDDL transformations, with particular reference to transformations in XSLT.

1. With unconstrained use of GRDDL, untrusted transformations may access URLs which the end-user has read or write permission, while the author of the transformation does not. This is particularly pertinent for URLs from the file: scheme; but many other schemes are also impacted. The untrusted code may, having read documents which the author did not have permission to access, transmit the content of the documents, to arbitrary Web servers by encoding the contents within a URL, that may be passed to the server.
2. Dangerous operations in the XSLT language include, but may not be limited to, the operations involving getting a URL: document(), doc(), unparsed-text() and unparsed-text-available(), and xsl:result-document which involves writing to a URL. xsl:include and xsl:import present fewer risks if they are processed before execution of the transformation, rather than during it.
3. Some transformation language implementations may provide facilities for loading and executing other programming language code. For example, an XSLT implementation may provide a method for executing Java code. Such facilities are obviously open to abuse. Designers of GRDDL transformations are advised against making use of such features. Besides being implementation-specific, they are more likely to be unavailable in secure implementations of the transformation language. The use of such operators in software executing GRDDL transformations should protect against such operators in case they are encountered.
4. XSLT implementations often provide their own extensions. Designers of GRDDL transformations are advised not make use of extensions because they are not guaranteed to be present in all implementations. Software executing GRDDL transformations should make sure that extensions are secure and do not present any kind of threat.
5. Since it is possible to write transformations that inordinately consume system resources or that loop indefinitely. Both types of transformations have the potential to cause damage if sent to unsuspecting recipients. Designers of GRDDL transformations are advised to avoid the construction and dissemination of such transformations. Software executing GRDDL transformations should provide appropriate mechanisms to abort processing after a reasonable amount of time has elapsed. In addition, GRDDL software should be limited to the consumption of only a reasonable amount of any given system resource.
6. Finally, bugs may exist in some interpreters of a transformation language which might be exploited to gain unauthorized access to a recipient's system. Apart from noting this possibility, no specific action is advised to take to prevent this aside from timely correction of such bugs as they are discovered.

9. The GRDDL Vocabulary

The following is excerpted from the GRDDL profile/namespace document:

This document, http://www.w3.org/2003/g/data-view, is a metadata profile in the sense of the HTML specification, in section 7.4.4.3 Meta data profiles.

The following term is introduced here as an XHTML link relationship name and RDF property name:

• transformation: relates a source document to a transformation, usually represented in XSLT, that relates the source document syntax to the RDF graph syntax. domain: RootNode; range: Transformation

The following terms are introduced here as RDF properties:

• namespaceTransformation: relates a namespace to a transformation for all documents in that namespace. range: Transformation
• profileTransformation: relates a profile document to a transformation for all documents bearing that profile. range: Transformation
• result: an RDF graph obtained from an information resource by directly parsing a representation in the standard RDF/XML syntax or indirectly by parsing some other dialect using a transformation nominated by the document. domain: InformationResource; range: RDFGraph
• transformationProperty relates a transformation to the algorithm specified by the property that computes an RDF graph from an XML document node domain: Transformation range: TransformationProperty
• Transformation an InformationResource that specifies a transformation from a set of XML documents to RDF graphs Each Transformation has at least one transformationProperty that is a TransformationProperty.
• TransformationProperty a FunctionalProperty that relates XML document root nodes to RDF graphs

The following terms are bound to concepts from existing standards:

• RootNode the root of the tree in the XPath data model, per section 5.1 Root Node in XML Path Language (XPath) Version 1.0
• RDFGraph a set of RDF triples, per definition in Resource Description Framework (RDF): Concepts and Abstract Syntax
• InformationResource A resource which has the property that all of its essential characteristics can be conveyed in a message, per definition in Architecture of the World Wide Web, Volume One

The namespace document includes RDF data about the terms in the GRDDL Vocabulary, but these RDF data do not include any triples whose predicate is grddl:profileTransformation.

In the section on Using GRDDL with XML Namespace Documents, only explicit grddl:namespaceTransformation triples satisfy the premise of the rule. Likewise, grddl:profileTransformation triples must be explicit in the GRDDL result of a profile document in order to satisfy the premise of the rule in the section on and on GRDDL for HTML Profiles. Authors of GRDDL source documents are advised against using RDFS or OWL expressions which imply such triples but do not explicitly state them.

10. References

Normative References

RFC3987

Internationalized Resource Identifiers (IRIs) Internet RFC 3987 January 2005. Duerst, Suignard

RFC3986

Uniform Resource Identifier (URI): Generic Syntax Internet RFC3986 January 2005. Berners-Lee, Fielding, Masinter

WEBARCH

Architecture of the World Wide Web, Volume One , N. Walsh, I. Jacobs, Editors, W3C Recommendation, 15 December 2004, http://www.w3.org/TR/2004/REC-webarch-20041215/ . Latest version available at http://www.w3.org/TR/webarch/ .

RDFC04

Resource Description Framework (RDF): Concepts and Abstract Syntax , G. Klyne, J. J. Carroll, Editors, W3C Recommendation, 10 February 2004, http://www.w3.org/TR/2004/REC-rdf-concepts-20040210/ . Latest version available at http://www.w3.org/TR/rdf-concepts/ .

RDF-MT

RDF Semantics , P. Hayes, Editor, W3C Recommendation, 10 February 2004, http://www.w3.org/TR/2004/REC-rdf-mt-20040210/ . Latest version available at http://www.w3.org/TR/rdf-mt/ .

RDFX

RDF/XML Syntax Specification (Revised), D. Beckett, Editor, W3C Recommendation, 10 February 2004, http://www.w3.org/TR/2004/REC-rdf-syntax-grammar-20040210/ . Latest version available at http://www.w3.org/TR/rdf-syntax-grammar .

XMLBASE

XML Base , J. Marsh, Editor, W3C Recommendation, 27 June 2001, http://www.w3.org/TR/2001/REC-xmlbase-20010627/ . Latest version available at http://www.w3.org/TR/xmlbase/ .

XHTML

Modularization of XHTML™ , S. Schnitzenbaumer, F. Boumphrey, T. Wugofski, S. McCarron, M. Altheim, S. Dooley, Editors, W3C Recommendation, 10 April 2001, http://www.w3.org/TR/2001/REC-xhtml-modularization-20010410/ . Latest version available at http://www.w3.org/TR/xhtml-modularization/ .

HTML4

HTML 4.01 Specification , D. Raggett, A. Le Hors, I. Jacobs, Editors, W3C Recommendation, 24 December 1999, http://www.w3.org/TR/1999/REC-html401-19991224 . Latest version available at http://www.w3.org/TR/html401 .

XPATH

XML Path Language (XPath) Version 1.0 , J. Clark, S. J. DeRose, Editors, W3C Recommendation, 16 November 1999, http://www.w3.org/TR/1999/REC-xpath-19991116 . Latest version available at http://www.w3.org/TR/xpath .

XSLT1

XSL Transformations (XSLT) Version 1.0 , J. Clark, Editor, W3C Recommendation, 16 November 1999, http://www.w3.org/TR/1999/REC-xslt-19991116 . Latest version available at http://www.w3.org/TR/xslt .

Informative references

The following documents provide additional background but are not part of this specification.

primer

GRDDL Primer , I. Davis, Editor, W3C Working Draft (work in progress), 2 October 2006, http://www.w3.org/TR/2006/WD-grddl-primer-20061002/ . Latest version available at http://www.w3.org/TR/grddl-primer/ .

usecases

GRDDL Use Cases: Scenarios of extracting RDF data from XML documents , F. Gandon, Editor, W3C Working Group Note, 6 April 2007, http://www.w3.org/TR/2007/NOTE-grddl-scenarios-20070406/ . Latest version available at http://www.w3.org/TR/grddl-scenarios/ .

GRDDL-TESTS

GRDDL Test Cases , C. Ogbuji, Editor, W3C Recommendation, 11 September 2007, http://www.w3.org/TR/2007/REC-grddl-tests-20070911/ . Latest version available at http://www.w3.org/TR/grddl-tests/ .

SPARQL

SPARQL Query Language for RDF , E. Prud'hommeaux, A. Seaborne, Editors, W3C Working Draft (work in progress), 26 March 2007, http://www.w3.org/TR/2007/WD-rdf-sparql-query-20070326/ . Latest version available at http://www.w3.org/TR/rdf-sparql-query/ .

XSLT2

XSL Transformations (XSLT) Version 2.0 , M. Kay, Editor, W3C Recommendation, 23 January 2007, http://www.w3.org/TR/2007/REC-xslt20-20070123/ . Latest version available at http://www.w3.org/TR/xslt20 .

RFC2731

J. Kunze Encoding Dublin Core Metadata in HTML in 1999

XFN

XFN: Introduction and Examples copyright GMPG 2003-2007. Eric, Tantek, and Matt

DCRDF

Expressing Simple Dublin Core in RDF/XML Beckett, Miller, Brickley 2002-07-31

P3P

The Platform for Privacy Preferences 1.0 (P3P1.0) Specification , M. Marchiori, Editor, W3C Recommendation, 16 April 2002, http://www.w3.org/TR/2002/REC-P3P-20020416/ . Latest version available at http://www.w3.org/TR/P3P/ .

STYPI

Associating Style Sheets with XML documents , J. Clark, Editor, W3C Recommendation, 29 June 1999, http://www.w3.org/1999/06/REC-xml-stylesheet-19990629 . Latest version available at http://www.w3.org/TR/xml-stylesheet .

XPROC

XProc: An XML Pipeline Language , N. Walsh, Editor, W3C Working Draft (work in progress), 28 September 2006, http://www.w3.org/TR/2006/WD-xproc-20060928/ . Latest version available at http://www.w3.org/TR/xproc/ .

MF-RDF-FAQ

Microformat FAQs for RDF Fans, last modified 17:57, 30 May 2006

Appendix: Transformations for Styling versus data extraction (Informative)

The xml-stylesheet processing instruction[STYPI] is generally deployed for automated presentation processing. This type of link is different from links to GRDDL transformation algorithms, which are intended to facilitate extracting data. Also, parsing the content of processing instructions is not supported by XML tools such as XSLT processors, and grounding processing instructions in URI space is not as straightforward as using namespaces with attributes.

Appendix: Base IRI considerations

In the Adding GRDDL to well-formed XML section, we have:

The base IRI for interpretting relative IRI references in a serialization of a graph produced by a GRDDL transformation is the base IRI of the source document.

This corresponds to RFC 3986, particularly section 5.1, which illustrates the identification of a base URI, with the following picture:

         .----------------------------------------------------------.
         |  .----------------------------------------------------.  |
         |  |  .----------------------------------------------.  |  |
         |  |  |  .----------------------------------------.  |  |  |
         |  |  |  |  .----------------------------------.  |  |  |  |
         |  |  |  |  |       <relative-reference>       |  |  |  |  |
         |  |  |  |  `----------------------------------'  |  |  |  |
         |  |  |  | (5.1.1) Base URI embedded in content   |  |  |  |
         |  |  |  `----------------------------------------'  |  |  |
         |  |  | (5.1.2) Base URI of the encapsulating entity |  |  |
         |  |  |         (message, representation, or none)   |  |  |
         |  |  `----------------------------------------------'  |  |
         |  | (5.1.3) URI used to retrieve the entity            |  |
         |  `----------------------------------------------------'  |
         | (5.1.4) Default Base URI (application-dependent)         |
         `----------------------------------------------------------'

During typical GRDDL processing, an intermediate RDF/XML serialization is produced as the output of a transform. To convert this serialization into an RDF graph, any relative references in the serialization are resolved to IRIs. To identify the appropriate base IRI for resolving a given relative reference, first check for a base URI embedded within this RDF/XML, following XML Base, as permitted by RDF Syntax. If there is no base URI embedded within this RDF/XML, then section 5.1.2 of RFC 3986 may apply, because the encapsulating entity of this serialization is the root element of the input document. If this element does not define a base URI, then its encapsulating entity, the input document, may define a base IRI.

The original document may be an XHTML family document, or it may be some other XML document.

The Base IRI of an XHTML Family document

For an XHTML family document, the base IRI of the input document may be specified as the value of the href attribute of the <base> element (if any). This is in accordance with section 5.1.1 of RFC 3986.

In many other cases, section 5.1.2 does not apply, and section 5.1.3 does apply. Section 5.1.3 specifies the use of the retrieval IRI as the base IRI. Furthermore, section 5.1.3 of RFC 3986 specifies that:

if the retrieval was the result of a redirected request, the last URI used (i.e., the URI that resulted in the actual retrieval of the representation) is the base URI.

The resulting IRI is used as the base IRI parameter for processing the intermediate RDF/XML serialization.

The base IRI of other XML documents

Other XML documents may use XML Base. This is only recommended when the specific document format permits the use of XML Base.

When an xml:base attribute is present on the root element of an XML document, this specifies the base IRI for that document, following section 5.1.1 of RFC 3986.

When there is no xml:base attribute on the root element, even if there is such an attribute on a descendent element, then section 5.1.1 of RFC 3986 does not apply.

As in the XHTML case, we then have to consider sections 5.1.2, 5.1.3 and 5.1.4 of RFC 3986.

Of these, sections 5.1.3 is the most common case, and the note about redirected retrieval also applies.

The base IRI in a processing pipeline

A GRDDL aware agent computes GRDDL results when

given a URI I of an information resource IR, and an XPath node N for a representation of IR

To use a GRDDL aware agent in a processing pipeline, as well as the XPath node N, it is also necessary to specify a corresponding IRI I. This is used as the base IRI when the other mechanisms do not apply. This corresponds to section 5.1.4 of RFC 3986. It is even possible for the default IRI used to bear no relationship with the XPath node N, but in such a case, we read:

As this definition is necessarily application-dependent, failing to define a base URI by using one of the other methods may result in the same content being interpreted differently by different types of applications.

A sender of a representation containing relative references is responsible for ensuring that a base URI for those references can be established.

Responsibilities for correct processing of base IRIs

Document authors, including profile and namespace documents

Document authors should, in general, include a base URI if the document is retrievable from some other URI.

For an XHTML family document[XHTML], this is done using the base element.

For other XML documents, if the format supports xml:base then this should be used. In general, experience suggests that there is least confusion when this is done on the root element. Document authors may also use xml:base attributes elsewhere in their documents, as permitted by the document format, with semantics as defined by XML Base[XMLBASE].

For XML documents in formats that do not support xml:base, and are not XHTML family documents, there is no support in GRDDL for specifying an in-line base URI.

When a profile or namespace document can be accessed via multiple URIs, for instance by a redirect, document authors should, in general, provide a GRDDL result that specifies profile transformations or namespace transformations for each of these URIs.

GRDDL aware agents

When a GRDDL result represented in RDF/XML using the rule for RDF/XML, a base URI may be needed for this representation, in order to convert it into a RDF Graph, following the rules in the RDF/XML Syntax Specification[RDFX].

GRDDL results represented in other ways may also need a base URI.

Following the analysis above, a base URI for resolving a relative reference is defined by following section 5.1 of RFC 3986.

In many applications, it is highly undesirable that GRDDL results may depend on an application default URI, from section 5.1.4 of RFC 3986, ; some GRDDL aware agents may treat this possibility as an error.

GRDDL transformation authors

In general, when writing a GRDDL transformation for an XHTML family document to RDF/XML the best advice is to ignore issues to do with the base URI. The easiest approach is to produce relative URIs in the output, corresponding to any relative URIs in the input, and absolute URIs corresponding to any concepts built into the transform. Such relative URIs will be resolved, during the processing performed by a GRDDL aware agent, against the correct base URI.

When writing a GRDDL transformation for an XML document format that does not support xml:base, and has no means to represent an in-line base URI, there is little choice but to ignore issues of the correct base.

When writing a GRDDL transformation for an XML document format, other than an XHTML family document, that does not support xml:base, but has some other means to represent an in-line base URI, then a GRDDL aware agent will be ignorant of this means, and a well-written GRDDL transformation will attempt to correct for this. When a base URI is specified in such a way, one approach is to insert the base URI into the RDF/XML output as the value of an xml:base attribute, so that the RDF/XML parser will resolve relative URIs against that base, and ignore the base URI passed by the GRDDL aware agent, which will have been computed ignoring the conventions specific to this format.

When writing a GRDDL transformation for an XML document format that does support xml:base, then it must be remembered that a GRDDL aware agent has responsibility to handle an xml:base on the root element. If there is such an xml:base attribute, then the simplest behaviour for a GRDDL transformation, is to ignore it.

However, other xml:base attributes, not on the root element, are the responsibility of the transform, since the GRDDL aware agent ignores these. Thus, these lower level xml:base attributes should be honored, most easily by copying them into the output graph in the appropriate place. However, in general, xml:base attributes on ancestor nodes also have to be taken into account, unless there is an intervening xml:base attribute with an absolute URI as its value. This is clearly non-trivial to get right: to assist, the GRDDL library provides a module to be imported into your stylesheet, see below.

In all cases, while often unnecessary, if a transform is aware of an absolute base URI, specified in its input, for the whole document, it is never incorrect to use this base URI as the base URI for the output, for example, by adding an appropriate xml:base attribute to the rdf:RDF element.

Transforms that do this need to guard against the possibly incorrect similar treatment of relative base URIs. For example a xml:base=".." on the root element might, in the interaction between a correct GRDDL aware agent and a poorly written transform, be applied twice, resulting in relative references being resolved at the wrong level in the directory hierarchy.