RDFa in XHTML: Syntax and Processing

The modern Web is made up of an enormous number of documents that have been created using HTML. These documents contain significant amounts of structured data, which is largely unavailable to tools and applications. When publishers can express this data more completely, and when tools can read it, a new world of user functionality becomes available, letting users transfer structured data between applications and web sites, and allowing browsing applications to improve the user experience. An event on a web page can be directly imported into a user's desktop calendar; a license on a document can be detected so that users can be informed of their rights automatically; a photo's creator, camera setting information, resolution, location and topic can be published as easily as the original photo itself, enabling structured search and sharing.

RDFa is a syntax for expressing this structured data in XHTML. The rendered, hypertext data of XHTML is reused by the RDFa markup, so that publishers don't repeat themselves. The underlying abstract representation is RDF, which lets publishers build their own vocabulary, extend others, and evolve their vocabulary with maximal interoperability over time. The expressed structure is closely tied to the data, so that rendered data can be copied and pasted along with its relevant structure.

The rules for interpreting the data are generic, so that there is no need for different rules for different formats; this allows authors and publishers of data to define their own formats without having to update software, register formats via a central authority, or worry that two formats may intere with each other.

How to Read this Document

If you are already familiar with RDFa, and you want to examine the processing rules--perhaps to create a parser--then you'll find section x of most interest. Each of the processing steps in this section is outlined in more detail in the sections after it.

If you are not familiar with RDFa, but you are familiar with RDF, then you might find reading the overview useful, since it gives a range of examples of XHTML mark-up that use RDFa. Seeing some examples first should make reading the processing rules easier.

If you are not familiar with RDF, then you might want to take a look at section y before trying to do too much with RDFa. Although RDFa is designed to be easy to author--and authors don't need to understand RDF to use it--anyone writing applications that consume RDFa will need to understand RDF. There is a lot of material on RDF on the web, and a growing range of tools that will support RDFa, so all we try to do in this document is provide enough background on RDF to make the goals of RDFa clearer.

And finally, if you are not familiar with either RDFa or RDF, and simply want to add RDFa to your documents, then you may find the RDFa Primer to be a better introduction.

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 an internal draft produced jointly by the Semantic Web Deployment Working Group [SWD-WG] and the XHTML 2 Working Group [XHTML2-WG]. Initial work on RDFa began in the XHTML 2 Working Group [XHTML2-WG].

This document has no official standing within the W3C. It is also a work in progress, which means it may change at any time, without warning, and you shouldn't rely on anything in this document.

1. Motivation

RDF/XML [RDF-SYNTAX] provides sufficient flexibility to represent all of the abstract concepts in RDF [RDF-CONCEPTS]. However, it presents two challenges; first it is difficult or impossible to validate documents that contain RDF/XML using XML Schemas or DTD's, which makes it difficult to import RDF/XML into other markup languages. Whilst newer schema languages such as RELAX NG [RELAXNG] do provide a way to validate documents that contain arbitrary RDF/XML, it will be a while before they gain wide support.

Second, even if one could add RDF/XML directly into an XML dialect like XHTML, there would be significant data duplication between the rendered data and the RDF/XML structured data. It would be far better to add RDF to a document without repeating the document's existing data. For example, an XHTML document that explicitly renders its author's name "Mark Birbeck" should not need to repeat this name for the RDF expression of the same concept: it should be possible to supplement the existing markup in such a way that it can also be interpreted as RDF, with minimal repetition of data.

Third, as users often want to transfer structured data from one application to another, sometimes to or from a non-web-based application, it is highly beneficial to express the web data's structure "in context." The user experience could then be enhanced, for example by providing contextual information about specific rendered data, perhaps when the user "right-clicks" on an item of interest.

In the past, many attributes were 'hard-wired' directly into the markup language to represent specific concepts. For example, in XHTML 1.1 [XHTML11] and HTML [HTML4] there is a cite attribute; the attribute allows an author to add information to a document which is used to indicate the origin of a quote.

However, these 'hard-wired' attributes make it difficult to define a generic process for extracting metadata from any document since a parser would need to know about each of the special attributes. One motivation for RDFa has been to devise a means by which documents can be augmented with metadata in a generic rather than hard-wired manner. This has been achieved by creating a fixed set of attributes and parsing rules, but allowing those attibutes to contain properties from any of a number of the growing range of available taxonomies. The values of those properties are in most cases the information that is already in an author's document.

RDFa takes the pressure off language authors to anticipate all the structural requirements users of their language might have, by outlining a new syntax for RDF that relies only on attributes. RDFa can be easily imported into other XML-based markup languages, as well as HTML, allowing any mark-up language to carry arbitrary RDF.

This specification deals specifically with the use of RDFa in HTML-based languages, such as HTML and XHTML, and defines an RDF mapping for a number of HTML attributes.

2. Orientation

NOTE: The idea of this section is to give the reader a quick intro to the features RDFa supports. It's not meant to be exhaustive, but if I have missed anything obvious then do let me know. The idea is to use a simple example, and to make them 'familiar', which is why I chose to use the mark-up from the Primer.

The following examples are intended to help readers who are not familiar with RDFa to quickly get a sense of how it works.

As an HTML author you will already be familiar with using meta and link to add additional information to your documents:

RDFa makes uses of this concept, enhancing it with the ability to make use of other vocabularies by using namespaces:

Although not widely used, HTML already supports the use of @rel and @rev on the a element. This becomes more useful in RDFa with the addition of namespace support:

Not only can URLs in the document be re-used to provide metadata, but so can inline text:

If some displayed text is different to the actual 'value' it represents, more precise values can be added, which can optionally include datatypes:

In many cases a block of mark-up will contain a number of properties that relate to the same item; it's possible with RDFa to indicate the type of that item:

The metadata features available in HTML only allow information to be expressed about the document itself. RDFa provides a means of referring to other documents and resources:

3. RDF Terminology

The previous section gave examples of typical mark-up in order to illustrate what RDFa in XHTML looks like. But what RDFa in XHTML represents is RDF. In order to author RDFa in XHTML you do not need to understood RDF at all, although it would certainly help. However, if you are building a system that consumes the RDF output of an RDFa in XHTML document you will obviously need to understand RDF--but you can also skip this section, since here we provide a very basic introduction to RDF terminology so that we can make use of it in the following sections.

3.1. Statements

The metadata information that RDFa provides access to is generally understood to be a collection of statements. A statement is a basic unit of information that has been constructed in a very specific way to make it easier to process. In turn, by breaking large sets of information down into a collection of statements, even very complex metadata can be made available for processing.

3.2. Triples

To make this information machine-processable RDF defines the structure of these statements very tightly. A statement is actually a triple, meaning that it is made up three components. The first is the subject of the statement, and is what we are making our statements about. In these examples the subject is always 'Albert'.

The second part of a triple is the property of the subject that we want to define. In the examples here, the properties would be 'was born on', 'was born in', and 'has a picture at'. These are more usually called predicates in RDF.

The final part of a triple is the value of the property, or the object. In the examples here the object values are 'March 14, 1879', 'Germany', and 'http://en.wikipedia.org/wiki/Image:Albert_Einstein_Head.jpg'.

3.3. URI references

Breaking complex information into manageable units obviously helps, but there is still ambiguity here. For example, which 'Albert' are we talking about? If some system has further facts--triples--about 'Albert' how could we know whether they are about the same person, and so add them to the list of things we know about that person? Also, if we wanted to find people born in Germany, how could we know that the predicate 'was born in' has the same purpose as the predicate 'birthplace' that exists in some other system? RDF solves this problem by replacing our vague terms with URI references.

URIs are most commonly used to identify web pages, but RDF makes use of them as a way to provide unique identifiers for concepts. For example, we could identify the subject of all of our statements by using the DBPedia URI for Albert Einstein:

URI references are also used to uniquely identify the objects in metadata statements (note that the picture of Einstein is already a URI):

And of course URI references are also used to ensure that predicates are unambiguous:

3.4. Plain literals

Although URI resources are always used for subjects and predicates, the object part of a triple can be either a URI or a literal. In the example triples, Einstein's name is represented by a plain literal, which means that it is a basic string with no type or language information:

3.5. Typed literals

Some literals, such as dates and numbers, have very specific meanings, so RDF provides a mechanism for indicating the type of a literal. A typed literal is indicated by attaching a URI to the end of a plain literal which indicates the literal's datatype. This URI is usually based on datatypes defined in the XML Schema Datatypes specification [XML SCHEMA DATATYPES REFERENCE / http://www.w3.org/TR/xmlschema-2/]. The following syntax would be used to unambiguously express Einstein's date of birth as a literal of type xsd:date:

3.6. N-Triples

RDF does not have one set way to express triples, since the key ideas of RDF are the triple and the use of URIs. However, a number of mechanisms are available, such as RDF/XML, N-Triples [N-TRIPLES], and of course RDFa. Most discussions of RDF make use of the N-Triple syntax to explain their ideas, since it's quite compact. The examples we have just seen are already using this syntax, and we'll continue to use it throughout this document, with a slight variation that long URIs can be abbreviated by using a URI mapping. This is indicated by removing the angle brackets from the URI, as follows:

3.7. Graphs

For more information on the concepts described above, see [RDF-CONCEPTS]. RDFa additionally defines the following terms:

3.7.1. Compact URIs

In order to allow for the compact expression of RDF statements, RDFa allows the contraction of all [URI reference]s into a form called a 'compact URI', or [CURIE]. Until recently QNames [QNames] have been the most common way to abbreviate URIs, but there is a well-known limitation that the syntax for QNames does not allow all possible [URI reference]s to be expressed. CURIEs have been specificially designed to look like QNames, but at the same time to get around their limitations.

Note that CURIEs are only used in the mark-up, and never appear in the generate [triple]s, which will always use [URI reference]s.

3.8. A description of RDFa in RDF terms

The following is a description of RDFa that uses RDF terminology, which may be useful to readers with an RDF background:

The aim of RDFa is to allow [RDF graph]s to be carried in XML documents of any type, although this specification deals only with RDFa in XHTML. An [RDF graph] comprises [node]s linked by relationships. The basic unit of a graph is a [triple], in which a subject [node] is linked to an object [node] via a [predicate]. The subject [node] is always either an [URI reference] or a [blank node], the predicate is always an [URI reference], and the object of a statement can be an [URI reference], a [literal], or a [blank node].

In RDFa, a subject [URI reference] is generally indicated using the attribute about, and predicates are represented using one of the attributes property,


instanceof

, rel, or rev. Objects which are [URI reference]s are represented using the attributes href, resource or src, whilst objects that are [literal]s are represented either with the attribute content (with an optional [datatype] expressed using the datatype attribute), or the content of the element in question.

4. Processing Model

This section looks at a generic set of processing rules for creating a set of triples that represent the metadata present in an XHTML+RDFa document. Processing need not follow the DOM traversal technique outlined here, although the effect of following some other manner of processing must be the same as if the processing outlined here were followed. The processing model is explained using the idea of DOM traversal which makes it easier to describe (particularly in relation to the 'evaluation context').

4.1. Overview

Parsing a document for RDFa triples is carried out by starting at the root element of the document, and visiting each of its child elements in turn, applying processing rules. Processing is recursive in that for each child element the processor also visits each of its child elements, and applies the same processing rules.

As processing continues, rules are applied which will either generate triples, or change the [evaluation context] information which will be used in subsequent processing. Some of the rules will be determined by the host language--in this case XHTML--and some of the rules will be part of RDFa.

Note that we don't say anything about what should happen to the triples generated, or whether more triples might be generated during processing than are outlined here. However, to be conformant, an RDFa processor needs to act as if at least the rules in this section are applied.

4.2. Evaluation Context

During processing, each rule is applied within an 'evaluation context'. Rules may further modify this evaluation context, or create triples that can be established by making use of this evaluation context. The context itself consists of the following pieces of information:

4.3. Processing

Processing would normally begin after the document to be parsed has been completely loaded. However, there is no requirement for this to be the case, and it is certainly possible to use a SAX-style processing model to extract the RDFa information. Note that if some approach other than the DOM traversal approach defined here is used, it is important to ensure that any meta or link elements processed in the head of the document honour any occurrences of base which may appear after those elements. (In other words, HTML processing rules must still be applied, even if document processing takes place in a non-HTML environment such as a search indexer.)

At the beginning of processing, the [current evaluation context] is initialised as follows:

Processing then begins with the root element, and all nodes in the tree are processed according to the following rules, depth-first:

Any changes to the [current evaluation context] are made first:

the [current element] is parsed for [URI mappings] and these are added to the [list of URI mappings]. Note that a [URI mapping] will simply overwrite any current mapping in the list that has the same name;
These mappings are provided by the xmlns atttribute. The value to be mapped is set by the XML namespace prefix, and the value to map is the value of the attribute--a URI. Note that the URI is not processed in any way; in particular if it is a relative path it is not resolved against the [current base]. Authors are advised to follow best practice for using namespaces, which includes not using relative paths. (See [xyz].)
the [current element] is parsed for any language information, and [language] is set in the [current evaluation context];
Language information can be provided using either the general-purpose XML attribute, xml:lang, or the HTML attribute lang.
the [current element] is parsed for any subject information, and it is used to set the [current resource] value, in the [current evaluation context];
The [current resource] can be set using @about. Note that the final value of the [current resource] is an absolute IRI, which means that if @about contains a relative path the value must be normalised against [base] in the [current evaluation context], using the algorithm defined in RFC 3986. The value can also be provided by a CURIE, and is processed as defined in section curie. Note that since this attribute can take both URIs and CURIEs, the latter will have been expressed using the [safe CURIE] syntax.
the [recurse] flag is set to true;
Processing will generally continue recursively through the entire tree of nodes available. However, if an author indicates that some branch of the tree should be treated as an XML literal, no further processing should take place on that branch. This flag is used to inhibit this processing.
the [chaining] flag is set to false;
If a collection of statements is contained by a [URI reference] then this may become the subject of further statements.

Once the [current evaluation context] has been set, object resolution is carried out, as follows:

the [current object resource] is established;
Since only one [current object resource] is set per element then some attributes will have a higher priority than others. The highest priority is given to the RDFa attribute resource. If there is no resource attribute then the HTML src attribute is used, and if that is not present, the HTML href attribute is used. If none of these are present then a unique identifier or [bnode] is created. Note that the final value of the [current object resource] is an absolute URI, which means that if any of these attributes contain relative paths they must be normalised against [base] in the [current evaluation context], using the algorithm defined in RFC 3986. Note also that since @resource can take both URIs and CURIEs, the latter will have been expressed using the [safe CURIE] syntax.
the [current object literal] is established;
The [current object literal] will be set as a [plain literal] if the content attribute is present, or the body of the [current element] contains only text (i.e., there are no child elements), or the body of the [current element] does have child elements but the datatype attribute has an empty value. Additionally, if there is a value for [current language] then the value of the [plain literal] should include this language information, as described here:???. The actual literal is either the value of the content attribute (if present) or a string created by concatenating the inner content of each of the children in turn, of the [current element].

The [current object literal] will be set as a [typed literal] if the datatype attribute is present, and does not have an empty value. The actual literal is either the value of the content attribute (if present) or a string created by concatenating the inner content of each of the children in turn, of the [current element]. The final string includes the datatype, as described here:???

The [current object literal] will be set as an [XML literal] if the [current element] has child elements, and the datatype attribute is not present. The value of the [XML literal] is a string created from the inner content of the [current element], i.e., not including the element itself.

Once object resolution is complete the processor will have two objects, one a resource and the other a literal. These objects can now be used to create triples with the [current resource], depending on the presence of other attributes. This is achieved using the following processing steps:

predicates for the [current object literal] are established;
Predicates for the [current object literal] can be set by using the property attribute. If present, the attribute must contain one or more [basic curies], each of which is converted to an absolute URI using CURIE processing rules, and then used to generate a triple as follows:

subject

[current resource]

predicate

expanded value from the basic curie

object

[current object literal]

Note that literal may include language and datatype information as discussed in the section on object resolution. Once the triple has been created, the [recurse] flag is set to false.
predicates for the [current object resource] are established;
Predicates for the [current object resource] can be set by using one or both of the @rel and @rev attributes.
If present, the rel attribute must contain one or more [basic curies], each of which is converted to an absolute URI using CURIE processing rules, and then used to generate a triple as follows:

subject

[current resource]

predicate

expanded value from the basic curie

object

[current object resource]

If present, the rev attribute must contain one or more [basic curies], each of which is converted to an absolute URI using CURIE processing rules, and then used to generate a triple as follows:

subject

[current object resource]

predicate

expanded value from the basic curie

object

[current resource]

If any triples are generated then the [chaining] flag is set to true.

If the [chaining] flag is set to true then the [current resource] is set to the value of the [current object resource], and the [chaining] flag is set to false.

type values for the [current object resource] are established;

One or more 'types' for the [current object resource] can be set by using the instanceof attribute. If present, the attribute must contain one or more [basic curies], each of which is converted to an absolute URI using CURIE processing rules, and then used to generate a triple as follows:

subject: [current object resource]
predicate: http://www.w3.org/1999/02/22-rdf-syntax-ns#type
object: expanded value from the basic curie

If any triples are generated then the [chaining] flag is set to true.

If the [recurse] flag is true, the [current evaluation context] is pushed onto a stack, and all nodes that are children of the [current element] are processed using the rules described here. Once all of the children have been processed then the [current evaluation context] is popped back off the stack.

5. RDFa in detail

This section provides an in-depth examination of the processing steps described in the previous section. It also includes examples which may help clarify some of the steps involved.

In the following examples, for brevity assume that the following namespace prefixes have been defined:

The key to processing is that a triple is generated whenever a predicate/object combination is detected. The actual triple generated will include a subject that may have been set previously, so this is tracked in the [current evaluation context] and is called the [current resource]. Since the subject will default to the current document if it hasn't been set explicitly, then a predicate/object combination is always enough to generate one or more triples.

The attributes for setting a predicate are rel, rev and property, whilst the attributes for setting an object are resource,


href

5.1. Changing the evaluation context

5.1.1. Setting the [current resource]

When triples are created they will always be in relation to the [current resource]. When parsing begins the [current resource] will be the URI of the document being parsed, or a value as set by base.

As processing progresses, any about attributes will change the [current resource]. The value of about is a URI or a CURIE. If it is a relative URI then it needs to be resolved against the current [base] value. In this mark-up the properties cal:summary and cal:dtstart becomes part of the 'event' object, rather than being to do with the document:

5.2. Object resolution

A [literal] object can be set by using the property attribute to express a [predicate], and then using either the attribute content, or the inline text of the element that property is on.

A [URI resource] object can be set using one of the attributes rel or rev to express a [predicate], and then using one of href, resource or src.

5.2.1. Literal object resolution

An [object literal] will be generated when the property attribute is present. The property attribute provides the predicate, and the following sections describe how the actual literal to be generated is determined.

5.2.1.1. Plain Literals

The value of the content attribute is given precedence over any element content, so the following would give exactly the same triple:

5.2.1.1.1. Language Tags

RDF allows [plain literal]s to have a language tag, as illustrated by the following example from [RDFTESTS-RDFMS-XMLLANG-TEST006]:

In RDFa the XML language attribute -- xml:lang -- is used to add this information, whether the plain literal is designated by the content attribute, or by the inline text of the element:

Note that the value can be inherited as defined in [XML-LANG], so the following syntax will give the same triple as above:

5.2.1.2. Typed literals

5.2.1.3. XML Literals

XML documents cannot contain XML mark-up in their attributes, which means it is not possible to represent XML within the content attribute. The following would cause an XML parser to generate an error:

It does not help to escape the content, since the output would simply be a string of text containing numerous ampersands:

RDF does, however, provide a datatype for indicating [XML literal]s. RDFa therefore adds this datatype to any [literal] that has child elements. For example:

There will be situations where the extra mark-up is not actually part of the meaning of the literal, and can be ignored. In this situation an empty datatype value can be used to override the XML literal behaviour:

Although the rendering of this page has highlighted the term the user searched for, setting datatype to nothing ensures that the data is interpreted as a plain literal, giving the following triples:

Note that the value of this [XML Literal] is the exclusive canonicalization of the RDFa element's value.

Although the RDFa processing model requires visiting each node in the tree, if the processor meets an [XML literal] then it shouldn't process any further down the tree. This is to prevent triples being generated from mark-up that is not actually in the hierarchy. For example, we might want to set the title of something to some XHTML that includes RDFa:

Note that this does effectively mean that the presence of @property inhibits any further processing, so authors should watch out for stray attributes, if they find they are getting fewer triples than they had expected.

5.2.2. URI object resolution

One or more [URI object]s is needed when the rel or rev attribute is present. Each attribute will cause triples to be generated when used with @href,


@resource

or @src.

The rel and rev attributes are essentially the inverse of each other; whilst rel establishes a relationship between the [current resource] as subject, and the [object resource] as the object, rev does the exact opposite, and uses the [object resource] as the subject, and the [current resource] as the object.

5.2.2.1. Using @resource to set the object

RDFa provides the resource attribute as a way to set the object of statements. This is particularly useful when referring to resources that are not themselves navigable links:

5.2.2.2. Using the href attribute

If no resource attribute is present, then @href can be used to set the object.

When a triple predicate has been expressed using the rel attribute, the href attribute on the [RDFa statement]'s element is used to indicate the object as a [URI reference]. Its type, just like that of the about attribute, is a URI:

It's also possible to use both rel and rev at the same time on an element. This is particularly useful when two things stand in two different relationships with each, for example when a picture is taken by Mark, but that picture also depicts him:

5.2.2.3. Using @src to set the object

The value of the about attribute sets the subject for any nested triples which means that the same triples can be expressed using this, more compact, syntax:

5.2.2.4. Using a bnode to set the object

When a triple predicate has been expressed using the rel attribute, and no href, src, or resource attribute exists on the same [RDFa element], then the CURIE represented by this element is used as the object. This CURIE is affected by the about attribute, but if none is present the object is a bnode (bnodes are discussed further in Section bnode [REF]). In all cases, the subject resolution for child elements is affected: where they do not override the subject, their subject is this same CURIE here resolved as the object.

Consider, for example, a simple fragment of HTML for describing the creator of a web page, with further information about the creator, including his name and email address:

5.2.2.4.1. Referencing Bnodes

To establish relationships between [blank node]s, the [unique anonymous ID] must be set explicity using a CURIE bnode as subject or object. For example, if our desired output is the following [triple]s:

5.3. The RDFa Attributes

RDFa in XHTML makes use of a number of XHTML attributes, as well as providing a few new ones:

5.3.1. Multiple Attribute Values

The rel, rev, and property attributes accept multiple space-separated CURIEs as a single attribute value. When there is more than one CURIE, then each expresses the exact same triples it would if it were the single CURIE in the attribute value. For example:

is interpreted by performing the normal subject and object resolutions dictated by the rel attribute on both the dc:creator and dc:publisher values. The resulting triples are:

6. CURIE Syntax Defnition

A basic CURIE is comprised of two components, a prefix and a reference. The prefix is separated from the reference by a colon (:). It is possible to omit both the prefix and the colon, or to omit just the prefix and leave the colon.

In some situations an attribute needs to be able to allow either a CURIE, or a normal IRI. Since it is difficult to distinguish between CURIEs and IRIs RDFa adds the notion of a [safe CURIE]. The syntax is simply to surround the CURIE with square brackets:

NOTE: The following language-independent prose will be removed shortly, once we have finalised this.

CURIEs can be used in any language, including non-XML languages. Any language that wishes to make use of CURIEs must provide a context which consists of:

A CURIE is a representation of a full IRI. This IRI is obtained by taking the currently in-scope mapping that is associated with prefix, and concatenating it with the reference. The result MUST be a syntactically valid IRI [IRI].

A. References

A.1. Related Specifications

A.2. Related Activities

B. Change History

2007-09-04: Migrated to XHTML 2 Working Group Publication System. Converted to a format that is consistent with REC-Track documents. Updated to reflect current processing model. Added normative definition of CURIEs. Started updating prose to be consistent with current task force agremeents. [ShaneMcCarron], [StevenPemberton], [MarkBirbeck]

2007-04-06: fixed some of the language to talk about "structure" rather than metadata. Added note regarding space-separated values in predicate-denoting attributes. [BenAdida]

2006-01-16: made the use of CURIE type for rel,rev,property consistent across document (particularly section 2.4 was erroneous). [BenAdida]

C. Acknowledgments

At the time of publication, the participants in the W3C XHTML 2 Working Group were:

cc:	http://creativecommons.org/ns#
dc:	http://purl.org/dc/elements/1.1/
ex:	http://example.org/
foaf:	http://xmlns.com/foaf/0.1/
rdf:	http://www.w3.org/1999/02/22-rdf-syntax-ns#
rdfs:	http://www.w3.org/2000/01/rdf-schema#
p:	http://dbpedia.org/property/
rdfa:	http://www.w3.org/ns/rdfa/
svg:	http://www.w3.org/2000/svg
xh11:	http://www.w3.org/1999/xhtml
xsd:	http://www.w3.org/2001/XMLSchema#
biblio:	http://example.org/biblio/0.1
taxo:	http://purl.org/rss/1.0/modules/taxonomy/