Provenance is information about entities, activities, and people involved in producing a piece of data or thing, which can be used to form assessments about its quality, reliability or trustworthiness. PROV-DM is the conceptual data model that forms a basis for the W3C provenance (PROV) family of specifications. PROV-DM distinguishes core structures, forming the essence of provenance information, from extended structures catering for more specific uses of provenance. PROV-DM is organized in six components, respectively dealing with: (1) entities and activities, and the time at which they were created, used, or ended; (2) derivations of entities from entities; (3) agents bearing responsibility for entities that were generated and activities that happened; (4) a notion of bundle, a mechanism to support provenance of provenance; and, (5) properties to link entities that refer to the same thing; (6) collections forming a logical structure for its members.

To provide examples of the PROV data model, the PROV notation (PROV-N) is introduced: aimed at human consumption, PROV-N allows serializations of PROV instances to be created in a compact manner. PROV-N facilitates the mapping of the PROV data model to concrete syntax, and is used as the basis for a formal semantics of PROV. The purpose of this document is to define the PROV-N notation.

The PROV Document Overview describes the overall state of PROV, and should be read before other PROV documents.

PROV Family of Documents

This document is part of the PROV family of documents, a set of documents defining various aspects that are necessary to achieve the vision of inter-operable interchange of provenance information in heterogeneous environments such as the Web. These documents are listed below. Please consult the [[PROV-OVERVIEW]] for a guide to reading these documents.

Endorsed By W3C

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.

Please Send Comments

Introduction

Provenance is a record that describes the people, institutions, entities, and activities, involved in producing, influencing, or delivering a piece of data or a thing in the world. Two companion specifications respectively define PROV-DM, a data model for provenance, allowing provenance descriptions to be expressed [[!PROV-DM]] and a set of constraints that provenance descriptions are expected to satisfy [[!PROV-CONSTRAINTS]].

Purpose of this Document and Target Audience

A key goal of PROV is the specification of a machine-processable data model for provenance. However, communicating provenance between humans is also important when teaching, illustrating, formalizing, and discussing provenance-related issues. With these two requirements in mind, this document introduces PROV-N, the PROV notation, a syntax designed to write instances of the PROV data model according to the following design principles: PROV-N has several known uses:

This document introduces the PROV-N grammar along with examples of its usage.

Its target audience is twofold:

Compliance with this Document

For the purpose of compliance, all sections of this document are normative, except Appendix A, Appendix B, and Appendix C.2.

Structure of this Document

This document is structured as follows.

Section 2 provides general consideration about the PROV-N grammar.

Section 3 presents the grammar of all expressions of the language grouped according to the PROV data model components.

Section 4 defines the grammar of document, a house-keeping construct of PROV-N capable of packaging up PROV-N expressions and namespace declarations.

Section 5 defines the extensibility mechanism for the PROV-N notation.

Section 6 defines media type for the PROV-N notation.

Notational Conventions

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [[!RFC2119]].

The following namespaces prefixes are used throughout this document.

Table 1: Prefix and Namespaces used in this specification
prefixnamespace uri definition
provhttp://www.w3.org/ns/prov#The PROV namespace (see Section 3.7.4)
xsdhttp://www.w3.org/2000/10/XMLSchema#XML Schema Namespace [[!XMLSCHEMA11-2]]
(others)(various)All other namespace prefixes are used in examples only.
In particular, URIs starting with "http://example.com" represent
some application-dependent URI [[!RFC3986]]

General grammar considerations

For convenience, all productions presented in this document have been grouped in a separate file.

Functional-style Syntax

PROV-N adopts a functional-style syntax consisting of a predicate name and an ordered list of terms.

All PROV data model types have an identifier. Furthermore, some expressions also admit additional elements that further characterize it.

The following expression should be read as "entity e1".

entity(e1)

The following expression should be read as "activity a2, which occurred between 2011-11-16T16:00:00 and 2011-11-16T16:00:01".

entity(e1)
activity(a2, 2011-11-16T16:00:00, 2011-11-16T16:00:01)

All PROV data model relations involve two primary elements, the subject and the object, in this order. Furthermore, some expressions also admit additional elements that further characterize it.

The following expression should be read as "e2 was derived from e1". Here e2 is the subject, and e1 is the object.

wasDerivedFrom(e2, e1)

The following expression expands the above derivation relation by providing additional elements the optional activity a, the generation g2, and the usage u1:

wasDerivedFrom(e2, e1, a, g2, u1)

EBNF Grammar

The grammar is specified using a subset of the Extended Backus-Naur Form (EBNF) notation, as defined in Extensible Markup Language (XML) 1.1 [[!XML11]] section 6 Notation.

The text below provides an introduction to the EBNF notation used in this document.

EBNF specifies a series of production rules (production). A production rule in the grammar defines a symbol expr (nonterminal symbol) using the following form:

expr  ::= term

Symbols are written with an initial capital letter if they are the start symbol of a regular language, otherwise with an initial lowercase letter. A production rule in the grammar defines a symbol <TERMINAL> (terminal symbol) using the following form:

<TERMINAL>  ::= term

Within the term on the right-hand side of a rule, the following terms are used to match strings of one or more characters:

Where suitable, the PROV-N grammar reuses production and terminal names of the SPARQL grammar [[!RDF-SPARQL-QUERY]].

Main Productions

Two productions are entry points to the grammar.

The production expression provides the structure for the core expressions of PROV-N.

Each of the symbols included in expression above, i.e., entityExpression, activityExpression etc., corresponds to one concept (e.g., Entity, Activity, etc.) of the PROV data model.

Alternatively, the production rule document provides the overall structure of PROV-N descriptions. It is a wrapper for a set of expressions, such that the text for an element matches the corresponding expression production, and some namespace declarations.

Optional terms in expressions

Some terms in an expression may be optional. For example:
wasDerivedFrom(e2, e1, a, g2, u1)
wasDerivedFrom(e2, e1)
In a derivation expression, the activity, generation, and usage are optional terms. They are specified in the first derivation, but not in the second.
activity(a2, 2011-11-16T16:00:00, 2011-11-16T16:00:01)
activity(a1)
The start and end times for an activity are optional. They are specified in the first expression, but not in the second.

The general rule for optionals is that, if none of the optionals are used in the expression, then they are simply omitted, resulting in a simpler expression as in the examples above.

However, it may be the case that only some of the optional terms are omitted. Because the position of the terms in the expression matters, an additional marker must be used to indicate that a particular term is not available. The symbol '-' is used for this purpose.

In the first expression below, all optionals are specified. However in the second and third, only one optional is specified, forcing the use of the marker for the missing terms.

wasDerivedFrom(e2, e1, a, g2, u1)
wasDerivedFrom(e2, e1, -, -, u1)
wasDerivedFrom(e2, e1, a, -, -)
Note that the more succinct form is just shorthand for a complete expression with all the markers specified:
activity(a1)
activity(a1, -, -)

Identifiers and attributes

Almost all expressions defined in the grammar include an identifier (see Section 3.7.1 for the full syntax of identifiers). Most expressions can also include a set of attribute-value pairs, delimited by square brackets. Identifiers are optional except for Entities, Activities, and Agents. Identifiers are always the first term in any expression. Optional identifiers MUST be separated using a semi-colon ';', but where the identifiers are required, a regular comma ',' MUST be used. This makes it possible to completely omit an optional identifier with no ambiguity arising. Also, if the set of attribute-value pairs is present, it is always the last term in any expression.

Derivation has an optional identifier. In the first expression, the identifier is not available, while it is explicit in the second. The third example shows that one can optionally indicate the missing identifier using the - marker. This is equivalent to the first expression.

wasDerivedFrom(e2, e1)
wasDerivedFrom(d; e2, e1)
wasDerivedFrom(-; e2, e1)
Lack of attributes can be equivalently expressed by omitting the list, or by using an empty list.

The first and second activity expressions do not specify any attributes, and are equivalent. The third activity expression specifies two attributes.

activity(ex:a1)
activity(ex:a1, [])
activity(ex:a1, [ex:param1="a", ex:param2="b"])

Comments

Comments in PROV-N take two forms:

Comments are treated as white space.

PROV-N Productions per Component

This section introduces grammar productions for each expression, followed by small examples of expressions illustrating the grammar. Strings conforming to the grammar are valid expressions in the PROV-N language.

Component 1: Entities and Activities

Entity

The following table summarizes how each constituent of a PROV-DM Entity maps to a PROV-N syntax element.

EntityNon-Terminal
ididentifier
attributesoptionalAttributeValuePairs
entity(tr:WD-prov-dm-20111215, [ prov:type="document" ])
Here tr:WD-prov-dm-20111215 is the entity identifier, and [ prov:type="document" ] groups the optional attributes, only one in this example, with their values.

entity(tr:WD-prov-dm-20111215)
Here, the optional attributes are absent.

Activity

The following table summarizes how each constituent of a PROV-DM Activity maps to a PROV-N syntax element.

ActivityNon-Terminal
ididentifier
startTimetimeOrMarker
endTimetimeOrMarker
attributesoptionalAttributeValuePairs
activity(ex:a10, 2011-11-16T16:00:00, 2011-11-16T16:00:01, [prov:type="createFile"])

Here ex:a10 is the activity identifier, 2011-11-16T16:00:00 and 2011-11-16T16:00:01 are the optional start and end times for the activity, and [prov:type="createFile"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
activity(ex:a10)
activity(ex:a10, -, -)
activity(ex:a10, -, -, [prov:type="edit"])
activity(ex:a10, -, 2011-11-16T16:00:00)
activity(ex:a10, 2011-11-16T16:00:00, -)
activity(ex:a10, 2011-11-16T16:00:00, -, [prov:type="createFile"])
activity(ex:a10, [prov:type="edit"])

Generation

The following table summarizes how each constituent of a PROV-DM Generation maps to a PROV-N syntax element.

GenerationNon-Terminal
idoptionalIdentifier
entityeIdentifier
activityaIdentifierOrMarker
timetimeOrMarker
attributesoptionalAttributeValuePairs
wasGeneratedBy(ex:g1; tr:WD-prov-dm-20111215, ex:edit1, 2011-11-16T16:00:00,  [ex:fct="save"])

Here ex:g1 is the optional generation identifier, tr:WD-prov-dm-20111215 is the identifier of the entity being generated, ex:edit1 is the optional identifier of the generating activity, 2011-11-16T16:00:00 is the optional generation time, and [ex:fct="save"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasGeneratedBy(e2, a1, -)
wasGeneratedBy(e2, a1, 2011-11-16T16:00:00)
wasGeneratedBy(e2, a1, -, [ex:fct="save"])     
wasGeneratedBy(e2, [ex:fct="save"])     
wasGeneratedBy(ex:g1; e)
wasGeneratedBy(ex:g1; e, a, tr:WD-prov-dm-20111215)

Additional semantic rules (Section 3.7.5) apply to generationExpression.

Usage

The following table summarizes how each constituent of a PROV-DM Usage maps to a PROV-N syntax element.

UsageNon-Terminal
idoptionalIdentifier
activityaIdentifier
entityeIdentifierOrMarker
timetimeOrMarker
attributesoptionalAttributeValuePairs
used(ex:u1; ex:act2, ar3:0111, 2011-11-16T16:00:00, [ex:fct="load"])
 

Here ex:u1 is the optional usage identifier, ex:act2 is the identifier of the using activity, ar3:0111 is the identifier of the entity being used, 2011-11-16T16:00:00 is the optional usage time, and [ex:fct="load"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
used(ex:act2)
used(ex:act2, ar3:0111, 2011-11-16T16:00:00)
used(a1,e1, -, [ex:fct="load"])
used(ex:u1; ex:act2, ar3:0111, -)

Additional semantic rules (Section 3.7.5) apply to usageExpression.

Communication

The following table summarizes how each constituent of a PROV-DM Communication maps to a PROV-N syntax element.

CommunicationNon-Terminal
idoptionalIdentifier
informedaIdentifier
informantaIdentifier
attributesoptionalAttributeValuePairs
wasInformedBy(ex:inf1; ex:a1, ex:a2, [ex:param1="a", ex:param2="b"])

Here ex:inf1 is the optional communication identifier, ex:a1 is the identifier of the informed activity, ex:a2 is the identifier of the informant activity, and [ex:param1="a", ex:param2="b"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasInformedBy(ex:a1, ex:a2)
wasInformedBy(ex:a1, ex:a2, [ex:param1="a", ex:param2="b"])
wasInformedBy(ex:i; ex:a1, ex:a2)
wasInformedBy(ex:i; ex:a1, ex:a2, [ex:param1="a", ex:param2="b"])

Start

The following table summarizes how each constituent of a PROV-DM Start maps to a PROV-N syntax element.

StartNon-Terminal
idoptionalIdentifier
activityaIdentifier
triggereIdentifierOrMarker
starteraIdentifierOrMarker
timetimeOrMarker
attributesoptionalAttributeValuePairs
wasStartedBy(ex:start; ex:act2, ex:trigger, ex:act1, 2011-11-16T16:00:00, [ex:param="a"])
  

Here start is the optional start identifier, ex:act2 is the identifier of the started activity, ex:trigger is the optional identifier for the entity that triggered the activity start, ex:act1 is the optional identifier for the activity that generated the (possibly unspecified) entity ex:trigger, 2011-11-16T16:00:00 is the optional start time, and [ex:param="a"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasStartedBy(ex:act2, -, ex:act1, -)
wasStartedBy(ex:act2, -, ex:act1, 2011-11-16T16:00:00)
wasStartedBy(ex:act2, -, -, 2011-11-16T16:00:00)
wasStartedBy(ex:act2, [ex:param="a"])
wasStartedBy(ex:start; ex:act2, e, ex:act1, 2011-11-16T16:00:00)

Additional semantic rules (Section 3.7.5) apply to startExpression.

End

The following table summarizes how each constituent of a PROV-DM End maps to a PROV-N syntax element.

EndNon-Terminal
idoptionalIdentifier
activityaIdentifier
triggereIdentifierOrMarker
enderaIdentifierOrMarker
timetimeOrMarker
attributesoptionalAttributeValuePairs
wasEndedBy(ex:end; ex:act2, ex:trigger, ex:act3,  2011-11-16T16:00:00, [ex:param="a"])
  

Here end is the optional end identifier, ex:act2 is the identifier of the ending activity, ex:trigger is the identifier of the entity that triggered the activity end, ex:act3 is the optional identifier for the activity that generated the (possibly unspecified) entity ex:trigger, 2011-11-16T16:00:00 is the optional usage time, and [ex:param="a"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasEndedBy(ex:act2, ex:trigger, -, -)
wasEndedBy(ex:act2, ex:trigger, -, 2011-11-16T16:00:00)
wasEndedBy(ex:act2, -, -, 2011-11-16T16:00:00)
wasEndedBy(ex:act2, -, -, 2011-11-16T16:00:00, [ex:param="a"])
wasEndedBy(ex:end; ex:act2)
wasEndedBy(ex:end; ex:act2, ex:trigger, -, 2011-11-16T16:00:00)

Additional semantic rules (Section 3.7.5) apply to endExpression.

Invalidation

The following table summarizes how each constituent of a PROV-DM Invalidation maps to a PROV-N syntax element.

InvalidationNon-Terminal
idoptionalIdentifier
entityeIdentifier
activityaIdentifierOrMarker
timetimeOrMarker
attributesoptionalAttributeValuePairs
wasInvalidatedBy(ex:inv; tr:WD-prov-dm-20111215, ex:edit1, 2011-11-16T16:00:00,  [ex:fct="save"])

Here ex:inv is the optional invalidation identifier, tr:WD-prov-dm-20111215 is the identifier of the entity being invalidated, ex:edit1 is the optional identifier of the invalidating activity, 2011-11-16T16:00:00 is the optional invalidation time, and [ex:fct="save"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasInvalidatedBy(tr:WD-prov-dm-20111215, ex:edit1, -)
wasInvalidatedBy(tr:WD-prov-dm-20111215, ex:edit1, 2011-11-16T16:00:00)
wasInvalidatedBy(e2, a1, -, [ex:fct="save"])     
wasInvalidatedBy(e2, -, -, [ex:fct="save"])     
wasInvalidatedBy(ex:inv; tr:WD-prov-dm-20111215, ex:edit1, -)
wasInvalidatedBy(tr:WD-prov-dm-20111215, ex:edit1, -)

Additional semantic rules (Section 3.7.5) apply to invalidationExpression.

Component 2: Derivations

Derivation

The following table summarizes how each constituent of a PROV-DM Derivation maps to a PROV-N syntax element.

DerivationNon-Terminal
idoptionalIdentifier
generatedEntityeIdentifier
usedEntityeIdentifier
activityaIdentifierOrMarker
generationgIdentifierOrMarker
usageuIdentifierOrMarker
attributesoptionalAttributeValuePairs
wasDerivedFrom(ex:d; e2, e1, a, g2, u1, [ex:comment="a righteous derivation"])
  

Here d is the optional derivation identifier, e2 is the identifier for the entity being derived, e1 is the identifier of the entity from which e2 is derived, a is the optional identifier of the activity which used/generated the entities, g2 is the optional identifier of the generation, u1 is the optional identifier of the usage, and [ex:comment="a righteous derivation"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasDerivedFrom(e2, e1)
wasDerivedFrom(e2, e1, a, g2, u1)
wasDerivedFrom(e2, e1, -, g2, u1)
wasDerivedFrom(e2, e1, a, -, u1)
wasDerivedFrom(e2, e1, a, g2, -)
wasDerivedFrom(e2, e1, a, -, -)
wasDerivedFrom(e2, e1, -, -, u1)
wasDerivedFrom(e2, e1, -, -, -)
wasDerivedFrom(ex:d; e2, e1, a, g2, u1)
wasDerivedFrom(-; e2, e1, a, g2, u1)

Revision

PROV-N provides no dedicated syntax for Revision. Instead, a Revision MUST be expressed as a derivationExpression with attribute prov:type='prov:Revision'.

wasDerivedFrom(ex:d; e2, e1, a, g2, u1, 
               [prov:type='prov:Revision', 
                ex:comment="a righteous derivation"])
  

Here, the derivation from Example REF is extended with a prov:type attribute and value prov:Revision. The expression 'prov:Revision' is convenienceNotation to denote a QUALIFIED_NAME literal (See Section 3.7.3. Literal).

Quotation

PROV-N provides no dedicated syntax for Quotation. Instead, a Quotation MUST be expressed as a derivationExpression with attribute prov:type='prov:Quotation'.

wasDerivedFrom(ex:quoteId1; ex:blockQuote,ex:blog, ex:act1, ex:g, ex:u,
               [ prov:type='prov:Quotation' ]) 
 

Here, the derivation is provided with a prov:type attribute and value prov:Quotation.

Primary Source

PROV-N provides no dedicated syntax for PrimarySource. Instead, a PrimarySource MUST be expressed as a derivationExpression with attribute prov:type='prov:Primary-Source'.

wasDerivedFrom(ex:sourceId1;  ex:e1, ex:e2, ex:act, ex:g, ex:u, 
               [ prov:type='prov:PrimarySource' ])

Here, the derivation is provided with a prov:type attribute and value prov:PrimarySource.

Component 3: Agents, Responsibility, and Influence

Agent

The following table summarizes how each constituent of a PROV-DM Agent maps to a PROV-N syntax element.

AgentNon-Terminal
ididentifier
attributesoptionalAttributeValuePairs

PROV-N provides no dedicated syntax for Person, Organization, SoftwareAgent. Instead, a Person, an Organization, or a SoftwareAgent MUST be expressed as an agentExpression with attribute prov:type='prov:Person', prov:type='prov:Organization', or prov:type='prov:SoftwareAgent', respectively.

agent(ex:ag4, [ prov:type='prov:Person', ex:name="David" ])

Here ag is the agent identifier, and [ prov:type='prov:Person', ex:name="David" ] is a list of optional attributes.

In the next example, the optional attributes are omitted.
agent(ex:ag4)

Attribution

The following table summarizes how each constituent of a PROV-DM Attribution maps to a PROV-N syntax element.

AttributionNon-Terminal
idoptionalIdentifier
entityeIdentifier
agentagIdentifier
attributesoptionalAttributeValuePairs
wasAttributedTo(ex:attr; e, ag, [ex:license='cc:attributionURL' ])
  

Here attr is the optional attribution identifier, e is an entity identifier, ag is the identifier of the agent to whom the entity is ascribed, and [ex:license='cc:attributionURL' ] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasAttributedTo(e, ag)
wasAttributedTo(e, ag, [ex:license='cc:attributionURL' ])

Association

The following table summarizes how each constituent of a PROV-DM Association maps to a PROV-N syntax element.

AssociationNon-Terminal
idoptionalIdentifier
activityaIdentifier
agentagIdentifierOrMarker
planeIdentifierOrMarker
attributesoptionalAttributeValuePairs

PROV-N provides no dedicated syntax for Plan. Instead, a Plan MUST be expressed as an entityExpression with attribute prov:type='prov:Plan'.

wasAssociatedWith(ex:assoc; ex:a1, ex:ag1, ex:e1, [ex:param1="a", ex:param2="b"])
  

Here ex:assoc is the optional attribution identifier, ex:a1 is an activity identifier, ex:ag1 is the optional identifier of the agent associated to the activity, ex:e1 is the optional identifier of the plan used by the agent in the context of the activity, and [ex:param1="a", ex:param2="b"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
wasAssociatedWith(ex:a1, -, ex:e1)
wasAssociatedWith(ex:a1, ex:ag1)
wasAssociatedWith(ex:a1, ex:ag1, ex:e1)
wasAssociatedWith(ex:a1, ex:ag1, ex:e1, [ex:param1="a", ex:param2="b"])
wasAssociatedWith(ex:assoc; ex:a1, -, ex:e1)

Additional semantic rules (Section 3.7.5) apply to associationExpression.

The following expression illustrates a plan.

  entity(ex:pl1,   [ prov:type='prov:Plan' ])

Delegation

The following table summarizes how each constituent of a PROV-DM Delegation maps to a PROV-N syntax element.

DelegationNon-Terminal
idoptionalIdentifier
delegateagIdentifier
responsibleagIdentifier
activityaIdentifierOrMarker
attributesoptionalAttributeValuePairs
actedOnBehalfOf(ex:del1; ex:ag2, ex:ag1, ex:a, [prov:type="contract"])
  

Here ex:del1 is the optional delegation identifier, ex:ag2 is the identifier for the delegate agent, ex:ag1 is the identifier of the responsible agent, ex:a is the optional identifier of the activity for which the delegation link holds, and [prov:type="contract"] is a list of optional attributes.

The remaining examples show cases where some of the optionals are omitted.
actedOnBehalfOf(ex:ag1, ex:ag2)
actedOnBehalfOf(ex:ag1, ex:ag2, ex:a)
actedOnBehalfOf(ex:ag1, ex:ag2, -, [prov:type="delegation"])
actedOnBehalfOf(ex:ag2, ex:ag3, ex:a, [prov:type="contract"])
actedOnBehalfOf(ex:del1; ex:ag2, ex:ag3, ex:a, [prov:type="contract"])

Influence

The following table summarizes how each constituent of a PROV-DM Influence maps to a PROV-N syntax element.

InfluenceNon-Terminal
idoptionalIdentifier
influenceeeIdentifier
influencereIdentifier
attributesoptionalAttributeValuePairs
wasInfluencedBy(ex:infl1;e2,e1,[ex:param="a"])
  

Here, ex:infl1 is the optional influence identifier, ex:e2 is an entity identifier, ex:e1 is the identifier for an ancestor entity that ex:e2 is influenced by, and [ex:param="a"] is the optional set of attributes.

The remaining examples show cases where some of the optionals are omitted.
wasInfluencedBy(ex:e2,ex:e1)
wasInfluencedBy(ex:e2,ex:e1,[ex:param="a"])
wasInfluencedBy(ex:infl1; ex:e2,ex:e1)

Component 4: Bundles

Bundle Constructor

Bundles cannot be nested. It is for this reason that a bundle is not defined as an expression, to prevent the occurrence of a bundle inside another bundle.

Each identifier occurring in a bundle, including the bundle identifier itself, MUST be interpreted with respect to the namespace declarations of that bundle, or if the identifier's prefix is not declared in the bundle, with respect to the namespace declarations in the document.

The following table summarizes how each constituent of a PROV-DM bundle maps to a PROV-N syntax element.

BundleNon-Terminal
ididentifier
descriptionsexpression
namespaceDeclarationnamespaceDeclarations
bundle ex:author-view
  prefix ex  <http://example.org/>
  agent(ex:Paolo,   [ prov:type='prov:Person' ])
  agent(ex:Simon,   [ prov:type='prov:Person' ])
  //...
endBundle
  

Here ex:author-view is the name of the bundle.

Bundle Type

When described, a Bundle MUST be expressed as an entityExpression with attribute prov:type='prov:Bundle'.

The bundle of Example REF can be referred to as an entity, and its provenance described.

  entity(ex:author-view,   [ prov:type='prov:Bundle' ])

Component 5: Alternate Entities

Alternate

The following table summarizes how each constituent of a PROV-DM Alternate maps to a PROV-N syntax element.

AlternateNon-Terminal
alternate1eIdentifier
alternate2eIdentifier
alternateOf(tr:WD-prov-dm-20111215,ex:alternate-20111215)
Here tr:WD-prov-dm-20111215 is alternate for ex:alternate-20111215.

Specialization

The following table summarizes how each constituent of a PROV-DM Specialization maps to a PROV-N syntax element.

SpecializationNon-Terminal
specificEntityeIdentifier
generalEntityeIdentifier
specializationOf(tr:WD-prov-dm-20111215,tr:prov-dm)
Here tr:WD-prov-dm-20111215 is a specialization of tr:prov-dm.

Component 6: Collections

Collection

PROV-N provides no dedicated syntax for Collection and EmptyCollection. Instead, a Collection or an EmptyCollection MUST be expressed as an entityExpression with attribute prov:type='prov:Collection', or prov:type='prov:EmptyCollection', respectively.

The following two expressions are about a collection and an empty collection, respectively.

  entity(ex:col1,   [ prov:type='prov:Collection' ])
  entity(ex:col2,   [ prov:type='prov:EmptyCollection' ])

Membership

The following table summarizes how each constituent of a PROV-DM Membership maps to a PROV-N syntax element.

MembershipNon-Terminal
collectioncIdentifier
entityeIdentifier
   hadMember(ex:c, ex:e1)   // ex:c contained ex:e1
   hadMember(ex:c, ex:e2)   // ex:c contained ex:e2 
  

Here ex:c is the identifier for the collection whose membership is stated, and ex:e1 and ex:e2 are the entities that are members of collection ex:c.

Further Expressions

This section defines further expressions of PROV-N.

Identifier

Various kinds of identifiers are used in productions.

A qualified name is a name subject to namespace interpretation. It consists of a namespace, denoted by an optional prefix, and a local name. The PROV data model stipulates that a qualified name can be mapped to an IRI by concatenating the IRI associated with the prefix and the local part. This section provides the exact details of this procedure for qualified names defined by PROV-N.

A qualified name's prefix is OPTIONAL. If a prefix occurs in a qualified name, the prefix MUST refer to a namespace declared in a namespace declaration. In the absence of prefix, the qualified name belongs to the default namespace.

A PROV-N qualified name (production QUALIFIED_NAME) has a more permissive syntax than XML's QName [[!XML-NAMES]] and SPARQL PrefixedName [[!RDF-SPARQL-QUERY]]. A QUALIFIED_NAME consists of a prefix and a local part. Prefixes follow the production PN_PREFIX defined by SPARQL [[!RDF-SPARQL-QUERY]]. Local parts have to be conformant with PN_LOCAL, which extends the original SPARQL PN_LOCAL definition by allowing further characters (see PN_CHARS_OTHERS):

  • an extra set of characters commonly encountered in IRIs;
  • %-escaped characters (see PERCENT) to be interpreted as per Section 3.1. Mapping of IRIs to URIs in [[!RFC3987]];
  • and \-escaped characters (see PN_CHARS_ESC).

Given that '=' (equal), ''' (single quote), '(' (left bracket), ')' (right bracket), ',' (comma), ':' (colon), ';' (semi-colon), '"' (double quote), '[' (left square bracket), ']' (right square bracket) are used by the PROV notation as delimiters, they are not allowed in local parts. Instead, among those characters, those that are permitted in SPARQL IRI_REF are also allowed in PN_LOCAL if they are escaped by the '\' (backslash character) as per production PN_CHARS_ESC. Furthermore, '.' (dot), ':' (colon), '-' (hyphen) can also be \-escaped.

A PROV-N qualified name QUALIFIED_NAME can be mapped to a valid IRI [[!RFC3987]] by concatenating the namespace denoted its local name PN_PREFIX to the local name PN_LOCAL, whose \-escaped characters have been unescaped by dropping the character '\' (backslash).

Examples of articles on the BBC Web site seen as entities.

document
  prefix bbc <http://www.bbc.co.uk/>
  prefix bbcNews <http://www.bbc.co.uk/news/>

  entity(bbc:)                          // bbc site itself
  entity(bbc:news/)                     // bbc news
  entity(bbc:news/world-asia-17507976)  // a given news article

  entity(bbcNews:)                      // an alternative way of referring to the bbc news site

endDocument

Examples of entities with declared and default namespace.

document
  default <http://example.org/2/>
  prefix ex <http://example.org/1/>

  entity(ex:a)     //  corresponds to IRI http://example.org/1/a
  entity(ex:a/)    //  corresponds to IRI http://example.org/1/a/
  entity(ex:a/b)   //  corresponds to IRI http://example.org/1/a/b
  entity(b)        //  corresponds to IRI http://example.org/2/b
  entity(ex:1234)  //  corresponds to IRI http://example.org/1/1234
  entity(4567)     //  corresponds to IRI http://example.org/2/4567
  entity(c/)       //  corresponds to IRI http://example.org/2/c/
  entity(ex:/)     //  corresponds to IRI http://example.org/1//
endDocument

Examples of \-escaped characters.

document
  prefix ex <http://example.org/>
  default <http://example.org/default>

  entity(ex:foo?a\=1)            // corresponds to IRI http://example.org/foo?a=1
  entity(ex:\-)                  // corresponds to IRI http://example.org/-
  entity(ex:?fred\=fish%20soup)  // corresponds to IRI http://example.org/?fred=fish%20soup

  used(-;a1,e1,-)                // identifier not specified for usage  
  used(\-;a1,e1,-)               // usage identifier corresponds to http://example.org/default-
endDocument

Note: The productions for the terminals QUALIFIED_NAME and PN_PREFIX are conflicting. Indeed, for a tokenizer operating independently of the parse tree, abc matches both QUALIFIED_NAME and PN_PREFIX. In the context of a namespaceDeclaration, a tokenizer should give preference to the production PN_PREFIX.

Attribute

The reserved attributes in the PROV namespace are the following. Their meaning is explained by [[PROV-DM]] (see Section 5.7.2: Attribute).

  1. prov:label
  2. prov:location
  3. prov:role
  4. prov:type
  5. prov:value

Literal

In production datatype, the QUALIFIED_NAME is used to denote a PROV data type [[PROV-DM]].

The non terminals STRING_LITERAL, INT_LITERAL, and QUALIFIED_NAME_LITERAL are syntactic sugar for quoted strings with datatype xsd:string, xsd:int, and prov:QUALIFIED_NAME respectively.

In particular, a Literal may be an IRI-typed string (with datatype xsd:anyURI); such IRI has no specific interpretation in the context of PROV.

Note: The productions for terminals QUALIFIED_NAME and INT_LITERAL are conflicting. Indeed, for a tokenizer operating independently of the parse tree, 1234 matches both INT_LITERAL and QUALIFIED_NAME (local name without prefix). In the context of a convenienceNotation, a tokenizer should give preference to the production INT_LITERAL.

The following examples illustrate convenience notations.

The two following expressions are strings; if datatype is not specified, it is xsd:string.

  "abc" %% xsd:string
  "abc"

The two following expressions are integers. For convenience, numbers, expressed as digits optionally preceded by a minus sign, can occur without quotes.

  "1234" %% xsd:integer
  1234
  "-1234" %% xsd:integer
  -1234

The two following expressions are qualified names. Values of type qualified name can be conveniently expressed within single quotes.

  "ex:value" %% prov:QUALIFIED_NAME
  'ex:value'

The following examples respectively are the string "abc", the string (in French) "bonjour", the integer number 1, and the IRI "http://example.org/foo".

  "abc"
  "bonjour"@fr
  "1" %% xsd:integer
  "http://example.org/foo" %% xsd:anyURI

The following examples respectively are the floating point number 1.01 and the boolean true.

  "1.01" %% xsd:float
  "true" %% xsd:boolean

Reserved Type Values

The reserved type values in the PROV namespace are the following. Their meaning is defined [[PROV-DM]] (see Section 5.7.2.4: prov:type).

  1. prov:Bundle
  2. prov:Collection
  3. prov:EmptyCollection
  4. prov:Organization
  5. prov:Person
  6. prov:Plan
  7. prov:PrimarySource
  8. prov:Quotation
  9. prov:Revision
  10. prov:SoftwareAgent

The agent ag is a person (type: prov:Person), whereas the entity pl is a plan (type: prov:Plan).

agent(ag,  [ prov:type='prov:Person' ])
entity(pl, [ prov:type='prov:Plan' ])

Time Values

Time instants are defined according to xsd:dateTime [[!XMLSCHEMA11-2]].

The third argument in the following usage expression is a time instance, namely 4pm on 2011-11-16.

used(ex:act2, ar3:0111, 2011-11-16T16:00:00)

Namespace Declaration

A namespaceDeclaration consists of a binding between a prefix and a namespace. Every qualified name with this prefix in the scope of this declaration belongs to this namespace. A defaultNamespaceDeclaration consists of a namespace. Every qualified name without prefix in the scope of this declaration belongs to this namespace. Scope of a prefix-namespace declaration is specified as follows:

  • The scope of a prefix-namespace declaration directly occurring in a bundle is the bundle itself.
  • The scope of a prefix-namespace declaration directly occurring in a document is the document including the bundles it contains but excluding those bundles that re-declare this prefix.

A set of namespace declarations namespaceDeclarations MUST NOT re-declare the same prefix.

A set of namespace declarations namespaceDeclarations occurring in a bundle MAY re-declare a prefix declared in a surrounding document.

A namespace declaration namespaceDeclaration MUST NOT declare prefixes prov and xsd (see Table 1 for their IRI).

The following example declares three namespaces, one default, and two with explicit prefixes ex1 and ex2.

document
  default <http://example.org/0/>
  prefix ex1 <http://example.org/1/>
  prefix ex2 <http://example.org/2/>
...
endDocument

In the following example, a document declares a default namespace and the occurrence of e001 directly occurring in the document refers to that namespace. A nested bundle also declares a default namespace, but with a different IRI. In that bundle, the occurrences of e001, including the bundle name, refer to the latest default namespace.

document
  default <http://example.org/1/>
  entity(e001)                             // IRI: http://example.org/1/e001

  bundle e001                              // IRI: http://example.org/2/e001
    default <http://example.org/2/>
    entity(e001)                           // IRI: http://example.org/2/e001
  endBundle
endDocument

In the following example, a document declares a namespace with prefix ex and the occurrence of ex:e001 directly occurring in the document refers to that namespace. In a nested bundle, the occurrence of ex:e001 also refers to the same namespace.

document
  prefix ex <http://example.org/1/>
  entity(ex:e001)                            // IRI: http://example.org/1/e001

  bundle b
    entity(ex:001)                           // IRI: http://example.org/1/e001
  endBundle
endDocument

Summary of additional semantic rules

Some of the grammar productions allow for expressions that are syntactically correct, and yet according to [[PROV-DM]] they are not acceptable, because additional semantic rules are defined for those expressions. The following table provides a summary of such expressions along with examples of syntactically correct but unacceptable forms, and the additional semantic rules.

Table 2: Summary of additional semantic rules for grammar productions
Production Examples of syntactically correct expressions Additional semantic rule
Generation expression wasGeneratedBy(e2, -, -)
wasGeneratedBy(-; e2, -, -)
At least one of id, activity, time, and attributes MUST be present.
Usage expression used(a2, -, -)
used(-; a2, -, -)
At least one of id, entity, time, and attributes MUST be present
Start expression wasStartedBy(e2, -, -, -)
wasStartedBy(-; e2, -, -, -)
At least one of id, trigger, starter, time, and attributes MUST be present
End expression wasEndedBy(e2, -, -, -)
wasEndedBy(-; e2, -, -, -)
At least one of id, trigger, ender, time, and attributes MUST be present
Invalidation expression wasInvalidatedBy(e2, -, -)
wasInvalidatedBy(-; e2, -, -)
At least one of id, activity, time, and attributes MUST be present
Association expression wasAssociatedWith(a, -, -)
wasAssociatedWith(-; a, -, -)
At least one of id, agent, plan, and attributes MUST be present

Document

A document is a house-keeping construct of PROV-N capable of packaging up PROV-N expressions and namespace declarations. A document forms a self-contained package of provenance descriptions for the purpose of exchanging them. A document may be used to package up PROV-N expressions in response to a request for the provenance of something ([[PROV-AQ]]).

Given its status of house-keeping construct for the purpose of exchanging provenance expressions, a document is not defined as a PROV-N expression (production expression).

A document's text matches the document production.

A document contains:

Thus, bundles MAY occur inside a document, but do not appear inside other bundles.

The following document contains expressions related to the provenance of entity e2.

document
  default <http://anotherexample.org/>
  prefix ex <http://example.org/>

  entity(e2, [ prov:type="File", ex:path="/shared/crime.txt", ex:creator="Alice", 
               ex:content="There was a lot of crime in London last month."])
  activity(a1, 2011-11-16T16:05:00, -, [prov:type="edit"])
  wasGeneratedBy(e2, a1, -, [ex:fct="save"])     
  wasAssociatedWith(a1, ag2, -, [prov:role="author"])
  agent(ag2, [ prov:type='prov:Person', ex:name="Bob" ])

endDocument

This container could, for instance, be returned as the result of a query to a provenance store for the provenance of entity e2 [[PROV-AQ]].

Extensibility

The PROV data model is extensible by means of attributes prov:type and prov:role allowing subtyping of expressions. For some applications, novel syntax may also be convenient. Hence, the normative requirements are as follow.

Expressions compatible with the extensibilityExpression production follow a general form of functional syntax, in which the predicate MUST be a qualifiedName with a non-empty prefix.

Collections are sets of entities, whose membership can be expressed using the hadMember relation. The following example shows how one can express membership for dictionaries, an illustrative extension of Collections consisting of sets of key-entity pairs, where a key is a literal. The notation is a variation of that used for Collections membership, allowing multiple member elements to be declared, and in which the elements are pairs. The name of the relation is qualified with the extension-specific namespace http://example.org/dictionaries.

  prefix dictExt <http://example.org/dictionaries#>
  dictExt:hadMembers(mId; d, {("k1",e1), ("k2",e2), ("k3",e3)}, []) 
Note that the generic extensibilityExpression production above allows for alternative notations to be used for expressing membership, if the designers of the extensions so desire. Here is an alternate syntax that is consistent with the productions:
  prefix dictExt <http://example.org/dictionaries#>
  dictExt:hadMembers(mid; d, dictExt:set(dictExt:pair("k1",e1), 
                                         dictExt:pair("k2",e2), 
                                         dictExt:pair("k3",e3)), 
                            [dictExt:uniqueKeys="true"]) 

Media Type

The media type of PROV-N is text/provenance-notation. The content encoding of PROV-N content is UTF-8.

Contact:
Ivan Herman
See also:
How to Register a Media Type for a W3C Specification
Internet Media Type registration, consistency of use
TAG Finding 3 June 2002 (Revised 4 September 2002)

The Internet Media Type / MIME Type for PROV-N is "text/provenance-notation".

It is recommended that PROV-N files have the extension ".provn" (all lowercase) on all platforms.

It is recommended that PROV-N files stored on Macintosh HFS file systems be given a file type of "TEXT".

The information that follows has been registered with IANA.

Type name:
text
Subtype name:
provenance-notation
Required parameters:
charset — the value of charset MUST always be UTF-8.
Optional parameters:
None
Encoding considerations: 8bit
The syntax of PROV-N is expressed over code points in Unicode [[!UNICODE]]. The encoding is always UTF-8 [[!UTF-8]].
Unicode code points may also be expressed using an \uXXXX (U+0 to U+FFFF) or \UXXXXXXXX syntax (for U+10000 onwards) where X is a hexadecimal digit [0-9A-F]
Security considerations:
PROV-N is a general-purpose language for describing the provenance of things; applications may evaluate given data to infer more descriptions or to dereference URIs, invoking the security considerations of the scheme for that URI. Note in particular, the privacy issues in [[!RFC3023]] section 10 for HTTP URIs. Data obtained from an inaccurate or malicious data source may lead to inaccurate or misleading conclusions, as well as the dereferencing of unintended URIs. Care must be taken to align the trust in consulted resources with the sensitivity of the intended use of the data.
PROV-N is used to express the provenance of arbitrary application data; security considerations will vary by domain of use. Security tools and protocols applicable to text (e.g. PGP encryption, MD5 sum validation, password-protected compression) may also be used on PROV-N documents. Security/privacy protocols must be imposed which reflect the sensitivity of the embedded information.
PROV-N can express data which is presented to the user, for example, by means of label attributes. Application rendering strings retrieved from untrusted PROV-N documents must ensure that malignant strings may not be used to mislead the reader. The security considerations in the media type registration for XML ([[!RFC3023]] section 10) provide additional guidance around the expression of arbitrary data and markup.
PROV-N is a language for describing the provenance of things, and therefore a PROV-N document is metadata for other resources. Untrusted PROV-N documents may mislead its consumers by indicating that a third-party resource has a reputable lineage, when it has not. Provenance of PROV-N document should be sought.
PROV-N uses qualified names mappable to IRIs as term identifiers. Applications interpreting data expressed in PROV-N should address the security issues of Internationalized Resource Identifiers (IRIs) [[!RFC3987]] Section 8, as well as Uniform Resource Identifier (URI): Generic Syntax [[!RFC3986]] Section 7.
Multiple IRIs may have the same appearance. Characters in different scripts may look similar (a Cyrillic "о" may appear similar to a Latin "o"). A character followed by combining characters may have the same visual representation as another character (LATIN SMALL LETTER E followed by COMBINING ACUTE ACCENT has the same visual representation as LATIN SMALL LETTER E WITH ACUTE). Any person or application that is writing or interpreting data in PROV-N must take care to use the IRI that matches the intended semantics, and avoid IRIs that make look similar. Further information about matching of similar characters can be found in Unicode Security Considerations [[UNISEC]] and Internationalized Resource Identifiers (IRIs) [[!RFC3987]] Section 8.
PROV-N offers an extensibility mechanism, which in turn may introduce additional security considerations. For example, predicates in extensibility expressions use qualified names, mappable to IRIs, and appropriate security considerations for IRIs apply too.
Interoperability considerations:
There are no known interoperability issues.
Published specification:
PROV-N: The Provenance Notation, Moreau, Missier, (eds), Cheney, Soiland-Reyes http://www.w3.org/TR/prov-n/, 2012.
Applications which use this media type:
It may be used by any application for publishing provenance information. This format is designed to be a human-readable form of provenance.
Additional information:
Magic number(s):
PROV-N documents may have the strings 'document' near the beginning of the document.
File extension(s):
".provn"
Base URI:
There are no constructs in the PROV-N Syntax to change the Base IRI.
Macintosh file type code(s):
"TEXT"
Person & email address to contact for further information:
Ivan Herman, ivan@w3.org
Intended usage:
COMMON
Restrictions on usage:
None
Author/Change controller:
The PROV-N specification is the product of the World Wide Web Consortium's Provenance Working Group. The W3C has change control over this specification.

Change Log

Changes since Proposed Recommendation

Changes since Candidate Recommendation

Changes since last call

Please see the Responses to Public Comments on the Last Call Working Draft for more details about the justification of these changes.

Acknowledgements

This document has been produced by the Provenance Working Group, and its contents reflect extensive discussion within the Working Group as a whole. The editors extend special thanks to Sandro Hawke (W3C/MIT) and Ivan Herman (W3C/ERCIM), W3C contacts for the Provenance Working Group.

The editors acknowledge valuable contributions from the following: Tom Baker, David Booth, Robert Freimuth, Satrajit Ghosh, Ralph Hodgson, Renato Iannella, Jacek Kopecky, James Leigh, Jacco van Ossenbruggen, Alan Ruttenberg, Reza Samavi, and Antoine Zimmermann.

Members of the Provenance Working Group at the time of publication of this document were: Ilkay Altintas (Invited expert), Reza B'Far (Oracle Corporation), Khalid Belhajjame (University of Manchester), James Cheney (University of Edinburgh, School of Informatics), Sam Coppens (iMinds - Ghent University), David Corsar (University of Aberdeen, Computing Science), Stephen Cresswell (The National Archives), Tom De Nies (iMinds - Ghent University), Helena Deus (DERI Galway at the National University of Ireland, Galway, Ireland), Simon Dobson (Invited expert), Martin Doerr (Foundation for Research and Technology - Hellas(FORTH)), Kai Eckert (Invited expert), Jean-Pierre EVAIN (European Broadcasting Union, EBU-UER), James Frew (Invited expert), Irini Fundulaki (Foundation for Research and Technology - Hellas(FORTH)), Daniel Garijo (Universidad Politécnica de Madrid), Yolanda Gil (Invited expert), Ryan Golden (Oracle Corporation), Paul Groth (Vrije Universiteit), Olaf Hartig (Invited expert), David Hau (National Cancer Institute, NCI), Sandro Hawke (W3C/MIT), Jörn Hees (German Research Center for Artificial Intelligence (DFKI) Gmbh), Ivan Herman, (W3C/ERCIM), Ralph Hodgson (TopQuadrant), Hook Hua (Invited expert), Trung Dong Huynh (University of Southampton), Graham Klyne (University of Oxford), Michael Lang (Revelytix, Inc.), Timothy Lebo (Rensselaer Polytechnic Institute), James McCusker (Rensselaer Polytechnic Institute), Deborah McGuinness (Rensselaer Polytechnic Institute), Simon Miles (Invited expert), Paolo Missier (School of Computing Science, Newcastle university), Luc Moreau (University of Southampton), James Myers (Rensselaer Polytechnic Institute), Vinh Nguyen (Wright State University), Edoardo Pignotti (University of Aberdeen, Computing Science), Paulo da Silva Pinheiro (Rensselaer Polytechnic Institute), Carl Reed (Open Geospatial Consortium), Adam Retter (Invited Expert), Christine Runnegar (Invited expert), Satya Sahoo (Invited expert), David Schaengold (Revelytix, Inc.), Daniel Schutzer (FSTC, Financial Services Technology Consortium), Yogesh Simmhan (Invited expert), Stian Soiland-Reyes (University of Manchester), Eric Stephan (Pacific Northwest National Laboratory), Linda Stewart (The National Archives), Ed Summers (Library of Congress), Maria Theodoridou (Foundation for Research and Technology - Hellas(FORTH)), Ted Thibodeau (OpenLink Software Inc.), Curt Tilmes (National Aeronautics and Space Administration), Craig Trim (IBM Corporation), Stephan Zednik (Rensselaer Polytechnic Institute), Jun Zhao (University of Oxford), Yuting Zhao (University of Aberdeen, Computing Science).