1EdTech Vocabulary Definition Exchange Information Model Version 1.0 Final Specification

Copyright © 2004 1EdTech Consortium, Inc. All Rights Reserved.
The 1EdTech Logo is a trademark of 1EdTech Consortium, Inc.
Document Name: 1EdTech Vocabulary Definition Exchange Information Model
Revision: 23 February 2004

IPR and Distribution Notices

Recipients of this document are requested to submit, with their comments, notification of any relevant patent claims or other intellectual property rights of which they may be aware that might be infringed by any implementation of the specification set forth in this document, and to provide supporting documentation.

1EdTech takes no position regarding the validity or scope of any intellectual property or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; neither does it represent that it has made any effort to identify any such rights. Information on 1EdTech's procedures with respect to rights in 1EdTech specifications can be found at the 1EdTech Intellectual Property Rights web page: http://www.imsglobal.org/ipr/imsipr_policyFinal.pdf.

Copyright © 2004 1EdTech Consortium. All Rights Reserved.

Permission is granted to all parties to use excerpts from this document as needed in producing requests for proposals.

Use of this specification to develop products or services is governed by the license with 1EdTech found on the 1EdTech website: http://www.imsglobal.org/license.html.

The limited permissions granted above are perpetual and will not be revoked by 1EdTech or its successors or assigns.

THIS SPECIFICATION IS BEING OFFERED WITHOUT ANY WARRANTY WHATSOEVER, AND IN PARTICULAR, ANY WARRANTY OF NONINFRINGEMENT IS EXPRESSLY DISCLAIMED. ANY USE OF THIS SPECIFICATION SHALL BE MADE ENTIRELY AT THE IMPLEMENTER'S OWN RISK, AND NEITHER THE CONSORTIUM, NOR ANY OF ITS MEMBERS OR SUBMITTERS, SHALL HAVE ANY LIABILITY WHATSOEVER TO ANY IMPLEMENTER OR THIRD PARTY FOR ANY DAMAGES OF ANY NATURE WHATSOEVER, DIRECTLY OR INDIRECTLY, ARISING FROM THE USE OF THIS SPECIFICATION.

Table of Contents

1. Introduction
     1.1 Overview and Relationship to Other Specifications
     1.2 1EdTech Vocabulary Definition Exchange Specification Components
           1.2.1 Information Model (this document)
           1.2.2 XML Binding
           1.2.3 Best Practice and Implementation Guide
           1.2.4 Conformance Requirements
     1.3 Nomenclature
           1.3.1 Abbreviated Terms
           1.3.2 Definitions
     1.4 References

2. Information Model
     2.1 Information Items Pertaining to the Vocabulary as a Whole
     2.2 Information Items Pertaining to Terms in the Vocabulary

3. VDEX Elements
     3.1 Langstring Elements
     3.2 Object Model

4. Semantics and Value Space for "Profile Type" Property of VDEX
Vocabularies
     4.1 Rationale and Introduction
     4.2 Definition of the Profile Types

5. Preferred Value Domains for Expressing Thesaurus Relationships

About This Document
     List of Contributors

Revision History

Index

1. Introduction

This section is not normative.

1.1 Overview and Relationship to Other Specifications

The 1EdTech Vocabulary Definition Exchange (VDEX) specification defines a grammar for the expression of value lists of various classes: collections often denoted "vocabulary". Section 1.3.2 provides some definitions of the classes of vocabulary that VDEX can be used for. The ISO prefers the terms "value domain" and "permitted value" for the data VDEX principally relates to.

VDEX does not seek to permit the expression of all possible vocabularies: rather to permit the light-weight expression of relatively simple value lists. Furthermore, VDEX is not a modelling language for vocabularies; there may be some properties of a term of interest to the vocabulary creator, an expert in a domain, but that are not required by the end user.

VDEX does permit the expression of simple machine-readable lists of human language terms together with information that may aid a human being in understanding the meaning or applicability of the various terms. Collections of such human language terms are often collected in the form of thesauri; VDEX enables thesauri to be expressed using arbitrary relationship types, inclusive of those specified in ISO 2788 and ISO 5964 standards.

VDEX may be also used to express valid data for use in instances of, for example: IEEE LOM, 1EdTech Meta-Data, 1EdTech Learner Information Package, ADL SCORM. In these specifications, the elements using value lists are often described as being of "vocabulary data type" and expressed through separate domain and value identifiers. The permitted values are formally codes to identify a concept and they may not be always be textually identical to human language terms

Whereas vocabulary data types typically have an associated unstructured list of terms, meta-data schemas often permit the expression of subject classifications from hierarchical vocabularies (taxonomies), for example Dewey Decimal (DDC) and specialist schemes such as the Medical Subject Headings (MeSH). VDEX can express strictly hierarchical schemes in a compact manner and allows for more loose networks of relationship to be expressed if required.

Polyhierarchical schemes and facetted schemes, such as Universal Decimal Classification (UDC), are a complex area. VDEX provides support for several use cases through the Information Model (this document) and makes some recommendations in the Best Practices and Implementation Guide. Experts in these fields with specialist use cases may need to apply techniques such as:

• defining a custom set of relationship types and using the VDEX model;
• using VDEX along side certain working practices;
• extending the VDEX model;
• embedding VDEX within a more elaborate model.

1.2 1EdTech Vocabulary Definition Exchange Specification Components

The VDEX documents permit simple collections of terms to be expressed and optionally exchanged. The specific types of terms are defined in section 1.3.2. This document is the 1EdTech Vocabulary Definition Exchange Specification version 1.0. As such it forms one of a set that comprise the specification, each with distinct scope:

1.2.1 Information Model (this document)

Describes the core aspects of the specification and contains parts that are normative for any binding claiming to use this information model. It contains details of: semantics, structure, data types, value spaces, multiplicity, and obligation (i.e., whether mandatory or optional).

1.2.2 XML Binding

Describes a binding of the Information Model to XML version 1.0 and is normative for any XML instance that claims to use this binding, whether by reference to the specification or by declaration of the namespace reserved by the specification. In cases of error or omission, the Information Model takes precedence. The VDEX XML Binding is released with a control document using W3C XML Schema 1.0 that should be used in implementations.

1.2.3 Best Practice and Implementation Guide

Provides non-normative guidance on application of the Information Model and XML Binding. This includes reference to existing practice in handling information that this specification seeks to support and guidance on practices that will promote interoperability and durability. It also includes examples to illustrate how the conceptual framework maps to practical uses and to identify the relationship between this specification and related 1EdTech specifications. Implementers are encouraged, but not required, to follow guidance in this part of the specification.

1.2.4 Conformance Requirements

Provides a set of testable statements that may be used in relation to applications of the information model and binding. These statements may form the basis of formal conformance testing and certification or informal assertions. This document makes no statement about the formal processes or method of certification or assertion; it only deals with criteria.

1.3 Nomenclature

1.3.1 Abbreviated Terms

1.3.2 Definitions

We adopt some widely accepted definitions for "vocabulary" and related terms. A useful collection of definitions is given in [CVLOM], drawing upon well-known authorities, and contains the following summary of various authorities:

Vocabulary:
"dictionary of terms related to a particular subject field." [ISO 5127/2-1983]1
"1.a A collection or list of words with brief explanations of their meanings; now esp. a list of this kind given in an elementary grammar or reading book of a foreign language. 2.a. The range of language of a particular person, class, profession, or the like. 3. The sum or aggregate of words composing a language."[OED]2


Classification:
"arrangement of concepts into classes (3) and their subdivisions to express the semantic relations between them; the classes are represented by means of a notation". [ISO 5127/6-1983]3
"1) The action of classifying or arranging in classes, according to common characteristics or affinities; assignment to the proper class. 2. The result of classifying; a systematic distribution, allocation, or arrangement, in the class or classes; esp. of things which form the subject-matter of a science or of a methodic enquiry."[OED]4


Taxonomy:
"1. Classification, esp. in relation to its general laws or principals; that department of science, or of a particular science or subject, which consists in or relates to classification; esp. the systematic classification of living things. 2. A classification of anything." [OED]5

 

Thesaurus:
"The vocabulary of a controlled indexing language, formally organized so that the a priori relationships between concepts (for example as broader and narrower) are made explicit." [ISO 2788:1986; and ISO 5964:1985] 6
"2.b. A collection of concepts or words arranged according to sense; also (U.S.) a dictionary of synonyms and antonyms. 2.c. A classified list of terms, esp. key-words, in a particular field, for us in indexing and information retrieval." 7


Glossary:
"dictionary of ancient, rare, little-known or new words in a given language." [ISO 5127/2-1983] 8
"A collection of glosses; a list with explanations of abstruse, antiquated, dialectical or technical terms; a partial dictionary." 9


Dictionary:
collection of words or a category of words from a language and explained or translated in one or more languages, arranged alphabetically or systematically. [ISO 5127/2-1983]10
"1.a. A book dealing with the individual words of a language (or certain specified classes of them), so as to set forth their orthography, pronunciation, signification, and use, their synonyms, derivation, and history, or at least some of these facts: for convenience of reference, the words are arranged in some stated order, now, in most languages, alphabetical; and in larger dictionaries the information given is illustrated by quotations from literature; a word-book, vocabulary of lexicon. 1.b. The vocabulary or whole list of words used or admitted by anyone. 2.a. By extension: A book of information or reference on any subject or branch of alphabetical knowledge, the items of which are arranged in alphabetical order; as an alphabetical encyclopaedia." [OED]11

 

1.4 References

2. Information Model

This section is normative. The terms defined in RFC 2119 are used with the meaning there defined.

The overall structure of the information model is shown in the following diagram and explained in the following subsections. The term structure may be repeated and nested. It should be noted that the ordering of elements at any level in the information model diagram and in the textual description of information model elements is not significant unless a statement is made to the contrary. NB: some bindings of the information model may require particular ordering.

2.1 Information Items Pertaining to the Vocabulary as a Whole

This section relates to properties of the collection of terms, the "value domain".

1. Identifier: A universally unique identifier controlled by registration authority or an unregistered identifier that the creator warrants to be persistent12 and takes reasonable care to make specific enough to avoid conflicts. The vocabulary identifier is formally a string but for exchange it is constrained to use the characters and encodings permitted by RFC 2396, as amended by RFC 2732, such that it is syntactically a URI. It must not contain a URI fragment reference. The VDEX Best Practices and Implementation Guide considers some application issues.
2. Registration Status: A simple flag to declare whether the identifier is registered or not. Meta-data could be used to identify the registration authority etc. Vocabularies are considered unregistered unless explicitly stated as being registered.
3. Name: Natural language name for the vocabulary. This can be expressed in any number of human languages. The name of a vocabulary should not be used as a surrogate for an identifier.
4. Profile Type: This property may be used to make explicit the class of vocabulary that an instance represents. Section 3.2 provides more information on the use of profile type. If no profile type is explicitly stated then it is assumed to be "lax".
5. Order Significance: This property is used to state whether the ordering of terms has any significance. In cases where there is a hierarchy of terms, it only applies to the ordering of top level terms. The default information is that order is not significant. It may not always be meaningful for order to be significant.
6. Default Language: This optional property may be used to declare a default language for all elements that represent human language strings. This may be overridden at the level of the individual language string.
7. Meta-data: Optional meta-data records that further describe the vocabulary. This is for convenience in cases where it is desirable to closely associate meta-data with the data it describes. More than one meta-data element may be used but each should only contain one meta-data record.

2.2 Information Items Pertaining to Terms in the Vocabulary

A vocabulary must consist of a non-empty set of permitted values. These are modeled as terms and relationships with properties as below.

1. Identifier: An identification string that must be unique among terms in the vocabulary. This may be used as the referent for relations defined from other terms (see "relationship" below).
2. Valid Index: An indication of whether or not this term is valid for use by indexing applications.
3. Caption: Natural language caption(s) that apply to the term.
4. Description: Textual notes for the use of the term or description of the term. These may be presented in any number of human languages, including languages not used for the caption element.
5. Media Descriptor: Media file (or files) to provide additional clarification of the description. This may also be used as a substitute for the textual description of a term.
6. Media Locator: A URL to locate the media file.
7. Interpretation Note: Textual information to explain how the media relates to the term. This note may appear in multiple human languages.
8. Order Significance: This property applies to child terms of the term on which it is declared. In the absence of a statement of order significance on a term then the order or children is not deemed significant. Note that there is no "inheritance" of this property down a hierarchy of terms.
9. Term: Terms may be arranged hierarchically in the information model - i.e., without the use of the relationship element. Such arrangements should be used for taxonomies where the only relation types are parent-child relations. They should not be used to express narrower/broader term relations in a thesaurus; the relationship element should be used for this purpose.
10. Meta-data: A container is available at the level of the vocabulary term as it is for the vocabulary as a whole.

Relationship:

1. Identifier: An optional identification string that (if it exists) must be unique among terms in the vocabulary.
2. Source Term: The identifier of the term that is the source of the relationship. The term identified by this property must exist but may be in another vocabulary. See the Binding document for more information.
3. Source Vocabulary: The identifier of the vocabulary containing the source term. If this information is not present then the {source|target} vocabulary is identical to the vocabulary containing the relationship.
4. Target Term: The identifier of the related term. The term identified by this property must exist but may be in another vocabulary. See the Binding document for more information.
5. Target Vocabulary: The identifier of the vocabulary containing the related term. If this information is not present then the {source|target} vocabulary is identical to the vocabulary containing the relationship.
6. Relationship Type: Relationship types are described using a sourced vocabulary. VDEX does not mandate any vocabulary for expressing relationship types but does make a recommendation in section 5 if a monolingual or multilingual thesaurus is being expressed.
7. Meta-data: A container is available at the level of the relationship as it is for the vocabulary as a whole.

3. VDEX Elements

This section is normative. The terms defined in RFC 2119 are used as defined therein.

The names used for elements in the VDEX information model are not normative names for any binding; named elements in a binding must only identify which of the elements described in this document they relate to.

Most of the VDEX information elements are optional to reflect the differing classes of vocabulary that VDEX can be used to express. Section 3.2 describes how the main classes handled by VDEX apply further constraints on the general grammar described in the following table.

3.1 Langstring Elements

Elements with type "langstring" and multiplicity "single" may only occur once but may contain multiple langstring constructs, each with an equivalent content. The SPM columns in Table 3.1 and Table 3.2 specify the individual constraints, which should be read in the context of the accompanying note on SPM notation and interpretation.

If an individual langstring element does not declare a language property then the Default Language of the instance must be taken to be the language of the langstring. If there is no Default Language then the language of the string is undefined.

Implementers must take care that:

• Two langstring constructs inside the same parent element must not have the same language property, including an undefined value;
• Undefined languages should be avoided;
• Users are discouraged from placing non-equivalent langstring constructs inside the same container.

The value space of the language property and the instance default language property is declared as "Language Code". This must conform to the requirements of RFC176613.

3.2 Object Model

This sub-section is not normative.

Figure 3.1 shows a class diagram following UML notation that may help some readers in interpreting the information model descriptions elsewhere in section. Types are not provided except where this specification defines a type: for vocabulary data types and langstring collections. This is an informative diagram in that it does not define the information model but supplements Table 3.1 and Table 3.2.

Relations are shown as always being binary even though the source or target terms may be in a different vocabulary to the relationship.

4. Semantics and Value Space for "Profile Type" Property of VDEX Vocabularies

4.1 Rationale and Introduction

This section is not normative.

The terminology "profile type" may be equivalently expressed as a "vocabulary type" but both terms run the risk of misinterpretation. We are not describing an application profile here; the VDEX "profile type" is more akin to a label for a subset of a more general information model. The subsets are intended to provide the necessary information elements for various well-known classes of vocabulary (see section 1.3.2). The VDEX "profile type" is not a property of the vocabulary to identify the class of vocabulary being expressed, hence we prefer not to use the terminology "vocabulary type" either.

The profile type is formally a property of the VDEX instance and makes explicit the subset of the VDEX grammar that is in use. This property is used to:

• Permit implementations to support minimal required subsets of the VDEX grammar in order to be as light-weight as reasonable;
• Promote interoperability by allowing exchangers of data to voluntarily reduce the degrees of freedom;
• Allow VDEX to provide a common grammar for several classes of vocabulary, each of which shares a great deal of information concepts but each of which has specific requirements. The alternative approaches of creating small models for each class or modularizing VDEX to a finer granularity are felt to be less maintainable, less intelligible and probably less interoperable than the proposed approach.

Profiling allows a single flexible information model with shared semantic units to be applied selectively if required. Making the profile type an explicit part of the model allows for easy profile-specific validity checking of, for example, XML instances by the use of more restrictive XML Schema control documents.

4.2 Definition of the Profile Types

This section is normative.

The profile types defined by VDEX are: "lax", "thesaurus", "flatTokenTerms", "hierarchicalTokenTerms", "glossaryOrDictionary". The non-profile, i.e., complete model, is identified by profileType="lax".

A profile type other than one of the specified profile types above should be declared as being "lax", even if only a small number of elements are used. If it is desired to express the identity of such a custom type then the information model should be extended to permit this information to be passed to systems that understand its meaning. Systems that do not understand the significance of a custom type would consider the type to be "lax" since that is the declared type as far as the VDEX information model is concerned.

Notes on using profiles: The most restrictive profile type should be declared that is sufficient to permit the elements in use.

5. Preferred Value Domains for Expressing Thesaurus Relationships

This section is normative. The terms defined in RFC 2119 are used as specified therein.

Monolingual thesauri expressed using VDEX should follow the conventions of d 2788. Multilingual thesauri should follow the recommendations of ISO 5964. Monolingual thesauri following ISO 2788 should use the following value domain to express the relationship types defined in ISO 2788:
Vocabulary Identifier: "http://www.imsglobal.org/vocabularies/iso2788_relations.xml"
Used for the "source" of a relationship type
Permitted terms: "USE", "UF", "RT", "BT", "NT", "TT"
Used for the "value" of a relationship type
Meaning of terms: Defined by ISO 2788

Multilingual thesauri following ISO 5964 should use the following value domain to express the degree of language equivalence: Vocabulary Identifier: "http://www.imsglobal.org/vocabularies/iso5964_equivalences.xml"
Used for the "source" of a relationship type
Permitted terms: "exact", "inexact", "partial", "singleToMultiple", "NonEquivalent"
Used for the "value" of a relationship type
Meaning of terms: Defined by ISO 5964

1EdTech will maintain a persistent copy of definitions of these value domains at a URL identical to the vocabulary identifiers above. These definitions will be expressed according to the most recent VDEX version XML binding.

VDEX instances using the above vocabulary identifiers must only use values from the relevant list of permitted terms. 1EdTech reserves all vocabulary identifiers starting "http://www.imsglobal.org/vocabularies/"; such identifiers may not be used without the authority of the Board of Directors of 1EdTech Consortium. Implementers should also refer to the "1EdTech Persistent, Location-Independent, Resource Identifier Implementation Handbook v1.0" [IMSPLID].

The preferred use of these value domains is described in the Vocabulary Definition Exchange Specification Best Practice and Implementation Guide.

Note on other relationship types:
The recommendations of this section apply only in relation to thesauri that follow the ISO standards mentioned. Other types of relationship models are permitted in VDEX and VDEX imposes no requirements as to the value domains and permitted terms for the associated relationship types. Specific mention is made of Z39.19 for monolingual thesauri since it extends the relationship indicators used in ISO 2788.

Note on multilingual thesauri and exact equivalence:
Multilngual thesauri that are essentially sets of equivalent terms and these sets are related in various ways should probably avoid using the Relationship element but make use of the fact that the Caption element may contain multiple language strings. This makes for greater clarity and economy in expressing the information.

About This Document

List of Contributors

The following individuals contributed to the development of this document:

Revision History

Index

G
grammar 1, 2, 3

I
identifier 1, 2, 3, 4, 5
1EdTech Specifications
     Content Packaging 1
     Learner Information Package 1
ISO 1, 2, 3
 

L
LOM 1, 2, 3

M
meta-data 1, 2, 3, 4

N
namespace 1
normative 1, 2, 3, 4, 5, 6

P
profile 1, 2, 3, 4

R
Relationship 1
relationship 1, 2, 3, 4, 5, 6, 7

S
schema 1, 2

T
taxonomy 1, 2, 3
term 1
thesauri 1, 2, 3
thesaurus 1, 2, 3, 4

U
URI 1, 2, 3, 4

V
valid 1, 2, 3

X
XML 1