Version 47 (modified by gordonrachar, 14 years ago)



Status of this document: Working Draft

This document is open for feedback, please post questions and comments in the forum at the bottom of this page. You will need a login to post in the forum.


  1. Introduction
  2. RDF (Resource Description Framework)
    1. Metadata
    2. Data Model
    3. Abstraction
    4. RDF
    5. Subject-Predicate-Object Triple Stores
    6. Uniform Resource Identifier
    7. More Information about RDF
    8. SPARQL
    9. OWL (Web Ontology Language)
  3. RDS/WIP (Reference Data System / Work In Progress)
    1. RDL (Reference Data Library)
  4. Gellish
  5. Ontology and Taxonomy: Difference Between
    1. Taxonomy
    2. Generalization/Specialization
    3. Subtype/Supertype
    4. Ontology
  6. Some More Definitions
    1. Semantic versus syntactic
    2. Semantic precision
    3. Semantic fidelity
    4. Reuse
    5. Encapsulate
  7. Next


There are a great many glossaries available, as well as on-line dictionaries and, of course, Wikipedia. Here are two:

So the world does not need another complete listing of computer terminology. But to save reader's time searching, the following are terms that are particularly interesting to the study of ISO 15926.

RDF (Resource Description Framework)

If you dig deeper under the hood of ISO 15926 you will soon run into this term because it is the means of storing the Part 4 definitions.

Wikipedia says that Resource Description Framework is a set of specifications originally designed as a metadata data model. But if you are like the author, this doesn't help at all, so we will deconstruct the definition.


  • Metatdata is data about data. For instance, one piece of metadata about the ISO 15926 Primer is that it was written on the POSC/Caesar wiki website.

Data Model

  • A data model is an abstract model that describes how data is represented and accessed.


  • Abstraction is a process of generalizing about something to reduce the information content about an object to only those attributes you are interested in. A typical abstraction is the answer 7600 Glover Road to the question "Where do you live?" You might live in a beautiful split level house with a wonderful view of the ocean framed by huge 100 year old pine trees but your questioner only wants to know where to have a package delivered. (On the other hand, yours could be a very ordinary house on a very ordinary road, but the city just wants your land for a freeway bypass and the friendly bulldozer operator needs to know where you live.)


Putting it all together, then, RDF is:

  • instructions on how to represent
  • just the bits of data you are interested in
  • that describes certain other bits of data
  • then access it easily

(Whew! I bet you thought that was going to be difficult!)

In particular, RDF makes statements about things, which it calls Resources, in the form of Subject-Predicate-Object expressions known as Triple Stores.

Subject-Predicate-Object Triple Stores

"The ISO 15926 Primer was written on the POSC/Caesar wiki" might be stored in the RDF as the triple:

  • the subject: ISO 15926 Primer
  • the predicate: was written on
  • the object: POSC/Caesar wiki

The each term in the subject-predicate-object triple may be explicitly named, as in the example above, or they could be in the form of a URI, a Uniform Resource Identifier.

Uniform Resource Identifier

You can think of a Uniform Resource Identifier as a website for a piece of information. This allows the same resource to be reliably referenced many times. So instead of writing the Subject-Predicate-Object triple as above, it could be rendered as:

And in fact we could carry this further by defining somewhere on the Internet the exact meaning of the phrase was written on, and put its URI in the predicate.

More Information about RDF

A good place to start if you want to know more about RDF is the RDF Primer written by the W3C. Be warned, it is not for the feint of heart. But if you can wade through it you will start to see what we mean when we say that "Everything, in the end, is reference data."



SPARQL, pronounced "sparkle", is a query language designed to be used with RDF triple stores. According to Wikipedia, the name stands for "SPARQL Protocol and RDF Query Language". The ISO 15926 RDS/WIP uses SPARQL.


OWL (Web Ontology Language)

OWL is actually a family languages for creating ontologies. It is fundamental to the Semantic Web. OWL ontologies are usually expressed using RDF/XML syntax.


RDS/WIP (Reference Data System / Work In Progress)

The RDS/WIP is several things:

  • a library of reference data for ISO 15926
  • a means of publishing core ISO 15926 definitions
  • a platform for developing new ISO 15926 definitions
  • a workspace for harmonizing other standards with ISO 15926 (or each other)

The RDS/WIP is a large triple store in the form of Subject-Predicate-Object. It uses semantic web technology (OWL, RDF, and SPARQL) over top of a conventional web technology such as HTTP to provide machine-oriented access to the stored definitions. A conventional HTML presentation is used to provide a human-oriented interface to the same system.

Anyone can search the RDS/WIP and find terms, much like in a dictionary. Accredited users can add information to the RDS/WIP.


RDL (Reference Data Library)

POSC Caesar has it's own library of reference data (hence: RDL) for ISO 15926-4.


Gellish (originally derived from General Engineering Language) is a language in which information can be expressed in a manner that is computer readable. It is used to make the ISO 15926-7 templates.


Ontology and Taxonomy: Difference Between

Taxonomy - Quick 'n Dirty

If you've ever made a classified list of all your CDs, you've made a taxonomy. (But if you're as old as the author, CDs are old hat. You learned how to do this years ago with your player piano rolls!") And if you've ever had to grapple with the question of where to classify Weird Al (under "Parody?", "Rock and Roll?", or "Idiot?"), you've come up against the idea of single or multiple inheritance!

Ontology - Quick 'n Dirty

If you've ever played the parlor game Twenty Questions, you intuitively understand ontology. In this game you more-or-less start with an Ontology-of-Everything-In-The-World, and with each successive question ("Is it a ...?") apply a more limited ontology as a filter (usually starting with "Is it an Animal, Vegetable, or Mineral?") The game ends when there is only one object left, The Answer, that satisfies membership (or non-membership in the case the answer to "Is it a ...?" is "No!") in all the ontologies.

Ontology and Taxonomy are both terms in a continuum that some information scientists call Knowledge Organization Systems (KOS). And just to confuse you some more, the continuum incudes Thesaurus, Controlled Vocabulary, and Faceted Classification among many other things. The bad news for those of you not used to dealing with ambiguity (All you mechanical engineers out there: Raise your hands!) is that there is a great deal of overlap in those terms. Even people who's job it is to know these things (All you mechanical engineers out there: Put your hands down!) can't give a short answer when asked where the boundaries are.


A taxonomy is a collection of terms that have explicit definitions that have been organized into a hierarchical structure. They tend to be organized in tree-like structures that are reasonably easy to understand, even by non-specialized people. Each term is related to its parent in a is-a-kind-of relationship.

For instance, a car is-a-kind-of automobile. But a car also is-a-kind-of machine, so if your taxonomy is concerned with machines, you should analyze the relative order of these three things. Depending on the purpose of your taxonomy, you will likely end up with:

  • car is-a-kind-of automobile, which is-a-kind-of machine.


The is-a-kind-of relationship is known as generalization/specialization. In the above example a car is a specialization of automobile; automobile is a generalization of car.


Subtype/supertype is just another way of saying generalization/specialization. So continuing the example above, car is a subytpe of automobile; automobile is a supertype of car. The understanding is that the subtype has all the constraints of the supertype, plus one or more additional constraints.


In the realm of philosophy, ontology is the study of being; the study of the things that are. In the realm of information science (which is where ISO 15926 firmly resides), ontology has a more formal meaning. Wikipedia says that an ontology is "a formal representation of a set of concepts within a domain and the relationships between those concepts."

Like Taxonomies, ontologies are also arranged in a is-a-kind-of relationship, but the relationships tend to be more richly defined. The difference is subtle. One commentator compared the difference between ontology and taxonomy to your computer hard disk. The taxonomy would be the directory structure without the files, while the ontology would be the files organized by the directory structure.

Earlier in this Primer, we talked about an Ontology of Things That Will Carry a Bicycle. This Ontology is the whole collection of things that will carry a bicycle in case the author's bicycle breaks down on the way to work. Each object in the ontology would have a Taxonomy that you could examine.


If you want to see how deep the Ontology rabbit hole does, Proffessor Dagobert Soergel, of the University of Maryland has written the following explanation of Knowledge Organization Systems.

  • ...

Some More Definitions

Semantic versus syntactic


Semantic precision


Semantic fidelity





Hiding complexity from users who really don't want to know any more.



You have no rights to see this discussion.

About PCA
Reference Data Services