Introduction

Below is the basic terminology used in the SuperSimpleSemantic system.

Glossery

Tripple

In any semantic system, the basic unit of knowledge is a tripple. This consists of a subject, a predicate and a value(sometimes also called object).

Subject

The item being talked about. "Granny Smiths are coloured green" In this sentance "Granny Smiths" are the subject.

In the SSS system, the Subject might be a file source; isGreen.txt colour green. In this case everything in the file specified can be taken as subjects with "colour green" following them.

Predicate

The specific type of property being talked about. In the last example this would be "coloured"

Object / Value

The value you are assigning to the property of the subject. In the above example this would be "green".

Property

A property is both the predicate and the value. "color=green" would be a property.

Subclass of

When something inherits all the property's of another thing, its considered a subclass of it. Thomas is a subclass of human. A human is a subclass of primate. etc.

SSSNode

A node is any predicate, subject or value. Its any single "thing" in the database.

URI

Uniform Resource Identifier is a string that uniquely identifys a node in the database. This is typically very much like a url, supplying both the domain the node is from, and its name. In SSSS if a URI isnt specified, it assumes the local domain, followed by a # followed by the name specified. eg, A list file containing just;

• orange
• lemon
• apple

Would create 3 nodes assigned default uris of

• examples/DefaultOntology.n3#orange
• examples/DefaultOntology.n3#lemon
• examples/DefaultOntology.n3#apple

if the file is hosted elsewhere its default might be;

http://www.atresica.nl/Meryll_wants_a_cookie/Semantics/DefaultOntology.n3#apple

If you wish any item in your database to be understood as the same as one elsewhere, you should use a full uri. Generally speaking, if you only plan to use the data locally to your project, you can just stick to the super-simple method of just having a list of words.

Label

A label is the human readable name for the node. Unlike a uri, a single node can have many lavels...normally one for each lanugage. Labels are in fact just like any other property an object has, the only difference is the interface use's them to make reading results friendly.

By default, if no label is specified, its the same as the URI.

Common Property List

A common property list is the basic unit of file storage in the SSS system. Its a list of all the nodes that have the same property. The list is nothing more then a text file, and a list of uri's. The index file is what tells the list what it is storing

An Index file

The SSS index files (.ntlist) are a list of triples that specify what types of things the common property list stores. It is somewhat close to the ".n3" semantic file format, except the subject node in a tipple can be a file url.

GWT

A method to compile Java code to Javascript, along with lots of browser-specific optimizations. SSS is fully compatible with it.

Fractular Dynamic Loading

This is a fancy sounding term to mean that the list file contents is loaded when needed, and if any list would indicate another list needs to load, then that one is too. In the fruit example, that might be a list of fruit containing "Apple", and because the index specify's a file containing apple types ("isApple.txt rdfs:subclassOf Apple"), it will load them too and add them to the internal fruit list.

[tbc]