The first step that made the Web into the World Wide Web was the hyperlink. The hyperlink connects documents in a web of information. You can refer the reader to a document giving a definition or background information on a term in your web page. This first step connects documents. What the link is supposed to link is left up to the reader.
The next step was to use links to link data on the web. By doing this we can give computers access to the information in the data on the web. To do this the links must give information about what the link says. An empty link like the WWW hyperlink merely points from one document to another, it lacks context. A semantic hyperlink also tells you what the link means. Data then becomes linked data. When you publish data you can link to the definition of the units you’re data is in. The definition of a unit can refer to other units, linking the defintion of degrees Celsius to degrees Fahrenheit for example.
Your data can thus become part of a world wide web of data by connecting you information to other information out there. Provided your information is findable and that it stays were it is. Each object of information can be linked by a relation to another object. Such a relation is called a predicate. You then make short sentences: subject - predicate - object like you would in a natural language. You use a standard dictionary of predicates: an ontology. The concept is easy but the implications are harder to grasp. It approaches information differently than we do on for example a database or in everyday language.
In a natural language we tend to describe an object by the parameters it has using adjectives. In a red shirt the fact that red is the colour of the shirt is implicit. It is a parameter of the shirt, like long can be a parameter. The way it would be described in linked data, on the semantic web would be shirt has the colour red , where has the colour is the predicate linking the objects shirt and red. So our languages differ from the semantic web, but not all languages do.
It is almost a cliché that relations are so important in China. But in the language they certainly are. Chinese doesn’t use adjectives like we do. Verbs are not conjugated but just stand there in sentences, being verbs. And of course a word is expressed by a character, the objects of the Chinese language.
Those character-objects have relations with other character objects. The aforementioned red shirt is a 红色的衬衫. Here are 红 (red, hóng) and 衬衫 (shirt, chènshān) the objects. But calling a red shirt a 红衬衫 would be simply wrong. The two objects must be connected by a predicate 色的 which can be translated as has colour. This predicate is formed by two characters 色 (colour, se) and 的 (has, de). The 的 is the ultimate connector. It translate it to has here but usually it just disappears when you translate a Chinese sentence to a European one, A verb like 是 (to be, shi) should also be seen as a predicate, not a verb. .
The linked data aspect of Chinese becomes evident when you try to translate from a western language to Chinese. We build long sentences in german and Dutch, consisting of sentences with sub-sentences, and long words consisting of other words. You just cannot translate a German or Dutch compound sentence into a single Chinese sentence. You must translate it into a series of short sentences all following the subject - predicate - object structure. In effect Chinese is linked data with it’s own ontology, ready to represent all Chinese knowledge on the World Wide Semantic Web.