The Facts database is a sort of combination of Wikipedia and other resources.
It relies on the basic trinary database that is often called the semantic web.
Each fact not only has information on what the fact is, but related facts and information on why people know the fact to be true or false.
This could be useful for organizing texts into a tree of facts. For instance, let's say you wanted to map all the prophecies in the Bible about the last days together, and see how they relate. You'd start with statements within the Bible, possible interpretations, and then map the interpretations to each other.
The Personalized Facts database would be Knowledge Vault. It would supplement the knowledge an individual has in their head.
Users can pick facts they want to remember and study them in a variety of methods. See Memorize.
Fact databases may share facts one with another much like USENET operated.
What is a Fact?
In simplest terms, it is this:
- An object ID that is the subject
- A predicate
- A object ID that is the object.
For all predicates except one, the object ID represents another object.
For the predicate "has the value", the object ID is not an object ID but a literal value.
Users are free to invent their own predicates. In cases of confusion, well, there's really not much to do there except to use IDs that point to distinguish predicates of the same name but different meaning.
Below, A is the left side, the subject. B is the right side, the object.
- "has the attribute X of", one for each attribute of interest.
- "has the name of"
- "has the title of"
- "has a weight of"
- "is a" : Infers a type relationship, where A is an instance of the class B in the classical object oriented paradigm.
- "inherits attributes from" : Infers a class relationship, where A is a subclass of B, in either the prototypical or classical object oriented paradigm.