Talk:SPARQL

From BITPlan Wiki
Jump to navigation Jump to search

Comment by Andries van Renssen 2018-01-16:

  1. The opening statement that a semantic statement has the form of subject-predicate-object is rather rigid. The statement neglects the relation of this statement to its inverse expression. The inverse expression expresses the same idea, although it is a different expression. I think it is important to clarify whether you treat the expression and its inverse expression as one idea or as two ideas that are equivalent. It also neglects that a semantic expression might also include two kinds of roles that are played by the two related objects. And that an intention might be explicitly added, instead of implicitly assuming that all expressions are (classified as) ‘statements’. Extended semantics include dialogues that can express statements, promises, denials, commands, questions, confirmations, etc.
  2. SparQL versus Gellish queries.
  3. SparQL is a query language combined with a reporting languages, whereas Gellish queries only express queries. Gellish queries have the same conventions as other Gellish expressions, but in queries unknowns are allowed.
  4. Conceptually, the SparQL query is very similar to Gellish. Both support the query:
    1. question: A is related to B.
    2. However, the syntax differs. 
In Gellish the software will detect that A is an unknown name and thus denotes one or more unknown concepts (including ‘things’).
  5. In SparQL the text string A, which denotes a concept C1, is not a free ‘name’, but has to satisfy specific criteria that bring specific semantics with it, such as prefix or a ? and a name that apparently classifies the unknown(s) denoted by the ?.
These conventions can be transformed to explicit expressions in Gellish. For example the expression ‘?car’ can be converted into ? <is classified as a> car or into ? <is a kind of> car, depending on the detection whether ? is an individual thing or a kind.
This makes that I think that it is possible to build a SparQL front end to a Gellish database system such as the Gellish Communicator software on GitHub.
  6. The information about the database is insufficient to judge how it compares to the Gellish object oriented database that is used in the Gellish Communicator.