Questions, we have questions - GlennClatworthy/kgc_discussion_group GitHub Wiki

In progress and not in final order:

  • What is the minimum "stack" (software and standards) that I could use to design, add data to, and query a linked data model?
  • What does RDFS (RDF Schema) add to RDF (the Resource Description Format)? Why would I use it in my ontology?
  • At what point in an ontology design does a class turn into a resource? That is, how do you know that you're referring not to a group of things but to a specific thing?
  • Is there a single "right" combination of properties, classes, and resources to describe what you're modeling, or can there be many different ways of describing a collection?
  • I'm frightened by the apparent complexity of OWL (the Web Ontology Language). What problems or challenges does it solve?
  • What is the difference between an RDFS class and an OWL class?
  • Many of the OWL-related references in particular seem to refer to logic. How much logic do I really need to know to design and use ontologies?
  • Can you give me an example of inference in action? Is it primarily used by ontology tools to check logic or can it be invoked using triple stores?
  • How can the same values or URIs be used both subjects and objects in triples? Aren't they either one thing or the other?
  • What is the difference between a URL, URI, and URN?
  • What is this hash vs. slash business?
    • How do I point to an RDF file in a URI without using an extension like “.rdf”?
    • What does it take to redirect browsers to different formats of my data?
  • If anyone can say anything about anything (the semantic web AAA slogan), how can I trust the results of combined data sources?
  • What is the difference between an RDF triple store and a property graph? Why would someone choose to use one over the other?
  • SPARQL:
    • What is SPARQL?
    • What are the basics of using SPARQL to query a graph database?
    • How is SPARQL similar to and different from SQL (the Structured Query Language)?
    • How do I filter on partial strings when I'm writing a SPARQL query?
    • How do I find SPARQL endpoints and connect to them in my queries? Is it better to download the RDF files and import them into my own triple store?
  • Do I need to add entity or language tags like “@en” to my values in RDF?
  • Why should I create my data in RDF? Isn't it just easier to import worksheet data into a knowledge graph application?
  • What is the difference between the various serialization formats? What are the strengths and weaknesses of each? Which is the best for beginners?
  • What is a "graph" within the context of an RDF file? Why would I need one and how would I create in RDF?
  • What is "dereferencing"? Can you show us an example of it in the wild?
  • Reification:
    • What is reification?
    • Why is reification important?
    • How would I implement reification in different graph systems?
  • Schema.org:
    • What is Schema.org?
    • How do I navigate Schema.org’s user interface?
    • How do I embed properties and classes from Schema.org into my own ontology?