Home‎ > ‎

The semanic.org Semantic Web

The Requirements for a Semantic Web
  • Ontological Schema/Definitions:  Ontologies or ontological schema provide chiefly the class and property definitions of the reference data and the semantic content of a semantic web.  In a closed semantic web such as "semanic.org" only "semanic.org" ontology files are used for its content and reference data.  This restriction gives the freedom of a seamless already integrated knowledge space.
  • Reference Data:  Reference data, in this case, is semantic data about common world knowledge such as geography, politics, or anything of general enough interest that it would be sharable over many semantic applications.
  • Semantic Content:  Content files contain definitions of the instances and RDF statements that comprise a network of meaning which models some subject.  Strictly speaking, this RDF content is probably contained in owl files and might use some OWL constructs such as "owl:sameAs".  A content file might be someone's OWL homepage or the more static semantic information from some companies semantic web site.
  • Semantic Store:  As with Google Search, user queries are presented to a store which returns results to the user.  In the case of a semantic store the store contains RDF statements or some equivalent, and returns successful query binding or graphs.  The semantic store serves both human and automated users in the form of web apps or server side applications
  • Semantic Web Crawler/Loader:  The semantic store must be loaded and kept updated with statements (such as triples) from the semantic content, reference data and ontologies from the semantic web.  This loading functionality can take various forms and operate at various levels
    • Fully Mature Crawler:  A full crawler might be expected to crawl the entire web and load an appropriate subset that can fit into the semantic store. 
    • Registration-based Loader:  This level of loader loads only semantic web pages that have been requested via registration.
    • Manual Loading of Store:  Presently, the semanic.org store is loading manually.
  • Semantic Web Browser:  Just as Google Search is performed by a web browser a semantic web must have a semantic web browser.  But rather than simple variants of keyword search, semantic web search would include:
    • SPARQL Search:  This search would return the usual bindings or graphs.
    • Prolog Semantic Search:  Prolog would be included as the next generation of semantic search.
  • Semantic Web HTTP Servers:  The semantic store must present itself as one or more HTTP web servers to serve semantic query results and manipulated results to conventional web clients (see below).
  • Web Client Apps:   A functioning semantic web environment must have conventional web client apps such as are written in JavaScript/HTML/CSS.  These clients would then communicate with the semanic web HTTP servers via JQuery calls, AJAX calls or some such.  A semantic web HTTP server can also communicate with IPhone and Android apps over their HTTP protocols.
Comments