: Semantic Web, RDF and SPARQL I'm new and kinda confused about this whole concept and would really appreciate it if you could help me out. The SEMFOX query parser disassembles the search

@Eichhorn148

RDF is a W3C standard for representing data using three-part { identifier, attribute name, attribute value} data structures. SPARQL is the query language for querying RDF data. The W3C's RDF Schema Language (RDFS) lets you declare classes, properties, and relationships between them using RDF, and their Web Ontology Language (OWL) lets you get fancier with how you describe classes, properties, and other resources, building on various knowledge representation traditions. As one example, OWL would let me say that the "spouse" property is symmetrical and that the "locatedIn" property is transitive.

This ability to store little bits of the meaning of terms like "spouse" and "locatedIn" to give queries more to work with is really what the semantic web is about, at least in the W3C sense of the term. As the term "semantics" became trendy, it has gotten used more and more broadly, especially when people advertise search engines and try to differentiate theirs from Google. I cannot find any connection between SEMFOX and these W3C standards, or even a mention of these standards, on www.semfox.com. It's very common for products to advertise "semantics" without having any particular connection to these W3C standards.

The software that people typically use to work with RDF includes what is known as a triplestore, which is essentially a database manager optimized around the triples model. Just as proper relational databases support SQL, proper triplestores support SPARQL, and they can act as "SPARQL endpoints"--that is, they support remote SPARQL queries delivered via HTTP. This is a common way to integrate triplestores with other systems.

Plenty of commercial and open source triplestores are available. RDF triplestores fall into the Graph category of NoSQL databases.

Vote Up Vote Down

@Jamie184

Oh Lord where to begin! The semantic web is not a thing. It is not something you do. As a term, it originally did not apply to the World Wide Web. It is a scientific principle.

You just create your web content for humans and not for machines. That is, for the most part, all you need to know. There are certain SEO tactics that take advantage of certain systemic design elements of search engines and I talk about them on this site quite often.

I am not sure what these people are selling, but I suspect they really loaded up the bull for market.

Semantic web, RDF, link graphs, trust networks are something I know extremely well and is nothing you have to get into. These are scientific terms that describe process theories. Search engines are built upon these principles since the advent of PageRank. However, these terms apply to social sciences and other scientific disciplines as well.

This is not something you need to know- certainly not as a beginner. And let me tell you, it is a huge topic!

Now on to SEMFox. This appears to be a search engine. But guess what. Search engines have used semantic web, RDF, link graphs, trust networks principles for the better part of 15 years and this is not a secret. Perhaps they can do a better job, but I will let you know that Google as a research project started with these principles in the form of PageRank and have built upon these principles in amazing ways. Do not think that Bing, Yahoo!, and other search engines are not based upon these principles. Heck! Harvest and Nutch which are free use them. So SEMFox would have some catching up to do and it does not sound like there is anything new to offer.

If you need to pay money for anything they are selling- I would advise keeping your wallet in your back pocket for now.

Vote Up Vote Down