Really?!? has this question not been asked here yet?
Quote W3C:
In addition to the classic “Web of documents” W3C is helping to build a technology stack to support a “Web of data,” the sort of data you find in databases
What always perplexes me is whether this was based on phenomenon or intention?
Ask someone to define the Semantic Web, and you'll get one of two things:
- an objective answer stating that it's Tim Berners-Lee's vision of a web—as defined in the Scientific American article—where machines can exchange and interact with data, and (partially) "understand" the meaning (the Semantics) of that data;
- after some squirming, the persons subjective vision of what the Semantic Web is and how it can be realised.
The former answer is more correct. The latter answer is often more "interesting".
You can tell a lot about a Semantic Web geek from asking them this question.
(In certain circles, it's almost taboo.)
I feel like an easy-yet-bad answer would be:
Content on the web that is stored in RDF and accessed with SPARQL
And that's a bad answer because neither RDF nor SPARQL are essential to the intent of the semantic web: it's not hard to imagine a semantic web designed and deployed by CYC. A better answer would be something like:
Content on the web that's stored in a format that's convenient for retrieval and analysis.
So, if there was a convenient format for grabbing and making inferences over CSV data, then it might turn out that Excel is our next de facto SemWeb development environment. And given what some of the people are doing with R and online CSV files these days, it's not such a crazy idea.
That said, there's a lot of stuff that's crazy-simple when done with a good knowledge representation framework like CYC, Common Logic, or RDF/SPARQL, like extracting interesting graph patterns from sparsely-populated data on highly-dimensional objects, or defining qualitative geometry calculi. But I'd still argue that those statistical computing guys working on Twitter data are an important part of our semantic web.and analysis.and analysis.and analysis.
I thought some members of the community (and a former British PM) gave interesting answers this question in video form as part of our "Elevator Pitch" challenge.
It's still an open challenge, by the way, if anyone here wants to give it a try!
The Semantic Web is one step in the evolution of the Web.
The answer in one sentence should be something like: "Is the web where the machines not only exchange data, but knowledge".
For me it has always been something like "data integration at Web-scale". At least to me it mostly revolves around data integration and I think that's something people might easily understand.
Sure there's a lot more to it, more technological details, more aspects, more things it can do. But do you really need to tell people about URIs, reasoning or the OWA to explain the basic idea behind the Semantic Web?
On the other hand it might sound a bit too innocent and boring, so I guess to really impress people and get them excited you need to start talking about use-cases.
For an alternative answer, I do believe that this set of slides gives a pretty all-encompassing answer:
http://www.w3.org/2007/Talks/1211-whit-tbl/
.. although the video to go with would be a fantastic utility to have.
I have made slides about Semantic Web: Intro if you'd like to see it. It will give you the feeling about what Semantic Web is, more or less. Anyway, Semantic Web is a way to use the Web as a semantic or knowledge store and analyzer.