Semantic Web is a concept that has been around about as long as XML and even HTML. The idea is that a new way to write data to the web will be created that will make it easier for a single user to find precisely the data they want.
The most common architectural framework we is Resource Description Framework or RDF. RDF aims to code data in statements of three terms in a relationship. A classic subject object predict kind of solution. A large group of these "triplets" creates a "triplestore". The notion is that a large triplestore will be able to draw conclusions beyond the data encoded.
For example you might have a triplestore with the following facts encoded:
- Coen Brothers directed True Grit (2010)
- Jeff Bridges stared in True Grit (2010)
- Bob Hooker did not see True Grit (2010)
- Bob Hooker saw Tron Legacy
- Jess Bridges stared in Tron Legacy
- Tron Legacy was released in 2010
- True Grit (2010) was released in 2010
Demos systems like this can then draw inspiring conclusions like Coen Brothers and Jeff Bridges worked together on a movie that came out in 2010 but Bob Hooker did not see it, but Bob Hooker did see a movie staring Jeff Bridges.
Okay well maybe something a bit more useful could come from it, the idea is the same.
As long as you pretty much know what you want to come out of a triplestore and build it correctly you can conclude fairly obvious things. If a triplestore has cured anything or come up with any theory yet we have not heard of it.
If the Semantic Web will someday scale up to actually being useful is a open question. It would be really cool if it did. But if it works or not it is to a central part of the changes that are going on it the web that can be meaningfully called Web 3.0.
We suspect Semantic Web will keep trying to be part of some future web event, and that it will be being pushed as a core part of Web 4.0. The idea of automated "thinking"has been the Holy Grail of Computer Science from before most people could access a computer, and will probably always be an attractive idea to computer researchers if not consumers.
What tend to define Web 3.0 as not semantic, but rather the extension of the Web 1.0 (content) and Web 2.0 (Social Graph) into the spatial domain. Web 3.0 web content and social nodes are both tagged with spatial relationships and able to form social relationships based on current location.
But more than that, Web 3.0 means that users have access to the Web just about where ever and whenever. Web 1.o answer what, Web 2.0 answer who, and Web 2.0 will answer where.
For use the key technology change that makes Web 3.0 a distinct and revolutionary phase in the evolution of the Web is the fact that the Internet is now with the user all the time. The line that broke in the world form online has been broken.
Just a few years ago a common way to insult people who posted online was to say they "should get a life" or "should get out more", but today people are posting on twitter on trains to work, while waiting in bars for friends, or from lounges of museums. Foursquare, Facebook places and Gowalla extent the activity of being online to the activity getting out and about in the real world.
This ability to do while you surf, to consume and produce content anywhere has changed radically the nature of the user experience of the Internet. It is this change in the lived experience of the Internet that makes the term Web 3.0 useful.
The change does not require the Semantic Web, nor would the Semantic Web cause this change. So we tend to see the Semantic Web as independent of the key event that we call web 3.0.
Web 1.0 put people in contact with a great deal of information. The stereo type of a Web 1.0 geek is someone consuming masses of pointless data pieces while not in social engagement. The truth is though there never was a Web 1.0, people who used the web a great deal would find forms like Usenet for form community on. People have always viewed the Internet as a social experience first and foremost.
Web 2.0 expanded the ability of people to join and form community online for just chat groups to Social Graphs. Web 2.0 stereotype geek is someone who lives their entire social life from their bedroom or Starbucks. A "Web 2.0 geek" has lots and lots of friends on Facebook but never goes out. This was very empowering for people disabled people who could not leave the house, but it tended to make the focus of social interaction the bedroom or study.
Web 3.0 takes the web in to the world. A Web 3.0 geek is someone who can't make eye contact because they are glued to their blackberry or iPhone all the time. They are online all the time, never more than a few minutes from checking their twitter account.
A more positive way to think about it would be:
Web 1.0 is reading about a protests on a group's web site before hand, going to it, and going home and looking information on the web about it in the news the next morning.
Web 2.0 is learning about a protest on Facebook, seeing the people who are going to go, IMing to people who will go on from you Social Graph, going to it, coming home and posting images to your blog or Facebook page that night.
Web 3.0 is learning where a surprise protest is going to meet 10 minutes before it happens on your blackberry, video tapping it on your iPhone and posting tweets and images on your phone from the event, reading the Facebook entries on the way home and once you get back not even bothering with you PC because you already know everything and are too tired. Rather watch TV while tweeting on your android.
This mobilization of the Internet experience has nothing to do with the Semantic web, it is entirely independent of it. The mobile post-PC Internet (with is the third wave or Web 3.0) will happen without Semantic Web like meaning, and Semantic Web like meaning, if it is possible, could happen without adding mobility to the Internet.
We at the Web 3.0 Lab thing that by adding more spatial dimensions you will get improved semantic understanding. Much of our social understanding is spatial. Reasoning that some people hope to get out of triplestores we think will emerge out of geo-tagging of information. Spatial arrangements of data will drive interesting conclusions about how that data relates to the real world, how it is used, and therefore what it means.
The Web 3.0 Lab has a working model of meaning that is a bit different than what the Semantic Web assumes. While the Semantic Web sees data as meaning what it is about, or what it captures or contains, we seeing the meaning of data coming from how it is used. We feel that this is a more mature understanding of language evolved over the past decades by linguist and social scientists.