Mini research: Semantic web

I've recently done some mini-research on Semantic Web and here what I found.

Wikipedia says: "The Semantic Web is an evolving extension of the World Wide Web in which the meaning of information and services on the web is defined, making it possible for the web to understand and satisfy the requests of people and machines to use the web content" (see more).

Not clear enough, I need details.

Key terms:

XML - The common language used for data storage and exchange (see wikipedia).

RDF (Resource description framework) - The XML-based language, used to describe everything. See wikipedia article for general information and official site for complete documentation if you need to.

OWL (Web Ontology Language) - Extended language, based on XML and RDF, used for describing objects and connections between them. Wikipedia article shows general info and there is official site for full specifications.

SPARQL (SPARQL Protocol and RDF Query Language [yes, it's recursive acronym]) - The RDF query language. See wikipedia and official site.

See full structure on diagram below:

How does it work?

Data on web sites is being described using languages above. There is software, that can read this language. By reading, it can classify and identify content on the pages. This makes software to "understand" the information as humans do.

A software may vary from web-based agents to desktop applications.

For example, you can describe your contact information using standard tags and commands of semantic languages and that will make you browser "understand" that there is a contact information on the page and that that particular number is the phone number of this contact name.

The browser is just one example of software, that can use this information. It is done on some level now and is planning to develop in the nearest future. For example there are add-ins for Firefox browser Operator and Tails Export and Internet Explorer 8 is claimed to support these features.

But browsers are just the beginning, the main goal is to connect web resources together using standard approach and protocols and transform the Web to a structured and classified database, which takes the web to another level - from the chaotic mix of information, where users have to distinguish things by themselves, to a world wide database, that can be queried and given automated results depending on previously made semantic descriptions. This gives to Web a totally new quality - integrity, where it can act as a single object, giving absolutely new product - query result.

Now when we know how it is planned to evolve, we can think do we like it or not.

What is going on now?

All these languages and protocols are mainly developed, supported and promoted at World Wide Web Consortium (W3) and it's director Tim Bernes-Lee, the inventor of World Wide Web.

W3 has developed a row of approaches and frameworks, called W3 Semantic Web Activity. As I can see the guys are working hard for making things done and changes to the Web are happening right now. More and more web sites are starting to support semantic tools and approaches. Even giants like Microsoft and Yahoo starting to deploy it in it's products (see article at ZDNET).

But despite all benefits of the approach, the frameworks are very complex and complicated to apply. That's why Web will probably need some time to adopt new principles. Simpler and more easy-to-deploy services will start to show up.

One of them, and probably the only really popular organization is Microformats (ZDNET article above shows us where it is being used now). It gives us a basic tools to easily deploy elements of semantic web to our web sites. The tools are standardized and supported by major software developers. You can read about how to use it on their site.

Using Microformats we can easily describe some of our web site elements, such as people, events, reviews, locations and tags and export them to semantic software (browser add-ins are among them, Microsoft products are getting ready to handle it too).

There is probably one, who could be unhappy about this trends - search advertising companies, since they use principles of "old" web, where human is the one, who distinguish things. When machines will learn to make difference between "apples" and "oranges" - humans will not read so much unwanted content. But I feel those companies will find their gaps ;)

So what to do?

Transforming the social Web2.0 to semantic Web3.0 and putting it on new rails will take time, so no need to panic, guys at W3 will make their job. But we must not ignore the new tendencies, we can and must prepare for new order.

Now it's great time to look at simple and easy-to-deploy, yet standardized approaches such as Microformats and use it as wide as we can. Wikipedia, Twitter, Yahoo Local and many other popular resources are already using elements of semantic web. If we want to get in stream - it's time now to relax about social and enjoy the new wave of semantic.

What I did - applied the hCard Microformat semantic code to contacts page on my web site - take a look ;)

See more

You might also want to see some extra links I visited today:
http://computer.howstuffworks.com/semantic-web1.htm
http://en.wikipedia.org/wiki/Dublin_core
http://infomesh.net/2001/swintro/
http://microformats.org/code/hcard/creator
http://microformats.org/wiki/get-started

Hope you enjoyed it as much as I did!
Denis Shapovalenko

Comments

[Eric Gagne]

Pretty good introduction. I incorporated your post in my blog, if you don't mind.

Cheers