DNHour

The current interest in social websites dubbed Web 2.0 as exemplified by MySpace, FaceBook and others has launched myriads of copycat sites. To me, the market is saturated and it has been difficult even for the big boys to convert their visitors.

Web 3.0, the next big thing?
Lets look forward to the “next big thing”. Named, imaginatively enough, Web 3.0, it is thought that this implementation will be all about data. The web currently is about documents where you find pages to read through or rather, skim through if you are the average website visitor.

Pages of information are in a fixed format laid out by the author and it takes a lot of work to pull together information from multiple sources to answer questions people may have. In fact much of this added value is lost in the current web.

It is all about the data
Data, on the other hand, can be structured or unstructured and can be relatively easily manipulated and reordered as needed. Data at least theoretically can be pulled from multiple sources and integrated to provide a specific answer.

Some personal finance programs already do this in a rudimentary way. If you get a printed statement from your bank and another from your investment company on your stock’s performance and so on, how easily can you find out what your total net worth is? Using commercial programs such as Money or Quicken, this data can be imported and tables and graphs quickly generated. Net worth becomes immediately obvious and easily updated as new data becomes available.

The semantic web
The bigger vision is to have a semantic web which allows machine readable data to be shared across the web in addition to the documents. There are major obstacles to this including issues of privacy and software engineering but I can see it come into play in some settings.

Already companies are integrating data from public databases and consortium databases with their own internal proprietary databases and this combined data allows conclusions to be reached that couldn’t be reached with any individual database.