I have been conscious of the term ‘semantic web’ for quite a few years now, and probably like most people it was all a bit fuzzy, didn’t seem to connect in my mind to anything solid. Well, two events in quick succession made it work for me, the first was hearing Sir Tim Berners-Lee describe it first hand, (WWW2006 conference, Edinburgh), and the second was seeing what advocates of JENA, (JENA Users event, Bristol), are doing right now. Amazingly, both events were in the UK where there seems to be some sort of nucleus of activity, though with world wide attendance. Sir Tim’s message was simple, the continued astronomical growth of the Web, including spin-offs such as Web Services into SOA means machines have to take some of the load currently falling on Humans; i.e. the ability to machine interpret content semantically, or with a human touch. I have blogged before on the way that e-mail is raising and the need for collaboration tools to take back some of this workload through improved semantic comprehension, but here is the real business issue. Metcalf’s Law says that the value of a Network increases logarithmically as the numbers connected rise. Put another way, there are more opportunities to identify, and react to optimise, events if you can communicate with those affected. We need Semantics to make sense of the rising tide of data from the vast numbers of sensor devices as well as just the people and computers we are dealing with now. The JENA Users event took a different direction, the semantic need was still the key, but the direction was down to the user level rather than up to the global level. JENA is an Open Source Java framework for building Semantic Web applications created and donated by HP Research Labs that has been growing in popularity for its ability to provide a programmatic environment for using XML with RDF, RDFS, OWL, SPARQL extensions Up till this point I had thought that this was strictly Enterprise regulated stuff on creation and maintenance of data, and adoption was going to be slow, but as real users stood on stage and described how they were building personal ‘graphs’ with JENA to provide their own ‘views’ of the mass of enterprise data available, but not aligned to their needs but I saw another example of user driven change. A variety of tools were demonstrated, all open source, together with some great work by HP and IBM that covered both users and enterprise solutions, but the real power was users testimonials and demonstrations of how even in large commercial enterprises they could make it work for them at a personal or department level. Put simply a ‘graph’ is a massively more effective version of a spreadsheet handling data elements instead of numbers, any change in remote data, changes a cell and this handles the overall relationship of the data within the total ‘view’. The ‘graph’ view is a big picture focussed cohesive information model that is unaffected by the external changes in data elements. Seems we are looking at the Semantic Web in the wrong way, just like the original Web its going to get driven by users, and their frustrations with corporate computing and innovation rates.