Daniel’s Blog

Exemplary Semantic Web Systems

I am surprised about how powerful and usable some of the Semantic Web Applications are out there. Things like:

• Zitgist Browser - Which is essentially a Semantic Web viewer
• RDF Browser - Which is similar to Zitgist, but has a few more features - and a different UI.
• DBpedia - A system which extracts information from wikipedia, provides it in a SemWeb format and allows you to query over it (using SPARQL). E.g. CLICK HERE for Get All “Birds” from wikipedia
• PingTheSemanticWeb - Can be seen as a Semantic Web version of something a bit like Technorati or Ping-o-Matic.
• Sindice - Is a Semantic Search Engine - worth checking this out, its cool.

All of the above are based on technology associated with or developed by a company called OpenLink Software (more on them in a later blog post, but I will say the following). OpenLink do provide an awesome data server which presents data in a completely Semantic Web fashion, and they take their experience with technologies such as ODBC and JDBC very seriously - they do this with innovation in mind and a completely different view on business that I have seen. Look out for Virtuoso and other OpenLink products.
There are some other Semantic Web services which are quite advanced and in a usable state, such as:

• The Talis Platform - A Semantic Data Store
  
• Facet by Leigh Dodds of Ingenta - An RDF/SPARQL Web UI engine
  
• Exhibit (part of the SIMILE project at MIT) - An AJAX/Javascript Semantic Web Browser “thingy”
• BBC Memoryshare: https://www.bbc.co.uk/memoryshare/ - A system to share memories of things past
  

We also have some smaller companies taking on RDF solutions for their projects, these include:

• GoNumber - a Business Directory
  
• Miomi - A User Generated History tool

All of these projects are successful through the innovative use of Simplicity and well researched products. It is a surprise that more web developers have not heard of them, and definitely a surprise that more people are not using a Semantic Web solution - especially with servers like Virtuoso around.

Its about the data

Tim Berners-Lee (creator of the Web) recently made a blog post about the Giant Global Graph, now this isn’t a renaming of the Semantic Web, it is purely saying:

We have the social graph already, just look at the Semantic Web. Semantic Web technology allows you to have things like FOAF, SKOS and SIOC to give you the powerful connections that you require.

I add these things are available now, the Semantic Web isn’t this mess or mash of “stuff”… its all well defined (well most of it), and its certainly more defined than most websites out there. Even websites that have been POSHed up! (well, at least Microformats are half way there)

Think of it this way, do you know what the first web system was… well I have already mentioned it was by TimBL. It was at CERN, the research labs in Switzerland. It was in fact a Social Web system, but it was a Semantic system too. This ENQUIRE system allowed the staff at CERN to have their own page, and to reference other pages and build up relationships between people. It was basically something that (TimBL, correct me if I am wrong):

• Share telephone numbers, office numbers
• Share current research trends
• Share documentation across teams

Now this is a simple, collaborative and social environment with a specific focus and connections are labelled with meaningful name (sounds very similar to the Web 3.0 post I made the other day doesn’t it?!?!). The most important thing here is the data.

Linking Open Data

Now that I have shown you that there are these wonderful robust tools, and these wonderful extensible vocabularies we have some more things to talk about. We have to Link all of this data, and there is an ever expanding task force that are trying to find standardised solutions for it, they are available here:

https://esw.w3.org/topic/SweoIG/TaskForces/CommunityProjects/LinkingOpenData

This comes back to what TimBL was talking about in his GGG blog post, linking these different files and vocabularies together - could also be called Schema Alignment (if we take some ideas from Relational Databases). This is an easier task than you can imagine, it is being done now, and nothing should stop you from using an RDF based solution (certainly hasn’t stopped companies like Adobe, Microsoft, BBC etc from using RDF solutions). More about Linked Data here: https://www.w3.org/DesignIssues/LinkedData.html

Conclusion

Today and yesterday has involved a lot of Semantic Web and Social Web conversation. My brain has digested a lot of new information, and I really want to share it with you all. I try my best to put things in terms that people can understand, but if you have any questions then please do shoot them at me.

Technorati Tags: semantic web, sweo, linkeddata, social web