WWW2008 Beijing: Day 1 – Linked Data on the Web (LDOW 2008) WorkshopBy David Peterson
Just arrived half way through the workshop… long day so far. Left from Sydney last night and haven’t slept yet, I hope I don’t nod off during the sessions.
Right now I listening to Tim Berners-Lee talk about linked data browsing and his frustration that the Web is a Read only medium. He and a small team (over the Christmas break) worked on a second version of their linked data browser Tabulator. This update took Tabulator from the Read/Browse only domain into the Write domain via a SPARQL update query.
Tim has always been very passionate about the ability to write back to the Web. The very first Web clients he wrote had the ability to write as well as read. This ability was lost in the early days and is just now surfacing again. He did admit that some would consider write is already available via blogs and wikis, but he won’t be happy until much more of the Web is writeable.
This stuff is all very new. SPARQL was just recently release as a W3C spec. The move is on to start work on SPARQL 2. One of the key ingredients is going to be the ability to perform updates. More on this later.
This editing/updating should be WYSIWG, if something is right on the screen you can just click on it and edit it. This of course is all in a structured, interlinked format of RDF. That is what makes it different to a wiki or a blog.
Tabulator looks at the whole global graph, not just single instances of data or files. So, if you are updating global data, where do your edits go? It will go back to the original source as long as it is writable.
The next steps they are working on is to integrate this with OpenID so that the security can be maintained and you can decide in a decentralised way.