After some weeks of silence, Sebastian asked for a pre-release meeting for the 0.2 release of xOperator which we held today. We updated the issue tracker and talked about future plans and ideas. For this, Jörg and Sebastian created presentations:

We plan the xOperator 0.2 release for the last week of August. Overall, this release includes all the features which I’ve presented at the ESWC 2008 Demo session. In addition to this, we will present a web-based configuration tool where users can easily create their own xOperator setup file.

I will give a talk about our xOperator project at the European Semantic Web Conference (ESWC2008) on Tenerife next Wednesday. The xOperator is an agent for xmpp / jabber network which queries and shares trusted resources (using RDF/SPARQL) for you and your jabber friends. In addition to the full paper, we got good reviews for the demo session too. So I present the xOperator in a more technical way at the demo session on Tuesday.

Another project which we started here is RDFAPI-JS. I present this work at the Scripting for the Semantic Web Workshop (SFSW2008) which is co-located with the ESWC. RDFAPI-JS is used for RDFa widgets which can be used to modify embedded RDF models.

Thanks to Jörg, Martin and Seppl (and of course Haschek for the awesome poster) which helped me preparing the demo and the presentations. Sorry for this last week …

For more information about the projects, have a look at the following publications:

• xOperator - Interconnecting the Semantic Web and Instant Messaging Networks (Sebastian Dietzold, Jörg Unbehauen and Sören Auer)
• Using JavaScript RDFa Widgets for Model/View Separation inside Read/Write Websites (Sebastian Dietzold, Sebastian Hellman and Martin Peklo)

The xOperator Project is moved to Google Code which is much faster (the tracker, not the subversion repository) and has some nice features (e.g. the customizable grid view).

Traditionally we use SourceForge.net for our open source project management. SF is a great side and I’m member of 9 projects there but it was time to move. Basically its a matter of speed. I work every day on tracker items like bug reports, feature requests and tasks for one of our projects and the sf-page is very slow in comparison to google code and also in comparison to our tracker for commercial projects (we use mantis for that).

Project resources of xOperator are now Issue Tracker, Downloads and Mailinglist (moved to google groups). Our homepage @ aksw.org does not move.

xOperator combines advantages of social network websites with instant messaging. It tries to be a semantic agent for xmpp / jabber network which finds and shares content about resources (using RDF/SPARQL) for you and your jabber friends.

The first public version of xOperator is now available for dowload. It includes the following features:

• basic xmpp account management (user account, proxy account)
• management of query templates, datastores and namespaces directly with your jabber client
• simple query templates for easy starting with xOperator
• query scripts in groovy for advanced queries and post processing

Links: Homepage, Changelog, Mailing list

Fresh from the subversion repository, our new semantic agent project xOperator is now able to play with my local OntoWiki knowledge bases. The initial development to this tool will be done by Jörg Unbehauen as part of his master thesis.

xOperator tries to create a decentralized network of trust. Information is shared in the form of resources, which can be anything (even the information that you are selling your old washing machine). In a later stage it enables you to find friends of friends like centralized services, can already. At the end of the road, xOperator might even be able to find the best product for you without the normally necessary tedious internet research. Companies will be able to offer semantic annotated information about their products, giving the customer a new way of yet unknown price transparency.

Although xOperator is able to use any SPARQL endpoint out there, our main use case is to marry it with OntoWiki. Users should be able to use the agent just as an additional view to the knowledge bases of their OntoWiki installation.

To do this, it will be possible not only to ask the agent with special query languages like SPARQL but also to ask him with natural language. To achieve this, xOperator will map templates based on the Artificial Intelligence Markup Language (AIML) to SPARQL queries and transform the XML result set to the instant message which will be send back to the user.

Implementation Status

The agent is able to log into existing accounts and can receive querying and configuration commands. The template based query generation is still a bit quirky. Queries can only be executed against locally configured data stores, the p2p part is not yet implemented.

If you’re interested in the project, please write to the public mailing list xoperator-public in our SourceForge area. The official homepage is http://aksw.org/Projects/xOperator