Did you know the French business, professional and individual phone directory had been semantified? The site,11800, is presented by an Internet marketing and technology company called SecondWeb with the help of the CubicWeb semantic web framework from France-based Logilab.
That’s probably the biggest public web site that’s build using the framework. But about 70 percent of Logilab’s business now is around using the framework it originally developed for internal use to build for its customers applications that rely on its object-oriented design model of using reusable data model and view components â€“ or ‘cubes’ â€“ that are their own entire applications providing data models, which then can be piled together in ‘constructions’ that integrate multiple types of sources and publish semantic data. The semantic views already integrated into the framework for publishing data include SIOC, OWL, FOAF, and DOAP ontologies.
All that’s needed to get started building a CubicWeb application is a data model described by its entity-relationship. Each CubicWeb application, explains Logilab software engineer Sandrine Ribeau, can use another CubicWeb application as a source of data for easier aggregation. It also can use an LDAP source and map the data to a CubicWeb data-model, and other SQL sources within an application. Data previously was queried only through its Relationship Query Language (RQL) since SPARQL wasn’t developed at the time the framework was developed, but it recently added a SPARQL interface.
“The idea behind this is to try to make the development of an application as easy as possible, so you only want to assemble different types of components,” she says. “It’s meant to be totally transparent for the user and the developerâ€”you can’t tell where data is coming from.”
One interesting thing about what Logilab has been doing is that it actually released the semantic web application framework under the LGPL as an open source project, and so developers have been creating what Ribeau says is a strong library of usable components. One project is to add RDF web sites sources like dbpedia. “The idea is to get more contributions and extend the community of users,” she says. She’s hoping to get more attention to the framework, which despite the contributions still doesn’t have a huge developer following.
One thing she hopes will spur more attention to the framework is that it’s about building your own web of data, using semantic data published on the web and publishing semantic data, while preserving your privacy in a permissions-based environment. “We feel it’s important to developers and users of applications to be able to say I am using this data but I don’t want everyone to know what I’m using and interested in,” she says. “I think it’s very important for developers to be able to set some rules on privacy. The framework is a permission-based environment based on content, on the type of data, on relations between data. It’s all done in the data model.” To find out more about the open source project, go here .