ann: Semantic Web Pipes

What are Semantic Web Pipes?

Using Semantic Web Pipes you can fetch, mix and process RDF files
published on the Web. As the output of a Pipe is an HTTP retrievable
RDF model, simple pipes can also work as inputs to more complex Pipes.

This is similar to what Yahoo Pipes does, but highly targeted to allow
live Semantic Web data recombination. We currently support 3 simple
RDF Pipes modules  (Fetch, Mix and SPARQL Transform)  but expect more
soon (including a novel "revocation" module for cooperative RDF

Pipes are written in a simple XML. E.G. a mixing 2 sources is as
simple as this:

     <source url=""/>
     <source url=""/>

The result can be displayed by invoking the pipe URI, in this case:

Pipe modules can be nested for more complex operations. For example
the following pipe will fetch Tim Berners-Lee's publications from DBLP
(linked data version), transform the RDF so to use use Tim
Berners-Lee's official URI and merge it with his FOAF file.

    <source url=""/>
     <source url=""/>
  <construct>  <![CDATA[

  CONSTRUCT {<> ?p ?o.
                  ?s2 ?p2
<>} where
{{<> ?p ?o}
{?s2 ?p2 <>}}


The result of this pipe can be displayed here

Please feel free to play with the online ajaxy editor (i.e. does some
XML copypasting for you) and create cool pipes for other to reuse
(some example provided)


All the implementation kudos go to Danh Le Puoch (DERI Galway);
Christian Morbidoni (SEMEDIA, Universita' Politecnica delle Marche,
Italy), Axel Pollares (DERI Galway) and me are to blame for the rest.


Received on Thursday, 1 November 2007 04:20:10 UTC