Hi Andrea,

is this an IKS or Stanbol question? I understand that you can not maintain
this component anymore. So we should look for other people/new
contributors. In my understanding this is the usual flow in an open-source
project. I don't see any reason to remove the component.

Best,
 - Fabian

Am Freitag, 16. Dezember 2011 schrieb Andrea Giovanni Nuzzolese :

> Hi all,
>
> I'd like to bring to your attention the current state of the Reengineer
> component and its next future (if any) in the project.
> Before being part of Stanbol, the design and the implementation of this
> component was started by the CNR team in the IKS project.
> Right now, we at CNR have no more resources in order to put effort in its
> maintenance. I am wondering if someone in Stanbol wants to continue to keep
> care and contribute to the evolution of the reengineer otherwise we should
> discuss about  its possible removal from Stanbol. My personal opinion is
> that it could be somehow merged with the cmsadapter.
> I remember you that the main functionality of the reengineer is to extract
> RDF triples from both relational databases and XML files. The
> transformation (reengineering) is performed according to an ontology [1,2]
> which describes the structure of the original source. In case of RDBMS, the
> output of the component is a RDF dataset described (rdf:type triples) in
> terms of tables, columns, raws. Primary keys and foreign keys allows to
> generate respectively URIs and relations between objects in the dataset. On
> the other hand in case of XMLs. the output is always a RDF dataset but the
> fact it is described in terms of elements, attributes and in general any
> descriptor part of an XML and XMLSchema.
>
> Note that parts of Stanbol Reengineer, namely the base API and the XML
> plugin, were extensively used in an important knowledge extraction
> component of the AmI Case demonstrator. This was billed as an effective
> usage example of the IKS, and one of the earliest at that, so at least
> those parts should survive in some form. Especially since the AmI
> environment will be setup again for the Y3 Review. Even having them as
> external bundles in IKS and not in Stanbol would be acceptable.
>
> --
> Andrea Giovanni Nuzzolese
> Semantic Technology Laboratory (STLab)
> Institute for Cognitive Science and Technology (ISTC)
> National Research Council  (CNR)
> Via Nomentana 56, Roma - Italy
>
> [1] http://ontologydesignpatterns.org/ont/iks/dbs_l1.owl
> [2] http://ontologydesignpatterns.org/ont/iks/oxml.owl
>
>

Reply via email to