W3C home > Mailing lists > Public > public-esw-thes@w3.org > November 2014

RE: ANN: VocBench 2.2 released

From: Armando Stellato <stellato@info.uniroma2.it>
Date: Mon, 10 Nov 2014 12:15:01 +0100
Message-ID: <DUB408-EAS2980D6D78A6525117241597A0800@phx.gbl>
To: "'Osma Suominen'" <osma.suominen@helsinki.fi>, <public-esw-thes@w3.org>
Dear Osma,

thanks for the congrats :) 
yes basically it is possible to add connectors for other triple stores,
which can be plugged as additional OSGi bundles.

Additional information can be found here:
http://semanticturkey.uniroma2.it/documentation/dev/modules/st-sesame2-ontma
nager/

Basically, we have a lightweight RDF layer (OWLART api:
http://art.uniroma2.it/owlart/ ) wrapping existing middlewares such as
Sesame or Jena. Then we have an OntologyManager
(http://semanticturkey.uniroma2.it/documentation/dev/modules/st-core-framewo
rk/apidocs/it/uniroma2/art/semanticturkey/ontology/STOntologyManager.html)
inside Semantic Turkey (that needs to be subclassed) in order to provide
additional features such as recursive import of ontologies, declaration of
"system" ontologies (e.g. data which shouldn't be shown to the user) etc...

Thus, writing a connector from scratch amounts to:
1) provide an OWLART implementation for a given middleware <MX>
2) provide the implementation of STOntologyManager for <MX>, using the
OWLART implementation above

For what concerns what we are maintaining/providing: there have been other
connectors in the past, just the only connector we maintain regularly is the
sesame one (maybe in the future we will add more to these "regularly
maintained").

So, if other triple stores can be queried through a Sesame2 remote
connection, things *should be* pretty easy, as it may require:
1) just using the sesame2 remote connection :-)
2) or writing only an extension for Sesame2 OntologyManager in case the
middeware requires a dedicated client (such as Allegro or Virtuoso), thus
reusing all of it code (except maybe the initialization) and only adding the
required client

I said *should be*, as API compliancy is often limited to being able to
process method calls and return "something". The problem is that
"something": there are differences on how OWLIM manages graphs wrt Sesame2
(see this msg on the ontotext mailing list [1]).

We have reported our experience with OWLIM-SE, in which we customized an
OWLIM rule-set (we had to customize it, see [2]) and managed to make
OWLIM-SE work with VB, with a plain sesame remote-connection.
Very recently, we have been working on building an analogous ruleset for
OWLIM-Lite: if you want, we can share privately our findings.

Cheers,

Armando

[1]
http://answers.ontotext.com/questions/1166/null-context-inference-and-explic
it-triples-from-other-graphs-inconsistency-with-sesame2-semantics
[2]
http://answers.ontotext.com/questions/1097/skosnotation-and-skos-label-prope
rties-classified-as-object-properties


> -----Original Message-----
> From: Osma Suominen [mailto:osma.suominen@helsinki.fi]
> Sent: Monday, November 10, 2014 7:58 AM
> To: public-esw-thes@w3.org
> Subject: Re: ANN: VocBench 2.2 released
> 
> Hi Armando!
> 
> Congratulations for the release!
> 
> I've been looking at installing VocBench 2.2 for a test drive and I'm
wondering
> about the triple store to use. I see that you recommend the use of OWLIM
> (nowadays also called GraphDB), but Sesame is also an alternative for
small
> datasets.
> 
> Am I right that OWLIM Lite (the free version) is not an option? I.e. the
> minimum is actually OWLIM SE? At least your install documentation only
> mentions SE.
> 
> What are the issues in supporting other triple stores, such as Jena
Fuseki,
> Virtuoso or Bigdata? Am I right that this depends mainly on triple store
support
> in Semantic Turkey? Do you think it would be easy to add support for other
> triple stores into VocBench and/or Semantic Turkey? For example, Fuseki,
> Virtuoso and Bigdata include all text index support (but all in their own
ways
> due to lack of standards).
> 
> I'm asking because OWLIM/GraphDB is a commercial, proprietary product and
> for our organization, a stack based on open source software would be more
> welcome.
> 
> Thanks,
> Osma
> 
> 
> On 04/11/14 18:10, Armando Stellato wrote:
> > *** apologies for cross-posting ***
> >
> > Version 2.2 of VocBench has been released today, and is available here:
> > http://vocbench.uniroma2.it/downloads/
> >
> > Since this is the first announcement for this platform (on
> > non-internal lists), here are few words for introducing it.
> >
> > VocBench is a web-based, multilingual, editing and workflow tool that
> > manages thesauri, authority lists and glossaries using SKOS-XL.
> > Designed to meet the needs of Semantic Web and Linked Data
> > environments, VocBench provides tools and functionalities that
> > facilitate both collaborative editing and multilingual terminology
> > development. It also includes administration and group management
> > features that permit flexible roles for maintenance, validation and
> publication.
> >
> > VocBench 2 is being developed in the context of a collaboration
> > between the Food and Agriculture Organization of the United Nations
> > (FAO) <http://www.fao.org/> [http://www.fao.org/] and the ART Group
> > <http://art.uniroma2.it/> of the University of Rome 'Tor Vergata'
> > <http://www.uniroma2.it/> [http://art.uniroma2.it/]
> >
> > Interested people can learn more about VocBench by visiting its
> > project
> > site: http://vocbench.uniroma2.it/ and/or by subscribing to its
> > user/developer discussion groups: http://vocbench.uniroma2.it/support/
> >
> > Cheers,
> >
> > The VocBench team
> >
> > http://art.uniroma2.it/
> >
> > http://aims.fao.org/
> >
> 
> 
> --
> Osma Suominen
> D.Sc. (Tech), Information Systems Specialist National Library of Finland
P.O. Box
> 26 (Teollisuuskatu 23)
> 00014 HELSINGIN YLIOPISTO
> Tel. +358 50 3199529
> osma.suominen@helsinki.fi
> http://www.nationallibrary.fi
Received on Monday, 10 November 2014 11:15:39 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 2 March 2016 13:32:22 UTC