Monthly Archives: October 2011

Helmut Nagy

Geological Survey Austria launches thesaurus project

by , Category: Companies & Institutions, Linked Data & Open Data

Throughout the last year the Semantic Web Company team has supported the Geological Survey of Austria (GBA) in setting up their thesaurus project. It started with a workshop in summer 2010 where we discussed use cases for using semantic web technologies as means to fulfill the INSPIRE directive. Now in fall 2011 GBA published their first thesauri as Linked Data using PoolParty’s new Linked Data front-end.

The Thesaurus Project of the GBA aims to create controlled vocabularies for the semantic harmonization of map-based geodata. The content-related realization of this project is governed by the Thesaurus Editorial Team, which consists of domain experts from the Geological Survey of Austria. With the development of semantically and technically interoperable geo-data the Geological Survey of Austria implements its legal obligation defined by the EU-Directive 2007/2/EC INSPIRE and the national “Geodateninfrastrukturgesetz” (GeoDIG), respectively.

Marcus Ebner, from the GBA Thesaurus Editorial Team

Marcus Ebner, from the GBA Thesaurus Editorial Team

The construction of the thesauri has been done using the PoolParty Thesaurus Manager so they all are based on SKOS and fully compliant to the Linked Data principles. Apart from the standard implementation of SKOS some additions were made to the data model using Dublin Core terms for extra metadata and custom sub properties of skos:related to give some semantic constraints to related properties. This basically means that a big effort was put into the integration of bibliographic references for every concept in the data set using dcterms:source. This aims at the requirements of reuse by the scientific community and incorporation in domain specific data sets. On the other hand rdfs:subProperityOf was used to express how international geologic time scales map on regional concepts.

Currently four thesauri have been published, all are available in English and German and can be used under the cc-by-sa license. Also mappings to DBpedia have been made:

With the new PoolParty Release (3.0) the Linked Data front-end has been redesigned and is now highly customizable and extendable. In the GBA Thesaurus Project it is used as an publishing interface for the created controlled vocabularies both for the machine readable RDF version and an custom HTML version for comfortable browsing and searching.

GBA Linked Data frontend

GBA Linked Data frontend

After all it’s satisfying to see a project we’ve supported and worked on for some time now come to live and now we are looking forward to the next steps that will be done in this project.

P.S.: Thanks to Marcus Ebner from GBA for his contribution to his blog post.

TassiloPellegrini

TimBL @ Hofreitschule in Vienna

by , Category: Conferences & Events, Miscellaneous, Uncategorized

Right now, 19:30, Tim Berners-Lee is giving a key note on the future of the internet at the Hofreitschule in Vienna, a marvellous, historic venue in the very city.

His talk is a plaidoyer for an open internet, that works independent from central control  and political implications, on top of open standards AND net neutrality. This is especially relevant when it comes to open data, where the social machinery of the web will help to improve many flaws democracy is facing today.

So, what are the implications: Study Web Science! And trigger gentle, non-violent change!

It was a pleasure to listen!

 

 

Enhanced by Zemanta
Andreas Blumauer

rNews and its benefits for publishers

by , Category: Internet & Media, Vocabularies & Languages

Last Wednesday at the Open House event of the Semantic Web Company in Vienna, Evan Sandhaus, Lead Semantic Architect at NY Times gave a comprehensive and entertaining introduction to rNews and its potential benefits for publishers.

Evan Sandhaus (f.l.t.r) busy preparing his talk in the kitchen of SWC, together with Andreas Blumauer (SWC) and Leo Sauermann (Gnowsis). Mr. Sandhaus in action.

rNews is a RDFa vocabulary, which is basically a carefully selected subset of the very rich IPTC vocabulary and some additional elements that came up during the standardization process. It is now available in version 1.0 and – according to Evan – actively supported by schema.org.

As showed above the data model of rNews is really simple and centered around two classes: the NewsItem and the Concept. This deliberate simplicity is a major advancement compared to standards like NewsML (whose complexity probably prohibited its critical uptake among the news industry). But due to the functional extensions attributed to RDFa, rNews might also be considered more complex than hNews, the microformat equivalent issued by the IPTC in 2009.

Evan mentioned three scenarios that might drive the uptake of rNews for the benefit of news publishers:

1) Better news search

rNews allows you to explicate and differentiate various documents elements like, title, author, text body, picture etc., thus giving the publisher better control of what to expose for indexers and web crawlers. This might not just improve the display of rich snippets in the search results of Google and other search engines, but also allow automated population of faceted search and metadata based similarity search.

2) Better ad placement

As rNews can be applied to any kind of news-relevant media irrespective of its format (grafics, audio, video, etc.) the metadata can be used to avoid “unfortunate juxtapositions” between editorial content and ads. Hence, media agencies could profit from this additional data by fuelling their matching algorithms and gain better insight into the context specificities of content items.

3) Better analytics

By improving the semantic granularity of a news item this additional information can be used to carry the web analytics beyond the page level and provide a better insight into usage patterns. The additional data can be applied for visualization and exploration purposes i.e. for search engine optimization, sentiment detection and many more.

This is just a small fraction of things rNews could be used for. All in all it is exciting to see that IPTC has finally started to provide publishers with a standard that is relatively easy to implement and help them to overcome the obstacles of existing technologies without disrupting existing publishing workflows. In multi-sided markets like the news industry this might be a crucial success factor!

 

Enhanced by Zemanta
Andreas Blumauer

rNews meets Wiener Schnitzel

by , Category: Conferences & Events

IPTC (International Press Telecommunications Council) has developed rNews, a set of specifications and best practices for using RDFa to embed news-specific metadata into HTML documents. Just recently the initiative received new momentum when schema.org announced that it has added support for rNews (see press release from Sep. 27, 2011).

Alltogether this is another very important building block for an even broader adoption of the semantic web – and the Semantic Web Company is proud of being able to welcome some of the proponents of rNews in Vienna.

Tomorrow at our first “Open House” event (which is proudly presented by the Vienna Semantic Web Meetup) not only our guest list but also the speakers list is very promising:

Live Stream on Justin.tv

We will report about the talks on “the day after” but we are sure that this meetup will bring a lot of light into the ongoing discussion why media and press companies finally should start to adopt semantic web standards.

 

Related articles
Enhanced by Zemanta