Over recent years the Linked
Open Data paradigm has gained momentum as a lightweight approach of
publishing data resources to and reusing data from the web - be it
for commercial or commons-oriented purposes.
Free Linked Open Data
Consultancy
PUBLINK is the Linked Open Data
Consultancy backed by the
consortia of the EU-FP7 LOD2 Integrated
Project. In order to lower the entrance barrier for potential
data publishers and tool providers, the LOD2 consortium offer the
free PUBLINK
Linked Open Data Consultancy to up to 5 selected organizations
supporting them with the publishing of Linked Open Data with an
overall effort of 10-20 days of support from highly skilled Linked
Data professionals.
- Are you a government or
government-related organization (administration, research center,
agency, …) and in possession of large amounts of data?
- Is your organization obliged to publish data?
- Do you have valuable data but lack the financial or
technical means to distribute it?
- Are you convinced that linking, merging or mapping your data
with other data sources will increase its value?
- Do you face inter-organizational data integration
challenges?
- Do you want or need your data to be re-used?
- Do you already publish tons of data in HTML, spreadsheets, PDFs
or proprietary formats?
If you answered
YES! to two or more of these questions, you have to continue
reading.
In addition
to data publishers, PUBLINK offers this year specific
support for Linked Data tool developers, who want to
integrate their tool with the Debian-based LOD2 Stack.
Why
the Linked Open Data
Starter Service?
- Is the basis for data
integration/mapping/linking on the Internet.
- Puts you in pole position for data
publishing.
- Empowers the world around you.
- Is the ultimate extension of your role as knowledge center,
regulator, publisher.
- Guarantees a huge leap forward in your data distribution
strategy without any risks.
You will benefit
from the technological, juridical, marketing, project and
communication experience of the multi-disciplinary LOD2 team to
make your first steps into the linked open data space a
success.
What
PUBLINK does for you.
It is a service provided by the
EU-supported project LOD2. The ultimate aim of the project is to
lift the emerging Web of Data to the next level. Besides research,
development and deployment in the areas linked data management,
interlinking, enrichment as well as browsing and authoring
LOD2 aims to substantially extend
the publication and use of Linked Data within and beyond the focus
domains of E-Government, Media and Enterprise Data. Through a
standardized phased approach it allows you to migrate your data to
the Internet in a smooth way at your pace.
PUBLINK Pathfinder
In a
one-day workshop the LOD2 team will crawl through your data
together with you. You should have your data or samples of it
available and a rough idea of what you want to do with it. At the
end of the day the LOD2 team will, together with you, have:
defined a high-level specification, a plan for next steps, a
high-level technical architecture, assessed risks and challenges,
identified possible non-technical issues to work on (copyright,
privacy, security, …).
PUBLINK Kickstart
In a 5-10-day mini-project
the LOD2 team will support you to bring (a) your data to the
internet as linked open data, or (b) integrate your tool with the
Debian-based LOD2 Stack. Basically, you have
the following options:
- PUBLINK RDF:
Linked Data at its core is in fact nothing more than the data
published as RDF and then linked to each other.
- PUBLINK SPARQL endpoint: SPARQL is the query language for RDF data, like SQL
for relational data bases.
To query an RDF Store, a
SPARQL endpoint is needed. This is the gateway to the RDF Store
through which queries are executed. The web service accepts other
formats as well and transforms them into SPARQL.
The SPARQL endpoint provides a means for machine-to-machine
communication so that the public presence of data can be fully
exploited, and foresees the needs of future web applications.
Here are examples of existing SPARQL
endpoints: http://esw.w3.org/topic/SparqlEndpoints
- PUBLINK tool: You are developing a tool for
Linked Data (or RDF) extraction, publishing, visualization,
exploration or consumption? The LOD2 Stack
is a integrated set of tools covering the whole life-cycle of
Linked Data and we welcome 3rd party contributions. Since the LOD2
Stack is based on the popular Debian software packaging system,
tools being integrated with the LOD2 Stack are extremly easy to
deploy and use. The main requirement for a tool to become part of
the LOD2 Stack is that it produces and/or consumes RDF, which can
be obtained or stored in a common SPARQL endpoint (Virtuoso is
primarily used in the LOD2 Stack for that purpose). In this variant
of PUBLINK we will support you with creating a Debian package of
your tool and integating it with other components of the stack (cf.
also our
HOWTO Contribute).
Who
should apply?
Target organizations for Linked
Open Data Starters include (but
are not limited to):
- Government administrations or
agencies, regional, national or European
- Knowledge center (health, infrastructure, chemical, tax,
financial, social, economic, biological, education, labour, …)
- Bureau for statistics
- Research centers
- Enterprises aiming to publish Linked Open Data
- RDF tool developers
Examples of Completed Publink
Projects
- In 2010 the European Commission Directorate General
CONNECT learned how to translate a PDF document with
performance indicators into a linked open data graphical
environment on the web for the Digital Agenda Scoreboard.
- In 2011 the Statistical Office of Serbia
learned how to transform their data into RDF and understood the
value of the Datacube standard for statistical information.
- In 2011 the Food and Agriculture Organization
of the United Nations learned how to manage equivalent
entities.
And many more organizations have been supported by the LOD2
consortium. Don't you also want to tap into the knowledge,
experience and expertise of this dynamic and pragmatic research
consortium?
How
In order to apply
for PUBLINK, please send en email with the following information
(max. 2 pages) to publink@lod2.eu:
- Brief profile of your
organization
- Description of the data
(quality, quantity, current format) or
tool (purpose, technology, interfaces) you aim to publish
- Potential stakeholders benefiting from publishing your data or
tool
- Envisioned involvement from your side - will you designate
staff for supporting the PUBLINK project and being responsible for
the maintenance of the data when the PUBLINK project will have
terminated
- Envisioned licensing model (in
general PUBLINK is aimed at publishing under open
licenses,
see http://opendefinition.org)
After assessment
of your application and a possible discussion with you on the value
and feasibility of your application, we will select up to five
PUBLINK projects for execution.
Timeline
- Application deadline: December 31st,
2012
- Notification of acceptance: January 30th, 2013
- PUBLINK pathfinder period: February & March 2013
- PUBLINK kickstart period: April till June 2013
For successful
PUBLINK projects we will, of course, offer individually negotiable
options for a continuing collaboration between (members of) the
LOD2 consortia and your organization.
What is Linked Open Data?
The Internet
previously was merely a virtual library where documents were linked
to each other, it was a web of documents for humans to read. The
W3C-hosted Linking Open Data initiative (LOD, http://esw.w3.org/SweoIG/TaskForces/CommunityProjects/LinkingOpenData)
now extended the 'web of documents' with a 'web of data' by
publishing and interlinking open data sources on the Web based on
well-established standards such as RDF and URIs.
A web comprising
linked data in addition to linked documents brings the advantage
that the content becomes machine-readable. Computer and software
can interpret the meaning of web content instead of just offering
it to the user to read. As a result, more intelligent search
engines (cf. schema.org or Google's knowledge graph) can combine
information from different sources, mash-ups integrating
heterogeneous information can be more easily built and many more,
currently unforeseen creative uses of information on the web become
possible.
With more than 30
billion facts already published as Linked Open Data (LOD) the Data
Web is not just a vision, but a reality right now. For example, all
BBC programming, Wikipedia as a structured knowledge base (DBpedia)
and statistical information from Eurostat and the US census are, in
addition to hundred of other datasets, readily available on the Web
of Data.