Typo cleansed edition :-)
Objectives
- Meet LOD Community Members
- Participate in Workshop
Meeting LOD Community Members
Although the Web continues to
shrink the planet by removing the restrictions of geopgrahic
location, meeting people face-to-face remains invaluable
(*priceless in Mastercard AD speak*). Naturally, meeting and
chatting with as many LOD community members as possible was high
up on my agenda.
Participate in Workshop
As one of the co-chairs of the Linking Open Data
Workshop (LODW), I had a 5 minute workshop opening slot
during which I spoke about the following:
Where we are today:
We have DBpedia as a major hub on the burgeoning
Linked Data Web. When OpenLink offered to host DBpedia (a combination of Virtuoso
DBMS Software and sizable backend Hardware infrastructure), it did
so knowing that such an effort would emphatically address the
"chicken and egg" conundrum that, prior to this undertaking,
stifled the ability to demonstrate practical utility of HTTP based Linked Data.
Today, the Linked Data bootstrap mission has been
accomplished.
Where we go next:
Although DBpedia is a hub (ground zero of Linked Data), we have to put it into
perspective in relation to a new set of needs and expectations
moving forward. Today, DBpedia is a Sun at the heart of a Solar
System within the Linked Data Galaxy. But unlike Space as we
know it, in Cyberspace we can have connectivity and
collaboration across Solar Systems -- life exists elsewhere and we
are part of a collaborative collective unimpeded by constraints of
space travel etc. Thus, expect to see the emergence of other Solar
Systems accessible to DBpedia and its collections of planets
(see. LOD diagram). Examples underway include UMBEL which will
serve the Linked Data planets from OpenCyc
(Subject Matter Concepts), Yago (Named Entities), and Bio2RDF (which
provides powerful Bio Informatics based Linked Data planet).
I urged the community to veer more aggressively towards
developing and demonstrating practical Linked Data driven solutions that are aligned
to well known problems. Of course, I encouraged all presenters to
make this an integral part of their presentations :-)
Workshop Summary:
The workshop was well attended and I found all the presentations
engaging and full of enthusiasm.
As the sessions progressed, it became clear during a number of
accompanying Q&A sessions that a new Linked Data exploitation frontier is emerging.
The frontier in question takes the form of a Linked Data substrate capable of addressing
the taxonomic needs of solutions aimed at automated Named Entity Extraction, Disambiguation,
Subject matter Concept alignment, transparently integrated with
existing Web Content. Thus, we are moving beyond the minting and
deployment of of dereferencable URIs and RDF data
sets to automagically associating existing Web Content with
Named Entities (People, Organizations,
Places, Events etc..) and Subject matter Concepts (Politics, Music,
Sports, and others) while remaining true to the Linking Open Data Community creed i.e.
ensuring the Named Entity and Subject matter Concept URIs are
available to user agents or users seeking to produce alternative
data views (i.e. Mesh-ups).
I will get to part 2 of this report once the actual workshop
sessions slides go live (*these are different from the pre-event PDFs links*).