The Web of Data: Closed or Open?

March 18th, 2016

Wouter Beek (wouter@triply.cc)

Problem 1

Not everyone has been able to publish something yet

Why does not everyone publish on the SW?

Metcalfe's Law

The value of a network is proportional to the square of the number of connected nodes

So... how many connected nodes are there?

Most nodes have low availability

Most nodes enforce restrictions

Growth is linear

Why?

[1] Allocative Efficiency

What the consumer pays should equal the marginal cost of production


[2] Multi-tenancy

Difficult in a decentralized deployment


Beek & Rietveld & Schlobach & Van Harmelen “LOD Laundromat: Why the Semantic Web Needs Centralization (Even If We Don't Like It)” IEEE Internet Computing 20 (2) p.78-81, 2016

Problem 2

Not everything that is published can also be accessed

AAA

Anyone can say anything about anything

Authority

IRI <auth:London> dereferences to auth's view on Lodon

What if a search for “U.S. foreign policy” would only show results published by the White House?

AAA

Even though everyone can say anything about anything, very few people are actually heard

But we can travel the graph, right?

“Follow your nose”

Fixing “Follow your nose”

LOD Laundromat


Beek & Rietveld & Bazoobandi & Wielemaker & Schlobach “LOD laundromat: A Uniform Way of Publishing Other People’s Dirty Data” ISWC 2014

Semantic Web Layer Cake

LOD Laundromat Layer Cake

Rietveld & Verborgh & Beek & Vander Sande & Schlobach, “Linked Data-as-a-Service: The Semantic Web Redeployed” ESWC 2015

Frank

Federated Resource Architecture for Networked Knowledge

https://github.com/LOD-Laundromat/Frank

W. Beek & L. Rietveld. “Frank: The LOD Cloud at your Fingertips” Extended Semantic Web Conference: Developers Workshop, 2015.
http://lodlaundromat.org/lodlab


Rietveld & Beek & Schlobach “LOD Lab: Experiments at LOD Scale” International Semantic Web Conference 2015 (Best Paper Award)

Problem 3

Anyone can say anything, but not everyone is a logician

bbc:sameAs

bbc:sameAs

owl:sameAs

?

AAA

Anyone can say that anything is identical to anything else
(and they do)

Leibniz's Law

$$a = b \leftrightarrow (\forall \phi \in \Psi)(\phi(a) = \phi(b))$$

Relatedness cannot replace identity

SKOS exactMatch indicates a high degree of confidence that two concepts can be used interchangeably across a wide range of information retrieval applications
From the SKOS standard

Problem 4

If you do not know SPARQL then you cannot tap into today's SW

Fixing the NatLang interface

  • Efficient computation through canonicity
  • Data enrichment by improved instance matching
  • User eXperience: language preference, “value labeling”
  • Improve NLP tasks with background knowledge


Ilievski & Beek & Van Erp & Rietveld & Schlobach “Literally Better: Analyzing and Improving the Quality of Literals” European Semantic Web Conference 2016 Beek & Ilievski & Debattista & Schlobach “Literally Better: Analyzing and Improving the Quality of Literals” Semantic Web Journal 2016 (under submission)

LOTUS

lotus.lodlaundromat.org

Datatypes

Language tags