Detailed session descriptions

Session 1 – Web of Data 101

Keynote: Yves Raimond, Keith Alexander
Session Chair: Michael Hausenblas

Abstract
This track can be understood as a base session where participants (i) learn the basics regarding the Web of Data, and (ii) get an overview of the achievements made so far.
As a starting point we will discuss basic issues regarding the Web of data including URI design, the RDF data model, (re)using vocabularies, and deployment technologies. Equipped with the linked data principles the participants will be able to understand how linked datasets are build and deployed. The discussion of the foundations is always performed in the context of real-world setups, that is, we use widely deployed and usable datasets such as DBpedia, Geonames, DBLP, etc. to elaborate on design issues and design decisions along with their impact. We will present and use tools suited for Web developers to integrate with their everyday environment, allowing them to create, to access and extend current linked datasets.
In the second part of the WoD101 we will focus on the achievement made so far. This discussion spans from early attempts of creating linked data (DBpedia) over linked-data applications (revyu.com, etc.); we will have a look at new ways of exposing linked data such as RDFa – for example in riese.joanneum.at or the Library of Congress Subject Headings – and highlight current work regarding dataset discovery (voiD), multimedia interlinking, and applications (e.g. DBpedia mobile).

Session 2 – Semantic Desktop Solutions

Keynote: Leo Sauermann, Brian Davis
Session Chair: Bernhard Schandl

Abstract
A large fraction of our daily lives deals with digitally represented information: we communicate via email, instant messaging, weblogs; we order products and services via the World Wide Web; we consume digital media like movies and music; we organize our last holiday's pictures on our computers, and much more information is managed digitally.
However, users of typical computer systems face similar problems as found on the Web: large amounts of information are stored in unstructured ways, and data are scattered across different systems. The problem increased over the last years through the plummeting cost of disk space, allowing us to store more and more information. Desktop operating systems provide no application-independent organizing scheme, the filesystem folder hierarchy is separated from the e-mail folders or contact categories. Thus it is hard to organize and retrieve personal information, both in the private and in the business domain.
The Semantic Desktop is an environment where personal data of different kinds is annotated and related with machine-interpretable information.
Semantic Web technology, like RDF, ontologies, and reasoning, are used to create a seamless personal space of information (PSI) that covers as many aspects of our digital life as possible, in order to make the personal computer a true extension to our brain.
The Semantic Desktop session within the WOD-PD will focus on recent progress in this area. It will discuss existing infrastructures for semantic desktop environments, and it will demonstrate how such components are integrated into existing operating systems and applications. It will show how ontologies, information fusion, and automatic classification can be applied in the domain of personal information management (PIM), and how they can help to convey the user's mental model into digital form. Finally it will analyze how semantics can be injected into the user's normal workflow, and how intuitive user interfaces for semantic technologies can be designed.

Session 3 – Bringing Legacy Data to the Web

Keynote: Sören Auer
Session Chair: Bernhard Haslhofer

Abstract
Currently, a vast amount of data of public interest are available in closed data silos and can be accessed and processed only be the organisations that host these data. The spectrum of public-interest data sources is broad and includes geographic information systems, domain-specific databases (e.g., movie or audio databases), or community driven encyclopaedias such as Wikipedia.
The major goal of the W3C’s Web of Data Initiative is to open these silos and publish the contained data on the Web so that other applications can access, process, and reuse these data in other contexts. This of course requires technologies and solutions that can expose existing, yet closed legacy data sources on the Web, in such a way that humans as well as machines can read and correctly interpret the exposed data.
The "Bringing Legacy Data on the Web" session is about solutions that fulfil that goal in respect to the various underlying technologies legacy data silos are based on. It will give an overview of the current possibilities of exposing data currently available in relational databases, present wrapper components for various Web-based protocols (e.g., OAI-PMH), and light-weight wrapper components for exposing (meta)data accessible via APIs provided by Online platforms such as Flickr or Google.
As soon as previously closed data are available on the Web, they can be interlinked and provide added value for higher-level applications. One could, for instance, interlink image metadata with geographic (GIS) datasets and provide location based search and discovery applications for end users. The "Bringing Legacy Data on the Web" session will also focus on this issue and present tools and heuristic methods for interlinking datasets.
As a result of this session, the visitor obtains an overview about the current possibilities of exposing data from various types of structured, semistructured, and unstructured data sources on the Web. In an accompanying hands-on session the visitors can experiment with available tools and expose sample data sets.

Session 4 – Using the Web of Data

Keynote: Alan Dix, Richard Cyganiak
Session Chair: Andreas Langegger

Links for the hands-on part of Alan and Richard

Abstract
In this session, the Web of Data will be discussed from two different usage perspectives, the programmer's point-of-view and the user's point-of-view. Additionally, it will highlight the fact that a huge amount of open data is already available in standardized formats. This data may be used not only for Web applications, but for any intelligent application requiring some generic or domainspecific knowledge. Regarding the programmer's perspective, several tools and current prototypes will be demonstrated to raise the awareness about what is already possible and what may be achieved in future.
From the end-user's point of view, the Web of Data is also a new paradigm of exploring and browsing data based on semantic links. Beside the availability of additional semantic links to related resources at any time and situation, user interfaces may provide much more intelligent methods of interaction based on semantics of the information being displayed.
Together with contextual information about the user himself (his/her geo position, time, interests, physical condition, etc.) user interfaces and applications will become more intelligent and selfadapting to the particular situation.

Session 5 – Social Aspects in the Web of Data

Keynote: Danny Ayers, Thomas Schandl
Session Chair: Wolfgang Halb

Abstract
The Social Aspects session of WOD-PD deals with the involvement of users in creating semantic information. Social networking web sites are one of the most typical representatives of Web 2.0 that have received an enormous uptake in the past months. Users of such platforms appreciate the possibility of socializing and staying in touch with others online. Several millions of users are part of social-networking services such as Facebook, MySpace, or StudiVZ and the number of users is expected to keep increasing over the coming years.
However, social networking sites are not the only way of socially interacting online. Extremely popular approaches like Wikipedia, weblogs, photo (flickr) and video (youtube) sharing platforms also engage users and user-communities to contribute to a collective knowledge space.
The Social Aspects session will give an insight to novel approaches for dealing with this strongly growing facet of social interactions in the light of the Web of Data. Approaches for converting the strong enthusiasm and user engagement from the Web 2.0 to the Web of Data environment such as the Semantic Mediawiki will be shown. The Friend-of-a-Friend (FOAF) vocabulary can be used to describe people's personal details which might include information about their professional and private live, their friends and all other social dispositions.
Several blogging and social networking sites like for instance LiveJournal already employ the FOAF vocabulary and provide semantically rich descriptions.
The Semantically-Interlinked Online Communities (SIOC) initiative also aims at integrating online community information by providing a Semantic Web ontology for representing rich data from the Social Web in RDF. An increasing number of applications such as the Wordpress blogging platform already supports this technology.
Attendees of the session will receive an overview about upcoming trends and technologies for using the benefits of semantically interlinked social networks and user contributed semantic content. Hands-on sessions will show practical aspects of implementing systems with high user involvement.

Session 6 – Multimedia in the Web of Data

Keynote: Ansgar Scherp, Yves Raimond
Session Chair: Tobias Bürger

Abstract
Since a few years from now the Web more and more turns into a multimedia environment, in which content is published, consumed and shared not only by commercial parties but also by end user contributing a significant amount of user generated content. What is often neglected in recent research of Web technologies is that the Web is a multimedia environment with complex semantics. This fact has however been acknowledged by numerous research projects on a European (e.g. ACEMEDIA, K-SPACE, SALERO, AIM@SHAPE, BOEMIE, …) or national (e.g. GRISINO, Smart Content Factory, ...) level and by W3C activities like the Multimedia Semantics Incubator Group or the recently initiated W3C Video Activity. These projects and initiatives both worked on solutions for the annotation, automated understanding, description, and retrieval of multimedia content.
The goal of this session is to introduce the recent advancements in the area of multimedia semantics introducing solutions for annotation and methods to support the lifecycle of both metadata and annotations.
The track rounds off the previous ones which focused on textual data. The attendees of this session will learn what kind of solutions are available for the annotation and description of multimedia data, suitable multimedia description standards and about solutions for semi-automatic multimedia content annotation and tagging.
The session will be concluded with a talk about interlinking of music on the Web of Data.
Topics which will be covered include

• Fundamentals about metadata for multimedia content and the introduction to relevant standards and ontologies
• The problem of annotation including the introduction of recent approaches in projects like xSMART and incentives for annotation
• The lifecycle of multimedia metadata and annotation
• Introduction of tools for content creation and annotation (like Kaliph and Emir or KAT), browsing and retrieval and presentation generation
• Emergent semantics in multimedia applications
• Interlinking Music on the Web of Data