Talia: a Research and Publishing Environment for Philosophy Scholars

Introduction
Talia is a distributed semantic digital library and publishing
system, which is specifi cally designed for the needs of the
scholarly research in the humanities. The system is designed
for the diverging needs of heterogeneous communities. It is
strictly based on Semantic Web technology1 and computational
ontologies for the organisation of knowledge and will help
the defi nition of a state-ofthe-art research and publishing
environment for the humanities in general and for philosophy
in particular.
Talia is developed within the Discovery project, which aims
at easing the work of philosophy scholars and researchers,
making a federation of distributed digital archives (nodes)
each of them dedicated to a specifi c philosopher: Pre-Socratic,
Diogenes Laertius, Leibniz, Nietzsche, Wittgenstein, Spinoza,
and others.
Use Case and Requirements
One of the use cases that Talia should satisfy is called
The Manuscripts Comparer.The archive contains digital
reproductions of handwritten manuscripts by one philosopher.
The scholar wants to browse them like books, navigating by
chapters and pages. The archive also contains transcriptions of
the different paragraphs, on which to click to see the correct
transcription.
Moreover, the scholar is interested in comparing different
versions of the same thought, placed by the philosopher in
different paragraphs of his works, so the scholar needs to fi nd
all those different versions. Finally, he/she can write his own
remarks on those paragraphs, so that they are published in the
archive, after a peer review process.
From this use case, we can devise some technical requirements
that drive the design of Talia:
1. Making metadata remotely available. The metadata content
of the archives shall be made available through interfaces
similar to those of URIQA2, which will allow automated
agents and clients to connect to a Talia node and ask
for metadata about a resource of interest to retrieve its
description and related metadata (e.g., the author, the
resource type, etc.). This type of services enable numerous
types of digital contents reuse.
2. Querying the system using standard query-languages. It shall
possible to directly query the distributed archives using the
SPARQL Semantic Web Query Language. SPARQL provides
a powerful and standard way to query RDF repositories but
also enables merging of remote datasets.
3. Transformation of encoded digital contents into RDF. The
semantic software layer shall provide tools to transform
textually encoded material, for example the transcription of
manuscripts in TEI3 format, into RDF.
4. Managing structural metadata. “Structural” metadata
describes the structure of a document (e.g., its format, its
publisher, etc.). It shall be necessary to provide tools to
manage these kinds of metadata.
5. Import and exporting RDF data. It shall be important to
include facilities in Talia to import and export RDF data
with standard and well know formats like RDF/XML and
N-Triples.
Talia should also provide facilities to enrich the content of the
archives with metadata, whose types and formats depend on
the kind of source and on the user community which works on
the data. For example, in some archives, different versions of
the same paragraph may exist in different documents. In order
to allow the users to follow the evolution of that particular
philosophical thought, the relations among these different
versions must be captured by the system.
An Overview of the Talia System
Talia is a distributed semantic digital library system which
combines the features of digital archives management with an
on-line peer-review system.
The Talia platform stores digital objects, called sources, which
are identifi ed by their unique and stable URI. Each source
represents either a work of a philosopher or a fragment of it
(e.g., a paragraph), and can have one or more data fi les (e.g.,
images or textual documents) connected to it.The system
also stores information about the sources, which can never be
removed once published and are maintained in a fi xed state.
Combined with other long-term preservation techniques, Talia
allows the scholars to reference their works and gives the
research community immediate access to new content. The most innovative aspect of Talia is that for the fi rst time,
at least in the fi eld of humanities, all the interfaces exposed
publicly will be based on proven Semantic Web standards
enabling data interoperability within the Talia federation, and
eases the data interchange with other systems. Two additional
features of Talia are the highly customisable Graphic User
Interface (GUI) and the dynamic contextualisation.
Figure 1: An Example of the Talia User Interface.
The web interface framework, based on modular elements
called widgets, allows to build GUIs according to requirements
coming from heterogeneous communities. Widgets can be
packaged independently and used as building blocks for
the application’s user interface. In particular, Talia provides
semantic widgets that interact with an RDF Knowledge Base.
To customise the site’s appearance, it also would be possible
to add custom HTML rendering templates. Figure 1 shows a
screenshot of a Talia’s GUI.
The dynamic contextualisation provides a means for data
exchange among different and distributed digital libraries
based on the Talia framework. The dynamic contextualisation
allows a Talia library (node) to share parts of its RDF graph in
a peer-to-peer network. By using this mechanism, a node may
notify another node that a semantic link between them exists,
and the other node may use this information to update its
own RDF graph and create a bidirectional connection.
Talia uses computational ontologies and Semantic Web
technology to help the creation of a state-of-the-art research
and publishing environment. Talia will provide an innovative
and adaptable system to enable and ease data interoperability
and new paradigms for information enrichment, dataretrieval,
and navigation.
Computational Ontologies
Ontologies have become popular in computer science as a
means for the organisation of information. This connotation of
ontology differs from the traditional use and meaning it has in
philosophy, where ontologies are considered as A system of
categories accounting for a certain vision of the world [2]
In computer science, the concept of (computational) ontology
evolved from the one fi rst provided by Gruber [3], who
defi ned an ontology as a specifi cation of a conceptualisation4
to a more precise one, extracted from Guarino’s defi nitions
[4]: A computational ontology is A formal, partial specifi cation
of a shared conceptualisation of a world (domain).
Intuitively, a computational ontology is a set of assumptions
that defi ne the structure of a given domain of interest (e.g.,
philosophy), allowing different people to use the same concepts
to describe that domain. Talia will use computational ontologies
to organise information about writings of a philosopher or
documents (manuscripts, essays, theses, and so on) of authors
concerning that philosopher.
Related Work
Talia is directly related to the Hyper Platform which was used
for the HyperNietzsche archive [1], a specialised solution,
designed for specifi c needs of the Nietzsche communities.
HyperNietzsche has a fi xed graphical user interface, it does
not use Semantic Web technology and it is not adaptable for
different communities with heterogeneous needs.
Talia shares some properties with other semantic digital
library systems like JeromeDL [5], BRICKS [6], and Fedora
[7]. However, these projects are mostly focused on the backend
technology and none of them offers a fl exible and highly
customisable research and publishing system like Talia.
Conclusion and Future Work
Talia is a novel semantic digital web library system, which aims
to improve scholarly research in the humanities, in particular
in the fi eld of philosophy. Thanks to the use of Semantic Web
technology, Talia represents a very adaptable state-of-the art
research and publishing system.
The ontologies used in Talia are currently being developed by
Discovery’s content partners, who are in charge of organising
their contributions. These ontologies will be part of the fi nal
core Talia application. At the moment, only a fi rst public demo
version is available5.
Although Talia is currently intended only for philosophy scholars,
it should be straightforward to adopt it for humanities, with
the help of suitably developed ontologies. Using dynamic Ruby
as programming language and the RubyOnRails framework,
Talia provides an ideal framework for the rapid development
of customised semantic digital libraries for the humanities.
Acknowledgements
This work has been supported by Discovery, an ECP 2005 CULT
038206 project under the EC eContentplus programme.
We thank Daniel Hahn and Michele Barbera for the fruitful
contributions to this paper. Notes
1 http://www.w3.org/2001/sw/
2 http://sw.nokia.com/uriqa/URIQA.html
3 http://www.tei-c.org/
4 A more up-to-date defi nition and additional readings can be found
at http://tomgruber.org/writing/ontologydefi nition-2007.htm
5 http://demo.talia.discovery-project.eu
References
[1] D’Iorio, P.: Nietzsche on new paths: The Hypernietzsche
project and open scholarship on the web. In Fornari, C.,
Franzese, S., eds.: Friedrich Nietzsche. Edizioni e interpretazioni.
Edizioni ETS, Pisa (2007)
[2] Calvanese, D., Guarino, N.: Ontologies and Description
Logics. Intelligenza Artifi ciale - The Journal of the Italian
Association for Artifi cial Intelligence, 3(1/2) (2006)
[3] Gruber, T.R.: A translation approach to portable ontology
specifi cations. Knowledge Acquisition 5(2) (1993) 199-220
[4] Guarino, N.: Formal Ontology and Information Systems.
In Guarino, N., ed.: 1st International Conference on Formal
Ontologies in Information Systems, IOS Press (1998) 3-15
[5] Kurk S., Woroniecki T., Gzella A., Dabrowski M. McDaniel
B.: Anatomy of a social semantic library. In: European Semantic
Web Conference. Volume Semantic Digital Library Tutorial. (2007).
[6] Risse T., Knezevic P., Meghini C., Hecht R., Basile F.:
The bricks infrastructure an overview. In: The International
Conference EVA, Moscow, 2005.
[7] Fedora Development Team: Fedora Open Source
Repository software. White Paper.