=Paper=
{{Paper
|id=None
|storemode=property
|title=Sharing Data on the Aquileia Heritage: Proposals for a Research Project
|pdfUrl=https://ceur-ws.org/Vol-806/paper3.pdf
|volume=Vol-806
|dblpUrl=https://dblp.org/rec/conf/aquileia/RobertoO11
}}
==Sharing Data on the Aquileia Heritage: Proposals for a Research Project==
https://ceur-ws.org/Vol-806/paper3.pdf
D-1
Sharing Data on the Aquileia Heritage:
Proposals for a Research Project
Vito Roberto and Paolo Omero
Department of Informatics, University of Udine, Italy
vito.roberto@uniud.it, paolo.omero@uniud.it
Abstract. Basic ideas are presented of a multi-national research project to
share data about the Roman city of Aquileia employing the Information and
Communication Technologies (ICT). A Consortium is proposed to manage the
project, adopting the Open-source approach to the software design. The
Consortium proposes a common vision of the data which we detail in the paper.
There results a shared vocabulary of terms and meanings, as well as standard
metadata formats to encode, classify and exchange data from whatever source.
A federated system of computer resources realizes and supports the project. We
also discuss the results to be realistically expected in short time of a low-cost,
joint research effort.
Keywords: Distributed Systems, Open Source, Web Services, Data
Repositories, Ontologies, Metadata.
1 Introduction and motivations
The extraordinary archaeological, historical, artistic heritage of Aquileia has attracted
scholars from several Countries, whose research efforts have contributed to the
studies since at least two centuries. There resulted another, invaluable heritage: the
data originating from surveys, excavations, maps, photos of the city and its large
commercial basin, including the Adriatic areas now pertaining to Italy, Slovenia and
Croatia, as well as districts of Austria, Croatia, Germany, Hungary, Slovenia.
Such data heritage is currently dispersed among numerous Institutions, which are
generally not aware of the data owned by the other ones. In addition, data are
recorded on different supports - manual, mechanic, electronic analog or digital ones,
which raises the problem of making them interoperable among the Institutions.
As a consequence, an invaluable amount of data is not adequately exploited, with
waste of human resources and research funds.
We believe it’s time to consider the opportunities offered by the Information and
Communication Technologies (ICT) to promote the exchange, interoperability and re-
use of data resources, namely: establishing a dedicated infrastructure based on the
D-2 V.Roberto, P.Omero
Internet and the Web; gathering data sets into shared repositories; proposing common
standards of digital encoding, indexing and database archiving.
In the domains of Cultural Heritage, data remain of interest for long time provided
they are accessible to a large number of users. They are validated by the association
and correlation with other data, so that pooling them in a consistent way allows to re-
use and better exploit their potentialities.
The paper presents a few proposals towards an international collaboration project. In
Section 2 we propose some basic technological principles of a federation agreement
among Institutions; Section 3 presents a federated architecture, i.e., how the
agreement can be put into practice by means of a distributed system of computer
resources; Section 4 addresses the issue of how to share a common vision of the data;
Section 5 contains our conclusions.
2 A federation agreement
The interested Institutions should agree on a few basic principles, under an ICT
perspective, inspiring a collaboration within a Consortium. Let us discuss some main
issues.
The Institution is willing to make available a part of its own data to other Partners,
to the extent dictated by its own policies;
The Institution remains the owner of the same data; the responsible of their
integrity, maintenance, updating and security;
The Institution is willing to adopt a unique semantic data model to be shared with
the Partners – i.e., a vocabulary of terms, concepts, taxonomies; a set of metadata
enabling the indexing and exchange over the Internet;
The Institution is willing to adopt Open Source software technologies in
accordance with its own policies, to support possible joint projects;
The Institution is willing to adopt technological standards of data encoding and
exchange over the Net.
The autonomy of each Institution is apparent: to preserve its investments on
hardware/software resources for data acquisition, storage, processing; to benefit from
the access to the partners’ data; to enable the better exploitation of its own data.
It should be emphasized the federated nature of the agreement. The Institutions are
peer partners; no one ‘rules’ the Consortium, nor has the power to prescribe technical
choices, if not agreed with the partners themselves.
It is also apparent that the federation needs a governance, e.g., a central authority in
charge to propose the joint projects; define and maintain the agreed data models;
propose novel technological standards; manage the facilities eventually owned by the
Consortium as a whole. Open source software technologies help in keeping minimal
the cost of the shared resources.
Sharing Data on the Aquileia Heritage… D-3
3 Federated computer resources
The Distributed Architectures (DA) were proposed in the domain of Computer
Systems already in the eighties of the past century, i.e., when the digital networks
were still being explored.
The idea is to assign the computations not to a single computer, but to a number of
connected, co-operating units called nodes. The nodes may be hardware components
(processors); sensors, like video-cameras or remotely-operated instruments; database
subsystems; entire computers; local computer networks,…..which gives room to a
wide range of real-life situations.
The advantage is the opportunity to manage huge amounts of information, not
affordable by a single units: large archives, just like the client accounts of a bank; data
flows, e.g., from a user to her/his personal account; computing speed of operations,….
Other chances are exploiting the decentralization, e.g., suggested by the logistics of an
enterprise; the operational independence, e.g., local teams manage local nodes
without relying on the intervention of externals.
Conflicting issues are the autonomy of the nodes vs. data sharing, and the co-
operation to achieve common goals. Therefore, the distributed architectures range
from the tightly-coupled ones -- lower autonomy of nodes, higher computational
efficiency, as in multi-processor systems -- up to the loosely-coupled systems, with
higher autonomy of the individual nodes and the need to coordinate them. The advent
of the Internet has boosted dramatically such architectures, and distributed computer
systems are now ubiquitous over the Net.
3.1 Federated architectures
Among the loosely-coupled ones, the federated architectures originated the federated
systems [1]. They foster the independence of the component nodes, even with
heterogeneous technological platforms, provided that adhere to an abstract vision of
how to share the information.
The federated framework applies to large enterprises or organizations, possibly on a
large spatial scale. They preserve the highest possible autonomy of each unit, and
achieve high flexibility, i.e., local resources and peculiarities are maintained, and
problems solved locally. Of crucial importance is achieving the interoperability of
data, by which we mean that data originally encoded and stored on a hardware-
software (HW-SW) platform, can be read, interpreted and processed elsewhere on a
different HW-SW platform.
As much as in a federation agreement, a governance is necessary for a federated
technological system, according to a vision of how to share information, as we shall
detail in the sequel.
D-4 V.Roberto, P.Omero
3.1 Semantic data models and exchanges
Sharing data is not merely accessing a site via the download/upload procedures: the
Net basic services would do the job with no need of complex architectures. Instead, a
crucial problem is sharing the meaning (semantics) of the data themselves, which is a
profound topic to be addressed.
First: the data are to be named according to a shared vocabulary of terms and
associated meanings. The Institutions should agree on such a vocabulary: concepts,
physical objects, relations among them,... Fortunately, this task is constrained by the
operational context – i.e., the archaeological research on Aquileia, in our case.
Second point: how do we describe each piece of data? We need to define – and agree
on – a description scheme, i.e., a structured record of metadata: data about acquisition
parameters; their technical features; classification; ownership; administration.
Therefore, two layers of a semantic data model are envisaged:
- A basic, conceptual one fixing the concepts, objects and terms to denote them;
- An upper layer, built on top of the former, containing structured records of data
descriptions, aimed at indexing them for efficient storage and retrieval.
Each federated Institution should adhere to the semantic model, in order to read,
interpret and process the data consistently with the other partners: it’s a fundamental
step towards the interoperability of the partner resources.
Once a semantic basis has been established, a main issue is how to exchange data.
Problems to be faced are the heterogeneous technological platforms of the nodes. On
general grounds, data are exchanged as messages by means of protocols, i.e., standard
formatting rules and procedures to transmit/receive bit packets over the Net.
Adequate technologies are the Web Services, international standards for data
exchange in distributed environments. Services are units on a computer to accomplish
definite tasks -- such as placing an on-line booking of an airline ticket. A Web service
is a scheme for organizing and using such units, that in principle can be controlled by
different technological and ownership domains.
Importing and exporting data from one domain to another is one such service:
therefore, Web services are candidate components of a federated architecture. An
amount of software design is required in order to develop such services.
Alternative solutions exist for exchanging data. In case a node hosts a database,
partial views of the latter can be defined in order to enable an external user to access
a portion of the local data. Another, simple-minded solution is uploading data files
from a local to a centralized repository: it should be operated under the responsibility
of the owner administration, and perhaps scheduled on a regular time basis, in order
to keep mutually aligned the two repositories.
Last - not least! - component of a federated system is the interface providing partners
with a unique access to the shared resources, according to the agreed data model,
which means, access to the available databases and exchanges with the partner data
repositories.
Sharing Data on the Aquileia Heritage… D-5
WEB SERVICE 1 LOCAL
UNIT 1
USER
INTERFACE LOCAL
UNIT 2
Final User
WEB SERVICE N LOCAL
UNIT N
Fig. 1. Abstract scheme of a federated computer architecture. N local units are connected to a
user interface via the Internet. The Units 1 and N exchange data via web services; Unit 2 is
connected simply by the basic Internet services.
4 Proposals for Aquileia
Specific technological solutions are needed for an integrated access to the Aquileia
data heritage. A large amount of information is made of spatial data, i.e., directly
referred to the terrain – georeferenced - by means of international standard coordinate
systems, like the UTM (Universal Transverse Mercator) or the GPS (Global
Positioning Satellite).
An appropriate choice is the GIS (Geographic Information System), i.e., a complex
software system – not merely an interface -- that displays georeferenced data;
processes them and extracts thematic layers; classifies and archives them; allows for
data query/retrieval by means of a search engine as in ordinary database systems.
4.1 Integrating sparse data
Three solutions support the effective data sharing among the partners:
The georeferencing mechanism: data referred to a spatial coordinate system can be
correlated and superimposed by the GIS into a unique working framework. As an
example, the home page of ANTEO, an open-source GIS operating on the Web has
been reported in Figure 3.
The central repository, i.e., a large disk hosting data that Institutions have made
available, that could not export from their own site. For example, historical maps to
be digitized, or data recorded on analogical supports. The central repository
ensures compliance to an agreed digital encoding standard, as well as a standard
D-6 V.Roberto, P.Omero
metadata classification scheme. Consequently, it guarantees a uniform access to
the data by all the partner nodes;
The query engine ensures the query/retrieval of data: either directly, via the central
repository (previous point), or via the other solutions (Section 3.2) to import/export
data from each Institution.
Stated differently, the technological governance of a federated system is grounded on
a GIS and a repository, that provide uniform access, indexing, processing, query and
retrieval of the shared data sources.
WEB SERVICE 1 LOCAL
UNIT 1
GIS
GEOGRAPHIC DATA REPOSITORY LOCAL
UNIT 2
INFORMATION
SYSTEM
Final User
QUERY ENGINE WEB SERVICE N LOCAL
UNIT N
Fig. 2. Proposal of a federated architecture for sharing the Aquileia data heritage, with
reference to Fig.1. A centralized data repository queries and retrieves data from each local unit.
It also directly hosts data after digitization and format conversion, in order to make them
available to all units. The access to the system is through a GIS.
Sharing Data on the Aquileia Heritage… D-7
Fig. 3. The ANTEO Open Geographic Information System [3]. Centre window: an aerial
image of the Aquileia (july 2007), with superimposed the grid of roads of the Roman city.
4.2 Encoding, archiving and beyond…
The heterogeneous, non-standard data on Aquileia are to be analyzed, and their
current status fixed. Problem is selecting international standard technologies for
encoding, indexing, archiving them, in order to ensure the interoperability of the
shared data sets.
Supposing that the digital encoding can be addressed by widespread technologies –
just a working hypothesis for the purposes of this paper – the indexing and archiving
problems deserve a major design effort.
Both issues pertain to what has been called ‘semantic data model ‘ in Section 3.2. In
order to share data, we must agree on a unique vocabulary of terms denoting the data
themselves along with their physical attributes and properties; the relations among
them; their purposes: what in the ICT is called an ontology [4].
Ontologies address the need of a semantic data model. They suggest descriptions of
the data themselves in terms of properties and parameters: the metadata, which
address the issue of classifying and indexing, to archive and retrieve data in a
database – just like the book cards in a public library.
Once more, problem is defining such ‘data card’ formats according to an international
standard scheme. A practical solution is given by the Dublin Core initiative [5].
D-8 V.Roberto, P.Omero
S IM P LE D UB LIN C OR E METAD ATA ELEM ENT SET (DCM ES)
1. Title 9. Format
2. Creator 10. Identifier
3. Subject 11. Source
4. Description 12. Language
5. Publisher 13. Relation
6. Contributor 14. Coverage
7. Date 15. Rights
8. Type
Fig. 4. A basic set of data descriptions according to the Dublin Core international initiative.
The Metadata Element Set is a vocabulary of fifteen properties for use in resource description.
Its elements are broad and generic, usable for describing a wide range of data resources.
We should keep in mind that local databases - with their own data description cards –
are likely to conflict with the common data model. Mapping the local format into the
shared one, and vice-versa, is a non-trivial issue to be addressed at a later time, for
each single unit, after a careful data analysis.
5 Conclusions
We have proposed some basic ideas of a project to share data on Aquileia and
overcome the current dispersions; to enable a better exploitation of the data heritage
and foster international collaborations.
A federation agreement among the interested Institutions should support such
perspectives, and inspire a research Consortium, grounded on the autonomy of each
Institution; the protection of the investments and technological choices of each; the
willingness to share a vision of the data, as well as exploit the opportunities of the
Open software and international standard solutions. A governance of the Consortium
ensures the proposal and maintenance of the agreed software standards.
On this basis, we propose a federation of computer systems, as a kind of distributed,
specialized computer network. A set of nodes, each pertaining to an Institution, are
connected via the Internet and the Web, and share a semantic data model.
Accordingly, well-experimented technologies ensure the exchange of data over the
Net with minimum cost. The technological governance based on Open-source
products ensures the uniform access to the federated resources through a GIS
platform.
An exploratory international project can perform a feasibility study, and a few results
are under reach with low cost in a reasonable time:
Sharing Data on the Aquileia Heritage… D-9
An inventory of the data made available by the Institutions, along with a
preliminary analysis of their features;
Proposals towards a semantic data model: an analysis of concepts and terms
towards a shared ontology;
Proposal of metadata formats;
An Open GIS platform for a uniform access to georeferenced data;
Experimental Web services to explore effective data exchanges among Institutions
through the Internet.
We believe that integrating local databases within more complex schemes – e.g.,
retrieving multiple pieces of data from different archives by joint queries, as in the
Online Public Access bibliographic Catalogs (Meta-OPAC) – is a difficult task, not
under reach until a careful analysis of the data resources has been carried out.
However, we are confident that the ICT solutions we have proposed open new
perspectives towards an international effort of data sharing on Aquileia.
The Institutions will be aware of the data hosted elsewhere. They will possibly access
pieces of data collected by other Institutions with minimum cost. Data will be
correlated within a unique GIS platform, enabling their joint interpretation. Novel
data clusters will be available and bring to the synthesis of new knowledge.
Another perspective opened by a federated system is the fast and consistent data
archiving: excavation, geophysical, aerial survey data can be registered and archived
by mobile devices according to standard formats. In this way, they will be made
available in short time by the owners to the users – researchers, schoolboys and girls,
tourists, citizens of any Country - upon authorization.
References
1. Heimbigner, D.M., McLeod, D.: A Federated Architecture for Information Management, ACM
Trans. On Information Systems (TOIS), Vol.3, 3, The ACM Press, New York, NY, USA (1985)
2. Geographic Information Systems (GIS), http://library.stanford.edu/depts/gis/index.html
3. Buora, M., Roberto, V.: New Work on the Plan of Aquileia based on Aerial Photographs and a
GIS Platform, Journal of Roman Archaeology, Vol.23, pp.320--334, USA (2010)
4. Gruber,T.: A Translation Approach to Portable Ontology Specification,
http://tomgruber.org/writing/ontolingua-kaj-1993.pdf (1993)
5. The Dublin Core Metadata Initiative, http://dublincore.org/documents/dces/