background image
DigiCULT
.
Info
5
H
IKING UP
M
OUNT
E
VEREST
to formalise domains inside cultural insti-
tutions like the Cité de Science et
Industrie (http://www.cite­sciences.fr) in
Paris.They are developing a portal that
accesses information sources on the human
genome project in a way that is under-
standable and meaningful for the user, in this
case the general public. Another project is
the Panopticon Lavoisier, which is aiming
to build a virtual museum, accessible via
the Internet, containing documents and
instruments of the French chemist
Lavoisier.The method we are using for
representation is based on the C-web data
model. Developed in Europe, it is a method
that allows you to create knowledge maps,
by defining and connecting neighbouring
knowledge domains.You could call it a
sort of scientific cartography.'
W
hat motivates a commercial
organisation to become involved
in a cultural project like MESMUSES
when the Semantic Web is still far away?
Meli: `The Semantic Web might be some-
thing of the future, but that does not mean
there is no use for concepts, methods and
tools that enhance the accessibility of data.
In that respect there is not much difference
between corporate publishing and opening
up our cultural heritage. Our cultural
heritage can be described as the memory
of our society, a very complex memory in
fact. Corporations also have a memory of
formal and tacit knowledge.They might
be a bit smaller and a bit less complex, but
to access both types of memory and to
retrieve the information they hold in a
meaningful way you need the same type
of tools.That is why EDW International
thinks it is important to be involved in
the development of the Semantic Web, far
away though it still may be.'
`A
t the moment there is no clear
vision of what the Semantic Web is
and what it could mean. Like the top of
Mount Everest it is shrouded in clouds.The
tools for developing the Semantic Web do
not come close yet to what we need. It is
like hiking up Mount Everest in a pair of
sneakers'. Marco Meli, co-founder and
director of EDW International, an Italian
company involved in corporate publishing
and developing content management
applications, is not so much sceptical about
the Semantic Web as realistic. A touch of
realism is needed, because a lot of people
tend to get carried away by the promises of
the Semantic Web. Meli: `Three years ago
no one had heard about the Semantic Web
or about XML. Now it seems that even
the coffee machine is fitted with XML to
pour you your favourite cup of coffee. But
I doubt very much that you can get to the
top of Mount Everest in a pair of sneakers.'
X
ML, eXtensible Markup Language, is
an important tool for creating the
Semantic Web at least in the vision of Tim
Berners Lee and other protagonists of a
World Wide Web, that lets machines com-
municate with each other to produce
meaningful and relevant data for different
kinds of users. Meli: `In my view XML as
it is used now is no more than a means to
exchange commands and visualise objects.
It is a major breakthrough, because it
allows strong interoperability among diffe-
rent applications from different vendors in
different operating systems.That is still a
long way from `semantics', i.e. meaningful
exchange of information. Some people
consider XML to be a form of syntax, an
agreement on the grammar.You definitely
need syntax but it won't get you anywhere
near a real Semantic Web. On top of that
you have to have an agreement on the
meaning of words and concepts, in short
an ontology.' An ontology can be defined
as `a set of knowledge terms, including
vocabulary, semantic interconnections and
rules of inference and logic for a particular
topic or domain'
1
.
I
n the short term the Semantic Web
will only be a new mechanism for
accessing data on the Web, Meli says.This
is a very useful target worth pursuing in
the short term. `Developing a Semantic
Web that produces meaningful informa-
tion is something for the future, because
we still have no ideas on how to deal with
meaning outside a very constricted
domain. It is as I said, a new field of research
and it will take some time to develop
useful applications.' Large companies such
as IBM and Microsoft have not said
which tools and formats should be used.
For example, they are not investing very
much money and effort in the develop-
ment of RDF, the Resource Description
Framework, supposedly (at least in the W3C
community) the basic language for repre-
senting semantics on the Web. Only Adobe
among the major software publishers
seems to be working on it, says Meli.
M
eli is nevertheless investing in the
Semantic Web: he is actively invol-
ved in MESMUSES (http://cweb.inria.fr/
Projects/Mesmuses/), a European project
that definitely has a semantic flavour to it.
The aim of the project is to develop tools,
such as an active memory manager, to
organise, structure and present scientific
and technical knowledge offered to the
public by science museums. Meli: `We try
I
NTERVIEW WITH
M
ARCO
M
ELI
, EDW I
NTERNATIONAL
.B
Y
J
OOST VAN
K
ASTEREN
1 Hendler, J., 2001. 'Agents and the Semantic Web,' IEEE
Intelligent Systems, vol. 16, no. 21, (March/April 2001), 30-37.