Introduction - Social Semantics: The Search for Meaning on the Web - page 5

nonetheless inherits problems regarding sense and reference from the philosophy of
natural language. So there is no easy way out of the hard question of representation.
Our proposed answer is then that only a theory of representation and semantics
that takes into account the socially grounded use of a multiplicity of representations
is sufficient to provide the meaning of a representation on the Web, from which
the meaning of a peculiar URI can be derived. In essence, we turn the question
on its head; instead of saying that a URI can have its meaning only by virtue of
what representations can be accessed from it, we instead say that the network of
representations and their use provides the meaning of a URI. Thanks must be given
to co-authors for letting me expand upon our earlier findings and re-use our earlier
words. The term 'we' is deployed in order to acknowledge their contributions.
Note that all previously published versions of work in this topic have been edited,
amended, and otherwise expanded.
In order to orient the reader to the Web, we give an extended introduction
to its history and its architecture in Chap. 2, while introducing the philosophical
terminology in concert with examples from the Web. In Chap. 3 we propose that
the Semantic Web, as embodied by the Resource Description Framework (RDF), is
a kind of URI-based knowledge representation language for data integration and
illustrate it by providing the elements of Web architecture in terms of a formal
Semantic Web ontology. The ontology in particular is joint work with Valentina
Presutti, whose latest version is published as The identity of resources on the
Web: An ontology for Web architecture in the journal Applied Ontology Halpin and
Presutti (2011). These works have in earlier forms been published as An Ontology
of Resources: Solving the Identity Crisis (Halpin and Presutti 2009) with Valentina
Presutti and my early essay The Semantic Web: The Origins of Artificial Intelligence
Redux (Halpin 2004).
In Chap. 4 we illustrate the crisis of the Semantic Web: There is no answer to
the aforementioned question of how to assign meaning to a URI. There are at least
two distinct positions to this question on the Semantic Web, each corresponding
to a distinct philosophical theory of semantics. The first response is the logicist
position , which states that the meaning of a URI is determined by whatever model(s)
satisfy the formal semantics of the Semantic Web (Hayes 2004). This answer is
identified with both the formal semantics of the Semantic Web itself and the
traditional Russellian theory of names and its descriptivist descendants (Russell
1905). While this answer may be sufficient for automated inference engines, this
answer is insufficient for humans, as it often crucially under-determines what kind of
things the URI identifies. As the prevailing position in early Semantic Web research,
this position has borne little fruit. Another response is the direct reference position
for the Web, which states that the meaning of a URI is whatever was intended by
the owner. This answer is identified with the intuitive understanding of many of the
original Web architects like Berners-Lee and a special case of Putnam's 'natural
kind' theory of meaning. This position is also a near relative to Kripke's famous
response to Russell (Kripke 1972; Putnam 1975). Further positions that have been
marginal to the debate on the Web, such as that of semiotics, are not explored.