Architecture of theWorld WideWeb - Social Semantics: The Search for Meaning on the Web - page 42

models of past and proposed systems is something which can be mapped onto a
concept of 'object' and some kind of name, address, or identifier for that object.
One can therefore define a set of name spaces in which these objects can be said to
exist. In order to abstract the idea of a generic object, the web needs the concepts
of the universal set of objects, and of the universal set of names or addresses of
objects” (1994a). The more informal notes of Berners-Lee are even more startling
in their claims for universality, stating that the first 'axiom' of Web architecture
is “universality” where “by 'universal' I mean that the Web is declared to be able
to contain in principle every bit of information accessible by networks” (1996b).
Although it appears he may be constraining himself to only talk about digital
'objects' that are accessible over the Internet in this early IETF RFCs, in later IETF
RFCs the principle quickly ran amok, as users of the Web wanted to use URIs to
refer to “human beings, corporations, and bound topics in a library” (Berners-Lee
et al. 1998).
There seems to be a certain way that web-pages are 'on the Web' in a way that
human beings, corporations, unicorns, and the Eiffel Tower are not. Accessing a
web-page in a browser means to receive some bits, while one cannot easily imagine
what accessing the Eiffel Tower itself or the concept of a unicorn in a browser
even means. This property of being 'on the Web' is a common-sense distinction
that separates things like a web-page about the Eiffel Tower from things like the
Eiffel Tower itself. This distinction is that between the use of URIs to access and
reference , between the local and the distal. The early notes of Berners-Lee that pre-
date the notion of URIs itself address this distinction between access and reference,
phrasing it as a distinction between locations and names. As Berners-Lee states,
“conventionally, a 'name' has tended to mean a logical way of referring to an
object in some abstract name space, while the term 'address' has been used for
something which specifies the physical location” (1991). So, a location is aterm
that can be used to access the thing , while a name is a term that can be used to
refer to a thing . Unlike access, reference is the use of an identifier for a thing to
which one is immediately causally disconnected . Access is the use of an identifier
to create immediately a causal connection to the thing identified (Hayes and Halpin
2008). The difference between the use of a URI to access a hypertext web-page
or other sort of information-based resource and the use of a URI to refer to some
non-Web accessible entity or concept ends up being quite important, as this ability
to representationally use URIs as 'stands-in' for referents forms the basis of the
distinction between the hypertext Web and the Semantic Web.
Names can serve as identifiers and even representations for distal things.
However, Berners-Lee immediately puts forward the hypothesis that “with wide-
area distributed systems, this distinction blurs” so that “things which at first look
like physical addresses. . . cease to give the actual location of the object. At the same
time,alogicalname...mustcontainsomeinformationwhichallowsthenameserver
to know where to start looking” (1991). He posits a third neutral term, “identifier”
that was “generally referred to a name which was guaranteed to be unique but had
little significance as regards the logical name or physical address” (Berners-Lee
1991). In other words, an identifier is a term that can be used to either access or