Architecture of theWorld WideWeb - Social Semantics: The Search for Meaning on the Web - page 29

the resource) such that “the resource is the conceptual mapping to an entity or
set of entities, not necessarily the entity which corresponds to that mapping at
any particular instance in time. Thus, a resource can remain constant even when
its content - the entities to which it currently corresponds - changes over time,
provided that the conceptual mapping is not changed in the process” (Berners-Lee
et al. 1998). This obviously begs an important question: If resources are identified
as conceptual mappings in the head of an individual(s), then how does an agent
know, given a URI, what the resource is? Is it our conceptual mapping, or the
conceptual mapping of the owner, or some consensus conceptual mapping? The
latest version of the URI specification deletes the confusing jargon of “conceptual
mappings” and instead re-iterates that URIs can also be things above and beyond
concrete individuals, for “abstract concepts can be resources, such as the operators
and operands of a mathematical equation” (Berners-Lee et al. 2005). After providing
a few telling examples of precisely how wide the notion of a resource is, the URI
specification finally ties the notion of resource directly to the act of identification
given by a URI, for “this specification does not limit the scope of what might be a
resource; rather, the term 'resource' is used in a general sense for whatever might
be identified by a URI” (Berners-Lee et al. 2005). Although this definition seems
at best tautological, the intent should be clear. A resource is any thing capable of
being content , or in other words, an 'identity' in a language. Since a sense is not
bound to particular encoding, in practice within certain protocols that allow access
to information, a resource is typically not a particular encoding of some content but
some content that can be given by many encodings . To rephrase in terms of sense,
the URI identifies content on a level of abstraction, not the encoding of the content .
So, a URI identifies the 'content' of the Eiffel Tower, not just a particular web-
page which is subject to change. However, there is nothing to forbid someone from
identifying a particular encoding of information with its own URI and resource. For
example, one could also have a distinct URI for a web-page about the Eiffel Tower in
English, or a web-page about the Eiffel Tower in English in HTML. In other words,
a resource can be given multiple URIs , each corresponding to a different encoding
or even different levels of abstraction. Furthermore, due to the decentralized nature
of URIs, often different agents create multiple URIs for the same content , which are
then called in Web architecture co-referential URIs .
We illustrate these distinctions in a typical HTTP interaction in Fig. 2.5 ,where
an agent via a web browser wants to access some information about the Eiffel Tower
via its URI. While on a level of abstraction a protocol allows a user-agent to identify
some resource, what the user-agent usually accesses concretely is some realization
of that resource in a particular encoding, such as a web-page in HTML or a picture
in the JPEG language (Pennebaker and Mitchell 1992). In our example, the URI is
resolved using the domain name system to an IP address of a concrete server, which
then transmits to the user-agent some concrete bits that realizes the resource, i.e.
that can be interpreted to the sense identified by the URI. In this example, all the
interactions are local, since the web-page encodes the content of the resource. This
HTTP entity can then be interpreted by a browser as a rendering on the screen of
Ralph's browser. Note this is a simplified example, as some status codes like 307