Architecture of theWorld WideWeb - Social Semantics: The Search for Meaning on the Web - page 23

information-bearing message is whatever is held in common between the source
and the receiver as a result of the conveyance of a particular message . While this is
similar to our definition of information itself, it is different. The content is whatever
is shared in common as a result of a particular message, such as the conveyance
of the sentence 'The Eiffel Tower is 300 m high.' The content of a message is
called the “facts” by Dretske, ( F ). This content is conveyed from the source ( s )
successfully to the receiver ( r ) when the content can be used by the receiver with
certainty, and that before the receipt of the message the receiver was not certain of
that particular content. Daniel can only successfully convey the content that 'Ralph
won a trip to Paris' if before receiving the message Amy does not know 'Ralph
won a trip to Paris' and after receiving the message Amy does know that fact.
Dretkse himself notes that information “does not mean that a signal must tell us
everything about a source to tell us something,” it just has to tell enough so that
the receiver is now certain about the content within the domain (1981). Millikan
rightfully notes that Dretske states his definition too strongly, for this probability
of 1 is just an approximation of a statistically “good bet” indexed to some domain
where the information was learned to be recognized (2004). For example, lightening
carries the content that “a thunderstorm is nearby” in rainy climes but in an arid
prairie lightning can convey a dust-storm. However, often the reverse is true, as
the same content is carried by messages in different encodings, like a web-page
about the Eiffel Tower being encoded in either English or French. These notions
of encoding and content are not strictly separable, which is why they together
compose the notion of information. An updated famous maxim of Hegel could
be applied: for information, there is no encoding without content, and no content
without encoding (1959).
The relationship of an encoding to its content ,isan interpretation . The inter-
pretation 'fills' in the necessary background left out of the encoding, and maps
the encoding to some content. In our previous example using binary digits as
an encoding scheme, a mapping could be made between the encoding 001 to
the content of the Eiffel Tower while the encoding 010 could be mapped to the
content of the Washington Monument. When the word 'interpretation' is used as
a noun, we mean the content given by a particular relationship between an agent
and an encoding, i.e. the interpretation. Usual definitions of “interpretation” tend to
conflate these issues. In formal semantics, the word “interpretation” often can be
used either in the sense of “an interpretation structure, which is a 'possible world'
considered as something independent of any particular vocabulary” (and so any
agent) or “an interpretation mapping from a vocabulary into the structure” or as
shorthand for both (Hayes 2004). The difference in use of the term seems somewhat
divided by fields. For example, computational linguistics often use “interpretation”
to mean what Hayes called the “interpretation structure.” In contrast, we use the term
'interpretation' to mean what Hayes called the “interpretation mapping,” reserving
the word 'content' for the “interpretation structure” or structures selected by a
particular agent in relationship to some encoding. Also, this quick aside into matters
of interpretation does not explicitly take on a formal definition of interpretation
as done in model theory, although our general definition has been designed to be
compatible with model-theoretic and other formal approaches to interpretation.