Architecture of theWorld WideWeb - Social Semantics: The Search for Meaning on the Web - page 14

identification scheme could be universal. In order to get the initiative of the Web off
the ground, Berners-Lee surrendered to the IETF and renamed URIs from Universal
Resource Identifiers (URIs) to Uniform Resource Locators (URLs) (Berners-Lee
2000). The Web begin growing at a prodigious rate once the employer of Berners-
Lee, CERN, released any intellectual property rights they had to the Web and after
Mosaic, the first graphical browser, was released. However, browser vendors started
adding supposed 'new features' that soon led to a 'lock-in' where certain sites could
only be viewed by one particular corporate browser. These 'browser wars' began to
fracture the rapidly growing Web into incompatible information spaces, thus nearly
defeating the proposed universality of the Web (Berners-Lee 2000).
Berners-Lee in particular realized it was in the long-term interest of the Web to
have a new form of standards body that would preserve its universality by allowing
corporations and others to have a more structured contribution than possible with the
IETF. With the informal position of merit Berners-Lee had as the supposed inventor
of the Web (although he freely admits that the invention of the Web was a collective
endeavour), he and others constituted the World Wide Web Consortium (W3C), a
non-profit dedicated to “leading the Web to its full potential by developing protocols
and guidelines that ensure long-term growth for the Web” (Jacobs 1999). In the
W3C, membership was open to any organization, commercial or non-profit. Unlike
the IETF, W3C membership came at a considerable membership fee. The W3C
is organized as a strict representative democracy, with each member organization
sending one member to the Advisory Committee of the W3C, although decisions
technically are always made by the Director, Berners-Lee himself. By opening up
a “vendor neutral” space, companies who previously were interested primarily in
advancing the technology for their own benefit could be brought to the table. The
primary product of the World Wide Web Consortium is a W3C Recommendation,
a standard for the Web that is explicitly voted on and endorsed by the W3C
membership. W3C Recommendations are thought to be similar to IETF RFCs, with
normative force due to the degree of formal verification given via voting by the
W3C Membership and a set number of implementations to prove interoperability.
A number of W3C Recommendations have become very well known technologies,
ranging from the vendor-neutral later versions of HTML (Raggett et al. 1999), which
stopped the fracture of the universal information space, to XML, which has become
a prominent transfer syntax for many types of data (Bray et al. 1998).
This topic will cite W3C Recommendations when appropriate, as these are
one of the main normative documents that define the Web. With IETF RFCs,
these normative standards collectively define the foundations of the Web. It is by
agreement on these standards that the Web functions as a whole. However, the
rough-and-ready process of the IETF and the more bureaucratic process of the W3C
has led to a terminological confusion that must be sorted in order to grasp the nature
of representations on the Web, causing even the most well-meaning of souls to fall
into a conceptual swamp of undefined and fuzzy terms. This is true in spades when
encountering the hotly-contested term 'representation.'