W3C home > Mailing lists > Public > www-tag@w3.org > September 2003

Re: Arch Doc: 26 September 2003 Editor's Draft

From: Roy T. Fielding <fielding@apache.org>
Date: Mon, 29 Sep 2003 15:59:17 -0700
Cc: www-tag@w3.org
To: "Ian B. Jacobs" <ij@w3.org>
Message-Id: <8DA514BA-F2D0-11D7-A778-000393753936@apache.org>

> See abstract proposed in 26 Sep draft.

Argh, sorry I didn't see the difference before the meeting.  The text
that was added to the abstract is over-constraining.

What I had was:

    The World Wide Web is a networked information space consisting
    of resources that are interconnected via hypertext links and
    descriptive metadata.

and that became

    The World Wide Web is an information space. This space consists of
    information objects collectively called "resources." Uniform Resource
    Identifiers (URIs) are used to identify these resources; URIs have
    global scope within this space. The information space is a network of
    Web resources that are interconnected via URIs and descriptive
    metadata.

which changes the meaning substantially.  The space consists of more
than just resources, and defining resources as information objects
is simply not appropriate.  URIs do not always have global scope in
the sense that you describe here, since URIs do not point to objects
but rather to resources. Finally, both descriptive metadata and links
use URIs to do their interconnecting.

Likewise, the additions of

    These systems include the "traditional"
    hyperlink Web (where users interact with resources via links)
    as well as emerging Semantic Web and Web Services technologies.

and

    They do so via representations of resource state, which are
    constructed from data formats such as HTML, and descriptive
    metadata.

again violate the split between information space and information
retrieval.  There are a lot more than just three information systems,
for example.  Likewise, please ignore all comments to the effect
that "hyperlink" is somehow better than "hypertext link", and there
is no need to define it here.  People who read this document are
expected to be familiar with the basic concepts of the Web, and
the details can be addressed in a glossary.

Please revert these changes -- they undo what I was trying to
accomplish by separating the web of resources from the web of
information retrieval actions.  Here is what I would write, taking
into account the editorial changes suggested:

    The World Wide Web is a networked information space consisting
    of resources that are interconnected via hypertext links and
    descriptive metadata.  This information space is the basis of,
    and is shared by, a number of information systems.  Within each
    of these systems, agents (e.g., browsers, servers, spiders, and
    proxies) provide, retrieve, create, analyze, and reason about
    resources.  Web architecture encompasses both protocols that define
    the information space, by way of identification and representation,
    and protocols that define the interaction of agents within an
    information system making use of the space. Web architecture
    is influenced by social requirements and software engineering
    principles, leading to design choices that constrain the behavior
    of systems using the Web in order to achieve desired properties:
    to be an efficient, scalable, shared information space that can
    continue to grow indefinitely across languages, cultures, and
    information media. This document explores the three dimensions
    of Web architecture: identification, interaction, and representation.

and that would address Dan's comment about length as well.

....Roy
Received on Monday, 29 September 2003 18:59:34 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:21 GMT