W3C home > Mailing lists > Public > www-tag@w3.org > January 2004

Web sites as resources

From: Tim Bray <tbray@textuality.com>
Date: Wed, 14 Jan 2004 22:33:16 -0800
Message-Id: <B369BB26-4724-11D8-ACAF-000A95A51C9E@textuality.com>
To: www-tag@w3.org <www-tag@w3.org>

I was thinking over Monday's discussion and the more I think about it, 
the more I find myself in agreement with Roy.

There is no doubt whatsoever that Web Sites are real things that exist. 
  Outside in the non-geek world, people talk about Web Sites more than 
Web Pages.  I operate a Web Site called 'ongoing'.  There is no 
confusion in my mind between the notion of the "site", which comprises 
several hundred resources (and in my mind, the linkages between them 
and the software that supports them), and the resource which is the 
site's Home Page, at http://www.tbray.org/ongoing/.

The difference is that whereas I have a URI for the Home Page, I don't 
have one for the site.  It would be nice for Sites, which are a common 
mental construct related to the Web, to be part of the Web.  Suppose I 
assert that there is a resource which is the "ongoing" Web Site and its 
URI is http://www.tbray.org/ongoing/Site, and then suppose that I start 
providing representations of that resource, and further that I start 
slinging around RDF assertions about the resource.  Then that resource 
for all practical purposes is my Site.

It may be an over-simplification to say that a Site is just a 
collection of Web Resources, i.e. the pages in the site; most mental 
models of a Web site, like mine, contain more than just an enumeration 
of URIs.  On the other hand, there clearly is a real-world "membership" 
relationship between resources and sites.

Web sites already exist and are important and are part of everyone's 
mental model of the Web information space.  Per the principles in 
Webarch, they should damn well be resources and have URIs. This is not 
just for reasons of aesthetic satisfaction, but because there are lots 
of per-site constructs and many different per-site autodiscovery 
requirements.  Yes, largely due to patterns of speech, there is 
potential for confusion between the notion of a Site and a Home Page, 
but this is at worst a minor difficulty.

All that's really left to argue about is

1. How to express the membership relation between resources and sites
2. What agents can expect to find in representations of sites
3. Whether the work required to agree on #1 and #2 is cost-effective.


Cheers, Tim Bray  http://www.tbray.org/ongoing/
Received on Thursday, 15 January 2004 01:33:13 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 22:56:03 UTC