W3C home > Mailing lists > Public > www-tag@w3.org > March 2002

Re: Summary: Section 2: What does a URI identify?

From: Mark Baker <distobj@acm.org>
Date: Sun, 17 Mar 2002 10:16:46 -0500 (EST)
Message-Id: <200203171516.KAA17007@markbaker.ca>
To: Norman.Walsh@Sun.COM (Norman Walsh)
Cc: www-tag@w3.org
Very nice.  My comments;

- re the editorial note about b-nodes.  The issue is not whether
something *has* an identifier, but whether it *could*.  Same goes for
last paragraph in sec 2.2.  I believe b-nodes could, but I'm no RDF

- I'd suggest avoiding the use of the term "home page", as I believe it
to be an artificial construction by those who don't recognize that a
URI can identify a real life thing.  e.g. "Mark's home page is
http://www.markbaker.ca" is equivalent to "http://www.markbaker.ca
identifies Mark".  In order to know what "http://www.w3.org" identifies,
you have to ask the W3C.  I assume that it identifies the organization.

- "Dereferencing a resource identifier yields a representation of the
current value of the referenced resource".  This is tricky, because
"current" can mean many things.  I'd suggest removing "current value of

- "or indeed after it has been retired".  I'm not so sure if this is the
case or not (though I would guess that it is not), but I don't believe
it relates to the statement about how the identity of a thing can be
assigned at the time of conceptualization, rather than the time of

- "a URI consists of [...] and an optional fragment identifier".  Oops!
8-)  A URI does not include a fragment identifier!  Ditto for 3.1;
"If a URI contains a[sic] sharp character [...] is a fragment

- sec 3.1 appears to be a bit schizophrenic. Fragment ids are for
identifying part of a resource representation, not a resource.  Also,
I don't consider them a "design flaw" - they do what they are intended
to do perfectly, people just use them in odd ways (RDF).  Plus you may
want to pick a different example than the HTML one, because of the name
vs. id issue.

- "This does not mean that the stream of bits associated with a URI
(if, in fact, there is one) can never change".  "the stream of bits"
appears to suggest a single stream, when many are possible.  I don't
believe "associated" captures the resource/representation distinction.
And re "if, in fact, there is one" - I believe it a property of all
resources, that there will always exist at least one representation
of each.

- "it is perfectly reasonable for a resource to be identified by several
different URIs".  True, but less is best.  How about "it is acceptable
for", or something less encouraging than "perfectly reasonable"

Mark Baker, Chief Science Officer, Planetfred, Inc.
Ottawa, Ontario, CANADA.      mbaker@planetfred.com
http://www.markbaker.ca   http://www.planetfred.com
Received on Sunday, 17 March 2002 10:12:06 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:30 UTC