W3C home > Mailing lists > Public > xml-uri@w3.org > May 2000

Re: the case of two bats

From: Dan Connolly <connolly@w3.org>
Date: Thu, 18 May 2000 09:47:39 -0500
Message-ID: <3924028B.F36E2181@w3.org>
To: "Simon St.Laurent" <simonstl@simonstl.com>
CC: xml-uri@w3.org
"Simon St.Laurent" wrote:
> 
> At 10:33 AM 5/16/00 -0400, Dan Connolly wrote:
> >I'd like to discuss this issue in more black-and-white terms;
> >here's an example that has clarified the situation for at
> >least a few folks:
> 
> I hate to be so blunt, but I think I'd fire anyone who produced a system
> where the actual contents of documents would change if files were moved
> from place to place.

Er.. you'd fire anybody who used relative URIs to link between
two documents in the same directory? Is that really what
you mean to say? I'm not sure what you mean by "actual contents".

I too would (like to) fire anybody who put an XHTML document at
	http://example.com/someplace/mydoc.html
that started with

	<!DOCTYPE html 
                       PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
                       "DTD/xhtml1-strict.dtd">

in stead of

	<!DOCTYPE html 
                       PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
         
"http://www.w3.org/TR/2000/REC-xhtml1-20000126/DTD/xhtml1-strict.dtd">

Actually, in the case of DTDs, if you make a copy of
http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd at
http://example.com/someplace/xhtml1-strict.dtd you can
get away with this, since the identifier of a DTD
doesn't have any impact on its semantics, and
W3C promises that
http://www.w3.org/TR/2000/REC-xhtml1-20000126/DTD/xhtml1-strict.dtd
doesn't have state that changes over time.

But links/annotations that apply to
http://www.w3.org/TR/2000/REC-xhtml1-20000126/DTD/xhtml1-strict.dtd
might not be known to apply to
http://example.com/someplace/xhtml1-strict.dtd
(or they might not apply, or ...).


Anyway... if somebody uses relative URI references when they should
have used an absolute URI reference, then yes, they should be fired.

But does that motivate a special exception to make namespace
identifier syntax different from identifiers for all
other Web resources?

i.e. do we find all the 'cause for termination' misuses of Web
technology and make exceptions in the specs to rule them out?
What happened to the principle of minimal constraint?

>  It might be convenient for the document creators, but
> I don't think I could possibly justify that in a production environment,
> even ones that weren't necessarily 'mission critical'.
> 
> >Web Architecture, as I understand it, says that no, the two
> ><bat/> elements are associated with different points
> >in the Web, and that the stylesheet should produce
> 
> What exactly is this 'Web Architecture' that keeps getting thrown around?

In this case, as I said, it's my own personal understanding of the Web.
It's based on my experience in the development of the URI, HTML, HTTP
specs and libwww, and all the various implementations I've looked at.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Thursday, 18 May 2000 10:47:55 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:32:42 UTC