- From: Dan Connolly <connolly@w3.org>
- Date: Thu, 18 May 2000 09:47:39 -0500
- 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