W3C home > Mailing lists > Public > www-rdf-interest@w3.org > February 2004

RE: pound sign vs. slash as final URI delimiter

From: Hammond, Tony (ELSLON) <T.Hammond@elsevier.com>
Date: Tue, 17 Feb 2004 12:23:05 -0000
Message-ID: <54A600C436EA694581B93E4BD4D4788A0A60D6AF@elslonexc004.eslo.co.uk>
To: 'Ron Daniel' <rdaniel@taxonomystrategies.com>, "DuCharme, Bob (LNG-CHO)" <bob.ducharme@lexisnexis.com>
Cc: www-rdf-interest@w3.org

Hi Ron:

I do have to query this statement

> 1) Fragment IDs imply downloading the source document, then 

Note that in the INFO URI scheme that has been proposed for identifying
information assets in public namesspaces we define a fragment component as a
regular part of the syntax,

	info-URI = info-scheme ":" info-identifier [ "#" fragment ] 

while at the same time asserting that INFO URIs are non-dereferenceable (and
hence there are no source documents to download). In discussing the use of
fragment components we use language which is closely aligned with RFC
2396bis to assert that:

	The (unescaped) values for the "fragment" component identify
secondary 
      information assets with respect to the primary information asset 
      which is referenced by the "info-identifier".

Tony

ps/
I can hear Patrick groaning in the background but it's anyway at least an
interesting architectural point (the use of fragment identifiers) worthy of
some discussion. ;) 


> -----Original Message-----
> From: www-rdf-interest-request@w3.org
> [mailto:www-rdf-interest-request@w3.org]On Behalf Of Ron Daniel
> Sent: 16 February 2004 16:08
> To: 'DuCharme, Bob (LNG-CHO)'
> Cc: www-rdf-interest@w3.org
> Subject: RE: pound sign vs. slash as final URI delimiter
> 
> 
> 
> Hi Bob,
> 
> My current rule of thumb is to use '/' unless there is some good
> reason not to. But this is not a strongly held belief.
> 
> Why do I prefer '/' over '#'?
> 1) Fragment IDs imply downloading the source document, then 
> picking through
>    it for the bit you need. For large vocabularies, like many 
> produced by
>    Government agencies, this would be a performance issue.  
> (Of course,
>    whether something is actually downloaded just because we 
> have used its
>    URL as a namespace ID is another issue.)
> 2) There are some people who are vociferous in maintaining 
> that there is a
>    very big difference between a resource and a fragment ID, 
> and that RDF is
>    about describing resources. I am not personally sure of 
> this, but don't
> see
>    much harm in using '/'.
> 
> Why I hesitate to categorically state that '/' should be used 
> instead of
> '#'?
> 1) Because # should fit a lot better with picking a predicate 
> out of an
>    XML document that specifies the namespace.
> 
> I'd appreciate it if people could clarify things.
> 
> Ron
> 
> > -----Original Message-----
> > From: www-rdf-interest-request@w3.org 
> > [mailto:www-rdf-interest-request@w3.org] On Behalf Of 
> > DuCharme, Bob (LNG-CHO)
> > Sent: Monday, February 16, 2004 7:15 AM
> > To: www-rdf-interest@w3.org
> > Subject: pound sign vs. slash as final URI delimiter
> > 
> > 
> > This feels like a beginner question, but after a few searches 
> > I can't find
> > any discussion of the issue. Let's say I have a namespace 
> > identified by the
> > URI http://www.example.com/pathname. To identify the name foo 
> > from that
> > namespace, what are the pros and cons of identifying it 
> with a URI of
> > http://www.example.com/pathname/foo as opposed to
> > http://www.example.com/pathname#foo? The pound sign seems to 
> > more clearly
> > indicate "the following is a name from the namespace named 
> up to this
> > point," but I see that most references to Dublin Core names (e.g.
> > http://purl.org/dc/elements/1.1/creator) use the slash. 
> > 
> > Perhaps the question is better framed without reference to 
> > syntax: is it
> > better for a name from a namespace to have it's own complete 
> > URI or for it
> > to be referenced using a fragment identifier appended to the 
> > URI for its
> > namespace?
> > 
> > thanks,
> > 
> > Bob
> > 
> > 
> 
Received on Tuesday, 17 February 2004 07:23:10 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:14:58 UTC