W3C home > Mailing lists > Public > www-tag@w3.org > June 2006

Re: CURIEs: A proposal

From: Steven Pemberton <steven.pemberton@cwi.nl>
Date: Tue, 27 Jun 2006 15:34:30 +0200
To: "Dan Connolly" <connolly@w3.org>
Cc: "Misha Wolf" <Misha.Wolf@reuters.com>, public-rdf-in-xhtml-tf@w3.org, semantic-web@w3.org, www-tag@w3.org
Message-ID: <op.tbs3zsd6smjzpq@acer3010.lan>

On Tue, 27 Jun 2006 14:52:39 +0200, Dan Connolly <connolly@w3.org> wrote:

>> >> 	CURIEs are a way of abbreviating URIs.
>> >
>> > Not as far as I can tell. From what I can tell, CURIEs
>> > are a collection of ways of abbreviating URIs. At
>> > least, that's what's proposed in
>> >  http://lists.w3.org/Archives/Public/www-tag/2006Jun/0007.html
>>
>> I think that depends on your point of view, and is not really an  
>> essential distinction.
>
> It's essential to my understanding.
>
> I'd like to get a clear answer to whether one could write one piece
> of software that implements CURIEs. With URI references, it's
> clear that one can. But if CURIEs are a cross-language thing,
> where each language specifies how to parse and handle them, then
> they're not something one could implement concretely, once and for all.

Actually, I think there's a lot of similarities with URI references. You  
can't fully interpret one until you know the media type of the resource  
pointed to by the URI part.

But still, the current CURIES are just

	CURIE ::= prefix ":" suffix
	prefix ::= NCName
	suffix::= URI-reference-tail

and the resultant URI is the concatenation of the URI identified by the  
prefix with the suffix. Misha would like this to be extended.

>>  Are URIs a method of assigning a name to a resource, or a
>> collection of methods? Depends on whether you look at the syntax of a  
>> URI
>> or what happens when the URI is dereferenced.
>
> Right; you can implement the generic URI syntax once and for all,
> but as new URI schemes come online, you have to stay tuned to the IANA
> registry and add more code.

and new media types. Misha's proposal would be similar: you can implement  
a generic CURIE syntax, but need to add knowledge about the context to how  
to expand the URI from it. I am personally not in favour of this.

>> Misha's proposal is at the moment just that: a proposal;
>
> Is there something else I should be looking at?

There is a draft but it is not yet official because it hasn't been  
reviewed by all parties yet (and it's member only, so I won't cite it  
here).

Steven
Received on Tuesday, 27 June 2006 13:34:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:41 GMT