W3C home > Mailing lists > Public > www-tag@w3.org > September 2005

Re: [Fwd: simple case of IRIs for Components in WSDL 2.0] (abstractComponentRefs-37 )

From: Dan Connolly <connolly@w3.org>
Date: Tue, 13 Sep 2005 01:36:40 -0500
To: noah_mendelsohn@us.ibm.com
Cc: www-tag@w3.org
Message-Id: <1126593400.4430.639.camel@dirk>

On Tue, 2005-09-13 at 02:05 -0400, noah_mendelsohn@us.ibm.com wrote:
> Dan Connolly wrote:
> 
> > Then we should be able to use  
> http://www.w3.org/2005/08/sparql-protocol-query#SparqlQuery
> 
> At the risk of putting salt into a still-open wound:  I think the TAG as a 
> whole is still, as of Basel, officially unresolved on whether there's a 
> preference for / or # in naming sub-resources.   Yes, our progress on 
> httpRange14 may take some pressure off, but I don't think we ever went 
> backi to / vs. # in particular, did we? Pending such resolution, is there 
> a reason that the above is to be recommended over 
> http://www.w3.org/2005/08/sparql-protocol-query/SparqlQuery ?

For one thing, because it's easier: you just stick one file in the
right place, at 2005/08/sparql-protocol-query , and it grounds
all the terms in the namespace.

In order to use the / version, I need to mess with .htaccess
and set up a redirect for each of the names in the namespace
so that they're not 404.

>    I'm not 
> arguing that one or the other is better, just that if we left the choice 
> pending for reasources in general, I'm curious why we would take an 
> earlier stand on namespaces in particular.  Both of the above look 
> plausible to me. 
> 
> For that matter, is it obvious that in all cases 
> http://www.w3.org/2005/08/sparql-protocol-query?name='SparqlQuery'
> is a mistake?  I certainly wouldn't push for this as good practice in most 
> cases, but I can't see why it's provably a mistake either.   What am I 
> missing? Thanks.

I don't think you're missing anything, but you're reading more
than what I wrote. I didn't say that was a mistake in all cases.
I'm just arguing that doc#name is a straightforward pattern that I
think is worth explicit support.

I've heard TimBL argue that this pattern is more than just convenient,
and it appeals to my intuitions, but the argument doesn't stick in
my head very long.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
D3C2 887B 0F92 6005 C541  0875 0F91 96DE 6E52 C29E
Received on Tuesday, 13 September 2005 06:36:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:36 GMT