W3C home > Mailing lists > Public > public-rdf-in-xhtml-tf@w3.org > July 2008

RE: RDFa test suite addition

From: Hausenblas, Michael <michael.hausenblas@joanneum.at>
Date: Thu, 31 Jul 2008 09:14:08 +0200
Message-ID: <768DACDC356ED04EA1F1130F97D2985201823F32@RZJC2EX.jr1.local>
To: "Mark Birbeck" <mark.birbeck@webbackplane.com>
Cc: "Manu Sporny" <msporny@digitalbazaar.com>, "RDFa mailing list" <public-rdf-in-xhtml-tf@w3.org>


Mark,

>I'm afraid I don't follow you. I'm not sure what you mean when you say
>that you agree with me BUT want to draw attention to the question of
>opacity, since I don't know which part of what I am saying detracts
>from opacity.
>
>Having said that, I'd also say that I don't see opacity of URIs as
>relevant to this discussion. :)

My fault. I should have been more explicit. Let me give an example that
demonstrates the issue.

As far as I understand it there is a basic assumption around that there
is a 1:1 mapping between a file system path and an URI structure. Though
this might be the case, this need not to be true. IMHO only under this
assumption the 'relative URI discussion' (as we have) makes sense.

Just to be crystal clear about what I talk:

http://example.com:8080/company/staff/person?id=mark#me

'http' ... scheme
'example.com:8080' ... authority
'company/staff/person' ... path
'id=mark' ... query
'me' ...  fragment

So, I'm talking about the URI path 'company/staff/person' 

Now, it is perfectly possible that there is somewhere a directory
structure on a file system where this resource is directly served from.
But what is about when mod_rewrite is used to generate cool URIs? That
is in fact we might have

http://example.com:8080/service.php?main=company&section=staff&person?id
=mark#me

and having a rewrite rule that allows to access that resource at 

http://example.com:8080/company/staff/person?id=mark#me

Thoughts?

Cheers,
	Michael


----------------------------------------------------------
 Michael Hausenblas, MSc.
 Institute of Information Systems & Information Management
 JOANNEUM RESEARCH Forschungsgesellschaft mbH
  
 http://www.joanneum.at/iis/
----------------------------------------------------------
 

>-----Original Message-----
>From: Mark Birbeck [mailto:mark.birbeck@webbackplane.com] 
>Sent: Wednesday, July 30, 2008 1:47 PM
>To: Hausenblas, Michael
>Cc: Manu Sporny; RDFa mailing list
>Subject: Re: RDFa test suite addition
>
>Hi Michael,
>
>> I do by and large agree with Mark (i.e. his excellent description,
>> below) BUT in the same moment I'd like to point out the 
>'Opacity Axiom'
>> [1], [2].
>
>I'm afraid I don't follow you. I'm not sure what you mean when you say
>that you agree with me BUT want to draw attention to the question of
>opacity, since I don't know which part of what I am saying detracts
>from opacity.
>
>Having said that, I'd also say that I don't see opacity of URIs as
>relevant to this discussion. :)
>
>
>> Please note as well that in RDF we talk about URIrefs [3] and
>> *relative URIs are not used in an RDF graph*
>
>That's right. As I said before, I made sure to make that point in the
>spec, and also pointed to the algorithm that should be used to turn
>relative paths into absolute ones. Following that algorithm for all
>URIs when parsing will remove *all* of the issues we have been
>discussing.
>
>
>> - FWIW, I'm happy to take
>> an action to evaluate how other RDF serialisations (e.g. RDF/XML, or
>> upcoming such as Turtle [4]) are dealing with this situation.
>
>I doubt anyone will stop you. :) But I'm not sure you'll find anything
>surprising; absolute URIs are fundamental to RDF, so they are used
>everywhere. Different serialisations allow different ways to
>abbreviate those URIs, but that doesn't change that the underlying
>identifier is an absolute URI.
>
>RDF/XML supports relative paths and QNames/namespaces as abbreviation
>mechanisms, whilst XHTML+RDFa supports relative paths and CURIEs as
>abbreviation mechanisms.
>
>But as far as RDF is concerned everything must be an absolute URI.
>
>I hope we don't get too side-tracked here; I think the issue is pretty
>straightforward, and the only question I think we have to answer is
>whether the spec should be more explicit, or whether it already says
>enough.
>
>Regards,
>
>Mark
>
>-- 
>Mark Birbeck, webBackplane
>
>mark.birbeck@webBackplane.com
>
>http://webBackplane.com/mark-birbeck
>
>webBackplane is a trading name of Backplane Ltd. (company number
>05972288, registered office: 2nd Floor, 69/85 Tabernacle Street,
>London, EC2A 4RR)
>
Received on Thursday, 31 July 2008 07:18:56 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 31 July 2008 07:18:57 GMT