W3C home > Mailing lists > Public > www-tag@w3.org > October 2004

RE: Towards a position of neutrality on issue httpRange-14 for AWWW (was RE: referendum on httpRange-14 (was RE: "information resource"))

From: <Patrick.Stickler@nokia.com>
Date: Thu, 21 Oct 2004 15:41:34 +0300
Message-ID: <1E4A0AC134884349A21955574A90A7A50ADD56@trebe051.ntc.nokia.com>
To: <skw@hp.com>, <www-tag@w3.org>



> -----Original Message-----
> From: ext Stuart Williams [mailto:skw@hp.com]
> Sent: 21 October, 2004 14:49
> To: Stickler Patrick (Nokia-TP-MSW/Tampere); www-tag@w3.org
> Subject: RE: Towards a position of neutrality on issue 
> httpRange-14 for
> AWWW (was RE: referendum on httpRange-14 (was RE: "information
> resource"))
> 
> 
> Hello Patrick,
> 
> Just lifting a few of the points from your message:
> 
> > I would be content, at this juncture, to see AWWW reach a 
> > stable, mature state without explicitly addressing issue 
> > httpRange-14, provided that AWWW does not otherwise include 
> > content which illustrates or favors one side of that issue 
> > but not the other.
> 
> FWIW that I think is what the TAG has indeed being trying to do for
> quite some time.
> [Stage direction: to be read in a tone of exasperation. :-)]

I can appreciate your saying that. Though given the fact
that stuff keeps creeping in here and there that clearly
reflect one side of that debate, it's hard to conclude
that the *entire* TAG is aiming for that goal ;-)


> Wrt the following:
> 
> > I propose modifying such text to make the examples neutral to 
> > the debate, e.g. for the example in section 4.5.3:
> > 
> >    "For flat namespaces, concatenation is one useful 
> mapping. If namespace 
> >     URIs that end with a hash ("#") are chosen, then simple 
> concatenation 
> >     of the namespace URI and the local name creates a URI 
> for a secondary 
> >     resource (the identified term). This technique is used 
> for many [RDFXML] 
> >     namespaces."
> > 
> > change the text to something akin to:
> > 
> >    "For flat namespaces, concatenation is one useful 
> mapping. The simple 
> >     concatenation of the namespace URI and the local name 
> creates a URI 
> >     for the identified term. This technique is used by [RDFXML]."
> > 
> > This revised text is compatible with either side of the 
> > httpRange-14 debate and hence fully neutral while still 
> > providing the intended utility.
> 
> I'd like to support the edit you suggest. Norm was actioned to revise 
> this text to remove an actual error that intimated that RDF actually 
> introduced '#' during the concatenation process [1]. Your replacement 
> edit could fall within the scope of that action.
> 
> [1] http://www.w3.org/2004/10/18-tagmem-irc#T19-59-57

Great.

> Wrt:
>  > The TAG may also wish to examine forms of expression such as
>  > in section 4.4:
>  >
>  >    "When one resource (representation) refers to another 
> resource with
>  >    a URI, this constitutes a link between the two resources."
>  >
>  > as potentially suggestive of the nature of the referring
>  > resource (since one can concieve of an information resource
>  > referring to another resource but not a non-information
>  > resource necessarily doing so).
> 
> Hmmm... I think that I can refer to a resource.

Yes, but can the planet Venus?

The point is that while one can conceive of any information
resource referring to another resource, one cannot easily
concieve of any arbitrary type of resource whatsoever
referring to another resource.

Some kinds of resources can, some can't.

> 
>  > As it is actually the representation that is doing the
>  > referring, stating this in a more neutral, and also a more
>  > precise, manner would be advisable.
> 
> Well that's a bit chicken and egg really. Why would a representation 
> contain a reference... because the resource it represents 
> refers to the 
> other resource.

Or because the owner of the URI chooses to publish a representation
of the resource which refers to another resource.

Remember, a representation is also a resource, (often) distinct
from the resource of which it is a representation.

Hence, some resources are able to refer to other resources.

All representations (which are resources) can refer to other resources.

For information resources, one could conclude that both the resource
and the representation refer to the other resource.


> IMO the links are between resources. 

I agree. And I believe my proposed alternate text reflects that.

> They may be manifest in 
> representations - an indeed some representations may supresses some 
> possible links.

Also true.

One could also say that a link in a representation R of some resource X
which refers to another resource Y that in addition to reflecting a relationship
between the resources X and Y, it reflects a relationship as well between
the resources R and Y, though that latter relationship is not as important
on the hypertext web.

> 
>  > E.g. change the above text to:
>  >
>  >    "When the representation of one resource refers to 
> another resource
>  >     with a URI, this constitutes a link between the two resources."
>  >
>  > etc.
> 
> I'm not persuaded on that one.

How about

   "When a representation of one resource contains a reference to
    another resource with a URI, this constitutes a link between
    the two resources".

???

This doesn't force one having to commit to which resource is referring
to which.

It may vary or overlap depending on the resources involved. I.e.
it may be an information resource doing the refering, or that
information resource may in fact be a poem and the author of
the representation is inserting a link not originally expressed
as part of the poem and hence the author is doing the referring,
or from the perspective of some software application, its the
representation itself that is doing the referring since the link
is expressed as part of the formal markup making up that 
representation, etc. etc.

In short, all we can know for sure for all cases is that if some 
representation of some resource *contains* a reference to
another resource, that there is a relation between (at least)
the two resources.

As to what/who is doing the referring, well, maybe the semantic
web can sort that out (eventually ;-)

Eh?

Patrick
Received on Thursday, 21 October 2004 12:41:52 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:32:43 UTC