Re: LDP-Server - Issue-57

On 13 Jun 2013, at 18:00, Arnaud Le Hors <lehors@us.ibm.com> wrote:

> Henry Story <henry.story@bblfish.net> wrote on 06/13/2013 08:44:07 AM:
> 
> > That is we can describe resources as being of certain types, and that
> > they allow certain types of interactions. We can declaratively describe
> > a resource as allowing certain actions. By doing that of course we
> > describe the web agent that the server we are interacting with is.
> > But we did not need to describe the web agent in full to do this, just
> > the resources we are interested in. 
> 
> We're mincing words here. When the description of the resources implies a specific behavior from the agent we're effectively describing the agent too. 
> 
> But as I said before I know you think the interaction is an integral part of the resource type and therefore the mere indication of the resource type in RDF is all one needs. We've heard others saying they don't see it that way though. 
> 
> Rather than continuing to argue over who's right I'm proposing we accept that there are two ways to look at this and accommodate both ways. This may be seen as introducing some redundancy but that's often how we get something that works for everyone in the standards world.

Well I am ok to go down this road as a thought experiment. It will help the discussion move along,
and clarify things.

> The header looks like this: Link: <http://www.w3.org/ns/ldp/profile>; rel=profile
> 
> As I said, all it means is that if a response you get from a server contains such a link header you can expect the server to be LDP compliant. What I mean by that is that the interaction model defined in the LDP spec applies. 


Ok if I turn this into turtle we then get the above to be semantically equivalent to 

   <> rel:profile ldp:profile .

( As it appears in the header, I am tempted to think of this as being a statement by the
server, but I am not sure of this intuition. The advantage of this is that we can then put
a bit of flesh on the idea of the server being an LDP server. Namely one that can truthfully
say the above, ie: follow up on doing the right things when the right messages are sent to 
it )

  Now you are saying that whenever we have the above statement in the header the
client knows that the given resource is a LDPR I think, right? So you are saying the 
following rule holds

{ ?resource rel:profile ldp:profile } => { ?resource a ldp:Resource }

we could have that as a definition of an ldp:Resource. But then could
we not also have the opposite, namely

{ ?resource rel:profile ldp:profile } <= { ?resource a ldp:Resource }

In which case we would have

{ ?resource rel:profile ldp:profile }  <=> { ?resource a ldp:Resource }

ie, they would be logically equivalent.

So if they are not logically equivalent, what is the difference?

This seems like a fair way to proceed, no?

Henry

Social Web Architect
http://bblfish.net/