W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > April to June 2010

Re: Base URI in updates?

From: Andy Seaborne <andy.seaborne@talis.com>
Date: Tue, 25 May 2010 11:20:21 +0100
Message-ID: <4BFBA465.2080206@talis.com>
To: Chimezie Ogbuji <ogbujic@ccf.org>
CC: Steve Harris <steve.harris@garlik.com>, SPARQL Working Group WG <public-rdf-dawg@w3.org>

On 21/05/2010 2:13 PM, Chimezie Ogbuji wrote:
> On 5/21/10 4:45 AM, "Andy Seaborne"<andy.seaborne@talis.com>  wrote:
>> What we decide really does need explaining the counter intuitive nature
>> in the doc.
> Yes.
>> I'd like to find a way to make the graph URI the base even if that means
>> contorting things a little - after all, this 3rd party service/graph
>> naming we are using isn't the primary design space of REST anyway.
> True
>> I'd argue that the use of the graph=<abs URI>  creates a new URI used to
>> retrieve the entity.
> What if the new URI isn't resolvable? I.e.,
> PUT /rdf-graphs/employees?graph=tag%3Aogbujic%40ccf.org%3A2010/consultant/56
> Host: example.com
> <?xml version='1.0' encoding='UTF-8'?>
> <rdf:RDF
>   .... no base named ....
> </rdf:RDF>
> We can't say that tag:ogbujic@ccf.org:2010/consultant/56 was used to
> 'retrieve' anything.

"Retrieve" isn't the best word, I grant you, but it is the URI used to 
access the graph within the collection of grapgs addressible at 
http://example.com/rdf-graphs/employees?graph=, so not "retreive" as in 
GET but in an abstract sense.

The same is true for

GET /rdf-graphs/employees?graph=tag%3Aogbujic%40ccf.org%3A2010/consultant/56

What word do you use to describe the action here of accessing 
tag:ogbujic@ccf.org:2010/consultant/56 ?

The choice of "retrieve" was motivated by the spirit, as I read it, of 
the text around 5.1.3:

    Note that
    if the retrieval was the result of a redirected request, the last URI
    used (i.e., the URI that resulted in the actual retrieval of the
    representation) is the base URI.

This does not directly apply but I think the intent is that the base URI 
is whatever was last in some access process.  This might be viewed as 
accessing an entity within a container.

Not ideal, but there is deployed experience using this and no 
contraindication that I know of so I'm minded to make it work by some 
means or other.

Maybe publishing, with a request for comments, is best and see what 
other opinions surface.

Received on Tuesday, 25 May 2010 10:20:55 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:01:00 UTC