W3C home > Mailing lists > Public > public-rdf-dawg@w3.org > January to March 2011

Dataset HTTP Protocol and Service Descriptions

From: Gregory Williams <greg@evilfunhouse.com>
Date: Mon, 28 Feb 2011 16:08:34 -0500
Message-Id: <3E78F44A-DD6C-48D5-BE39-11DB9F88CBE0@evilfunhouse.com>
Cc: Chimezie Ogbuji <chimezie@gmail.com>
To: SPARQL Working Group <public-rdf-dawg@w3.org>
Chime, all,

As part of ACTION-396 I'm trying to sort out the current state of the interaction between the RDF Dataset HTTP Protocol and  Service Descriptions. I think your most recent message on this is here:

http://lists.w3.org/Archives/Public/public-rdf-dawg/2011JanMar/0197.html

You also linked to a message from last May in which you suggest some overlap between the dataset protocol and the service description vocabulary:

http://lists.w3.org/Archives/Public/public-rdf-dawg/2010AprJun/0202.html

It seem to me that the easiest path forward might be adding a subclass of sd:Dataset that indicates that the dataset can be accessed via the dataset protocol. So instead of this:

[] a sd:Service ;
    sd:defaultDatasetDescription </dataset> .
</dataset> a sd:Dataset .

you'd have:

[] a sd:Service ;
    sd:defaultDatasetDescription </dataset> .
</dataset> a sd:RESTDataset .

In your email you had suggested sd:RESTDatasetService, but "Service" seems odd to me as that URI identifies the dataset, right? I'm not tied to "RESTDataset", but want to make sure that whatever we end up with makes sense.

This addition seems like the path of least resistance to me, so I'd like to hear your thoughts about it. Would it satisfy your needs for being able to describe datasets made available via the dataset protocol?

thanks,
.greg
Received on Monday, 28 February 2011 21:09:06 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:15:45 GMT