W3C home > Mailing lists > Public > public-rdf-wg@w3.org > February 2013

Re: Blank Node Identifiers and RDF Dataset Normalization

From: Manu Sporny <msporny@digitalbazaar.com>
Date: Tue, 26 Feb 2013 21:46:33 -0500
Message-ID: <512D7389.5050308@digitalbazaar.com>
To: "public-rdf-wg@w3.org" <public-rdf-wg@w3.org>
CC: "public-linked-json@w3.org" <public-linked-json@w3.org>
On 02/26/2013 10:44 AM, Bauer, Herbert S. (Scott) wrote:
> I've shared Manu's blog post with some of my counterparts here at 
> Mayo, and the response from at least one person is that JSON-LD has 
> no real need for bnodes as these can be supported through nested 
> structures.

That assumes that there is only one incoming edge in the graph to the
node that is being nested. If there is more than one incoming edge, and
the object doesn't have an IRI identifier, then what's the solution?

In other words, JSON-LD wouldn't need blank node identifiers if we
restricted the number of incoming edges to an unlabeled node in the
graph to 1. That restriction would take out a chunk of very important
use cases.

> There an even more emphatic concern that much damage could be done
> to the RDF effort if blank nodes are allowed to be used as
> predicates.

There are certainly down-sides for blank node predicates... however, you
will have to be more specific about what the "damage" you're talking
about entails.

-- manu

-- 
Manu Sporny (skype: msporny, twitter: manusporny, G+: +Manu Sporny)
Founder/CEO - Digital Bazaar, Inc.
blog: Aaron Swartz, PaySwarm, and Academic Journals
http://manu.sporny.org/2013/payswarm-journals/
Received on Wednesday, 27 February 2013 02:47:06 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 16:25:54 GMT