RE: RDF-ISSUE-140 (dataset-comparison): RDF Dataset Comparison (Ivan Herman) [RDF Concepts]

On Friday, August 09, 2013 4:17 PM, Ivan Herman wrote:
> On Aug 9, 2013, at 11:43 , Markus Lanthaler <>
>> On Friday, August 09, 2013 11:24 AM, Ivan Herman wrote:
>>> We had a long discussion some times ago and we concluded that graphs in
>>> a dataset share bnodes. As a consequence, I believe Gavin's statement
>>> seems to be the proper conclusion...
>> Yes, the graphs share bnodes but I'm not sure how that relates to the
>> names. So you could as well argue that there are two sets of blank node
>> identifiers and that in the examples below the mappings are
>> Example 1: _:y -> _:x (nodes) | _:y -> _:x (graphs)
>> Example 2: _:y -> _:y (nodes) | _:y -> _:x (graphs)
>> Or do I miss something? As far as I understand it, there's no
>> between a blank node identifier used as graph name and a blank node
>> identifier used as node (you could say they are in different scopes) from
>> which I conclude that the same bnode id mappings can be mapped
> Yes, we could do that. But that seems to be confusing, at least to me.

Yes, it's confusing. But I think it is a consequence of the decision to not
define any dataset semantics. Blank nodes used as graph names do not denote
the graph.

> Is there a use case for the separation of the different scopes?

Well, ask the people who voted against letting bnodes denote the graph.

> It
> looks way more obvious to me to consider a bnode as a label and a bnode
> in one of the graphs as being identical...

Fully agreed, but I think under the current semantics they are not. Actually
the same is true for IRIs but since their scope is global the difference
doesn't matter.

Markus Lanthaler

Received on Friday, 9 August 2013 14:54:24 UTC