Improve explanation of linkset directionality

[GoogleCodeExporter]

As evident per discussion in [1], the modeling of linkset directionality can be 
quite confusing. There are two notions of directionality: What's the subject 
and object of the link triples; and which dataset contains the link triples. 
The former is expressed via void:subjects/objectsTarget. The latter via 
void:subset inclusion of the linkset in one of the datasets. The text in the 
relevant section [2] has a note that comments on this, but this should be made 
even clearer.

[1] http://www.w3.org/2001/sw/interest/ED-void-20101216/#target
[2] 
http://groups.google.com/group/void-discussion/browse_thread/thread/42af457f244d
2f76#

Original issue reported on code.google.com by richard....@gmail.com on 3 Feb 2011 at 11:05

[GoogleCodeExporter]

Original comment by richard....@gmail.com on 9 Feb 2011 at 1:37

• Added labels: Milestone-Release2.0

[GoogleCodeExporter]

Section modified as follows in r186 with a lengthy note:

---- begin quote ----
Note: There are two different notions of “directionality” for RDF links:

1. Which dataset provides the subjects of the triples, and which the objects?
2. Which dataset contains the links? (Who published them?)

The void:subjectsTarget and void:objectsTarget properties express the first 
notion, while not stating anything about containment of the links. The second 
notion is expressed by making the linkset a void:subset of the respective 
target datasets.

Especially when referring to owl:sameAs links, usually the second notion is 
intended. The property is symmetric, so their subjects and objects are 
exchangeable. The question is usually which publisher made the links available 
as part of their dataset.
---- end quote ----

Original comment by richard....@gmail.com on 16 Feb 2011 at 12:03

• Changed state: Fixed