Re: Blank Nodes Re: Toward easier RDF: a proposal

On 11/24/18 2:08 PM, Pat Hayes wrote:
 > On 11/22/18 5:49 PM, David Booth wrote:
 > . . .
 >> A blank node . . . asserts that there *exists*
 >> a thing, as explained in the RDF Semantics:
 >> https://www.w3.org/TR/rdf11-mt/#blank-nodes In contrast,
 >> an IRI represents *a* thing.  I'm sorry to be pedantic
 >> here, but I mention it because it underscores my point:
 >> the semantics of blank nodes really *are* subtle -- at
 >> least to *average* developers.
 >
 > Is this idea really hard for anyone?

Short answer: Yes!

 > If URIs are names, then blank nodes are pronouns, like
 > 'anyone' in the previous sentence. People don't seem to find
 > pronouns hard or subtle or confusing, or complain that they
 > have devious semantics.

Longer answer: I think an average developer certainly could
understand that concept if it were isolated and explained
that way.  But it isn't.  It is only one of *several* concepts
that are used in *combination* in RDF.  And the implications
of the resulting combination are *not* so easy to grok.
(Case in point: No middle 33% developer is going to grok the RDF
Semantics document, though it is based on simple concepts.)
This principle shows up all the time in every scientific
field -- mathematics, biology, etc. -- where a few "simple"
concepts combine to have consequences that are amazingly hard
to understand.

But this "existence" issue is a bit of a red herring, because it
is not the most significant stumbling block with blank nodes.
The most significant stumbling block is around blank node
*identifiers".

 > . . .
 > However, I agree with your point about bnode
 > *identifiers*. . . . There is something inherently
 > contradictory in having an identifier for something which,
 > by definition, is something which does not identify.

Agreed.

David Booth

Received on Monday, 26 November 2018 03:13:50 UTC