- From: Peter F. Patel-Schneider <pfpschneider@gmail.com>
- Date: Thu, 24 Sep 2015 19:39:52 -0700
- To: Holger Knublauch <holger@topquadrant.com>, public-data-shapes-wg@w3.org
There are a number of consequences of permitting literals to be focus nodes. Whether these are desirable or reasonable consequences is probably a matter for debate. Presumably literal nodes will satisfy some inverse property constraints so they can be traversed. Traversing literal nodes makes the syntactic status of literal nodes more visible. For example, "01"^^xsd:integer, "1"^^xsd:integer, and "1"^^xsd:int are all different nodes. Some literal nodes (e.g., "0"^^xsd:integer) may be part of very many triples. This could inadvertently make checking some constraints very expensive. peter On 09/24/2015 06:57 PM, Holger Knublauch wrote: > PROPOSAL: Close ISSUE-90 stating that literals can also serve as focus nodes. > > Holger > > > On 9/21/2015 10:44, RDF Data Shapes Working Group Issue Tracker wrote: >> shapes-ISSUE-90 (Literal focus nodes): Can the focus node be a literal? >> [SHACL Spec] >> >> http://www.w3.org/2014/data-shapes/track/issues/90 >> >> Raised by: Holger Knublauch >> On product: SHACL Spec >> >> The spec is vague on that topic, but we need to clarify whether the focus >> node can also be a literal. My assumption from the SPIN days is that they >> would only be IRIs and blank nodes, but now I see no real reason for such a >> restriction. But I have not tried it out yet, so I may be missing some >> consequences. Does anyone see problems? >> >> A use case would be to allow sh:valueShape to apply to literals too, e.g. to >> check its own datatype or length, e.g. for qualified constraints. >> >> If we change this, then >> http://w3c.github.io/data-shapes/shacl/#AllObjectsScope should be generalized. >> >> >> > >
Received on Friday, 25 September 2015 02:40:29 UTC