RIF data types and built-ins [1] has all the functions you should need in everyday use. No need to make up new IRIs. <rant>Semantic web tooling and uptake would be a lot further along if people complained less and grokked more of the specs.</rant> [1] http://www.w3.org/TR/2013/REC-rif-dtb-20130205/ > On Oct 13, 2017, at 10:30, Gregory Williams <greg@evilfunhouse.com> wrote: > > /cc Ivan Herman > >> On Oct 13, 2017, at 8:14 AM, Andreas Harth <andreas@harth.org> wrote: >> >> The lack of URIs for SPARQL functions has irked me for a while. >> >> We have been using N3 as rule syntax in our Linked Data-Fu system, >> and have found that the CWM builtins [1] are problematic. Some really >> useful functions are missing (such as isLiteral(), isURI, datatype()), >> and some of the existing functions are underspecified concerning the >> support for datatypes. >> >> SPARQL has a much more complete set of functions and it would be nice to >> be able to use those in N3. Defining a prefix for the SPARQL builtin >> functions (and operators) should be easy. I wonder why nobody has done >> that already. > > During the SPARQL 1.1 WG, it was always my intention to define IRIs for all the built-in functions. We even set up a namespace for it[1]. Alas, it seems it was never properly populated, as BOUND is the only function with an assigned IRI. Maybe it would still be possible to get those all assigned—I still have local working copies of what I think should be in /ns/sparql.. > > Thanks, > Greg > > [1] https://www.w3.org/ns/sparql
Received on Friday, 13 October 2017 16:27:55 UTC