W3C

– DRAFT –
Data Shapes telecon 2025-05-26

26 May 2025

Attendees

Present
ajnelson-nist, bergos, caribou, David, elianaP, simonstey, YoucTagh
Regrets
Andy, Holger
Chair
Nick
Scribe
elianaP

Meeting minutes

Data Shapes telecon 2025-05-26

New readme

Nick: please look at the new readme, it has been updated to better reflect our work

Meeting times

Nick: 2PM CEST proposed instead of 9PM

(no objections)

<simonstey> https://www.timeanddate.com/worldclock/meetingdetails.html?year=2025&month=5&day=26&hour=12&min=0&sec=0&p1=259&p2=47&p3=179&p4=224

Decision: Trial 2PM CEST

PR 335 Profiles doc & auto-publishing??

SimonStey: may be an issue without consensus, depending on the contents

Nick: Will approve the shell document. We can vote for the FPWD and then move it to auto-publishing. Applies for profiles and UI documents.

Usage of SHACL in the TREE CG

<bergos> Tree spec: https://treecg.github.io/specification/

<bergos> proposal to switch to shapes in tree cg: TREEcg/specification#115

<simonstey> GeoSHACL: https://github.com/chrdebru/chrdebru.github.io/blob/master/ontologies/geoshacl/ontology.ttl ( https://ceur-ws.org/Vol-2977/paper8.pdf )

bergos: no specific action point for TreeCG, just informing the group

GeoSHACL

<bergos> Shape Topologies specification: https://treecg.github.io/specification/shape-topologies.html

bergos: if more people are interested, we might consider getting involved
… This is about constraints you can add to your shape

Nick: GeoSPARQL already contains shapes
… might serve as a starting point for this. These 43 shapes are generic.

<nicholascar> Here is the GeoSPARQL 1.1 validator in SHACL: https://github.com/opengeospatial/ogc-geosparql/blob/master/vocabularies/validator.ttl

Nick: SHACL shapes using GeoSPARQL should work fine as long as the latter is enabled at the specific endpoint.

<simonstey> maybe relevant as well opengeospatial/ogc-geosparql#173

bergos: Does any SHACL engine provide this kind of capability out of the box? Would this be able to run without a SPARQL endpoint?

Nick: For the shapes likely to be written by people, a spatial index (therefore geoSPARQL) would be necessary.

<bergos> JS library for geo functions: https://turfjs.org/

bergos: is anyone in touch with the people working on these projects?

<simonstey> https://www.timeanddate.com/worldclock/meetingdetails.html?year=2025&month=5&day=26&hour=12&min=0&sec=0&p1=259&p2=47&p3=179&p4=224 here again

Minutes manually created (not a transcript), formatted by scribe.perl version 244 (Thu Feb 27 01:23:09 2025 UTC).

Diagnostics

Maybe present: Decision, Nick

All speakers: bergos, Decision, Nick, SimonStey

Active on IRC: ajnelson-nist, bergos, caribou, David, elianaP, nicholascar, simonstey, YoucTagh