Pieter Colpaert
banner
pietercolpaert.be
Pieter Colpaert
@pietercolpaert.be
620 followers 230 following 140 posts
Professor #interoperability and #knowledgegraphs at #Ghent University Building Linked Data Event Streams #ldes Personal website: https://pietercolpaert.be Team website: https://knows.idlab.ugent.be/
Posts Media Videos Starter Packs
Last but not least, Gertjan De Mulder shows how to query Verifiable Credentials via SPARQL in OID4VP flows. RDF graphs enable selective, privacy-preserving presentations.
🔗 ceur-ws.org/Vol-4085/pap...
#ISWC2025
Wout Slabbinck and Beatriz Esteves present LOAMA — a low-code ODRL tool for access management. Visual policy editing for non-technical users & semantic integration.
🔗 ceur-ws.org/Vol-4085/pap...
#ISWC2025
Ieben Smessaert and Julián Rojas will demo SPARQL time functions — treating partial & floating dates as intervals for better temporal queries. Implemented in Comunica.
🔗 smessie.github.io/Article-ISWC...
#ISWC2025
Julián Rojas presents our #Trustflows model: combining CQRS + Event Sourcing with Linked Data to decouple reads/writes and add trust contexts. Validated in the PACSOI health use case.
🔗 ceur-ws.org/Vol-4085/pap...
#ISWC2025
Julián Rojas, Arthur Vercruysse and Ieben Smessaert lead the RDF-Connect tutorial — building provenance-aware, multilingual streaming pipelines. Let’s create a weather KG for Nara 🇯🇵!
🔗 rdf-connect.github.io/Tutorial-ISW...
#ISWC2025
Jonni Hanski and Ruben Taelman explore automated client-side federation for Wikidata after its graph split. Shows when automation can replace manual SERVICE clauses.

🔗 www.rubensworks.net/raw/publicat...
#ISWC2025
Ruben Taelman also leads a tutorial on RDF & SPARQL 1.2 — a hands-on intro to the first spec update in a decade! Covers motivations, RDF-star, triple-terms & reification.
🔗 www.w3.org/Talks/2025/i...
#ISWC2025
Ruben Eschauzier and Ruben Taelman present “Revisiting Link Prioritization…”:evaluating link prioritization in #Solid environments. Defines the R3 metric, tests in Comunica, finds no gain over FIFO.
🔗 rubeneschauzier.github.io/r3-metric-da...
#ISWC2025
Our Knowledge on Web-Scale team is at #ISWC2025! 🎉
We’re sharing work on link prioritization, time functions in SPARQL, low-code ODRL tools, Trustflows, queryable VCs, client-side federation, plus tutorials on RDF/SPARQL 1.2 and RDF-Connect pipelines.
Op zoek naar een locatie in centrum Gent waar je reservatie kan doen om te eten/drinken met 75-tal mensen, waarbij iedere tafel apart afrekent. Patrick Foley‘s geen optie: willen enkel in 1 keer afrekenen.
Sadly there are only 25 hours in a day
Haha! I must confess I also do this
How do people with kids get anything done, ever?
I’m serious, whenever I see someone roughly my age accomplish anything these days, first thing I do is check whether they have kids. Written a book? Cool. Written a book *while having kids*? You’re a GOD.
The NextTrain Pebble app provides a countdown to your train you intended to catch in Belgium and the NL. UX-wise it is fabulous. I maintain the API behind it that has been online since 2010 and still is. So much looking fwd to using NextTrain again!
I don’t think both ways of seeing named graphs are in conflict: you either choose what you do in your store, but I also see ways in which they can co-exist in 1 store if that’s what you want to do.
They are still named, just with local identifiers that cannot conflict with any other external source
Thanks for recommending it! Happy to open the discussion on blank node graphs, I brought my arguments in favor ;)
In the attached image, I’m using named graphs on the one hand as a contextual assertion, but also to bundle triples together as an RDF Message. For the latter, I believe we should have another syntactic feature (cfr. NDJSON or Jelly frames).
Named graphs in #RDF (2014) are still controversial after all this time and I don’t think they should be.

While their semantics were left open, functionally we all use them as contextual assertions anyway. Prove me wrong 🤓

👉 pietercolpaert.be/linkeddata/2...

#linkeddata #knowledgegraphs
the expert group on the Next Generation European Interoperability Framework (EIF) I’m part of is launching a survey to take your ideas into account. Do you have ideas about interoperability governance? Help us out: interoperable-europe.ec.europa.eu/collection/c...
Wikidata, Wikimedia Commons, OpenStreetMap, UniProt, PubChem, DBLP, public transit schedules, our world in data, Europeana, data.europa.eu (the meta dataset), national statistics, ...
On my way home after a fruitful #semantics2025! During the closing ceremony I had to honour to invite the audience to #semantics2026 in one of the most beautiful cities in the world. See you next year in Ghent!
As Linked Data practitioners we know vocabularies and shapes, but that’s not enough. Full interoperability needs two more artefacts: reusable interaction patterns across domains, and implementation guides to help developers avoid endless choices.

Long read: pietercolpaert.be/interoperabi...
Tomorrow it’s #semantics2025! Looking fwd to the keynote of Hannah Bast, the developer workshop, the RDF-Connect tutorial, the #sem4tra2025 workshop on semantics in transport, and the next generation data governance workshop! 🚀
Reposted by Pieter Colpaert
OpenHPS @openhps.org · Apr 22
Our latest #OpenHPS update enables developers to stream processed (location) data to #Solid using Linked Data Event Streams ( #LDES ). Simply adding a sink to an OpenHPS model is all it takes to enable this in existing positioning systems.

Read more in our blog post ...
openhps.org/posts/2025/0...
Stream Location Data to Solid | OpenHPS
OpenHPS is a framework for processing sensor data to create a positioning system. In this blog post, we will demonstrate how OpenHPS can be used to stream location data directly to Solid to offer tran...
openhps.org