


Implementing and rolling out data models, processes and tooling for product traceability, product information management and publication. Co-founder @Semaku


I'm at the Declarative Amsterdam conference later this week. Time to get my geek on: declarative.amsterdam/

"Renaissance Developer" sounds much better than "Jack of all trades"

upstream connect error or disconnect/reset before headers. reset reason: overflow

Is @Bitbucket down?


Mmmm... crumpets

Expired certificate on prefix.cc anyone know who's taking care of that site these days? @namedgraph @danbri @kidehen


Looking at capturing provenance information about data processed by code running on AWS Lambda. Wondering whether to use prov:SoftwareAgent to represent the function or a specific execution environment. Ideas welcome!

I'm using a versioned S3 bucket to store the documents, so can either have each patch with a distinct object key, or use the same key for all patches to same resource and stack up the patches as versions of that object... choices 🤔

I've got some RDF Patch documents describing changes to some resource over time. If I want to make those patches available, should each patch document get a distinct URL, and/or are those different states of the same (abstract) patch resource?

Are there other approaches to consider?

2) Construct/insert statements with the constraint baked in and then do path across those statements: construct { ?s ex:knowsFemale ?o } where { ?s schema:knows ?o . ?o schema:gender schema:Female . }

1) Enumerate all possible path lengths with the constraint expressed in the graph pattern: select * where { { ?s schema:knows ?o } union {?s schema:knows [ schema:gender schema:Female ; schema:knows ?o ] } # extend as needed ?o schema:gender schema:Female . }

My initial impression is that it is not possible to express as a property path. A couple of possible approaches are: